ATMEL AT89LP6440-20JU

Features
• 8-bit Microcontroller Compatible with MCS®51 Products
Enhanced 8051 Architecture
– Single-clock Cycle per Byte Fetch
– Up to 20 MIPS Throughput at 20 MHz Clock Frequency
– Fully Static Operation: 0 Hz to 20 MHz
– On-chip 2-cycle Hardware Multiplier
– 16x16 Multiply–Accumulate Unit
– 256x8 Internal RAM
– 4096x8 Internal Extra RAM
– Dual Data Pointers
– 4-level Interrupt Priority
Nonvolatile Program and Data Memory
– 64K Bytes of In-System Programmable (ISP) Flash Program Memory
– 8K Bytes of Flash Data Memory
– Endurance: Minimum 100,000 Write/Erase Cycles
– Serial Interface for Program Downloading
– 64-byte Fast Page Programming Mode
– 256-Byte User Signature Array
– 2-level Program Memory Lock for Software Security
– In-Application Programming of Program Memory
Peripheral Features
– Three 16-bit Enhanced Timer/Counters
– Two 8-bit PWM Outputs
– 4-Channel 16-bit Compare/Capture/PWM Array
– Enhanced UART with Automatic Address Recognition and Framing
Error Detection
– Enhanced Master/Slave SPI with Double-buffered Send/Receive
– Master/Slave Two-Wire Serial Interface
– Programmable Watchdog Timer with Software Reset
– Dual Analog Comparators with Selectable Interrupts and Debouncing
– 8-channel 10-bit ADC/DAC
– 8 General-purpose Interrupt Pins
Special Microcontroller Features
– Two-wire On-chip Debug Interface
– Brown-out Detection and Power-on Reset with Power-off Flag
– Active-low External Reset Pin
– Internal RC Oscillator
– Low Power Idle and Power-down Modes
– Interrupt Recovery from Power-down Mode
I/O and Packages
– Up to 38 Programmable I/O Lines
– 40-lead PDIP or 44-lead TQFP/PLCC or 44-pad VQFN/MLF
– Configurable I/O Modes
• Quasi-bidirectional (80C51 Style)
• Input-Only (Tristate)
• Push-pull CMOS Output
• Open-drain
Operating Conditions
– 2.4V to 3.6V VDD Voltage Range
– -40° C to 85°C Temperature Range
– 0 to 20 MHz @ 2.4–3.6V
8-bit
Microcontroller
with 64K Bytes
In-System
Programmable
Flash
AT89LP6440 Preliminary
3706A–MICRO–9/09
1. Pin Configurations
1.1
40P6: 40-lead PDIP
T2/P1.0
T2EX/P1.1
SDA/P1.2
SCL/P1.3
SS/P1.4
MOSI/P1.5
MISO/P1.6
SCK/P1.7
RST/P4.2
RXD/P3.0
TXD/P3.1
INT0/P3.2
INT1/P3.3
T0/P3.4
T1/P3.5
WR/P3.6
RD/P3.7
XTAL2/P4.1
XTAL1/P4.0
GND
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
VDD
P0.0/AD0
P0.1/AD1
P0.2/AD2
P0.3/AD3
P0.4/AD4
P0.5/AD5
P0.6/AD6
P0.7/AD7
P4.3
P4.4/ALE
P4.5
P2.7/AIN3/A15
P2.6/AIN2/A14
P2.5/AIN1/A13
P2.4/AIN0/A12
P2.3/A11/CCD
P2.2/A10/CCC
P2.1/A9/CCB
P2.0/A8/CCA
44A: 44-lead TQFP (Top View)
44
43
42
41
40
39
38
37
36
35
34
P1.4/SS
P1.3/SCL
P1.2/SDA
P1.1/T2EX
P1.0/T2
VDD
AVDD
P0.0/AD0
P0.1/AD1
P0.2/AD2
P0.3/AD3
1.2
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
33
32
31
30
29
28
27
26
25
24
23
1
2
3
4
5
6
7
8
9
10
11
P0.4/AD4
P0.5/AD5
P0.6/AD6
P0.7/AD7
P4.3
GND
P4.4/ALE
P4.5
P2.7/AIN3/A15
P2.6/AIN2/A14
P2.5/AIN1/A13
WR/P3.6
RD/P3.7
XTAL2/P4.1
XTAL1/P4.0
GND
GND
CCA/A8/P2.0
CCB/A9/P2.1
CCC/A10/P2.2
CCD/A11/P2.3
A12/AIN0/P2.4
12
13
14
15
16
17
18
19
20
21
22
MOSI/P1.5
MISO/P1.6
SCK/P1.7
RST/P4.2
RXD/P3.0
VDD
TXD/P3.1
INT0/P3.2
INT1/P3.3
T0/P3.4
T1/P3.5
2
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
44J: 44-lead PLCC
39
38
37
36
35
34
33
32
31
30
29
18
19
20
21
22
23
24
25
26
27
28
7
8
9
10
11
12
13
14
15
16
17
P0.4/AD4
P0.5/AD5
P0.6/AD6
P0.7/AD7
P4.3
GND
P4.4/ALE
P4.5
P2.7/AIN3/A15
P2.6/AIN2/A14
P2.5/AIN1/A13
WR/P3.6
RD/P3.7
XTAL2/P4.1
XTAL1/P4.0
GND
GND
CCA/A8/AIN0/P2.0
CCB/A9/P2.1
CCC/A10/P2.2
CCD/A11/P2.3
A12/AIN0/P2.4
MOSI/P1.5
MISO/P1.6
SCK/P1.7
RST/P4.2
RXD/P3.0
VDD
TXD/P3.1
INT0/P3.2
INT1/P3.3
T0/P3.4
T1/P3.5
6
5
4
3
2
1
44
43
42
41
40
P1.4/SS
P1.3/SCL
P1.2/SDA
P1.1/T2EX
P1.0/T2
VDD
AVDD
P0.0/AD0
P0.1/AD1
P0.2/AD2
P0.3/AD3
1.3
44M1: 44-pad VQFN/MLF
44
43
42
41
40
39
38
37
36
35
34
P1.4/SS
P1.3/SCL
P1.2/SDA
P1.1/T2EX
P1.0/T2
VDD
AVDD
P0.0/AD0
P0.1/AD1
P0.2/AD2
P0.3/AD3
1.4
33
32
31
30
29
28
27
26
25
24
23
NOTE:
Bottom pad
should be
soldered to ground
P0.4/AD4
P0.5/AD5
P0.6/AD6
P0.7/AD7
P4.3
GND
P4.4/ALE
P4.5
P2.7/AIN3/A15
P2.6/AIN2/A14
P2.5/AIN1/A13
12
13
14
15
16
17
18
19
20
21
22
1
2
3
4
5
6
7
8
9
10
11
WR/P3.6
RD/P3.7
XTAL2/P4.1
XTAL1/P4.0
GND
GND
CCA/A8/P2.0
CCB/A9/P2.1
CCC/A10/P2.2
CCD/A11/P2.3
A12/AIN0/P2.4
MOSI/P1.5
MISO/P1.6
SCK/P1.7
RST/P4.2
RXD/P3.0
VDD
TXD/P3.1
INT0/P3.2
INT1/P3.3
T0/P3.4
T1/P3.5
3
3706A–MICRO–9/09
1.5
Pin Description
Table 1-1.
AT89LP6440 Pin Description
Pin Number
TQFP
1
PLCC
7
PDIP
6
VQFN
1
Symbol
P1.5
Type
I/O
I/O
I
2
8
7
2
P1.6
I/O
I/O
I
3
9
8
3
P1.7
I/O
I/O
I
4
10
9
4
P4.2
I/O
I
I
5
P3.0
I/O
I
6
VDD
I
11
7
P3.1
14
12
8
9
15
13
10
16
11
5
11
6
12
7
13
8
10
I/O
P1.5: User-configurable I/O Port 1 bit 5.
MOSI: SPI master-out/slave-in. When configured as master, this pin is an output.
When configured as slave, this pin is an input.
GPI5: General-purpose Interrupt input 5.
P1.6: User-configurable I/O Port 1 bit 6.
MISO: SPI master-in/slave-out. When configured as master, this pin is an input. When
configured as slave, this pin is an output.
GPI6: General-purpose Interrupt input 6.
P1.7: User-configurable I/O Port 1 bit 7.
SCK: SPI Clock. When configured as master, this pin is an output. When configured
as slave, this pin is an input.
GPI7: General-purpose Interrupt input 7.
P4.2: User-configurable I/O Port 4 bit 2 (if Reset Fuse is disabled).
RST: External Active-Low Reset input (if Reset Fuse is enabled. See “External
Reset” on page 34.).
DCL: Serial Clock input for On-Chip Debug Interface when OCD is enabled.
P3.0: User-configurable I/O Port 3 bit 0.
RXD: Serial Port Receiver Input.
Supply Voltage
O
P3.1: User-configurable I/O Port 3 bit 1.
TXD: Serial Port Transmitter Output.
P3.2
I/O
I
P3.2: User-configurable I/O Port 3 bit 2.
INT0: External Interrupt 0 Input or Timer 0 Gate Input.
9
P3.3
I/O
I
P3.3: User-configurable I/O Port 3 bit 3.
INT1: External Interrupt 1 Input or Timer 1 Gate Input
14
10
P3.4
I/O
P3.4: User-configurable I/O Port 3 bit 4.
T1: Timer/Counter 0 External input or PWM output.
17
15
11
P3.5
12
18
16
12
P3.6
13
19
17
13
P3.7
I/O
I/O
I/O
I/O
O
I/O
O
I/O
O
14
20
18
14
P4.1
O
I/O
I/O
I
15
21
19
15
P4.0
I/O
16
4
Description
22
N/A
16
GND
I
P3.5: User-configurable I/O Port 3 bit 5.
T1: Timer/Counter 1 External input or PWM output.
P3.6: User-configurable I/O Port 3 bit 6.
WR: External memory interface Write Strobe (active-low).
P3.7: User-configurable I/O Port 3 bit 7.
RD: External memory interface Read Strobe (active-low).
P4.1: User-configurable I/O Port 4 bit 1.
XTAL2: Output from inverting oscillator amplifier. It may be used as a port pin if the
internal RC oscillator is selected as the clock source.
CLKOUT: When the internal RC oscillator is selected as the clock source, may be
used to output the internal clock divided by 2.
DDA: Serial Data input/output for On-Chip Debug Interface when OCD is enabled and
the external clock is selected as the clock source.
P4.0: User-configurable I/O Port 4 bit 0.
XTAL1: Input to the inverting oscillator amplifier and internal clock generation circuits.
It may be used as a port pin if the internal RC oscillator is selected as the clock
source.
DDA: Serial Data input/output for On-Chip Debug Interface when OCD is enabled and
the internal RC oscillator is selected as the clock source.
Ground
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Table 1-1.
AT89LP6440 Pin Description
Pin Number
TQFP
PLCC
PDIP
VQFN
Symbol
Type
17
23
20
17
GND
I
18
24
21
18
P2.0
I/O
I/O
O
P2.0: User-configurable I/O Port 2 bit 0.
CCA: Timer 2 Channel A Compare Output or Capture Input.
A8: External memory interface Address bit 8.
19
25
22
19
P2.1
I/O
I/O
O
P2.1: User-configurable I/O Port 2 bit 1.
CCB: Timer 2 Channel B Compare Output or Capture Input.
A9: External memory interface Address bit 9.
P2.1
I/O
I/O
O
O
P2.2: User-configurable I/O Port 2 bit 2.
CCC: Timer 2 Channel C Compare Output or Capture Input.
A10: External memory interface Address bit 10.
DA-: DAC negative differential output.
P2.3: User-configurable I/O Port 2 bit 3.
CCD: Timer 2 Channel D Compare Output or Capture Input.
A11: External memory interface Address bit 11.
D+-: DAC positive differential output.
20
26
23
20
Description
Ground
21
27
24
21
P2.3
I/O
I/O
O
O
22
28
25
22
P2.4
I/O
I
O
P2.4: User-configurable I/O Port 2 bit 5.
AIN0: Analog Comparator Input 0.
A12: External memory interface Address bit 12.
23
29
26
23
P2.5
I/O
I
O
P2.5: User-configurable I/O Port 2 bit 5.
AIN1: Analog Comparator Input 1.
A13: External memory interface Address bit 13.
24
30
27
24
P2.6
I/O
I
O
P2.6: User-configurable I/O Port 2 bit 6.
AIN2: Analog Comparator Input 2.
A14: External memory interface Address bit 14.
25
31
28
25
P2.7
I/O
I
O
P2.7: User-configurable I/O Port 2 bit 7.
AIN3: Analog Comparator Input 3.
A15: External memory interface Address bit 15.
26
32
29
26
P4.5
I/O
P4.5: User-configurable I/O Port 4 bit 5.
27
33
30
27
P4.4
I/O
O
P4.4: User-configurable I/O Port 4 bit 4.
ALE: External memory interface Address Latch Enable.
28
34
28
GND
I
29
35
31
29
P4.3
I/O
I/O
P4.3: User-configurable I/O Port 4 bit 3.
DDA: Serial Data input/output for On-Chip Debug Interface when OCD is enabled and
the Crystal oscillator is selected as the clock source.
30
36
32
30
P0.7
I/O
O
I
P0.7: User-configurable I/O Port 0 bit 7.
AD7: External memory interface Address/Data bit 7.
ADC7: ADC analog input 7.
31
37
33
31
P0.6
I/O
O
I
P0.6: User-configurable I/O Port 0 bit 6.
AD6: External memory interface Address/Data bit 6.
ADC6: ADC analog input 6.
32
38
34
32
P0.5
I/O
O
I
P0.5: User-configurable I/O Port 0 bit 5.
AD5: External memory interface Address/Data bit 5.
ADC5: ADC analog input 5.
33
39
35
33
P0.4
I/O
O
I
P0.4: User-configurable I/O Port 0 bit 4.
AD4: External memory interface Address/Data bit 4.
ADC4: ADC analog input 4.
34
40
36
34
P0.3
I/O
O
I
P0.3: User-configurable I/O Port 0 bit 3.
AD3: External memory interface Address/Data bit 3.
ADC3: ADC analog input 3.
Ground
5
3706A–MICRO–9/09
Table 1-1.
AT89LP6440 Pin Description
Pin Number
TQFP
PLCC
PDIP
VQFN
Symbol
Type
Description
35
41
37
35
P0.2
I/O
O
I
P0.2: User-configurable I/O Port 0 bit 2.
AD2: External memory interface Address/Data bit 2.
ADC2: ADC analog input 2.
36
42
38
36
P0.1
I/O
O
I
P0.1: User-configurable I/O Port 0 bit 1.
AD1: External memory interface Address/Data bit 1.
ADC1: ADC analog input 1.
37
43
39
37
P0.0
I/O
O
I
P0.0: User-configurable I/O Port 0 bit 0.
AD0: External memory interface Address/Data bit 0.
ADC0: ADC analog input 0.
38
44
40
38
AVDD
I
Analog Supply Voltage
39
1
39
VDD
I
Supply Voltage
40
2
1
40
P1.0
I/O
I/O
I
P1.0: User-configurable I/O Port 1 bit 0.
T2: Timer 2 External Input or Clock Output.
GPI0: General-purpose Interrupt input 0.
41
3
2
41
P1.1
I/O
I
I
P1.1: User-configurable I/O Port 1 bit 1.
T2EX: Timer 2 External Capture/Reload Input.
GPI1: General-purpose Interrupt input 1
42
4
3
42
P1.2
I/O
I
P1.2: User-configurable I/O Port 1 bit 2.
GPI2: General-purpose Interrupt input 2.
43
5
4
43
P1.3
I/O
I
P1.3: User-configurable I/O Port 1 bit 3.
GPI3: General-purpose Interrupt input 3.
44
6
5
44
P1.4
I/O
I
I
P1.4: User-configurable I/O Port 1 bit 4.
SS: SPI Slave-Select.
GPI6: General-purpose Interrupt input 4.
2. Overview
The AT89LP6440 is a low-power, high-performance CMOS 8-bit microcontroller with 64K bytes
of In-System Programmable Flash program memory and 8K bytes of Flash data memory. The
device is manufactured using Atmel®'s high-density nonvolatile memory technology and is compatible with the industry-standard 8051 instruction set. The AT89LP6440 is built around an
enhanced CPU core that can fetch a single byte from memory every clock cycle. In the classic
8051 architecture, each fetch requires 6 clock cycles, forcing instructions to execute in 12, 24 or
48 clock cycles. In the AT89LP6440 CPU, standard instructions need only 1 to 4 clock cycles
providing 6 to 12 times more throughput than the standard 8051. Seventy percent of instructions
need only as many clock cycles as they have bytes to execute, and most of the remaining
instructions require only one additional clock. The enhanced CPU core is capable of 20 MIPS
throughput whereas the classic 8051 CPU can deliver only 4 MIPS at the same current consumption. Conversely, at the same throughput as the classic 8051, the new CPU core runs at a
much lower speed and thereby greatly reducing power consumption and EMI.
The AT89LP6440 provides the following standard features: 64K bytes of In-System Programmable Flash program memory, 8K bytes of Flash data memory, 4352 bytes of RAM, up to 38 I/O
lines, three 16-bit timer/counters, up to six PWM outputs, a programmable watchdog timer, two
analog comparators, a 10-bit ADC/DAC with 8 input channels, a full-duplex serial port, a serial
peripheral interface, a two-wire serial interface, an internal RC oscillator, on-chip crystal oscillator, and a four-level, twelve-vector interrupt system. A block diagram is shown in Figure 2-1.
6
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Timer 0 and Timer 1 in the AT89LP6440 are enhanced with two new modes. Mode 0 can be
configured as a variable 9- to 16-bit timer/counter and Mode 1 can be configured as a 16-bit
auto-reload timer/counter. In addition, the timer/counters may each independently drive an 8-bit
precision pulse width modulation output.
Timer 2 on the AT89LP6440 serves as a 16-bit time base for a 4-channel Compare/Capture
Array with up to four multi-phasic, variable precision (up to 16-bit) PWM outputs.
The enhanced UART of the AT89LP6440 includes Framing Error Detection and Automatic
Address Recognition. In addition, enhancements to Mode 0 allow hardware accelerated emulation of half-duplex SPI or Two Wire interfaces.
The I/O ports of the AT89LP6440 can be independently configured in one of four operating
modes. In quasi-bidirectional mode, the ports operate as in the classic 8051. In input-only mode,
the ports are tristated. Push-pull output mode provides full CMOS drivers and open-drain mode
provides just a pull-down. In addition, all 8 pins of Port 1 can be configured to generate an interrupt using the general-purpose interrupt interface.
2.1
Block Diagram
Figure 2-1.
AT89LP6440 Block Diagram
64K Bytes
Flash Code
8K Bytes
Flash Data
256 Bytes
RAM
4K Bytes
ERAM
XRAM
Interface
8051 Single Cycle CPU
Crystal or
Resonator
Port 0
Configurable I/O
Watchdog
Timer
UART
Port 1
Configurable I/O
General-purpose
Interrupt
SPI
Port 2
Configurable I/O
Dual Data
Pointers
TWI
Port 3
Configurable I/O
Multiply
Accumulate
(16 x 16)
Timer 0
Timer 1
Port 4
Configurable I/O
POR
BOD
Timer 2
Configurable
Oscillator
Internal
RC Oscillator
Compare/
Capture Array
Dual Analog
Comparators
On-Chip
Debug
8-channel 10-bit
ADC/DAC
8
7
3706A–MICRO–9/09
2.2
Comparison to Standard 8051
The AT89LP6440 is part of a family of devices with enhanced features that are fully binary compatible with the 8051 instruction set. In addition, most SFR addresses, bit assignments, and pin
alternate functions are identical to Atmel's existing standard 8051 products. However, due to the
high performance nature of the device, some system behaviors are different from those of
Atmel's standard 8051 products such as AT89S52 or AT89C2051. The major differences from
the standard 8051 are outlined in the following paragraphs and may be useful to users migrating
to the AT89LP6440 from older devices.
2.2.1
System Clock
The maximum CPU clock frequency equals the externally supplied XTAL1 frequency. The oscillator is not divided by 2 to provide the internal clock and X2 mode is not supported. The System
Clock Divider can scale the CPU clock versus the oscillator source (See Section 6.5 on page
31).
2.2.2
Reset
The RST pin of the AT89LP6440 is active-LOW as compared with the active-high reset in the
standard 8051. In addition, the RST pin is sampled every clock cycle and must be held low for a
minimum of two clock cycles, instead of 24 clock cycles, to be recognized as a valid reset.
2.2.3
Instruction Execution with Single-cycle Fetch
The CPU fetches one code byte from memory every clock cycle instead of every six clock
cycles. This greatly increases the throughput of the CPU. As a consequence, the CPU no longer
executes instructions in 12, 24 or 48 clock cycles. Each standard instruction executes in only 1
to 4 clock cycles. See “Instruction Set Summary” on page 143 for more details. Any software
delay loops or instruction-based timing operations may need to be retuned to achieve the
desired results.
2.2.4
Interrupt Handling
The interrupt controller polls the interrupt flags during the last clock cycle of any instruction. In
order for an interrupt to be serviced at the end of an instruction, its flag needs to have been
latched as active during the next to last clock cycle of the instruction, or in the last clock cycle of
the previous instruction if the current instruction executes in only a single clock cycle.
The external interrupt pins, INT0 and INT1, are sampled at every clock cycle instead of once
every 12 clock cycles. Coupled with the shorter instruction timing and faster interrupt response,
this leads to a higher maximum rate of incidence for the external interrupts.
The Serial Peripheral Interface (SPI) has a dedicated interrupt vector. The SPI no longer shares
its interrupt with the Serial Port and the ESPI (IE2.2) bit replaces SPIE (SPCR.7).
2.2.5
Timer/Counters
By default Timer0, Timer 1 and Timer 2 are incremented at a rate of once per clock cycle. This
compares to once every 12 clocks in the standard 8051. A common prescaler is available to
divide the time base for all timers and reduce the increment rate. The TPS3-0 bits in the CLKREG
SFR control the prescaler (Table 6-2 on page 32). Setting TPS3-0 = 1011B will cause the timers
to count once every 12 clocks.
The external Timer/Counter pins, T0, T1, T2 and T2EX, are sampled at every clock cycle instead
of once every 12 clock cycles. This increases the maximum rate at which the Counter modules
may function.
8
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
There is no difference in counting rate between Timer 2’s Auto-Reload/Capture and Baud
Rate/Clock Out modes. All modes increment the timer once per clock cycle. Timer 2 in AutoReload/Capture mode increments at 12 times the rate of standard 8051s. Setting TPS3-0 =
1101B will force Timer 2 to count every twelve clocks. Timer 2 in Baud Rate or Clock Out mode
increments at twice the rate of standard 8051s. Setting TPS3-0 = 0001B will force Timer 2 to
count every two clocks.
2.2.6
Serial Port
The baud rate of the UART in Mode 0 defaults to 1/4 the clock frequency, compared to 1/12 the
clock frequency in the standard 8051. In should also be noted that when using Timer 1 to generate the baud rate in UART Modes 1 or 3, the timer counts at the clock frequency and not at 1/12
the clock frequency. To maintain the same baud rate in the AT89LP6440 while running at the
same frequency as a standard 8051, the time-out period must be 12 times longer. Mode 1 of
Timer 1 supports 16-bit auto-reload to facilitate longer time-out periods for generating low baud
rates.
Timer 2 generated baud rates are twice as fast in the AT89LP6440 than on standard 8051s
when operating at the same frequency. The Timer Prescaler can also scale the baud rate to
match an existing application.
2.2.7
SPI
The Serial Peripheral Interface (SPI) has a dedicated interrupt vector. The ESPI (IE2.2) bit
replaces SPIE (SPCR.7). SPCR.7 (TSCK) now enables timer-generated baud rate.
The SPI includes Mode Fault detection. If multiple-master capabilities are not required, SSIG
(SPSR.2) must be set to one for master mode to function correctly when SS (P1.4) is a general
purpose I/O.
2.2.8
Watchdog Timer
The Watchdog Timer in AT89LP6440 counts at a rate of once per clock cycle. This compares to
once every 12 clocks in the standard 8051. A common prescaler is available to divide the time
base for all timers and reduce the counting rate.
2.2.9
I/O Ports
The I/O ports of the AT89LP6440 may be configured in four different modes. By default all the
I/O ports revert to input-only (tristated) mode at power-up or reset. In the standard 8051, all ports
are weakly pulled high during power-up or reset. To enable 8051-like ports, the ports must be
put into quasi-bidirectional mode by clearing the P1M0, P2M0, P3M0 and P4M0 SFRs. The user
can also configure the ports to start in quasi-bidirectional mode by disabling the Tristate-Port
User Fuse. When this fuse is disabled, P1M0, P2M0, P3M0 and P4M0 will reset to 00h instead
of FFh and the ports will be weakly pulled high. Port 0 and the upper nibble of Port 2 always
power up tristated regardless of the fuse setting due to their analog functions.
2.2.10
External Memory Interface
The AT89LP6440 does not support external program memory. The PSEN and EA functions are
not supported and those pins are replaced with general purpose I/O. The ALE strobe does not
toggle continuously and cannot be used as a board-level clock.
9
3706A–MICRO–9/09
3. Memory Organization
The AT89LP6440 uses a Harvard Architecture with separate address spaces for program and
data memory. The program memory has a regular linear address space with support for 64K
bytes of directly addressable application code. The data memory has 256 bytes of internal RAM
and 128 bytes of Special Function Register I/O space. The AT89LP6440 supports external data
memory with portions of the external data memory space implemented on chip as Extra RAM
and nonvolatile Flash data memory. External program memory is not supported. The memory
address spaces of the AT89LP6440 are listed in Table 3-1.
Table 3-1.
3.1
AT89LP6440 Memory Address Spaces
Name
Description
Range
DATA
Directly addressable internal RAM
00H–7FH
IDATA
Indirectly addressable internal RAM and stack space
00H–FFH
SFR
Directly addressable I/O register space
80H–FFH
EDATA
On-chip Extra RAM and extended stack space
0000H–0FFFH
FDATA
On-chip nonvolatile Flash data memory
1000H–2FFFH
XDATA
External data memory
3000H–FFFFH
CODE
On-chip nonvolatile Flash program memory
0000H–FFFFH
SIG
On-chip nonvolatile Flash signature array
0000H–01FFH
Program Memory
The AT89LP6440 contains 64K bytes of on-chip In-System Programmable Flash memory for
program storage. The Flash memory has an endurance of at least 100,000 write/erase cycles
and a minimum data retention time of 10 years. The reset and interrupt vectors are located
within the first 83 bytes of program memory (refer to Table 9-1 on page 40). Constant tables can
be allocated within the entire 64K program memory address space for access by the MOVC
instruction. The AT89LP6440 does not support external program memory. A map of the
AT89LP6440 program memory is shown in Figure 3-1.
3.1.1
SIG
In addition to the 64K code space, the AT89LP6440 also supports a 256-byte User Signature
Array and a 128-byte Atmel Signature Array that are accessible by the CPU. The Atmel Signature Array is initialized with the Device ID in the factory. The second page of the User Signature
Array (0180H–01FFH) is initialized with analog configuration data including the Internal RC
Oscillator calibration byte. The User Signature Array is available for user identification codes or
constant parameter data. Data stored in the signature array is not secure. Security bits will disable writes to the array; however, reads by an external device programmer are always allowed.
In order to read from the signature arrays, the SIGEN bit (DPCF.3) must be set (See Table 5-5
on page 27). While SIGEN is one, MOVC A,@A+DPTR will access the signature arrays. The
User Signature Array is mapped from addresses 0100h to 01FFh and the Atmel Signature Array
is mapped from addresses 0000h to 007Fh. SIGEN must be cleared before using MOVC to
access the code memory. The User Signature Array may also be modified by the In-Application
Programming interface. When IAP = 1 and SIGEN = 1, MOVX @DPTR instructions will access
the array (See Section 3.5 on page 20).
10
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Figure 3-1.
Program Memory Map
AT89LP6440
01FF
User Signature Array
0100
SIGEN=1
007F
Atmel Signature Array
0000
FFFF
Program Memory
SIGEN=0
0000
3.2
Internal Data Memory
The AT89LP6440 contains 256 bytes of general SRAM data memory plus 128 bytes of I/O
memory mapped into a single 8-bit address space. Access to the internal data memory does not
require any configuration. The internal data memory has three address spaces: DATA, IDATA
and SFR; as shown in Figure 3-2. Some portions of external data memory are also implemented
internally. See “External Data Memory” below for more information.
Figure 3-2.
Internal Data Memory Map
FFH
FFH
IDATA
ACCESSIBLE
BY INDIRECT
ADDRESSING
ONLY
UPPER
128
80H
7FH
0
3.2.1
80H
DATA/IDATA
ACCESSIBLE
BY DIRECT
AND INDIRECT
ADDRESSING
LOWER
128
SFR
ACCESSIBLE
BY DIRECT
ADDRESSING
SPECIAL
FUNCTION
REGISTERS
PORTS
STATUS AND
CONTROL BITS
TIMERS
REGISTERS
STACK POINTER
ACCUMULATOR
(ETC.)
DATA
The first 128 bytes of RAM are directly addressable by an 8-bit address (00H–7FH) included in
the instruction. The lowest 32 bytes of DATA memory are grouped into 4 banks of 8 registers
each. The RS0 and RS1 bits (PSW.3 and PSW.4) select which register bank is in use. Instructions using register addressing will only access the currently specified bank. The lower 128 bit
addresses are also mapped into DATA addresses 20H—2FH.
11
3706A–MICRO–9/09
3.2.2
IDATA
The full 256 byte internal RAM can be indirectly addressed using the 8-bit pointers R0 and R1.
The first 128 bytes of IDATA include the DATA space. The hardware stack is also located in the
IDATA space when XSTK = 0.
3.2.3
SFR
The upper 128 direct addresses (80H–FFH) access the I/O registers. I/O registers on AT89LP
devices are referred to as Special Function Registers. The SFRs can only be accessed through
direct addressing. All SFR locations are not implemented. See Section 4. for a listed of available
SFRs.
3.3
External Data Memory
AT89LP microcontrollers support a 16-bit external memory address space for up to 64K bytes of
external data memory (XDATA). The external memory space is accessed with the MOVX
instructions. Some internal data memory resources are mapped into portions of the external
address space as shown in Figure 3-3. These memory spaces may require configuration before
the CPU can access them. The AT89LP6440 includes 4K bytes of on-chip Extra RAM (EDATA)
and 8K bytes of nonvolatile Flash data memory (FDATA).
Figure 3-3.
FFFF
External Data Memory Map
FFFF
FFFF
External Data
(XDATA: 52KB)
External Data
(XDATA: 64KB)
External Data
(XDATA: 60KB)
3000
2FFF
Flash Data
(FDATA: 8KB)
1000
0FFF
Extra RAM
(EDATA: 4KB)
1000
0FFF
Extra RAM
(EDATA: 4KB)
0000
EXRAM = 1
3.3.1
EXRAM = 0
DMEN = 0
EXRAM = 0
DMEN = 1
XDATA
The external data memory space can accommodate up to 64KB of external memory. The
AT89LP6440 uses the standard 8051 external memory interface with the upper address byte on
Port 2, the lower address byte and data in/out multiplexed on Port 0, and the ALE, RD and WR
strobes. MOVX instructions targeted to XDATA require a minimum of 4 clock cycles. XDATA can
be accessed with both 16-bit (MOVX @DPTR) and 8-bit (MOVX @Ri) addresses. See Section
3.3.4 on page 16 for more details of the external memory interface.
12
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Some internal data memory spaces are mapped into portions of the XDATA address space. In
this case the lower address ranges will access internal resources instead of external memory.
Addresses above the range implemented internally will default to XDATA. The AT89LP6440
supports up to 52K or 60K bytes of external memory when using the internally mapped memories. Setting the EXRAM bit (AUXR.1) to one will force all MOVX instructions to access the entire
64KB XDATA regardless of their address (See “AUXR – Auxiliary Control Register” on page 17).
3.3.2
EDATA
The Extra RAM is a portion of the external memory space implemented as an internal 4K byte
auxiliary RAM. The Extra RAM is mapped into the EDATA space at the bottom of the external
memory address space, from 0000H to 0FFFH. MOVX instructions to this address range will
access the internal Extra RAM. EDATA can be accessed with both 16-bit (MOVX @DPTR) and
8-bit (MOVX @Ri) addresses. When 8-bit addresses are used, the PAGE register (086H) supplies the upper address bits. The PAGE register breaks EDATA into sixteen 256-byte pages. A
page cannot be specified independently for MOVX @R0 and MOVX @R1. Setting PAGE above
0FH enables XDATA access, but does not change the value of Port 2. When 16-bit addresses
are used (DPTR), the IAP bit (MEMCON.7) must be zero to access EDATA. MOVX instructions
to EDATA require a minimum of 2 clock cycles.
Table 3-2.
PAGE – EDATA Page Register
PAGE = 86H
Reset Value = 0000 0000B
Not Bit Addressable
PAGE.7
PAGE.6
PAGE.5
PAGE.4
PAGE.3
PAGE.2
PAGE.1
PAGE.0
7
6
5
4
3
2
1
0
Bit
Symbol
Function
PAGE7-0
Selects which 256-byte page of EDATA is currently accessible by MOVX @Ri instructions when PAGE < 10H. Any PAGE
value between 10H and FFH will selected XDATA; however, this value will not be output on P2.
3.3.3
FDATA
The Flash Data Memory is a portion of the external memory space implemented as an internal
nonvolatile data memory. Flash Data Memory is enabled by setting the DMEN bit (MEMCON.3)
to one. When IAP = 0 and DMEN = 1, the Flash Data Memory is mapped into the FDATA space,
directly above the EDATA space near the bottom of the external memory address space, from
1000H to 2FFFH. (See Figure 3-3). MOVX instructions to this address range will access the
internal nonvolatile memory. FDATA is not accessible while DMEN = 0. FDATA can be
accessed only by 16-bit (MOVX @DPTR) addresses. MOVX @Ri instructions to the FDATA
address range will access external memory. Addresses above the FDATA range are mapped to
XDATA. MOVX instructions to FDATA require a minimum of 4 clock cycles.
3.3.3.1
Write Protocol
The FDATA address space accesses an internal nonvolatile data memory. This address space
can be read just like EDATA by issuing a MOVX A,@DPTR; however, writes to FDATA require a
more complex protocol and take several milliseconds to complete. The AT89LP6440 uses an
idle-while-write architecture where the CPU is placed in an idle state while the write occurs.
When the write completes, the CPU will continue executing with the instruction after the
MOVX @DPTR,A instruction that started the write. All peripherals will continue to function during
the write cycle; however, interrupts will not be serviced until the write completes.
13
3706A–MICRO–9/09
To enable write access to the nonvolatile data memory, the MWEN bit (MEMCON.4) must be set
to one. When MWEN = 1 and DMEN = 1, MOVX @DPTR,A may be used to write to FDATA.
FDATA uses flash memory with a page-based programming model. Flash data memory differs
from traditional EEPROM data memory in the method of writing data. EEPROM generally can
update a single byte with any value. Flash memory splits programming into write and erase
operations. A Flash write can only program zeroes, i.e change ones into zeroes ( 1 →0 ). Any
ones in the write data are ignored. A Flash erase sets an entire page of data to ones so that all
bytes become FFH. Therefore after an erase, each byte in the page can be written only once
with any possible value. Bytes can not be overwritten once they are changed from the erased
state without possibility of corrupting the data. Therefore, if even a single byte needs updating;
then the contents of the page must first be saved, the entire page must be erased and the zero
bits in all bytes (old and new data combined) must be written. Avoiding unnecessary page
erases greatly improves the endurance of the memory.
The AT89LP6440 includes 64 data pages of 128 bytes each. One or more bytes in a page may
be written at one time. The AT89LP6440 includes a temporary page buffer of 64 bytes, or half of
a page. Because the page buffer is 64 bytes long, the maximum number of bytes written at one
time is 64. Therefore, two write cycles are required to fill the entire 128-byte page, one for the
low half page (00H–3FH) and one for the high half page (40H–7FH) as shown in Figure 3-4.
Figure 3-4.
Page Programming Structure
00
3F
Page Buffer
Data Memory
Low Half Page
00
High Half Page
3F 40
7F
The LDPG bit (MEMCON.5) allows multiple data bytes to be loaded to the temporary page buffer. While LDPG = 1, MOVX @DPTR,A instructions will load data to the page buffer, but will not
start a write sequence. Note that a previously loaded byte must not be reloaded prior to the write
sequence. To write the half page into the memory, LDPG must first be cleared and then a
MOVX @DPTR,A with the final data byte is issued. The address of the final MOVX determines
which half page will be written. If a MOVX @DPTR,A instruction is issued while LDPG = 0 without loading any previous bytes, only a single byte will be written. The page buffer is reset after
each write operation. Figures 3-5 and Figure 3-6 on page 15 show the difference between byte
writes and page writes.
Figure 3-5.
FDATA Byte Write
DMEN
MWEN
LDPG
IDLE
tWC
tWC
MOVX
14
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Figure 3-6.
FDATA Page Write
DMEN
MWEN
LDPG
IDLE
tWC
MOVX
The auto-erase bit AERS (MEMCON.6) can be set to one to perform a page erase automatically
at the beginning of any write sequence. The page erase will erase the entire page, i.e. both the
low and high half pages. However, the write operation paired with the auto-erase can only program one of the half pages. A second write cycle without auto-erase is required to update the
other half page.
Frequently just a few bytes within a page must be updated while maintaining the state of the
other bytes. There are two options for handling this situation that allow the Flash Data memory
to emulate a traditional EEPROM memory. The simplest method is to copy the entire page into a
buffer allocated in RAM, modify the desired byte locations in the RAM buffer, and then load and
write back first the low half page (with auto-erase) and then the high half page to the Flash memory. This option requires that at least one page size of RAM is available as a temporary buffer.
The second option is to store only one half page in RAM. The unmodified bytes of the other page
are loaded directly into the Flash memory’s temporary load buffer before loading the updated
values of the modified bytes. For example, if just the low half page needs modification, the user
must first store the high half page to RAM, followed by reading and loading the unaffected bytes
of the low half page into the page buffer. Then the modified bytes of the low half page are stored
to the page buffer before starting the auto-erase sequence. The stored value of the high half
page must be written without auto-erase after the programming of the low half page completes.
This method reduces the amount of RAM required; however, more software overhead is needed
because the read-and-load-back routine must skip those bytes in the page that need to be
updated in order to prevent those locations in the buffer from being loaded with the previous
data, as this will block the new data from being loaded correctly.
A write sequence will not occur if the Brown-out Detector is active, even if the BOD reset has
been disabled. In cases where the BOD reset is disabled, the user should check the BOD status
by reading the WRTINH bit in MEMCON. If a write currently in progress is interrupted by the
BOD due to a low voltage condition, the ERR flag will be set. FDATA can always be read regardless of the BOD state.
For more details on using the Flash Data Memory, see the application note titled “AT89LP Flash
Data Memory”. FDATA may also be programmed by an external device programmer (See Section 25. on page 157).
15
3706A–MICRO–9/09
Table 3-3.
MEMCON – Memory Control Register
MEMCON = 96H
Reset Value = 0000 00XXB
Not Bit Addressable
Bit
IAP
AERS
LDPG
MWEN
DMEN
ERR
–
WRTINH
7
6
5
4
3
2
1
0
Symbol
Function
IAP
In-Application Programming Enable. When IAP = 1 and the IAP Fuse is enabled, programming of the CODE/SIG space
is enabled and MOVX @DPTR instructions will access CODE/SIG instead of EDATA or FDATA. Clear IAP to disable
programming of CODE/SIG and allow access to EDATA and FDATA.
AERS
Auto-Erase Enable. Set to perform an auto-erase of a Flash memory page (CODE, SIG or FDATA) during the next write
sequence. Clear to perform write without erase.
LDPG
Load Page Enable. Set to this bit to load multiple bytes to the temporary page buffer. Byte locations may not be loaded
more than once before a write. LDPG must be cleared before writing.
MWEN
Memory Write Enable. Set to enable programming of a nonvolatile memory location (CODE, SIG or FDATA). Clear to
disable programming of all nonvolatile memories.
DMEN
Data Memory Enable. Set to enable nonvolatile data memory and map it into the FDATA space. Clear to disable
nonvolatile data memory.
ERR
Error Flag. Set by hardware if an error occurred during the last programming sequence due to a brownout condition (low
voltage on VDD). Must be cleared by software.
WRTINH
Write Inhibit Flag. Cleared by hardware when the voltage on VDD has fallen below the minimum programming voltage.
Set by hardware when the voltage on VDD is above the minimum programming voltage.
3.3.4
External Memory Interface
The AT89LP6440 uses the standard 8051 external memory interface with the upper address on
Port 2, the lower address and data in/out multiplexed on Port 0, and the ALE, RD and WR
strobes. The interface may be used in two different configurations depending on which type of
MOVX instruction is used to access XDATA.
Figure 3-7 shows a hardware configuration for accessing up to 64K bytes of external RAM using
a 16-bit linear address. Port 0 serves as a multiplexed address/data bus to the RAM. The
Address Latch Enable strobe (ALE) is used to latch the lower address byte into an external register so that Port 0 can be freed for data input/output. Port 2 provides the upper address byte
throughout the operation. The MOVX @DPTR instructions use Linear Address mode
Figure 3-7.
External Memory 16-bit Linear Address Mode
EXTERNAL
DATA
MEMORY
DATA
AT89LP
P0
P1
ALE
LATCH
ADDR
P2
RD
WR
16
P3
WE
OE
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Figure 3-8 shows a hardware configuration for accessing 256-byte blocks of external RAM using
an 8-bit paged address. Port 0 serves as a multiplexed address/data bus to the RAM. The ALE
strobe is used to latch the address byte into an external register so that Port 0 can be freed for
data input/output. The Port 2 I/O lines (or other ports) can provide control lines to page the memory; however, this operation is not handled automatically by hardware. The software application
must change the Port 2 register when appropriate to access different pages. The MOVX @Ri
instructions use Paged Address mode.
Figure 3-8.
External Memory 8-bit Paged Address Mode
EXTERNAL
DATA
MEMORY
DATA
AT89LP
P0
P1
ALE
RD
WR
Table 3-4.
LATCH
ADDR
PAGE
BITS
WE
P3 P2
I/O
OE
AUXR – Auxiliary Control Register
CLKREG = 8EH
Reset Value = xxx0 0000B
Not Bit Addressable
Bit
Symbol
–
–
–
XSTK
WS1
WS0
EXRAM
ALES
7
6
5
4
3
2
1
0
Function
XSTK
Extended Stack Enable. When XSTK = 0 the stack resides in IDATA and is limited to 256 bytes. Set XSTK = 1 to place
the stack in EDATA for up to 4K bytes of extended stack space. All PUSH, POP, CALL and RET instructions will incur a
one or two cycle penalty when accessing the extended stack.
WS[1-0]
Wait State Select. Determines the number of wait states inserted into external memory accesses.
WS1
WS0
Wait States
RD / WR Strobe Width
0
0
0
1 x tCYC
0
1
1
2 x tCYC
1
0
2
3 x tCYC
1
1
3
4 x tCYC
EXRAM
External RAM Enable. When EXRAM = 0, MOVX instructions can access the internally mapped portions of the address
space. Accesses to addresses above internally mapped memory will access external memory. Set EXRAM = 1 to
bypass the internal memory and map the entire address space to external memory.
ALES
ALE Idle State. When ALES = 0 the idle polarity of ALE is high (active). When ALES = 1 the idle polarity of ALE is low
(inactive). The ALE strobe pulse is always active high. ALES must be zero in order to use P4.4 as a general I/O.
Note that prior to using the external memory interface, Port 2, WR (P3.6), RD (P3.7) and ALE
(P4.4) must be configured as outputs. See Section 10.1 “Port Configuration” on page 44. Port 0
is configured automatically to push-pull output mode when outputting address or data and is
17
3706A–MICRO–9/09
automatically tristated when inputting data regardless of the Port 0 configuration. The Port 0
configuration will determine the idle state of Port 0 when not accessing the external memory.
Figure 3-9 and Figure 3-10 show examples of external data memory write and read cycles,
respectively. The address on P0 and P2 is stable at the falling edge of ALE. The idle polarity of
ALE is controlled by ALES (AUXR.0). When ALES = 0 the idle polarity of ALE is high (active).
When ALES = 1 the idle polarity of ALE is low (inactive). The ALE strobe pulse is always active
high. Unlike standard 8051s, ALE will not toggle continuously when not accessing external
memory. ALES must be zero in order to use P4.4 as a general-purpose I/O. The WS bits in
AUXR can extended the RD and WR strobes by 1, 2 or 3 cycles as shown in Figures 3-11, 3-12
and 3-13. If a longer strobe is required, the application can scale the system clock with the clock
divider to meet the requirements (See Section 6.5 on page 31).
Figure 3-9.
External Data Memory Write Cycle (WS = 00B)
S1
S2
S3
S4
CLK
ALES = 1
ALE
ALES = 0
WR
P0
P0 SFR
P2
P2 SFR
DPL or Ri OUT
DATA OUT
P0 SFR
DPH or P2 OUT
Figure 3-10. External Data Memory Read Cycle (WS = 00B)
S1
S2
S3
P2 SFR
S4
CLK
ALES = 1
ALE
ALES = 0
RD
DATA SAMPLED
18
P0
P0 SFR
P2
P2 SFR
DPL or Ri OUT
FLOAT
DPH or P2 OUT
P0 SFR
P2 SFR
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Figure 3-11. MOVX with One Wait State (WS = 01B)
S1
S2
S3
W1
S4
CLK
ALE
P2
P2 SFR
DPH or P2 OUT
P2 SFR
WR
P0
P0 SFR
DPL OUT
P0 SFR
DPL OUT
DATA OUT
P0 SFR
RD
P0
FLOAT
P0 SFR
Figure 3-12. MOVX with Two Wait States (WS = 10B)
S1
S2
S3
W1
W2
S4
CLK
ALE
P2
P2 SFR
DPH or P2 OUT
P2 SFR
WR
P0
P0 SFR
DPL OUT
P0 SFR
DPL OUT
DATA OUT
P0 SFR
RD
P0
FLOAT
P0 SFR
Figure 3-13. MOVX with Three Wait States (WS = 11B)
S1
S2
S3
W1
W2
W3
S4
CLK
ALE
P2
P2 SFR
DPH or P2 OUT
P2 SFR
WR
P0
P0 SFR
DPL OUT
P0 SFR
DPL OUT
DATA OUT
P0 SFR
RD
P0
3.4
FLOAT
P0 SFR
Extended Stack
The AT89LP6440 provides an extended stack space for applications requiring additional stack
memory. By default the stack is located in the 256-byte IDATA space of internal data memory.
The IDATA stack is referenced solely by the 8-bit Stack Pointer (SP: 81H). Setting the XSTK bit
in AUXR enables the extended stack. The extended stack resides in the EDATA space for up to
4KB of stack memory. The extended stack is referenced by a 12-bit pointer formed from SP and
the four LSBs of the Extended Stack Pointer (SPX: 9EH) as shown in Figure 3-14. SP is shared
between both stacks. Note that the standard IDATA stack will not overflow to the EDATA stack
or vice versa. The stack and extended stack are mutually exclusive and SPX is ignored when
XTSK = 0. An application choosing to switch between stacks by toggling XSTK must maintain
19
3706A–MICRO–9/09
separate copies of SP for use with each stack space. Interrupts should be disabled while swapping copies of SP in such an application to prevent illegal stack accesses.
All interrupt calls and PUSH, POP, ACALL, LCALL, RET and RETI instructions will incur a one
or two-cycle penalty while the extended stack is enabled, depending on the number of stack
access in each instruction. The extended stack may only exist within the internal EDATA space;
it cannot be placed in XDATA. The stack will continue to use EDATA even if EDATA is disabled
by setting EXRAM = 1.
Figure 3-14. Stack Configurations
FFFh
EDATA
(4K)
FFh
7
IDATA
(256)
0
3 07
SPX
0
SP
SP
00h
00h
XSTK = 0
3.5
XSTK = 1
In-Application Programming (IAP)
The AT89LP6440 supports In-Application Programming (IAP), allowing the program memory to
be modified during execution. IAP can be used to modify the user application on the fly or to use
program memory for nonvolatile data storage. The same page structure write protocol for
FDATA also applies to IAP (See Section 3.3.3.1 “Write Protocol” on page 13). The CPU is
always placed in idle while modifying the program memory. When the write completes, the CPU
will continue executing with the instruction after the MOVX @DPTR,A instruction that started the
write.
To enable access to the program memory, the IAP bit (MEMCON.7) must be set to one and the
IAP User Fuse must be enabled. The IAP User Fuse can disable all IAP operations. When this
fuse is disabled, the IAP bit will be forced to 0. While IAP is enabled, all MOVX @DPTR instructions will access the CODE space instead of EDATA/FDATA/XDATA. IAP also allows
reprogramming of the User Signature Array when SIGEN = 1. The IAP access settings are summarized in Table 3-5.
Table 3-5.
20
IAP Access Settings
IAP
SIGEN
DMEN
MOVX @DPTR
MOVC @DPTR
0
0
0
EDATA (0000–0FFFH)
CODE (0000–FFFFH)
0
0
1
FDATA (1000–2FFFH)
CODE (0000–FFFFH)
0
1
0
EDATA (0000–0FFFH)
SIG (0000–01FFH)
0
1
1
FDATA (1000–2FFFH)
SIG (0000–01FFH)
1
0
X
CODE (0000–FFFFH)
CODE (0000–FFFFH)
1
1
X
SIG (0000–01FFH)
SIG (0000–01FFH)
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
4. Special Function Registers
A map of the on-chip memory area called the Special Function Register (SFR) space is shown in
Table 4-1. See also “Register Index” on page 153.
Note that not all of the addresses are occupied, and unoccupied addresses may not be implemented on the chip. Read accesses to these addresses will in general return random data, and
write accesses will have an indeterminate effect. User software should not write to these unlisted
locations, since they may be used in future products to invoke new features.
Table 4-1.
AT89LP6440 SFR Map and Reset Values
8
9
A
B
C
D
E
F
0F8H
0FFH
0F0H
B
0000 0000
0E8H
SPSR
000x x000
SPCR
0000 0000
SPDR
xxxx xxxx
0E0H
ACC
0000 0000
AX
0000 0000
DSPR
0000 0000
DADC
0000 0000
DADI
0000 0000
PSW
0000 0000
T2CCA
0000 0000
T2CCL
0000 0000
T2CON
0000 0000
T2MOD
0000 0000
0D8H
0D0H
0C8H
0C0H
P4
xx11 1111
0B8H
IP
0000 0000
0B0H
P3
1111 1111
0A8H
IE
0000 0000
0A0H
BX
0000 0000
SADEN
0000 0000
SADDR
0000 0000
P2
1111 1111
0F7H
0EFH
MACL
0000 0000
MACH
0000 0000
0E7H
DADL
0000 0000
DADH
0000 0000
0DFH
T2CCH
0000 0000
T2CCC
0000 0000
T2CCF
0000 0000
0D7H
RCAP2L
0000 000
RCAP2H
0000 0000
TL2
0000 000
TH2
0000 0000
0CFH
P1M0(2)
P1M1
0000 0000
P2M0(2)
P2M1
0000 0000
P0M0
1111 1111
P0M1
0000 0000
TWCR
0000 0000
FIRD
0000 0000
TWSR
0000 0000
P3M0(2)
P3M1
0000 0000
0C7H
P4M0(2)
P4M1
xx00 0000
0BFH
IE2
xxxx x000
IP2
xxxx x000
IP2H
xxxx x000
IPH
0000 0000
0B7H
TWAR
0000 0000
TWDR
0000 0000
TWBR
0000 0000
AREF
0000 0000
0AFH
WDTRST
(write-only)
WDTCON
0000 x000
0A7H
DPCF
0000 00x0
98H
SCON
0000 0000
SBUF
xxxx xxxx
GPMOD
0000 0000
GPLS
0000 0000
GPIEN
0000 0000
GPIF
0000 0000
SPX
xxxx 0000
ACSRB
1100 0000
9FH
90H
P1
1111 1111
TCONB
0010 0100
RL0
0000 0000
RL1
0000 0000
RH0
0000 0000
RH1
0000 0000
MEMCON
0000 00xx
ACSRA
0000 0000
97H
88H
TCON
0000 0000
TMOD
0000 0000
TL0
0000 0000
TL1
0000 0000
TH0
0000 0000
TH1
0000 0000
AUXR
0000 0000
CLKREG
0000 x000
8FH
SP
0000 0111
DP0L
0000 0000
DP0H
0000 0000
DP1L
0000 0000
DP1H
0000 0000
PAGE
0000 0000
PCON
0000 0000
87H
1
2
3
4
5
6
7
80H
0
Notes:
1. All SFRs in the left-most column are bit-addressable.
2. Reset value is 1111 1111B when Tristate-Port Fuse is enabled and 0000 0000B when disabled.
21
3706A–MICRO–9/09
5. Enhanced CPU
The AT89LP6440 uses an enhanced 8051 CPU that runs at 6 to 12 times the speed of standard
8051 devices (or 3 to 6 times the speed of X2 8051 devices). The increase in performance is
due to two factors. First, the CPU fetches one instruction byte from the code memory every clock
cycle. Second, the CPU uses a simple two-stage pipeline to fetch and execute instructions in
parallel. This basic pipelining concept allows the CPU to obtain up to 1 MIPS per MHz. A simple
example is shown in Figure 5-1.
The 8051 instruction set allows for instructions of variable length from 1 to 3 bytes. In a singleclock-per-byte-fetch system this means each instruction takes at least as many clocks as it has
bytes to execute. The majority of instructions in the AT89LP6440 follow this rule: the instruction
execution time in clock cycles equals the number of bytes per instruction, with a few exceptions.
Branches and Calls require an additional cycle to compute the target address and some other
complex instructions require multiple cycles. See “Instruction Set Summary” on page 143. for
more detailed information on individual instructions. Figures 5-2 and 5-3 show examples of 1and 2-byte instructions.
Figure 5-1.
Parallel Instruction Fetches and Executions
Tn
Tn+1
Fetch
Execute
Tn+2
System Clock
nth Instruction
(n+1)th Instruction
Fetch
(n+2)th Instruction
Figure 5-2.
Execute
Fetch
Single-cycle ALU Operation (Example: INC R0)
T1
T2
T3
System Clock
Total Execution Time
Register Operand Fetch
ALU Operation Execute
Result Write Back
Fetch Next Instruction
22
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Figure 5-3.
Two-cycle ALU Operation (Example: ADD A, #data)
T1
T2
T3
System Clock
Total Execution Time
Fetch Immediate Operand
ALU Operation Execute
Result Write Back
Fetch Next Instruction
5.1
Multiply–Accumulate Unit (MAC)
The AT89LP6440 includes a multiply and accumulate (MAC) unit that can significantly speed up
many mathematical operations required for digital signal processing. The MAC unit includes a
16-by-16 bit multiplier and a 40-bit adder that can perform integer or fractional multiply-accumulate operations on signed 16-bit input values. The MAC unit also includes a 1-bit arithmetic
shifter that will left or right shift the contents of the 40-bit MAC accumulator register (M).
A block diagram of the MAC unit is shown in Figure 5-4. The 16-bit signed operands are provided by the register pairs (AX,ACC) and (BX,B) where AX (E1H) and BX (F7H) hold the higher
order bytes. The 16-by-16 bit multiplication is computed through partial products using the
AT89LP6440’s 8-bit multiplier. The 32-bit signed product is added to the 40-bit M accumulator
register. The MAC operation is summarized as follows:
M ←M + {AX, ACC} × {BX, B}
MAC AB:
All computation is done in signed two’s complement form.
Figure 5-4.
Multiply–Accumulate Unit
AX
SMLA
SMLB
ACC
BX
B
8 x 8-bit Signed MULT
40-bit ADD
PSW
MRW
M4
M3
M2
Shifter
M1
M0
MACH
MACL
23
3706A–MICRO–9/09
The MAC operation is performed by executing the MAC AB (A5 A4H) extended instruction. This
two-byte instruction requires nine clock cycles to complete. The operand registers are not modified by the instruction and the result is stored in the 40-bit M register. MAC AB also updates the
C and OV flags in PSW. C represents the sign of the MAC result and OV is the two’s complement overflow. Note that MAC AB will not clear OV if it was previously set to one.
Three additional extended instructions operate directly on the M register. CLR M (A5 E4H)
clears the entire 40-bit register in two clock cycles. LSL M (A5 23H) and ASR (A5 03H) shift M
one bit to the left and right respectively. Right shifts are done arithmetically, i.e. the sign is
preserved.
The 40-bit M register is accessible 16-bits at a time through a sliding window as shown in Figure
5-5. The MRW1-0 bits in DSPR (Table 5-1) select which 16-bit segment is currently accessible
through the MACL and MACH addresses. For normal fixed point operations the window can be
fixed to the rank of interest. For example, multiplying two 1.15 format numbers places a 2.30 format result in the M register. If MRW is set to 10B, a 1.15 value is obtained after performing a
single LSL M.
Figure 5-5.
M Register with Sliding Window
M
Byte 4
Byte 3
Byte 2
Byte 1
Byte 0
39 – 32
31 – 24
23 – 16
15 – 8
7–0
MACH
MACL
MACH
MACH
MACH
MACL
MACL
MACL
MRW1-0 = 00B
MRW1-0 = 01B
MRW1-0 = 10B
MRW1-0 = 11B
As a consequence of the MAC unit, the standard 8x8 MUL AB instruction can support signed
multiplication. The SMLA and SMLB bits in DSPR control the multiplier’s interpretation of the
ACC and B registers, allowing any combination of signed and unsigned operand multiplication.
These bits have no effect on the MAC operation which always multiplies signed-by-signed.
5.2
Enhanced Dual Data Pointers
The AT89LP6440 provides two 16-bit data pointers: DPTR0 formed by the register pair DPOL
and DPOH (82H an 83H), and DPTR1 formed by the register pair DP1L and DP1H (84H and
85H). The data pointers are used by several instructions to access the program or data memories. The Data Pointer Configuration Register (DPCF) controls operation of the dual data
pointers (Table 5-5 on page 27). The DPS bit in DPCF selects which data pointer is currently referenced by instructions including the DPTR operand. Each data pointer may be accessed at its
respective SFR addresses regardless of the DPS value. The AT89LP6440 provides two methods for fast context switching of the data pointers:
• Bit 2 of DPCF is hard-wired as a logic 0. The DPS bit may be toggled (to switch data pointers)
simply by incrementing the DPCF register, without altering other bits in the register
unintentionally. This is the preferred method when only a single data pointer will be used at
one time.
EX:
24
INC
DPCF
; Toggle DPS
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Table 5-1.
DSPR – Digital Signal Processing Configuration Register
DSPR = E2H
Reset Value = 0000 0000B
Not Bit Addressable
Bit
MRW1
MRW0
SMLB
SMLA
CBE1
CBE0
MVCD
DPRB
7
6
5
4
3
2
1
0
Symbol
Function
MRW1-0
M Register Window. Selects which pair of bytes from the 5-byte M register is accessible through MACH (E5H) and
MACL (E4H) as shown in Figure 5-5. For example, MRW = 10B for normal 16-bit fixed-point operations where the lowest
order portion of the fractional result is discarded.
SMLB
Signed Multiply Operand B. When SMLB = 0, the MUL AB instruction treats the contents of B as an unsigned value.
When SMLB = 1, the MUL AB instruction interprets the contents of B as a signed two’s complement value. SMLB does
not affect the MAC operation.
SMLA
Signed Multiply Operand A. When SMLA = 0, the MUL AB instruction treats the contents of ACC as an unsigned value.
When SMLA = 1, the MUL AB instruction interprets the contents of ACC as a signed two’s complement value. SMLA
does not affect the MAC operation.
CBE1
DPTR1 Circular Buffer Enable. Set CBE1 = 1 to configure DPTR1 for circular addressing over the two circular buffer
address ranges. Clear CBE1 for normal DPTR operation.
CBE0
DPTR0 Circular Buffer Enable. Set CBE0 = 1 to configure DPTR0 for circular addressing over the two circular buffer
address ranges. Clear CBE0 for normal DPTR operation.
MVCD
MOVC Index Disable. When MVCD = 0, the MOVC A, @A+DPTR instruction functions normally with indexed
addressing. Setting MVCD = 1 disables the indexed addressing mode such that MOVC A, @A+DPTR functions as
MOVC A, @DPTR.
DPRB
DPTR1 Redirect to B. DPRB selects the source/destination register for MOVC/MOVX instructions that reference DPTR1.
When DPRB = 0, ACC is the source/destination. When DPRB = 1, B is the source/destination. DPRB does not change
the index register for MOVC instructions.
• In some cases, both data pointers must be used simultaneously. To prevent frequent toggling
of DPS, the AT89LP6440 supports a prefix notation for selecting the opposite data pointer
per instruction. All DPTR instructions, with the exception of JMP @A+DPTR, when prefixed
with an 0A5H opcode will use the inverse value of DPS (DPS) to select the data pointer.
Some assemblers may support this operation by using the /DPTR operand. For example, the
following code performs a block copy within EDATA:
COPY:
MOV
DPCF, #00H
; DPS = 0
MOV
DPTR, #SRC
; load source address to dptr0
MOV
/DPTR, #DST
; load destination address to dptr1
MOV
R7, #BLKSIZE
; number of bytes to copy
MOVX A, @DPTR
; read source (dptr0)
INC
; next src (dptr0+1)
DPTR
MOVX @/DPTR, A
; write destination (dptr1)
INC
; next dst (dptr1+1)
/DPTR
DJNZ R7, COPY
For assemblers that do not support this notation, the 0A5H prefix must be declared in-line:
EX:
DB
0A5H
INC
DPTR
; equivalent to INC /DPTR
25
3706A–MICRO–9/09
A summary of data pointer instructions with fast context switching is listed inTable 5-2.
Table 5-2.
Data Pointer Instructions
Operation
5.2.1
Instruction
DPS = 0
DPS = 1
JMP @A+DPTR
JMP @A+DPTR0
JMP @A+DPTR1
MOV DPTR, #data16
MOV DPTR0, #data16
MOV DPTR1, #data16
MOV /DPTR, #data16
MOV DPTR1, #data16
MOV DPTR0, #data16
INC DPTR
INC DPTR0
INC DPTR1
INC /DPTR
INC DPTR1
INC DPTR0
MOVC A,@A+DPTR
MOVC A,@A+DPTR0
MOVC A,@A+DPTR1
MOVC A,@A+/DPTR
MOVC A,@A+DPTR1
MOVC A,@A+DPTR0
MOVX A,@DPTR
MOVX A,@DPTR0
MOVX A,@DPTR1
MOVX A,@/DPTR
MOVX A,@DPTR1
MOVX A,@DPTR0
MOVX @DPTR, A
MOVX @DPTR0, A
MOVX @DPTR1, A
MOVX @/DPTR, A
MOVX @DPTR1, A
MOVX @DPTR0, A
Data Pointer Update
The Dual Data Pointers on the AT89LP6440 include two features that control how the data
pointers are updated. The data pointer decrement bits, DPD1 and DPD0 in DPCF, configure the
INC DPTR instruction to act as DEC DPTR. The resulting operation will depend on DPS as
shown in Table 5-3.
Table 5-3.
Data Pointer Decrement Behavior
Equivalent Operation for INC DPTR and INC /DPTR
DPS = 0
DPS = 1
DPD1
DPD0
INC DPTR
INC /DPTR
INC DPTR
INC /DPTR
0
0
INC DPTR0
INC DPTR1
INC DPTR1
INC DPTR0
0
1
DEC DPTR0
INC DPTR1
INC DPTR1
DEC DPTR0
1
0
INC DPTR0
DEC DPTR1
DEC DPTR1
INC DPTR0
1
1
DEC DPTR0
DEC DPTR1
DEC DPTR1
DEC DPTR0
The data pointer update bits, DPU1 and DPU0, allow MOVX @DPTR and MOVC @DPTR
instructions to update the selected data pointer automatically in a post-increment or post-decrement fashion. The direction of update depends on the DPD1 and DPD0 bits as shown in Table
5-4.
Table 5-4.
Data Pointer Auto-Update
Update Operation for MOVX and MOVC (DPU1 = 1 & DPU0 = 1)
DPS = 0
26
DPS = 1
DPD1
DPD0
DPTR
/DPTR
DPTR
/DPTR
0
0
DPTR0++
DPTR1++
DPTR1++
DPTR0++
0
1
DPTR0--
DPTR1++
DPTR1++
DPTR0--
1
0
DPTR0++
DPTR1--
DPTR1--
DPTR0++
1
1
DPTR0--
DPTR1--
DPTR1--
DPTR0--
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Table 5-5.
DPCF – Data Pointer Configuration Register
DPCF = A2H
Reset Value = 0000 00X0B
Not Bit Addressable
Bit
DPU1
DPU0
DPD1
DPD0
SIGEN
0
–
DPS
7
6
5
4
3
2
1
0
Symbol
Function
DPU1
Data Pointer 1 Update. When set, MOVX @DPTR and MOVC @DPTR instructions that use DPTR1 will also update
DPTR1 based on DPD1. If DPD1 = 0 the operation is post-increment and if DPD1 = 1 the operation is post-decrement.
When DPU1 = 0, DPTR1 is not updated.
DPU0
Data Pointer 0 Update. When set, MOVX @DPTR and MOVC @DPTR instructions that use DPTR0 will also update
DPTR0 based on DPD0. If DPD0 = 0 the operation is post-increment and if DPD0 = 1 the operation is post-decrement.
When DPU0 = 0, DPTR0 is not updated.
DPD1
Data Pointer 1 Decrement. When set, INC DPTR instructions targeted to DPTR1 will decrement DPTR1. When cleared,
INC DPTR instructions will increment DPTR1. DPD1 also determines the direction of auto-update for DPTR1 when
DPU1 = 1.
DPD0
Data Pointer 0 Decrement. When set, INC DPTR instructions targeted to DPTR0 will decrement DPTR0. When cleared,
INC DPTR instructions will increment DPTR0. DPD0 also determines the direction of auto-update for DPTR0 when
DPU0 = 1.
SIGEN
Signature Enable. When SIGEN = 1 all MOVC @DPTR instructions and all IAP accesses will target the signature array
memory. When SIGEN = 0, all MOVC and IAP accesses target CODE memory.
DPS
Data Pointer Select. DPS selects the active data pointer for instructions that reference DPTR. When DPS = 0, DPTR will
target DPTR0 and /DPTR will target DPTR1. When DPS = 1, DPTR will target DPTR1 and /DPTR will target DPTR0.
5.2.2
5.2.2.1
Data Pointer Operating Modes
The Dual Data Pointers on the AT89LP6440 include three additional operating modes that affect
data pointer based instructions. These modes are controlled by bits in DSPR.
DPTR Redirect
The Data Pointer Redirect to B bit, DPRB (DSPR.0), allows MOVX and MOVC instructions to
use the B register as the data source/destination when the instruction references DPTR1 as
shown in Table 5-6 and Table 5-7. DPRB can improve the efficiency of routines that must fetch
multiple operands from different RAM locations.
Table 5-6.
MOVX @DPTR Operating Modes
Equivalent Operation for MOVX
MOVX A, @DPTR
MOVX @DPTR, A
DPRB
DPS
DPTR
/DPTR
DPTR
/DPTR
0
0
MOVX
A, @DPTR0
MOVX
A, @DPTR1
MOVX
@DPTR0, A
MOVX
@DPTR1, A
0
1
MOVX
A, @DPTR1
MOVX
A, @DPTR0
MOVX
@DPTR1, A
MOVX
@DPTR0, A
1
0
MOVX
A, @DPTR0
MOVX
B, @DPTR1
MOVX
@DPTR0, A
MOVX
@DPTR1, B
1
1
MOVX
B, @DPTR1
MOVX
A, @DPTR0
MOVX
@DPTR1, B
MOVX
@DPTR0, A
27
3706A–MICRO–9/09
5.2.2.2
Index Disable
The MOVC Index Disable bit, MVCD (DSPR.1), disables the indexed addressing mode of the
MOVC A, @A+DPTR instruction. When MVCD = 1, the MOVC instruction functions as
MOVC A, @DPTR with no indexing as shown in Table 5-7. MVCD can improve the efficiency of
routines that must fetch multiple operands from program memory. DPRB can change the MOVC
destination register from ACC to B, but has no effect on the MOVC index register.
Table 5-7.
MOVC @DPTR Operating Modes
Equivalent Operation for MOVC A, @A+DPTR
DPS = 0
5.2.2.3
DPS = 1
MVCD
DPRB
DPTR
/DPTR
DPTR
/DPTR
0
0
MOVC
A, @A+DPTR0
MOVC
A, @A+DPTR1
MOVC
A, @A+DPTR1
MOVC
A, @A+DPTR0
0
1
MOVC
A, @A+DPTR0
MOVC
B, @A+DPTR1
MOVC
B, @A+DPTR1
MOVC
A, @A+DPTR0
1
0
MOVC
A, @DPTR0
MOVC
A, @DPTR1
MOVC
A, @DPTR1
MOVC
A, @DPTR0
1
1
MOVC
A, @DPTR0
MOVC
B, @DPTR1
MOVC
B, @DPTR1
MOVC
A, @DPTR0
Circular Buffers
The CBE0 and CBE1 bits in DSPR can configure DPTR0 and DPTR1, respectively, to operate in
circular buffer mode. The AT89LP6440 maps circular buffers into two identically sized regions of
EDATA/XDATA. These buffers can speed up convolution computations such as FIR and IAR
digital filters. The length of the buffers are set by the value of the FIRD (E3H) register for up to
256 entries. Buffer A is mapped from 0000H to FIRD and Buffer B is mapped from 0100H to
100H+FIRD as shown in Figure 5-6. Both data pointers may operate in either buffer. When circular buffer mode is enabled, updates to a data pointer referencing the buffer region will follow
circular addressing rules. If the data pointer is equal to FIRD or 100H+FIRD any increment will
cause it to overflow to 0000H or 0100H respectively. If the data pointer is equal to 0000H or
0100H any decrement will cause it to underflow to FIRD or 100H+FIRD respectively. In this
mode, updates can be either an explicit INC DPTR or an automatic update using DPUn where
the DPDn bits control the direction. The data pointer will increment or decrement normally at any
other addresses. Therefore, when circular addressing is in use, the data pointers can still operate as regular pointers in the FIRD+1 to 00FFH and greater than 100H+FIRD ranges.
Figure 5-6.
Circular Buffer Mode
DPDn = 1
DPDn = 1
B
DPDn = 0
DPTR
DPDn = 0
100h + FIRD
0100h
FIRD
A
DPTR
0000h
28
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
5.3
Instruction Set Extensions
Table 5-8 lists the additions to the 8051 instruction set that are supported by the AT89LP6440.
For more information on the instruction set see Section 22. “Instruction Set Summary” on page
143. For detailed descriptions of the extended instructions see Section 22.1 “Instruction Set
Extensions” on page 147.
Table 5-8.
Opcode
AT89LP6440 Extended Instructions
Mnemonic
Description
Bytes
Cycles
A5 00
BREAK
Software breakpoint
2
2
A5 03
ASR M
Arithmetic shift right of M register
2
2
A5 23
LSL M
Logical shift left of M register
2
2
A5 73
JMP @A+PC
Indirect jump relative to PC
2
3
A5 90
MOV /DPTR, #data16
Move 16-bit constant to alternate data
pointer
4
4
A5 93
MOVC A, @A+/DPTR
Move code location to ACC relative to
alternate data pointer
2
4
A5 A3
INC /DPTR
Increment alternate data pointer
2
3
A5 A4
MAC AB
Multiply and accumulate
2
9
A5 B6
CJNE A, @R0, rel
Compare ACC to indirect RAM and
jump if not equal
3
4
A5 B7
CJNE A, @R1, rel
Compare ACC to indirect RAM and
jump if not equal
3
4
A5 E0
MOVX A, @/DPTR
Move external to ACC; 16-bit address
in alternate data pointer
2
3/5
A5 E4
CLR M
Clear M register
2
2
A5 F0
MOVX @/DPTR, A
Move ACC to external; 16-bit address
in alternate data pointer
2
3/5
• The /DPTR instructions provide support for the dual data pointer features described above
(See Section 5.2).
• The ASR M, LSL M, CLR M and MAC AB instructions are part of the Multiply-Accumulate
Unit (See Section 5.1).
• The JMP @A+PC instruction supports localized jump tables without using a data pointer.
• The CJNE A, @Ri, rel instructions allow compares of array values with non-constant values.
• The BREAK instruction is used by the On-Chip Debug system. See Section 24. on page 155.
29
3706A–MICRO–9/09
6. System Clock
The system clock is generated directly from one of three selectable clock sources. The three
sources are the on-chip crystal oscillator, external clock source, and internal RC oscillator. The
on-chip crystal oscillator may also be configured for low or high speed operation. The clock
source is selected by the Clock Source User Fuses as shown in Table 6-1. See “User Configuration Fuses” on page 164. By default, no internal clock division is used to generate the CPU clock
from the system clock. However, the system clock divider may be used to prescale the system
clock. The choice of clock source also affects the start-up time after a POR, BOD or Powerdown event (See “Reset” on page 32 or “Power-down Mode” on page 36)
Table 6-1.
6.1
Clock Source Settings
Clock Source
Fuse 1
Clock Source
Fuse 0
0
0
High Speed Crystal Oscillator (f > 500 kHz)
0
1
Low Speed Crystal Oscillator (f ≤100 kHz)
1
0
External Clock on XTAL1
1
1
Internal 8 MHz RC Oscillator
Selected Clock Source
Crystal Oscillator
When enabled, the internal inverting oscillator amplifier is connected between XTAL1 and
XTAL2 for connection to an external quartz crystal or ceramic resonator. The oscillator may
operate in either high-speed or low-speed mode. Low-speed mode is intended for 32.768 kHz
watch crystals and consumes less power than high-speed mode. The configuration as shown in
Figure 6-1 applies for both high and low speed oscillators. Note that the internal structure of the
device adds about 10 pF of capacitance to both XTAL1 and XTAL2, so that in some cases an
external capacitor may NOT be required. It is recommended that a resistor R1 be connected to
XTAL1, instead of load capacitor C1, for improved startup performance. The total capacitance
on XTAL1 or XTAL2, including the external load capacitor plus internal device load, board trace
and crystal loadings, should not exceed 20 pF. When using the crystal oscillator, P4.0 and P4.1
will have their inputs and outputs disabled. Also, XTAL2 in crystal oscillator mode should not be
used to directly drive a board-level clock without a buffer.
Figure 6-1.
Crystal Oscillator Connections
C2
~10 pF
R1
~10 pF
Note:
1.
C1
R1
30
= 0–10 pF for Crystals
= 0–10 pF for Ceramic Resonators
= 4–5 MΩ
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
6.2
External Clock Source
The external clock option disables the oscillator amplifier and allows XTAL1 to be driven directly
by an external clock source as shown in Figure 6-2. XTAL2 may be left unconnected, used as
general purpose I/O P4.1, or configured to output a divided version of the system clock.
Figure 6-2.
External Clock Drive Configuration
NC, GPIO, or
CLKOUT
XTAL2 (P4.1)
EXTERNAL
OSCILLATOR
SIGNAL
XTAL1 (P4.0)
GND
6.3
Internal RC Oscillator
The AT89LP6440 has an Internal RC oscillator (IRC) tuned to 8.0 MHz ±2.5%. When enabled
as the clock source, XTAL1 and XTAL2 may be used as P4.0 and P4.1 respectively. XTAL2
may also be configured to output a divided version of the system clock. The frequency of the
oscillator may be adjusted within limits by changing the RC Calibration Byte stored at byte 128 of
the User Signature Array. This location may be updated using the IAP interface (location 0180H
in SIG space) or by an external device programmer (UROW location 0080H). See Section 25.8
“User Signature and Analog Configuration” on page 165. A copy of the factory calibration byte is
stored at byte 8 of the Atmel Signature Array (0008H in SIG space).
6.4
System Clock Out
When the AT89LP6440 is configured to use either an external clock or the internal RC oscillator,
the system clock divided by 2 may be output on XTAL2 (P4.1). The clock out feature is enabled
by setting the COE bit in CLKREG. For example, setting COE = “1” when using the internal oscillator will result in a 4.0 MHz (±2.5%) clock output on P4.1. P4.1 must be configured as an output
in order to use the clock out feature.
6.5
System Clock Divider
The CDV2-0 bits in CLKREG allow the system clock to be divided down from the selected clock
source by powers of 2. The clock divider provides users with a greater frequency range when
using the Internal RC Oscillator. For example, to achieve a 1 MHz system frequency when using
the IRC, CDV2-0 should be set to 011B for divide-by-8 operation. The divider can also be used to
reduce power consumption by decreasing the operational frequency during non-critical periods.
The resulting system frequency is given by the following equation:
f OSC
f SYS = -----------CDV
2
where fOSC is the frequency of the selected clock source. The clock divider will prescale the clock
for the CPU and all peripherals. The value of CDV may be changed at any time without interrupt-
31
3706A–MICRO–9/09
ing normal execution. Changes to CDV are synchronized such that the system clock will not
pass through intermediate frequencies. When CDV is updated, the new frequency will take
affect within a maximum period of 128 x tOSC.
CLKREG – Clock Control Register
Table 6-2.
CLKREG = 8FH
Reset Value = 0000 0000B
Not Bit Addressable
Bit
Symbol
TPS[3-0]
TPS3
TPS2
TPS1
TPS0
CDV2
CDV1
CDV0
COE
7
6
5
4
3
2
1
0
Function
Timer Prescaler. The Timer Prescaler selects the time base for Timer 0, Timer 1, Timer 2 and the Watchdog Timer. The
prescaler is implemented as a 4-bit binary down counter. When the counter reaches zero it is reloaded with the value
stored in the TPS bits to give a division ratio between 1 and 16. By default the timers will count every clock cycle (TPS =
0000B). To configure the timers to count at a standard 8051 rate of once every 12 clock cycles, TPS should be set to
1011B.
System Clock Division. Determines the frequency of the system clock relative to the oscillator clock source.
CDIV2
CDIV1
CDIV0
System Clock Frequency
0
0
0
fOSC/1
0
0
1
fOSC/2
0
1
0
fOSC/4
0
1
1
fOSC/8
1
0
0
fOSC/16
1
0
1
fOSC/32
1
1
0
fOSC/64
1
1
1
fOSC/128
CDV[2-0]
COE
Clock Out Enable. Set COE to output the system clock divided by 2 on XTAL2 (P4.1). The internal RC oscillator or
external clock source must be selected in order to use this feature and P4.1 must be configured as an output.
7. Reset
During reset, all I/O Registers are set to their initial values, the port pins are tristated, and the
program starts execution from the Reset Vector, 0000H. The AT89LP6440 has five sources of
reset: power-on reset, brown-out reset, external reset, watchdog reset, and software reset.
7.1
Power-on Reset
A Power-on Reset (POR) is generated by an on-chip detection circuit. The detection level VPOR
is nominally 1.4V. The POR is activated whenever VDD is below the detection level. The POR circuit can be used to trigger the start-up reset or to detect a supply voltage failure in devices
without a brown-out detector. The POR circuit ensures that the device is reset from power-on. A
power-on sequence is shown in Figure 7-1 on page 33. When VDD reaches the Power-on Reset
threshold voltage VPOR, an initialization sequence lasting tPOR is started. When the initialization
sequence completes, the start-up timer determines how long the device is kept in POR after VDD
rise. The POR signal is activated again, without any delay, when V DD falls below the POR
threshold level. A Power-on Reset (i.e. a cold reset) will set the POF flag in PCON. The internally
32
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
generated reset can be extended beyond the power-on period by holding the RST pin low longer
than the time-out.
Figure 7-1.
Power-on Reset Sequence (BOD Disabled)
VPOR
VDD
VPOR
tPOR + tSUT
Time-out
RST
(RST Tied to VCC)
Internal
Reset
RST
VIH
(RST Controlled Externally)
Internal
Reset
tRHD
If the Brown-out Detector (BOD) is also enabled, the start-up timer does not begin counting until
after VDD reaches the BOD threshold voltage VBOD as shown in Figure 7-2. However, if this event
occurs prior to the end of the initialization sequence, the timer must first wait for that sequence to
complete before counting.
Figure 7-2.
Power-on Reset Sequence (BOD Enabled)
VBOD
VDD
Time-out
VPOR
tPOR
tSUT
RST
(RST Tied to VCC)
Internal
Reset
RST
(RST Controlled Externally)
Internal
Reset
Note:
VIH
tRHD
tPOR is approximately 143 µs ± 5%.
The start-up timer delay is user-configurable with the Start-up Time User Fuses and depends on
the clock source (Table 7-1). The Start-Up Time fuses also control the length of the start-up time
after a Brown-out Reset or when waking up from Power-down during internally timed mode. The
start-up delay should be selected to provide enough settling time for VDD and the selected clock
source. The device operating environment (supply voltage, frequency, temperature, etc.) must
33
3706A–MICRO–9/09
meet the minimum system requirements before the device exits reset and starts normal operation. The RST pin may be held low externally until these conditions are met.
Table 7-1.
Start-up Timer Settings
SUT Fuse 1
SUT Fuse 0
0
0
Clock Source
tSUT (± 5%) µs
Internal RC/External Clock
0
Crystal Oscillator
1024
Internal RC/External Clock
512
Crystal Oscillator
2048
Internal RC/External Clock
1024
Crystal Oscillator
4096
Internal RC/External Clock
4096
Crystal Oscillator
16384
1
1
0
1
7.2
16
1
Brown-out Reset
The AT89LP6440 has an on-chip Brown-out Detection (BOD) circuit for monitoring the VDD level
during operation by comparing it to a fixed trigger level. The trigger level VBOD for the BOD is
nominally 2.0V. The purpose of the BOD is to ensure that if VDD fails or dips while executing at
speed, the system will gracefully enter reset without the possibility of errors induced by incorrect
execution. A BOD sequence is shown in Figure 7-3. When VDD decreases to a value below the
trigger level VBOD, the internal reset is immediately activated. When VDD increases above the
trigger level plus about 200 mV of hysteresis, the start-up timer releases the internal reset after
the specified time-out period has expired (Table 7-1). The Brown-out Detector must be enabled
by setting the BOD Enable Fuse. (See “User Configuration Fuses” on page 164.)
Figure 7-3.
VDD
Brown-out Detector Reset
VPOR
Time-out
VBOD
tSUT
Internal
Reset
The AT89LP6440 allows for a wide VDD operating range. The on-chip BOD may not be sufficient
to prevent incorrect execution if VBOD is lower than the minimum required VDD range, such as
when a 3.6V supply is coupled with high frequency operation. In such cases an external Brownout Reset circuit connected to the RST pin may be required.
7.3
External Reset
The P4.2/RST pin can function as either an active-LOW reset input or as a digital generalpurpose I/O, P4.2. The Reset Pin Enable Fuse, when set to “1”, enables the external reset input
function on P4.2. (See “User Configuration Fuses” on page 164.) When cleared, P4.2 may be
used as an input or output pin. When configured as a reset input, the pin must be held low for at
least two clock cycles to trigger the internal reset. The RST pin includes an on-chip pull-up resistor tied to VDD. The pull-up is disabled when the pin is configured as P4.2.
34
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Note:
7.4
During a power-up sequence, the fuse selection is always overridden and therefore the pin will
always function as a reset input. An external circuit connected to this pin should not hold this
pin LOW during a power-on sequence if the pin will be configured as a general I/O, as this
will keep the device in reset until the pin transitions high. After the power-up delay, this input
will function either as an external reset input or as a digital input as defined by the fuse bit. Only a
power-up reset will temporarily override the selection defined by the reset fuse bit. Other sources
of reset will not override the reset fuse bit. P4.2/RST also serves as the In-System Programming
(ISP) enable. ISP is enabled when the external reset pin is held low. When the reset pin is disabled by the fuse, ISP may only be entered by pulling P4.2 low during power-up.
Watchdog Reset
When the Watchdog times out, it will generate an internal reset pulse lasting 16 clock cycles.
Watchdog reset will also set the WDTOVF flag in WDTCON. To prevent a Watchdog reset, the
watchdog reset sequence 1EH/E1H must be written to WDTRST before the Watchdog times
out. See “Programmable Watchdog Timer” on page 141. for details on the operation of the
Watchdog.
7.5
Software Reset
The CPU may generate an internal 16-clock cycle reset pulse by writing the software reset
sequence 5AH/A5H to the WDRST register. A software reset will set the SWRST bit in WDTCON. See “Software Reset” on page 142 for more information on software reset. Writing any
sequences other than 5AH/A5H or 1EH/E1H to WDTRST will generate an immediate reset and
set both WDTOVF and SWRST to flag an error.
8. Power Saving Modes
The AT89LP6440 supports two different power-reducing modes: Idle and Power-down. These
modes are accessed through the PCON register. Additional steps may be required to achieve
the lowest possible power consumption while using these modes.
8.1
Idle Mode
Setting the IDL bit in PCON enters idle mode. Idle mode halts the internal CPU clock. The CPU
state is preserved in its entirety, including the RAM, stack pointer, program counter, program
status word, and accumulator. The Port pins hold the logic states they had at the time that Idle
was activated. Idle mode leaves the peripherals running in order to allow them to wake up the
CPU when an interrupt is generated. The timers, UART, SPI, TWI, comparators, ADC, GPI and
CCA peripherals continue to function during Idle. If these functions are not needed during idle,
they should be explicitly disabled by clearing the appropriate control bits in their respective
SFRs. The watchdog may be selectively enabled or disabled during Idle by setting/clearing the
WDIDLE bit. The Brown-out Detector, if enabled, is always active during Idle. Any enabled interrupt source or reset may terminate Idle mode. When exiting Idle mode with an interrupt, the
interrupt will immediately be serviced, and following RETI the next instruction to be executed will
be the one following the instruction that put the device into Idle.
The power consumption during Idle mode can be further reduced by prescaling down the system
clock using the System Clock Divider (Section 6.5 on page 31). Be aware that the clock divider
will affect all peripheral functions except the ADC. Therefore baud rates or PWM periods may
need to be adjusted to maintain their rate with the new clock frequency.
35
3706A–MICRO–9/09
.
Table 8-1.
PCON – Power Control Register
PCON = 87H
Reset Value = 000X 0000B
Not Bit Addressable
Bit
SMOD1
SMOD0
PWDEX
POF
GF1
GF0
PD
IDL
7
6
5
4
3
2
1
0
Symbol
Function
SMOD1
Double Baud Rate bit. Doubles the baud rate of the UART in Modes 1, 2, or 3.
SMOD0
Frame Error Select. When SMOD0 = 1, SCON.7 is SM0. When SMOD0 = 1, SCON.7 is FE. Note that FE will be set after
a frame error regardless of the state of SMOD0.
PWDEX
Power-down Exit Mode. When PWDEX = 1, wake up from Power-down is externally controlled. When PWDEX = 1, wake
up from Power-down is internally timed.
POF
Power Off Flag. POF is set to “1” during power up (i.e. cold reset). It can be set or reset under software control and is not
affected by RST or BOD (i.e. warm resets).
GF1, GF0
General-purpose Flags
PD
Power-down bit. Setting this bit activates power-down operation. The PD bit is cleared automatically by hardware when
waking up from power-down.
IDL
Idle Mode bit. Setting this bit activates Idle mode operation. The IDL bit is cleared automatically by hardware when
waking up from idle
8.2
Power-down Mode
Setting the Power-down (PD) bit in PCON enters Power-down mode. Power-down mode stops
the oscillator, disables the BOD and powers down the Flash memory in order to minimize power
consumption. Only the power-on circuitry will continue to draw power during Power-down. During Power-down, the power supply voltage may be reduced to the RAM keep-alive voltage. The
RAM contents will be retained, but the SFR contents are not guaranteed once VDD has been
reduced. Power-down may be exited by external reset, power-on reset, or certain enabled
interrupts.
8.2.1
Interrupt Recovery from Power-down
Three external interrupt sources may be configured to terminate Power-down mode: external
interrupts INT0 (P3.2) and INT1 (P3.3); and the general-purpose interrupts (GPI). To wake up by
external interrupt INT0 or INT1, that interrupt must be enabled by setting EX0 or EX1 in IE and
must be configured for level-sensitive operation by clearing IT0 or IT1. Any General-purpose
interrupt on Port 1 (GPI7-0) can also wake up the device. The GPI pin must be enabled in GPIEN
and configured for level-sensitive detection, and EGP in IE2 must be set in order to terminate
Power-down.
When terminating Power-down by an interrupt, two different wake-up modes are available.
When PWDEX in PCON is zero, the wake-up period is internally timed as shown in Figure 8-1.
At the falling edge on the interrupt pin, Power-down is exited, the oscillator is restarted, and an
internal timer begins counting. The internal clock will not be allowed to propagate to the CPU
until after the timer has timed out. After the time-out period the interrupt service routine will
begin. The time-out period is controlled by the Start-up Timer Fuses (see Table 7-1 on page 34).
The interrupt pin need not remain low for the entire time-out period.
36
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Figure 8-1.
Interrupt Recovery from Power-down (PWDEX = 0)
PWD
XTAL1
tSUT
INT1
Internal
Clock
When PWDEX = “1”, the wake-up period is controlled externally by the interrupt. Again, at the
falling edge on the interrupt pin, power-down is exited and the oscillator is restarted. However,
the internal clock will not propagate until the rising edge of the interrupt pin as shown in Figure 82. The interrupt pin should be held low long enough for the selected clock source to stabilize.
After the rising edge on the pin the interrupt service routine will be executed.
Figure 8-2.
Interrupt Recovery from Power-down (PWDEX = 1)
PWD
XTAL1
INT1
Internal
Clock
8.2.2
8.3
Reset Recovery from Power-down
The wake-up from Power-down through an external reset is similar to the interrupt with
PWDEX = “0”. At the falling edge of RST, Power-down is exited, the oscillator is restarted, and
an internal timer begins counting as shown in Figure 8-3. The internal clock will not be allowed to
propagate to the CPU until after the timer has timed out. The time-out period is controlled by the
Start-up Timer Fuses. (See Table 7-1 on page 34). If RST returns high before the time-out, a two
clock cycle internal reset is generated when the internal clock restarts. Otherwise, the device will
remain in reset until RST is brought high.
Reducing Power Consumption
Several possibilities need consideration when trying to reduce the power consumption in an
AT89LP-based system. Generally, Idle or Power-down mode should be used as much as possible. All unneeded functions should be disabled. In particular, the following modules may need
special consideration when trying to achieve the lowest possible power consumption.
8.3.1
Brown-out Detector
If the Brown-out Detector is not needed by the application, this module should be turned off. If
the Brown-out Detector is enabled by the BOD Enable Fuse, it will be enabled in all modes
except Power-down. See Section 25.7 “User Configuration Fuses” on page 164.
37
3706A–MICRO–9/09
Figure 8-3.
Reset Recovery from Power-down
PWD
XTAL1
tSUT
RST
Internal
Clock
Internal
Reset
8.3.2
Analog Comparators
The comparators will operate during Idle mode if enabled. To save power, the comparators
should be disabled before entering Idle mode if possible. When the comparators are turned off
and on again, some settling time is required for the analog circuits to stabilize. If the comparators
are enabled, they will consume the least power when using an external reference, RFA1-0 = 00B
and RFB1-0 = 00B.
8.3.3
Analog-to-Digital Converter
The DADC will operate during Idle mode if enabled. To save power, the DADC should be disabled before entering Idle mode if possible. When the DADC is turned off and on again, some
settling time is required for the analog circuits to stabilize. If the DADC is enabled, it will consume the least power when configured to use the system clock instead of the internal RC
oscillator (unless the IRC is the system clock source) and when the internal reference is disabled
(IREF = 0). The DADC must always be disabled before entering power-down.
9. Interrupts
The AT89LP6440 provides 12 interrupt sources: two external interrupts, three timer interrupts, a
serial port interrupt, an analog comparator interrupt, a general-purpose interrupt, a compare/capture interrupt, a two-wire interrupt, an ADC interrupt and an SPI interrupt. These
interrupts and the system reset each have a separate program vector at the start of the program
memory space. Each interrupt source can be individually enabled or disabled by setting or clearing a bit in the interrupt enable registers IE and IE2. The IE register also contains a global
disable bit, EA, which disables all interrupts.
Each interrupt source can be individually programmed to one of four priority levels by setting or
clearing bits in the interrupt priority registers IP, IPH, IP2 and IP2H. IP and IP2 hold the low order
priority bits and IPH and IP2H hold the high priority bits for each interrupt. An interrupt service
routine in progress can be interrupted by a higher priority interrupt, but not by another interrupt of
the same or lower priority. The highest priority interrupt cannot be interrupted by any other interrupt source. If two requests of different priority levels are pending at the end of an instruction, the
request of higher priority level is serviced. If requests of the same priority level are pending at
the end of an instruction, an internal polling sequence determines which request is serviced. The
polling sequence is based on the vector address; an interrupt with a lower vector address has
higher priority than an interrupt with a higher vector address. Note that the polling sequence is
only used to resolve pending requests of the same priority level.
38
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
The IPxD bits located at the seventh bit of IP, IPH, IP2 and IP2H can be used to disable all interrupts of a given priority level, allowing software implementations of more complex interrupt
priority handling schemes such as level-based round-robin scheduling.
The External Interrupts INT0 and INT1 can each be either level-activated or edge-activated,
depending on bits IT0 and IT1 in Register TCON. The flags that actually generate these interrupts are the IE0 and IE1 bits in TCON. When the service routine is vectored to, hardware clears
the flag that generated an external interrupt only if the interrupt was edge-activated. If the interrupt was level activated, then the external requesting source (rather than the on-chip hardware)
controls the request flag.
The Timer 0 and Timer 1 Interrupts are generated by TF0 and TF1, which are set by a rollover in
their respective Timer/Counter registers (except for Timer 0 in Mode 3). When a timer interrupt is
generated, the on-chip hardware clears the flag that generated it when the service routine is
vectored to. The Timer 2 Interrupt is generated by a logic OR of bits TF2 and EXF2 in register
T2CON. Neither of these flags is cleared by hardware when the CPU vectors to the service routine. The service routine normally must determine whether TF2 or EXF2 generated the interrupt
and that bit must be cleared by software.
The Serial Port Interrupt is generated by the logic OR of RI and TI in SCON. Neither of these
flags is cleared by hardware when the CPU vectors to the service routine. The service routine
normally must determine whether RI or TI generated the interrupt and that bit must be cleared by
software.
The Serial Peripheral Interface Interrupt is generated by the logic OR of SPIF, MODF and TXE
in SPSR. None of these flags is cleared by hardware when the CPU vectors to the service routine. The service routine normally must determine which bit generated the interrupt and that bit
must be cleared by software.
A logic OR of all eight flags in the GPIF register causes the General-purpose Interrupt. None of
these flags is cleared by hardware when the service routine is vectored to. The service routine
must determine which bit generated the interrupt and that bit must be cleared in software. If the
interrupt was level activated, then the external requesting source must de-assert the interrupt
before the flag may be cleared by software.
The CFA and CFB bits in ACSRA and ACSRB respectively generate the Comparator Interrupt.
The service routine must normally determine whether CFA or CFB generated the interrupt, and
the bit must be cleared by software. The DAC/ADC Conversion Interrupt is generated by ADIF in
DADC. On-chip hardware clears the ADIF flag when vectoring to the service routine.
A logic OR of the four least significant bits in the T2CCF register causes the Compare/Capture
Array Interrupt. None of these flags is cleared by hardware when the service routine is vectored
to. The service routine must determine which bit generated the interrupt and that bit must be
cleared in software.
The Two-Wire Interface Interrupt is generated by TWIF in TWCR. The flag is not cleared by
hardware when the CPU vectors to the service routine. The service routine normally must determine the status in TWSR and respond accordingly before the bit is cleared by software.
All of the bits that generate interrupts can be set or cleared by software, with the same result as
though they had been set or cleared by hardware. That is, interrupts can be generated and
pending interrupts can be canceled in software.
39
3706A–MICRO–9/09
Table 9-1.
9.1
Interrupt Vector Addresses
Interrupt
Source
Vector Address
System Reset
RST or POR or BOD
0000H
External Interrupt 0
IE0
0003H
Timer 0 Overflow
TF0
000BH
External Interrupt 1
IE1
0013H
Timer 1 Overflow
TF1
001BH
Serial Port Interrupt
RI or TI
0023H
Timer 2 Interrupt
TF2 or EXF2
002BH
Analog Comparator Interrupt
CFA or CFB
0033H
General-purpose Interrupt
GPIF7-0
003BH
Compare/Capture Array Interrupt
T2CCF3-0
0043H
Serial Peripheral Interface Interrupt
SPIF or MODF or TXE
004BH
ADC Interrupt
ADIF
0053H
Two-Wire Interface Interrupt
TWIF
005BH
Interrupt Response Time
The interrupt flags may be set by their hardware in any clock cycle. The interrupt controller polls
the flags in the last clock cycle of the instruction in progress. If one of the flags was set in the
preceding cycle, the polling cycle will find it and the interrupt system will generate an LCALL to
the appropriate service routine as the next instruction, provided that the interrupt is not blocked
by any of the following conditions: an interrupt of equal or higher priority level is already in progress; the instruction in progress is RETI or any write to the IE, IP, IPH, IE2, IP2 or IP2H registers;
the CPU is currently forced into idle by an IAP or FDATA write. Each of these conditions will
block the generation of the LCALL to the interrupt service routine. The second condition ensures
that if the instruction in progress is RETI or any access to IE, IP, IPH, IE2, IP2 or IP2H, then at
least one more instruction will be executed before any interrupt is vectored to. The polling cycle
is repeated at the last cycle of each instruction, and the values polled are the values that were
present at the previous clock cycle. If an active interrupt flag is not being serviced because of
one of the above conditions and is no longer active when the blocking condition is removed, the
denied interrupt will not be serviced. In other words, the fact that the interrupt flag was once
active but not serviced is not remembered. Every polling cycle is new.
If a request is active and conditions are met for it to be acknowledged, a hardware subroutine
call to the requested service routine will be the next instruction executed. The call itself takes
four cycles. Thus, a minimum of five complete clock cycles elapsed between activation of an
interrupt request and the beginning of execution of the first instruction of the service routine. A
longer response time results if the request is blocked by one of the previously listed conditions. If
an interrupt of equal or higher priority level is already in progress, the additional wait time
depends on the nature of the other interrupt's service routine. If the instruction in progress is not
in its final clock cycle, the additional wait time cannot be more than 8 cycles, since the longest
instruction is 9 cycles long. If the instruction in progress is RETI with XSTK, the additional wait
time cannot be more than 14 cycles (a maximum of 5 more cycles to complete the instruction in
progress, plus a maximum of 9 cycles to complete the next instruction). Thus, in a single-inter-
40
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
rupt system, the response time is always more than 5 clock cycles and less than 21 clock cycles.
See Figure 9-1 and Figure 9-2.
Figure 9-1.
Minimum Interrupt Response Time
Clock Cycles
1
5
INT0
IE0
Instruction
Figure 9-2.
Ack.
Cur. Instr.
LCALL
1st ISR Instr.
Maximum Interrupt Response Time
Clock Cycles
1
6
15
21
INT0
Ack.
IE0
Instruction
Table 9-2.
RETI
MAC AB
LCALL
1st ISR Instr.
IE – Interrupt Enable Register
IE = A8H
Reset Value = 0000 0000B
Bit Addressable
Bit
EA
EC
ET2
ES
ET1
EX1
ET0
EX0
7
6
5
4
3
2
1
0
Symbol
Function
EA
Global enable/disable. All interrupts are disabled when EA = 0. When EA = 1, each interrupt source is enabled/disabled
by setting /clearing its own enable bit.
EC
Comparator Interrupt Enable
ET2
Timer 2 Interrupt Enable
ES
Serial Port Interrupt Enable
ET1
Timer 1 Interrupt Enable
EX1
External Interrupt 1 Enable
ET0
Timer 0 Interrupt Enable
EX0
External Interrupt 0 Enable
.
41
3706A–MICRO–9/09
Table 9-3.
IE2 – Interrupt Enable 2 Register
IE = B4H
Reset Value = xxxx x000B
Not Bit Addressable
Bit
–
–
–
ETWI
EADC
ESPI
ECC
EGP
7
6
5
4
3
2
1
0
Symbol
Function
ETWI
Two-Wire Interface Interrupt Enable
EADC
ADC Interrupt Enable
ESPI
Serial Peripheral Interface Interrupt Enable
ECC
Compare/Capture Array Interrupt Enable
EGP
General-purpose Interrupt Enable
Table 9-4.
IP – Interrupt Priority Register
IP = B8H
Reset Value = 0000 0000B
Bit Addressable
Bit
IP0D
PC
PT2
PS
PT1
PX1
PT0
PX0
7
6
5
4
3
2
1
0
Symbol
Function
IP0D
Interrupt Priority 0 Disable. Set IP0D to 1 to disable all interrupts with priority level zero. Clear to 0 to enable all interrupts
with priority level zero when EA = 1.
PC
Comparator Interrupt Priority Low
PT2
Timer 2 Interrupt Priority Low
PS
Serial Port Interrupt Priority Low
PT1
Timer 1 Interrupt Priority Low
PX1
External Interrupt 1 Priority Low
PT0
Timer 0 Interrupt Priority Low
PX0
External Interrupt 0 Priority Low
Table 9-5.
IP2 – Interrupt Priority 2 Register
IP = B5H
Reset Value = 0xxx x000B
No Bit Addressable
Bit
IP2D
–
–
PTWI
PADC
PSP
PCC
PGP
7
6
5
4
3
2
1
0
Symbol
Function
IP2D
Interrupt Priority 2 Disable. Set IP2D to 1 to disable all interrupts with priority level two. Clear to 0 to enable all interrupts
with priority level two when EA = 1.
PTWI
Two-wire Interface Interrupt Priority Low
PADC
ADC Interrupt Priority Low
42
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Symbol
Function
PSP
Serial Peripheral Interface Interrupt Priority Low
PCC
Compare/Capture Array Interrupt Priority Low
PGP
General-purpose Interrupt 0 Priority Low
Table 9-6.
IPH – Interrupt Priority High Register
IPH = B7H
Reset Value = 0000 0000B
Not Bit Addressable
Bit
IP1D
PCH
PT2H
PSH
PT1H
PX1H
PT0H
PX0H
7
6
5
4
3
2
1
0
Symbol
Function
IP1D
Interrupt Priority 1 Disable. Set IP1D to 1 to disable all interrupts with priority level one. Clear to 0 to enable all interrupts
with priority level one when EA = 1.
PCH
Comparator Interrupt Priority High
PT2H
Timer 2 Interrupt Priority High
PSH
Serial Port Interrupt Priority High
PT1H
Timer 1 Interrupt Priority High
PX1H
External Interrupt 1 Priority High
PT0H
Timer 0 Interrupt Priority High
PX0H
External Interrupt 0 Priority High
Table 9-7.
IP2H – Interrupt Priority 2 High Register
IPH = B6H
Reset Value = 0xxx x000B
Not Bit Addressable
Bit
IP3D
–
–
PTWH
PADH
PSPH
PCCH
PGPH
7
6
5
4
3
2
1
0
Symbol
Function
IP3D
Interrupt Priority 3 Disable. Set IP3D to 1 to disable all interrupts with priority level three. Clear to 0 to enable all
interrupts with priority level three when EA = 1.
PTWH
Two-Wire Interface Interrupt Priority High
PADH
ADC Interrupt Priority High
PSPH
Serial Peripheral Interface Interrupt Priority High
PCCH
Compare/Capture Array Interrupt Priority High
PGPH
General-purpose Interrupt 0 Priority High
43
3706A–MICRO–9/09
10. I/O Ports
The AT89LP6440 can be configured for between 35 and 38 I/O pins. The exact number of I/O
pins available depends on the clock and reset options as shown in Table 10-1.
Table 10-1.
I/O Pin Configurations
Clock Source
Reset Option
Number of I/O Pins
External Crystal or
Resonator
External RST Pin
35
No external reset
36
External RST Pin
36
No external reset
37
External RST Pin
37
No external reset
38
External Clock
Internal RC Oscillator
10.1
Port Configuration
All port pins on the AT89LP6440 may be configured to one of four modes: quasi-bidirectional
(standard 8051 port outputs), push-pull output, open-drain output, or input-only. Port modes may
be assigned in software on a pin-by-pin basis as shown in Table 10-2 using the registers listed in
Table 10-3. The Tristate-Port User Fuse determines the default state of the port pins. When the
fuse is enabled, all port pins default to input-only mode after reset. When the fuse is disabled, all
port pins, with the exception of the analog inputs, P0.7-0, P2.4, P2.5, P2.6 and P2.7, default to
quasi-bidirectional mode after reset and are weakly pulled high. The analog input pins always
reset to input-only (tristate) mode. Each port pin also has a Schmitt-triggered input for improved
input noise rejection. During Power-down all the Schmitt-triggered inputs are disabled with the
exception of P3.2 (INT0), P3.3 (INT1), P4.2 (RST), P4.0 (XTAL1) and P4.1 (XTAL2) which may
be used to wake up the device. Therefore, P3.2, P3.3, P4.2, P4.0 and P4.1 should not be left
floating during Power-down. In addition any pin of Port 1 configured as a General-Purpose interrupt input will also remain active during Power-down to wake-up the device. These interrupt pins
should either be disabled before entering Power-down or they should not be left floating.
.
Table 10-2.
PxM0.y
PxM1.y
0
0
Quasi-bidirectional
0
1
Push-pull Output
1
0
Input Only (High Impedance)
1
1
Open-Drain Output
.
Table 10-3.
44
Configuration Modes for Port x, Bit y
Port Mode
Port Configuration Registers
Port
Port Data
Port Configuration
0
P0 (80H)
P0M0 (BAH), P0M1 (BBH)
1
P1 (90H)
P1M0 (C2H), P1M1 (C3H)
2
P2 (A0H)
P2M0 (C4H), P2M1 (C5H)
3
P3 (B0H)
P3M0 (C6H), P3M1 (C7H)
4
P4 (C0H)
P4M0 (BEH), P4M1 (BFH)
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
10.1.1
Quasi-bidirectional Output
Port pins in quasi-bidirectional output mode function similar to standard 8051 port pins. A Quasibidirectional port can be used both as an input and output without the need to reconfigure the
port. This is possible because when the port outputs a logic high, it is weakly driven, allowing an
external device to pull the pin low. When the pin is driven low, it is driven strongly and able to
sink a large current. There are three pull-up transistors in the quasi-bidirectional output that
serve different purposes.
One of these pull-ups, called the “very weak” pull-up, is turned on whenever the port latch for the
pin contains a logic “1”. This very weak pull-up sources a very small current that will pull the pin
high if it is left floating.
A second pull-up, called the “weak” pull-up, is turned on when the port latch for the pin contains
a logic “1” and the pin itself is also at a logic “1” level. This pull-up provides the primary source
current for a quasi-bidirectional pin that is outputting a “1”. If this pin is pulled low by an external
device, this weak pull-up turns off, and only the very weak pull-up remains on. In order to pull the
pin low under these conditions, the external device has to sink enough current to overpower the
weak pull-up and pull the port pin below its input threshold voltage.
The third pull-up is referred to as the “strong” pull-up. This pull-up is used to speed up low-tohigh transitions on a quasi-bidirectional port pin when the port latch changes from a logic “0” to a
logic “1”. When this occurs, the strong pull-up turns on for two CPU clocks quickly pulling the
port pin high. The quasi-bidirectional port configuration is shown in Figure 10-1.
Figure 10-1. Quasi-bidirectional Output
1 Clock Delay
(D Flip-Flop)
VCC
VCC
VCC
Strong
Very
Weak
Weak
Port
Pin
From Port
Register
Input
Data
PWD
10.1.2
Input-only Mode
The input only port configuration is shown in Figure 10-2. The output drivers are tristated. The
input includes a Schmitt-triggered input for improved input noise rejection. The input circuitry of
P3.2, P3.3, P4.2, P4.0 and P4.1 is not disabled during Power-down (see Figure 10-3) and therefore these pins should not be left floating during Power-down when configured in this mode.
Figure 10-2. Input Only
Input
Data
Port
Pin
PWD
45
3706A–MICRO–9/09
Figure 10-3. Input Circuit for P3.2, P3.3, P4.0, P4.1 and P4.2
Input
Data
10.1.3
Port
Pin
Open-drain Output
The open-drain output configuration turns off all pull-ups and only drives the pull-down transistor
of the port pin when the port latch contains a logic “0”. To be used as a logic output, a port configured in this manner must have an external pull-up, typically a resistor tied to VDD. The pulldown for this mode is the same as for the quasi-bidirectional mode. The open-drain port configuration is shown in Figure 10-4. The input circuitry of P3.2, P3.3, P4.0, P4.1 and P4.2 is not
disabled during Power-down (see Figure 10-3) and therefore these pins should not be left floating during Power-down when configured in this mode.
Figure 10-4. Open-Drain Output
Port
Pin
From Port
Register
Input
Data
PWD
10.1.4
Push-pull Output
The push-pull output configuration has the same pull-down structure as both the open-drain and
the quasi-bidirectional output modes, but provides a continuous strong pull-up when the port
latch contains a logic “1”. The push-pull mode may be used when more source current is needed
from a port output. The push-pull port configuration is shown in Figure 10-5.
Figure 10-5. Push-pull Output
VCC
Port
Pin
From Port
Register
Input
Data
PWD
46
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
10.2
Port Analog Functions
The AT89LP6440 incorporates two analog comparators and an 8-channel analog-to-digital converter. In order to give the best analog performance and minimize power consumption, pins that
are being used for analog functions must have both their digital outputs and digital inputs disabled. Digital outputs are disabled by putting the port pins into the input-only mode as described
in “Port Configuration” on page 44. The analog input pins will always default to input-only mode
after reset regardless of the state of the Tristate-Port Fuse.
Digital inputs on P2.4, P2.5, P2.6 and P2.7 are disabled whenever an analog comparator is
enabled by setting the CENA or CENB bits in ACSRA and ACSRB and that pin is configured for
input-only mode. To use an analog input pin as a high-impedance digital input while a comparator is enabled, that pin should be configured in open-drain mode and the corresponding port
register bit should be set to 1.
Digital inputs on Port 0 are disabled for each pin configured for input-only mode whenever the
ADC is enabled by setting the ADCE bit and clearing the DAC bit in DADC. To use any Port 0
input pin as a high-impedance digital input while the ADC is enabled, that pin should be configured in open-drain mode and the corresponding port register bit should be set to 1. When DAC
mode is enabled, P2.2 and P2.3 are forced to input-only mode.
10.3
Port Read-Modify-Write
A read from a port will read either the state of the pins or the state of the port register depending
on which instruction is used. Simple read instructions will always access the port pins directly.
Read-modify-write instructions, which read a value, possibly modify it, and then write it back, will
always access the port register. This includes bit write instructions such as CLR or SETB as they
actually read the entire port, modify a single bit, then write the data back to the entire port. See
Table 10-4 for a complete list of Read-Modify-Write instruction which may access the ports.
Table 10-4.
Port Read-Modify-Write Instructions
Mnemonic
Instruction
Example
ANL
Logical AND
ANL P1, A
ORL
Logical OR
ORL P1, A
XRL
Logical EX-OR
XRL P1, A
JBC
Jump if bit set and clear bit
JBC P3.0, LABEL
CPL
Complement bit
CPL P3.1
INC
Increment
INC P1
DEC
Decrement
DEC P3
DJNZ
Decrement and jump if not zero
DJNZ P3, LABEL
MOV PX.Y, C
Move carry to bit Y of Port X
MOV P1.0, C
CLR PX.Y
Clear bit Y of Port X
CLR P1.1
SETB PX.Y
Set bit Y of Port X
SETB P3.2
47
3706A–MICRO–9/09
10.4
Port Alternate Functions
Most general-purpose digital I/O pins of the AT89LP6440 share functionality with the various
I/Os needed for the peripheral units. Table 10-6 lists the alternate functions of the port pins.
Alternate functions are connected to the pins in a logic AND fashion. In order to enable the
alternate function on a port pin, that pin must have a “1” in its corresponding port register bit,
otherwise the input/output will always be “0”. However, alternate functions may be temporarily
forced to “0” by clearing the associated port bit, provided that the pin is not in input-only mode.
Furthermore, each pin must be configured for the correct input/output mode as required by its
peripheral before it may be used as such. Table 10-5 shows how to configure a generic pin for
use with an alternate function.
Table 10-5.
Alternate Function Configurations for Pin y of Port x
PxM0.y
PxM1.y
Px.y
0
0
1
bidirectional (internal pull-up)
0
1
1
output
1
0
X
input
1
1
1
bidirectional (external pull-up)
Table 10-6.
Port Pin Alternate Functions
Configuration Bits
48
I/O Mode
Port Pin
PxM0.y
PxM1.y
P0.0
P0M0.0
P0M1.0
P0.1
P0M0.1
P0M1.1
P0.2
P0M0.2
P0M1.2
P0.3
P0M0.3
P0M1.3
P0.4
P0M0.4
P0M1.4
P0.5
P0M0.5
P0M1.5
P0.6
P0M0.6
P0M1.6
P0.7
P0M0.7
P0M1.7
P1.0
P1M0.0
P1M1.0
P1.1
P1M0.1
P1M1.1
Alternate
Function
AD0
ADC0
AD1
ADC1
AD2
ADC2
AD3
ADC3
AD4
ADC4
AD5
ADC5
AD6
ADC6
AD7
ADC7
Notes
Automatic configuration
input-only
Automatic configuration
input-only
Automatic configuration
input-only
Automatic configuration
input-only
Automatic configuration
input-only
Automatic configuration
input-only
Automatic configuration
input-only
Automatic configuration
input-only
T2
GPI0
T2EX
GPI1
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Table 10-6.
Port Pin Alternate Functions
Configuration Bits
Alternate
Function
Port Pin
PxM0.y
PxM1.y
P1.2
P1M0.2
P1M1.2
P1.3
P1M0.3
P1M1.3
P1.4
P1M0.4
P1M1.4
P1.5
P1M0.5
P1M1.5
P1.6
P1M0.6
P1M1.6
P1.7
P1M0.7
P1M1.7
P2.0
P2M0.0
P2M1.0
CCA
P2.1
P2M0.1
P2M1.1
CCB
P2.2
P2M0.2
P2M1.2
P2.3
P2M0.3
P2M1.3
P2.4
P2M0.4
P2.5
SDA
Notes
open-drain
GPI2
SCL
open-drain
GPI3
SS
GPI4
MOSI
GPI5
MISO
GPI6
SCK
GPI7
CCC
DA+
input-only
CCD
DA-
input-only
P2M1.4
AIN0
input-only
P2M0.5
P2M1.5
AIN1
input-only
P2.6
P2M0.6
P2M1.6
AIN2
input-only
P2.7
P2M0.7
P2M1.7
AIN3
input-only
P3.0
P3M0.0
P3M1.0
RXD
P3.1
P3M0.1
P3M1.1
TXD
P3.2
P3M0.2
P3M1.2
INT0
P3.3
P3M0.3
P3M1.3
INT1
P3.4
P3M0.4
P3M1.4
T0
P3.5
P3M0.5
P3M1.5
T1
P3.6
P3M0.6
P3M1.6
WR
P3.7
P3M0.7
P3M1.7
RD
P4.2
P3M0.5
P3M1.5
RST
RST must be disabled to use P4.2
P4.6
not configurable
CMPA
Pin is tied to comparator output
P4.7
not configurable
CMPB
Pin is tied to comparator output
49
3706A–MICRO–9/09
11. Enhanced Timer 0 and Timer 1 with PWM
The AT89LP6440 has two 16-bit Timer/Counters, Timer 0 and Timer 1, with the following
features:
• Two 16-bit timer/counters with 16-bit reload registers
• Two independent 8-bit precision PWM outputs with 8-bit prescalers
• UART or SPI baud rate generation using Timer 1
• Output pin toggle on timer overflow
• Split timer mode allows for three separate timers (2 8-bit, 1 16-bit)
• Gated modes allow timers to run/halt based on an external input
Timer 0 and Timer 1 have similar modes of operation. As timers, the timer registers increase
every clock cycle by default. Thus, the registers count clock cycles. Since a clock cycle consists
of one oscillator period, the count rate is equal to the oscillator frequency. The timer rate can be
prescaled by a value between 1 and 16 using the Timer Prescaler (see Table 6-2 on page 32).
Both Timers share the same prescaler.
As counters, the timer registers are incremented in response to a 1-to-0 transition at the corresponding input pins, T0 or T1. The external input is sampled every clock cycle. When the
samples show a high in one cycle and a low in the next cycle, the count is incremented. The new
count value appears in the register during the cycle following the one in which the transition was
detected. Since 2 clock cycles are required to recognize a 1-to-0 transition, the maximum count
rate is 1/2 of the oscillator frequency. There are no restrictions on the duty cycle of the input signal, but it should be held for at least one full clock cycle to ensure that a given level is sampled at
least once before it changes.
Furthermore, the Timer or Counter functions for Timer 0 and Timer 1 have four operating modes:
variable width timer, 16-bit auto-reload timer, 8-bit auto-reload timer, and split timer. The control
bits C/T in the Special Function Register TMOD select the Timer or Counter function. The bit
pairs (M1, M0) in TMOD select the operating modes.
Table 11-1.
50
Timer 0/1 Register Summary
Name
Address
Purpose
Bit-Addressable
TCON
88H
Control
Y
TMOD
89H
Mode
N
TL0
8AH
Timer 0 low-byte
N
TL1
8BH
Timer 1 low-byte
N
TH0
8CH
Timer 0 high-byte
N
TH1
8DH
Timer 1 high-byte
N
TCONB
91H
Mode
N
RL0
92H
Timer 0 reload low-byte
N
RL1
93H
Timer 1 reload low-byte
N
RH0
94H
Timer 0 reload high-byte
N
RH1
95H
Timer 1 reload high-byte
N
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
11.1
Mode 0 – Variable Width Timer/Counter
Both Timers in Mode 0 are 8-bit Counters with a variable prescaler. The prescaler may vary from
1 to 8 bits depending on the PSC bits in TCONB, giving the timer a range of 9 to 16 bits.
By default the timer is configured as a 13-bit timer compatible to Mode 0 in the standard 8051.
Figure 11-1 shows the Mode 0 operation as it applies to Timer 1 in 13-bit mode. As the count
rolls over from all “1”s to all “0”s, it sets the Timer interrupt flag TF1. The counter input is enabled
to the Timer when TR1 = 1 and either GATE1 = 0 or INT1 = 1. Setting GATE1 = 1 allows the
Timer to be controlled by external input INT1, to facilitate pulse width measurements. TR1 is a
control bit in the Special Function Register TCON. GATE1 is in TMOD. The 13-bit register consists of all 8 bits of TH1 and the lower 5 bits of TL1. The upper 3 bits of TL1 are indeterminate
and should be ignored. Setting the run flag (TR1) does not clear the registers.
PSC0 + 1
Mode 0:
Note:
256 × 2
Time-out Period = ------------------------------------------------------- × ( TPS + 1 )
Oscillator Frequency
RH1/RL1 are not required by Timer 1 during Mode 0 and may be used as temporary storage
registers.
Figure 11-1. Timer/Counter 1 Mode 0: Variable Width Counter
OSC
÷TPS
C/T = 0
TL1
(8 Bits)
C/T = 1
T1 Pin
Control
PSC1
TR1
TH1
(8 Bits)
GATE1
TF1
Interrupt
INT1 Pin
Mode 0 operation is the same for Timer 0 as for Timer 1, except that TR0, TF0, GATE0 and
INT0 replace the corresponding Timer 1 signals in Figure 11-1. There are two different C/T bits,
one for Timer 1 (TMOD.6) and one for Timer 0 (TMOD.2).
11.2
Mode 1 – 16-bit Auto-Reload Timer/Counter
In Mode 1 the Timers are configured for 16-bit auto-reload. The Timer register is run with all
16 bits. The 16-bit reload value is stored in the high and low reload registers (RH1/RL1). The
clock is applied to the combined high and low timer registers (TH1/TL1). As clock pulses are
received, the timer counts up: 0000H, 0001H, 0002H, etc. An overflow occurs on the FFFFH-to0000H transition, upon which the timer register is reloaded with the value from RH1/RL1 and the
overflow flag bit in TCON is set. See Figure 11-2. The reload registers default to 0000H, which
gives the full 16-bit timer period compatible with the standard 8051. Mode 1 operation is the
same for Timer/Counter 0.
Mode 1:
( 65536 – {RH0, RL0} )
Time-out Period = --------------------------------------------------------- × ( TPS + 1 )
Oscillator Frequency
51
3706A–MICRO–9/09
Figure 11-2. Timer/Counter 1 Mode 1: 16-bit Auto-Reload
RL1
(8 Bits)
RH1
(8 Bits)
÷TPS
OSC
Reload
C/T = 0
TL1
(8 Bits)
TH1
(8 Bits)
Interrupt
TF1
C/T =1
T1 Pin
Control
TR1
GATE1
INT1 Pin
11.3
Mode 2 – 8-bit Auto-Reload Timer/Counter
Mode 2 configures the Timer register as an 8-bit Counter (TL1) with automatic reload, as shown
in Figure 11-3. Overflow from TL1 not only sets TF1, but also reloads TL1 with the contents of
TH1, which is preset by software. The reload leaves TH1 unchanged. Mode 2 operation is the
same for Timer/Counter 0.
Mode 2:
( 256 – TH0 )
Time-out Period = ------------------------------------------------------- × ( TPS + 1 )
Oscillator Frequency
Figure 11-3. Timer/Counter 1 Mode 2: 8-bit Auto-Reload
OSC
÷TPS
C/T = 0
TL1
(8 Bits)
TF1
Interrupt
C/T = 1
Control
T1 Pin
Reload
TR1
TH1
(8 Bits)
GATE1
INT0 Pin
Note:
11.4
RH1/RL1 are not required by Timer 1 during Mode 2 and may be used as temporary storage
registers.
Mode 3 – 8-bit Split Timer
Timer 1 in Mode 3 simply holds its count. The effect is the same as setting TR1 = 0. Timer 0 in
Mode 3 establishes TL0 and TH0 as two separate counters. The logic for Mode 3 on Timer 0 is
shown in Figure 11-4. TL0 uses the Timer 0 control bits: C/T, GATE0, TR0, INT0, and TF0. TH0
is locked into a timer function (counting clock cycles) and takes over the use of TR1 and TF1
from Timer 1. Thus, TH0 now controls the Timer 1 interrupt. While Timer 0 is in Mode 3, Timer 1
will still obey its settings in TMOD but cannot generate an interrupt.
52
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Mode 3 is for applications requiring an extra 8-bit timer or counter. With Timer 0 in Mode 3, the
AT89LP6440 can appear to have four Timer/Counters. When Timer 0 is in Mode 3, Timer 1 can
be turned on and off by switching it out of and into its own Mode 3. In this case, Timer 1 can still
be used by the serial port as a baud rate generator or in any application not requiring an
interrupt.
Figure 11-4. Timer/Counter 0 Mode 3: Two 8-bit Counters
÷TPS
C/T = 0
C/T =1
T0 Pin
(8 Bits)
Interrupt
(8 Bits)
Interrupt
Control
GATE0
INT0 Pin
÷TPS
Control
Note:
.
Table 11-2.
RH0/RL0 are not required by Timer 0 during Mode 3 and may be used as temporary storage
registers.
TCON – Timer/Counter Control Register
TCON = 88H
Reset Value = 0000 0000B
Bit Addressable
Bit
TF1
TR1
TF0
TR0
IE1
IT1
IE0
IT0
7
6
5
4
3
2
1
0
Symbol
Function
TF1
Timer 1 overflow flag. Set by hardware on Timer/Counter overflow. Cleared by hardware when the processor vectors
to interrupt routine.
TR1
Timer 1 run control bit. Set/cleared by software to turn Timer/Counter on/off.
TF0
Timer 0 overflow flag. Set by hardware on Timer/Counter overflow. Cleared by hardware when the processor vectors
to interrupt routine.
TR0
Timer 0 run control bit. Set/cleared by software to turn Timer/Counter on/off.
IE1
Interrupt 1 edge flag. Set by hardware when external interrupt edge detected. Cleared when interrupt processed.
IT1
Interrupt 1 type control bit. Set/cleared by software to specify falling edge/low level triggered external interrupts.
IE0
Interrupt 0 edge flag. Set by hardware when external interrupt edge detected. Cleared when interrupt processed.
IT0
Interrupt 0 type control bit. Set/cleared by software to specify falling edge/low level triggered external interrupts.
53
3706A–MICRO–9/09
Table 11-3.
TMOD – Timer/Counter Mode Control Register
TMOD Address = 089H
Reset Value = 0000 0000B
Not Bit Addressable
Bit
GATE1
C/T1
T1M1
T1M0
GATE0
C/T0
T0M0
T0M1
7
6
5
4
3
2
1
0
Symbol
Function
GATE1
Timer 1 Gating Control. When set, Timer/Counter 1 is enabled only while INT1 pin is high and TR1 control pin is set.
When cleared, Timer 1 is enabled whenever TR1 control bit is set.
C/T1
Timer or Counter Selector 1. Cleared for Timer operation (input from internal system clock). Set for Counter operation
(input from T1 input pin). C/T1 must be zero when using Timer 1 in PWM mode.
T1M1
T1M0
Timer 1 Operating Mode
Mode
T1M1
T1M0
Operation
0
0
0
Variable 9–16-bit Timer Mode. 8-bit Timer/Counter TH1 with TL1 as 1–8-bit prescaler.
1
0
1
16-bit Auto-Reload Mode. TH1 and TL1 are cascaded to form a 16-bit Timer/Counter
that is reloaded with RH1 and RL1 each time it overflows.
2
1
0
8-bit Auto Reload Mode. TH1 holds a value which is reloaded into 8-bit Timer/Counter
TL1 each time it overflows.
3
1
1
Timer/Counter 1 is stopped
GATE0
Timer 0 Gating Control. When set, Timer/Counter 0 is enabled only while INT0 pin is high and TR0 control pin is set.
When cleared, Timer 0 is enabled whenever TR0 control bit is set.
C/T0
Timer or Counter Selector 0. Cleared for Timer operation (input from internal system clock). Set for Counter operation
(input from T0 input pin). C/T0 must be zero when using Timer 0 in PWM mode.
T0M1
T0M0
54
Timer 0 Operating Mode
Mode
T0M1
T0M0
Operation
0
0
0
Variable 9–16-bit Timer Mode. 8-bit Timer/Counter TH0 with TL0 as 1–8-bit prescaler.
1
0
1
16-bit Auto-Reload Mode. TH0 and TL0 are cascaded to form a 16-bit Timer/Counter
that is reloaded with RH0 and RL0 each time it overflows.
2
1
0
8-bit Auto Reload Mode. TH0 holds a value which is reloaded into 8-bit Timer/Counter
TL0 each time it overflows.
3
1
1
Split Timer Mode. TL0 is an 8-bit Timer/Counter controlled by the standard Timer 0
control bits. TH0 is an 8-bit timer only controlled by Timer 1 control bits.
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Table 11-4.
TCONB – Timer/Counter Control Register B
TCONB = 91H
Reset Value = 0010 0100B
Not Bit Addressable
PWM1EN
PWM0EN
PSC12
PSC11
PSC10
PSC02
PSC01
PSC00
7
6
5
4
3
2
1
0
Bit
Symbol
Function
PWM1EN
Configures Timer 1 for Pulse Width Modulation output on T1 (P3.5).
PWM0EN
Configures Timer 0 for Pulse Width Modulation output on T0 (P3.4).
PSC12
PSC11
PSC10
Prescaler for Timer 1 Mode 0. The number of active bits in TL1 equals PSC1 + 1. After reset PSC1 = 100B which
enables 5 bits of TL1 for compatibility with the 13-bit Mode 0 in AT89S2051.
PSC02
PSC01
PSC00
Prescaler for Timer 0 Mode 0. The number of active bits in TL0 equals PSC0 + 1. After reset PSC0 = 100B which
enables 5 bits of TL0 for compatibility with the 13-bit Mode 0 in AT89C52.
11.5
Pulse Width Modulation
On the AT89LP6440, Timer 0 and Timer 1 may be independently configured as 8-bit asymmetrical (edge-aligned) pulse width modulators (PWM) by setting the PWM0EN or PWM1EN bits in
TCONB, respectively. In PWM Mode the generated waveform is output on the timer's input pin,
T0 or T1. Therefore, C/Tx must be set to “0” when in PWM mode and the T0 (P3.4) and T1
(P3.5) must be configured in an output mode. The Timer Overflow Flags and Interrupts will continue to function while in PWM Mode and Timer 1 may still generate the baud rate for the UART.
The timer GATE function also works in PWM mode, allowing the output to be halted by an external input. Each PWM channel has four modes selected by the mode bits in TMOD.
An example waveform for Timer 0 in PWM Mode 0 is shown in Figure 11-5. TH0 acts as an 8-bit
counter while RH0 stores the 8-bit compare value. When TH0 is 00H the PWM output is
set high. When the TH0 count reaches the value stored in RH0 the PWM output is set low.
Therefore, the pulse width is proportional to the value in RH0. To prevent glitches, writes to
RH0 only take effect on the FFH to 00H overflow of TH0. Setting RH0 to 00H will keep the PWM
output low.
Figure 11-5. 8-bit Asymmetrical Pulse Width Modulation
TH0
TF0 Set
FFH
RH0
00H
time
(P3.4)T0
55
3706A–MICRO–9/09
11.5.1
Mode 0 – 8-bit PWM with 8-bit Logarithmic Prescaler
In Mode 0, TLx acts as a logarithmic prescaler driving 8-bit counter THx (see Figure 11-6). The
PSCx bits in TCONB control the prescaler value. On THx overflow, the duty cycle value in RHx
is transferred to OCRx and the output pin is set high. When the count in THx matches OCRx, the
output pin is cleared low. The following formulas give the output frequency and duty cycle for
Timer 0 in PWM Mode 0. Timer 1 in PWM Mode 0 is identical to Timer 0.
Oscillator Frequency
1
- × --------------------f out = -----------------------------------------------------PSC0 + 1
TPS
+1
256 × 2
Mode 0:
RH0
Duty Cycle % = 100 × -----------256
Figure 11-6. Timer/Counter 1 PWM Mode 0
RH1
(8 Bits)
TL1
(8 Bits)
÷TPS
OSC
OCR1
Control
=
TR1
T1
PSC1
GATE1
TH1
(8 Bits)
INT1 Pin
11.5.2
Mode 1 – 8-bit PWM with 8-bit Linear Prescaler
In Mode 1, TLx provides linear prescaling with an 8-bit auto-reload from RLx (see Figure 11-7 on
page 57). On TLx overflow, TLx is loaded with the value of RLx. THx acts as an 8-bit counter. On
THx overflow, the duty cycle value in RHx is transferred to OCRx and the output pin is set high.
When the count in THx matches OCRx, the output pin is cleared low. The following formulas
give the output frequency and duty cycle for Timer 0 in PWM Mode 1. Timer 1 in PWM Mode 1 is
identical to Timer 0.
Oscillator Frequency
1
f out = ------------------------------------------------------- × --------------------256 × ( 256 – RL0 )
TPS + 1
Mode 1:
11.5.3
Mode 2 – 8-bit Frequency Generator
Timer 0 in PWM Mode 2 functions as an 8-bit Auto-Reload timer, the same as normal Mode 2,
with the exception that the output pin T0 is toggled at every TL0 overflow (see Figure 11-8 and
Figure 11-9 on page 57). Timer 1 in PWM Mode 2 is identical to Timer 0. PWM Mode 2 can be
used to output a square wave of varying frequency. THx acts as an 8-bit counter. The following
formula gives the output frequency for Timer 0 in PWM Mode 2.
Mode 2:
56
RH0
Duty Cycle % = 100 × -----------256
Oscillator Frequency
1
f out = ------------------------------------------------------- × --------------------2 × ( 256 – TH0 )
TPS + 1
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Figure 11-7. Timer/Counter 1 PWM Mode 1
RH1
(8 Bits)
RL1
(8 Bits)
OCR1
=
T1
OSC
TH1
(8 Bits)
TL1
(8 Bits)
÷TPS
Control
TR1
GATE1
INT1 Pin
Figure 11-8. Timer/Counter 1 PWM Mode 2
TH1
(8 Bits)
OSC
TL1
(8 Bits)
÷TPS
T1
Control
TR1
GATE1
INT1 Pin
Note:
{RH0 & RL0}/{RH1 & RL1} are not required by Timer 0/Timer 1 during PWM Mode 2 and may be
used as temporary storage registers.
Figure 11-9. PWM Mode 2 Waveform
FFh
THx
Tx
57
3706A–MICRO–9/09
11.5.4
Mode 3 – Split 8-bit PWM
Timer 1 in PWM Mode 3 simply holds its count. The effect is the same as setting TR1 = 0.
Timer 0 in PWM Mode 3 establishes TL0 and TH0 as two separate PWM counters in a manner
similar to normal Mode 3. PWM Mode 3 on Timer 0 is shown in Figure 11-10. Only the Timer
Prescaler is available to change the output frequency during PWM Mode 3. TL0 can use the
Timer 0 control bits: GATE, TR0, INT0, PWM0EN and TF0. TH0 is locked into a timer function
and uses TR1, PWM1EN and TF1. RL0 provides the duty cycle for TL0 and RH0 provides the
duty cycle for TH0.
PWM Mode 3 is for applications requiring a single PWM channel and two timers, or two PWM
channels and an extra timer or counter. With Timer 0 in PWM Mode 3, the AT89LP6440 can
appear to have four Timer/Counters. When Timer 0 is in PWM Mode 3, Timer 1 can be turned on
and off by switching it out of and into its own Mode 3. In this case, Timer 1 can still be used by
the serial port as a baud rate generator or in any application not requiring an interrupt. The following formulas give the output frequency and duty cycle for Timer 0 in PWM Mode 3.
Oscillator Frequency
1
f out = ------------------------------------------------------- × --------------------256
TPS + 1
Mode 3:
Mode 3, T0:
RL0
Duty Cycle % = 100 × ----------256
Mode 3, T1:
RH0
Duty Cycle % = 100 × -----------256
Figure 11-10. Timer/Counter 0 PWM Mode 3
RL0
(8 Bits)
OCR0
=
T0
TL0
(8 Bits)
÷TPS
OSC
Control
RH0
(8 Bits)
TR0
GATE0
OCR1
INT0 Pin
=
T1
÷TPS
OSC
TH0
(8 Bits)
TR1
58
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
12. Enhanced Timer 2
The AT89LP6440 includes a 16-bit Timer/Counter 2 with the following features:
• 16-bit timer/counter with one 16-bit reload/capture register
• One external reload/capture input
• Up/Down counting mode with external direction control
• UART baud rate generation
• Output-pin toggle on timer overflow
• Dual slope symmetric operating modes
Timer 2 is a 16-bit Timer/Counter that can operate as either a timer or an event counter. The
type of operation is selected by bit C/T2 in the SFR T2CON. Timer 2 has three operating modes:
capture, auto-reload (up or down counting), and baud rate generator. The modes are selected
by bits in T2CON and T2MOD, as shown in Table 12-3. Timer 2 also serves as the time base for
the Compare/Capture Array (See Section 13. “Compare/Capture Array” on page 68).
Timer 2 consists of two 8-bit registers, TH2 and TL2. In the Timer function, the register is incremented every clock cycle. Since a clock cycle consists of one oscillator period, the count rate is
equal to the oscillator frequency. The timer rate can be prescaled by a value between 1 and 16
using the Timer Prescaler (see Table 6-2 on page 32).
In the Counter function, the register is incremented in response to a 1-to-0 transition at its corresponding external input pin, T2. In this function, the external input is sampled every clock cycle.
When the samples show a high in one cycle and a low in the next cycle, the count is incremented. The new count value appears in the register during the cycle following the one in which
the transition was detected. Since two clock cycles are required to recognize a 1-to-0 transition,
the maximum count rate is 1/2 of the oscillator frequency. To ensure that a given level is sampled at least once before it changes, the level should be held for at least one full clock cycle.
Table 12-1.
Timer 2 Operating Modes
RCLK + TCLK
CP/RL2
DCEN
T2OE
TR2
MODE
0
0
0
0
1
16-bit Auto-reload
0
0
1
0
1
16-bit Auto-reload Up-Down
0
1
X
0
1
16-bit Capture
1
X
X
X
1
Baud Rate Generator
X
X
X
1
1
Frequency Generator
X
X
X
X
0
(Off)
The following definitions for Timer 2 are used in the subsequent paragraphs:
Table 12-2.
Symbol
Timer 2 Definitions
Definition
MIN
0000H
MAX
FFFFH
BOTTOM
16-bit value of {RCAP2H,RCAP2L} (standard modes)
TOP
16-bit value of {RCAP2H,RCAP2L} (enhanced modes)
59
3706A–MICRO–9/09
12.1
Timer 2 Registers
Control and status bits for Timer 2 are contained in registers T2CON (see Table 12-3) and
T2MOD (see Table 12-4). The register pair {TH2, TL2} at addresses 0CDH and 0CCH are the
16-bit timer register for Timer 2. The register pair {RCAP2H, RCAP2L} at addresses 0CBH and
0CAH are the 16-bit Capture/Reload register for Timer 2 in capture and auto-reload modes.
Table 12-3.
T2CON – Timer/Counter 2 Control Register
T2CON Address = 0C8H
Reset Value = 0000 0000B
Bit Addressable
Bit
TF2
EXF2
RCLK
TCLK
EXEN2
TR2
C/T2
CP/RL2
7
6
5
4
3
2
1
0
Symbol
Function
TF2
Timer 2 overflow flag set by a Timer 2 overflow and must be cleared by software. TF2 will not be set when either
RCLK = 1 or TCLK = 1.
EXF2
Timer 2 external flag set when either a capture or reload is caused by a negative transition on T2EX and EXEN2 = 1.
When Timer 2 interrupt is enabled, EXF2 = 1 will cause the CPU to vector to the Timer 2 interrupt routine. EXF2 must be
cleared by software. EXF2 does not cause an interrupt in up/down counter mode (DCEN = 1) or dual-slope mode.
RCLK
Receive clock enable. When set, causes the serial port to use Timer 2 overflow pulses for its receive clock in serial port
Modes 1 and 3. RCLK = 0 causes Timer 1 overflows to be used for the receive clock.
TCLK
Transmit clock enable. When set, causes the serial port to use Timer 2 overflow pulses for its transmit clock in serial port
Modes 1 and 3. TCLK = 0 causes Timer 1 overflows to be used for the transmit clock.
EXEN2
Timer 2 external enable. When set, allows a capture or reload to occur as a result of a negative transition on T2EX if
Timer 2 is not being used to clock the serial port. EXEN2 = 0 causes Timer 2 to ignore events at T2EX.
TR2
Start/Stop control for Timer 2. TR2 = 1 starts the timer.
C/T2
Timer or counter select for Timer 2. C/T2 = 0 for timer function. C/T2 = 1 for external event counter (falling edge
triggered).
CP/RL2
Capture/Reload select. CP/RL2 = 1 causes captures to occur on negative transitions at T2EX if EXEN2 = 1. CP/RL2 = 0
causes automatic reloads to occur when Timer 2 overflows or negative transitions occur at T2EX when EXEN2 = 1. When
either RCLK or TCLK = 1, this bit is ignored and the timer is forced to auto-reload on Timer 2 overflow.
Table 12-4.
T2MOD – Timer 2 Mode Control Register
T2MOD Address = 0C9H
Reset Value = 0000 0000B
Not Bit Addressable
Bit
PHSD
PHS2
PHS1
PHS0
T2CM1
T2CM0
T2OE
DCEN
7
6
5
4
3
2
1
0
Symbol
Function
PHSD
CCA Phase Direction. For phase modes with 3 or 4 channels, PHSD determines the direction that the channels are
cycled through. PHSD also determines the initial phase relationship for 2 phase modes.
PHSD
60
Direction
0
A →B →A →B
or
A →B →C →A →B →C
or
A →B →C →D →A →B →C →D
1
B →A →B →A
or
C →B →A →C →B →A
or
D →C →B →A →D →C →B →A
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Symbol
Function
PHS [2-0]
CCA Phase Mode. PWM channels may be grouped by 2, 3 or 4 such that only one channel in a group produces a pulse
in any one period. The PHS[2-0] bits may only be written when the timer is not active, i.e. TR2 = 0.
T2CM
[1-0]
PHS2
PHS1
PHS0
Phase Mode
0
0
0
Disabled, all channels active
0
0
1
2-phase output on channels A & B
0
1
0
3-phase output on channels A, B & C
0
1
1
4-phase output on channels A, B, C & D
1
0
0
Dual 2-phase output on channels A & B and C & D
1
0
1
reserved
1
1
0
reserved
1
1
1
reserved
Timer 2 Count Mode.
T2CM1
T2CM0
Count Mode
0
0
Standard Timer 2 (up count: BOTTOM →MAX )
0
1
Clear on RCAP compare (up count: MIN →TOP )
1
0
Dual-slope with single update (up-down count:
1
1
Dual-slope with double update (up-down count:
MIN →TOP →MIN )
MIN →TOP →MIN )
T2OE
Timer 2 Output Enable. When T2OE = 1 and C/T2 = 0, the T2 pin will toggle after every Timer 2 overflow.
DCEN
Timer 2 Down Count Enable. When Timer 2 operates in Auto-Reload mode and EXEN2 = 1, setting DCEN = 1 will cause
Timer 2 to count up or down depending on the state of T2EX.
12.2
Capture Mode
In the Capture mode, Timer 2 is a fixed 16-bit timer or counter that counts up from MIN to MAX.
An overflow from MAX to MIN sets bit TF2 in T2CON. If EXEN2 = 1, a 1-to-0 transition at external input T2EX also causes the current value in TH2 and TL2 to be captured into RCAP2H and
RCAP2L, respectively. In addition, the transition at T2EX causes bit EXF2 in T2CON to be set.
The EXF2 and TF2 bits can generate an interrupt. Capture mode is illustrated in Figure 12-1.
The Timer 2 overflow rate in Capture mode is given by the following equation:
Capture Mode:
65536
Time-out Period = ------------------------------------------------------- × ( TPS + 1 )
Oscillator Frequency
61
3706A–MICRO–9/09
Figure 12-1. Timer 2 Diagram: Capture Mode
OSC
÷TPS
C/T2 = 0
TL2
TH2
TF2
OVERFLOW
C/T2 = 1
TR2
CAPTURE
T2 PIN
RCAP2L
RCAP2H
TRANSITION
DETECTOR
TIMER 2
INTERRUPT
T2EX PIN
EXF2
EXEN2
12.3
Auto-Reload Mode
Timer 2 can be programmed to count up or down when configured in its 16-bit auto-reload
mode. This feature is invoked by the DCEN (Down Counter Enable) bit located in the SFR
T2MOD (see Table 12-4). Upon reset, the DCEN bit is set to 0 so that timer 2 will default to
count up. When DCEN is set, Timer 2 can count up or down, depending on the value of the
T2EX pin. The overflow and reload values depend on the Timer 2 Count Mode bits, T2CM1-0 in
T2MOD. A summary of the Auto-Reload behaviors is listed in Table 12-5.
Table 12-5.
12.3.1
Summary of Auto-Reload Modes
T2CM1-0
DCEN
T2EX
Direction
Behavior
00
0
X
Up
BOTTOM →MAX reload to BOTTOM
00
1
0
Down
MAX →BOTTOM underflow to MAX
00
1
1
Up
BOTTOM →MAX overflow to BOTTOM
01
0
X
Up
MIN →TOP reload to MIN
01
1
0
Down
01
1
1
Up
10
X
X
Up-Down
MIN →TOP →MIN and repeat
11
X
X
Up-Down
MIN →TOP →MIN and repeat
TOP →MIN underflow to TOP
MIN →TOP overflow to MIN
Up Counter
Figure 12-2 shows Timer 2 automatically counting up when DCEN = 0 and T2CM1-0 = 00B. In
this mode Timer 2 counts up to MAX and then sets the TF2 bit upon overflow. The overflow also
causes the timer registers to be reloaded with BOTTOM, the 16-bit value in RCAP2H and
RCAP2L. If EXEN2 = 1, a 16-bit reload can be triggered either by an overflow or by a 1-to-0 transition at external input T2EX. This transition also sets the EXF2 bit. Both the TF2 and EXF2 bits
can generate an interrupt. The Timer 2 overflow rate for this mode is given in the following
equation:
Auto-Reload Mode:
65536 – {RCAP2H , RCAP2L}
Time-out Period = ------------------------------------------------------------------------------- × ( TPS + 1 )
Oscillator Frequency
DCEN = 0, T2CM = 00B
Timer 2 may also be configured to count from MIN to TOP instead of BOTTOM to MAX by setting T2CM1-0 = 01B. In this mode Timer 2 counts up to TOP, the 16-bit value in RCAP2H and
62
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
RCAP2L and then overflows. The overflow sets TF2 and causes the timer registers to be
reloaded with MIN. If EXEN2 = 1, a 1-to-0 transition on T2EX will clear the timer and set EXF2.
The Timer 2 overflow rate for this mode is given in the following equation:
Auto-Reload Mode:
{RCAP2H , RCAP2L} + 1
Time-out Period = ------------------------------------------------------------------ × ( TPS + 1 )
Oscillator Frequency
DCEN = 0, T2CM = 01B
Timer 2 Count Mode 1 is provided to support variable precision asymmetrical PWM in the CCA.
The value of TOP stored in RCAP2H and RCAP2L is double-buffered such that a new TOP
value takes affect only after an overflow. The behavior of Count Mode 0 versus Count Mode 1 is
shown in Figure 12-3.
Figure 12-2. Timer 2 Diagram: Auto-Reload Mode (DCEN = 0)
÷TPS
TL2
TH2
Figure 12-3. Timer 2 Waveform: Auto-Reload Mode (DCEN = 0)
TF2 Set
MAX
BOTTOM
MIN
MAX
T2CM1-0 = 00B, DCEN = 0
TF2 Set
TOP
MIN
T2CM1-0 = 01B, DCEN = 0
12.3.2
Up or Down Counter
Setting DCEN = 1 enables Timer 2 to count up or down, as shown in Figure 12-4. In this mode,
the T2EX pin controls the direction of the count (if EXEN2 = 1). A logic 1 at T2EX makes Timer 2
count up. When T2CM1-0 = 00B, the timer will overflow at MAX and set the TF2 bit. This overflow
also causes BOTTOM, the 16-bit value in RCAP2H and RCAP2L, to be reloaded into the timer
63
3706A–MICRO–9/09
registers, TH2 and TL2, respectively. A logic 0 at T2EX makes Timer 2 count down. The timer
underflows when TH2 and TL2 equal BOTTOM, the 16-bit value stored in RCAP2H and
RCAP2L. The underflow sets the TF2 bit and causes MAX to be reloaded into the timer registers. The EXF2 bit toggles whenever Timer 2 overflows or underflows and can be used as a 17th
bit of resolution. In this operating mode, EXF2 does not flag an interrupt.
When T2EX = 1 and T2CM1-0 = 01B, the timer will overflow at TOP and set the TF2 bit. This
overflow also causes MIN to be reloaded into the timer registers. A logic 0 at T2EX makes Timer
2 count down. The timer underflows when TH2 and TL2 equal MIN. The underflow sets the TF2
bit and causes TOP to be reloaded into the timer registers. The behavior of Count Mode 0 versus Count Mode 0 when DCEN is enabled is shown in Figure 12-6.
Figure 12-4. Timer 2 Diagram: Auto-Reload Mode (T2CM1-0 = 00B, DCEN = 1)
÷TPS
The timer overflow/underflow rate for up-down counting mode is the same as for up counting
mode, provided that the count direction does not change. Changes to the count direction may
result in longer or shorter periods between time-outs.
Figure 12-5. Timer 2 Diagram: Auto-Reload Mode (T2CM1-0 = 01B, DCEN = 1)
÷TPS
64
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Figure 12-6. Timer 2 Waveform: Auto-Reload Mode (DCEN = 1)
TF2 Set
MAX
BOTTOM
MIN
T2CM1-0 = 00B, DCEN = 1
T2EX
EXF2
MAX
TF2 Set
TOP
MIN
T2CM1-0 = 01B, DCEN = 1
12.3.3
Dual Slope Counter
When Timer 2 Auto-Reload Mode uses Count Mode 2 (T2CM 1-0 = 10B) or Count Mode 3
(T2CM1-0 = 11B), the timer operates in a dual slope fashion. The timer counts up from MIN to
TOP and then counts down from TOP to MIN, following a sawtooth waveform as shown in Figure 12-7. The EXF2 bit is set/cleared by hardware to reflect the current count direction (Up =
0and Down = 1). The value of TOP stored in RCAP2H and RCAP2L is double-buffered such that
a new TOP value takes affect only after an underflow. The only difference between Mode 2 and
Mode 3 is when the interrupt flag is set. In Mode 2 TF2 is set once per count period when the
timer underflows at MIN. In Mode 3 TF2 is set twice per count period, once when the timer overflows at TOP and once when the timer underflows at MIN. The interrupt service routine can
check the EXF2 bit to determine if TF2 was set at TOP or MIN. These count modes are provided
to support variable precision symmetrical PWM in the CCA. DCEN has no effect when using
dual slope operation. The Timer 2 overflow rate for this mode is given in the following equation:
Auto-Reload Mode:
{RCAP2H , RCAP2L} × 2
Time-out Period = ------------------------------------------------------------------- × ( TPS + 1 )
Oscillator Frequency
DCEN = 0, T2CM = 10B
65
3706A–MICRO–9/09
Figure 12-7. Timer 2 Waveform: Dual Slope Modes
MAX
T2CM1-0 = 10B
TF2 Set
TOP
MIN
EXF2
MAX
T2CM1-0 = 11B
TF2 Set
TOP
MIN
12.4
Baud Rate Generator
Timer 2 is selected as the baud rate generator by setting TCLK and/or RCLK in T2CON (Table
12-3). Note that the baud rates for transmit and receive can be different if Timer 2 is used for the
receiver or transmitter and Timer 1 is used for the other function. Setting RCLK and/or TCLK
puts Timer 2 into its baud rate generator mode, as shown in Figure 12-8.
The baud rate generator mode is similar to the auto-reload mode, in that a rollover in TH2
causes the Timer 2 registers to be reloaded with the 16-bit value in registers RCAP2H and
RCAP2L, which are preset by software.
The baud rates in UART Modes 1 and 3 are determined by Timer 2’s overflow rate according to
the following equation.
2 Overflow RateModes 1 and 3 Baud Rates = Timer
----------------------------------------------------------16
The Timer can be configured for either timer or counter operation. In most applications, it is configured for timer operation (CP/T2 = 0). The baud rate formulas are given below.
T2CM = 00B
T2CM = 01B
Modes 1, 3
Baud Rate
Modes 1, 3
Baud Rate
Oscillator Frequency
= --------------------------------------------------------------------------------------------------------------------------------16 × ( TPS + 1 ) × [ 65536 – ( RCAP2H,RCAP2L ) ]
Oscillator Frequency
= -----------------------------------------------------------------------------------------------------------------16 × ( TPS + 1 ) × [ (RCAP2H,RCAP2L) + 1 ]
where (RCAP2H, RCAP2L) is the content of RCAP2H and RCAP2L taken as a 16-bit unsigned
integer.
Timer 2 as a baud rate generator is shown in Figure 12-8. This figure is valid only if RCLK or
TCLK = 1 in T2CON. Note that a rollover in TH2 does not set TF2 and will not generate an interrupt. Note too, that if EXEN2 is set, a 1-to-0 transition in T2EX will set EXF2 but will not cause a
reload from (RCAP2H, RCAP2L) to (TH2, TL2). Thus when Timer 2 is in use as a baud rate gen-
66
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
erator, T2EX can be used as an extra external interrupt. Also note that the Baud Rate and
Frequency Generator modes may be used simultaneously.
Figure 12-8. Timer 2 in Baud Rate Generator Mode
TIMER 1 OVERFLOW
÷2
"0"
OSC
÷TPS
"1"
SMOD1
C/T2 = 0
"1"
TL2
"0"
TH2
RCLK
÷16
TR2
Rx
CLOCK
C/T2 = 1
"1"
"0"
T2 PIN
RCAP2L
TCLK
RCAP2H
TRANSITION
DETECTOR
÷ 16
T2EX PIN
EXF2
Tx
CLOCK
TIMER 2
INTERRUPT
EXEN2
12.5
Frequency Generator (Programmable Clock Out)
Timer 2 can generate a 50% duty cycle clock on T2 (P1.0), as shown in Figure 12-9. This pin,
besides being a regular I/O pin, has two alternate functions. It can be programmed to input the
external clock for Timer/Counter 2 or to toggle its output at every timer overflow. To configure
the Timer/Counter 2 as a clock generator, bit C/T2 (T2CON.1) must be cleared and bit T2OE
(T2MOD.1) must be set. Bit TR2 (T2CON.2) starts and stops the timer.
The clock-out frequency depends on the oscillator frequency and the reload value of Timer 2
capture registers (RCAP2H, RCAP2L), as shown in the following equations.
T2CM = 00B
Oscillator Frequency
1
Clock Out Frequency = -------------------------------------------------------------------------------------------- × --------------------2 × [ 65536 – ( RCAP2H,RCAP2L ) ] TPS + 1
T2CM = 01B
Oscillator Frequency
1
Clock Out Frequency = ------------------------------------------------------------------------------- × --------------------2 × [ ( RCAP2H,RCAP2L ) + 1 ] TPS + 1
In the frequency generator mode, Timer 2 roll-overs will not generate an interrupt. This behavior
is similar to when Timer 2 is used as a baud-rate generator. It is possible to use Timer 2 as a
baud-rate generator and a clock generator simultaneously. Note, however, that the baud-rate
and clock-out frequencies cannot be determined independently from one another since they
both use RCAP2H and RCAP2L.
67
3706A–MICRO–9/09
Figure 12-9. Timer 2 in Clock-out Mode
OSC
÷TPS
TL2
TH2
RCAP2L
RCAP2H
TR2
C/T2
T2OE
÷2
T2 PIN
T2EX PIN
EXF2
TRANSITION
DETECTOR
TIMER 2
INTERRUPT
EXEN2
13. Compare/Capture Array
The AT89LP6440 includes a four channel Compare/Capture Array (CCA) that performs a variety
of timing operations including input event capture, output compare waveform generation and
pulse width modulation (PWM). Timer 2 serves as the time base for the four 16-bit compare/capture modules. The CCA has the following features:
• Four 16-bit Compare/Capture channels
• Common time base provided by Timer 2
• Selectable external and internal capture events including pin change, timer overflow and
comparator output change
• Symmetric/Asymmetric PWM with selectable polarity
• Multi-phasic PWM outputs
• One interrupt flag per channel with a common interrupt vector
The block diagram of the CCA is given in Figure 13-1. Each channel consists of an 8-bit control
register and a 16-bit data register. The channel registers are not directly accessible. The CCA
address register T2CCA provides an index into the array. The control, data low and data high
bytes of the currently indexed channel are accessed through the T2CCC, T2CCL and T2CCH
registers respectively.
Each channel can be individually configured for capture or compare mode. Capture mode is the
default setting. During capture mode the current value of the time base is copied into the channel’s data register when the specified external or internal event occurs. An interrupt flag is set at
the same time and the time base may be optionally cleared. To enable compare mode, the
CCMx bit in the channel’s control register (CCCx) should be set to 1. In compare mode an interrupt flag is set and an output pin is optionally toggled when the value of the time base matches
the value of the channel’s data register. The time base may also be optionally cleared on a compare match.
Timer 2 must be running (TR2 = 1) in order to perform captures or compares with the CCA.
However, when TR2 = 0 the external capture events will still set their associated flags and may
be used as additional external interrupts.
68
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Figure 13-1. Compare/Capture Array Block Diagram
OSC
÷TPS
RCAP2L
RCAP2H
TL2
TH2
C/T2 = 0
(P1.0) T2
C/T2 =1
TF2
T2CCF
Timer 2 Interrupt
CCA Interrupt
TR2
CCCA
CCAL
CCAH
CCA (P2.0)
CCCB
CCBL
CCBH
CCB (P2.1)
CCCC
CCCL
CCCH
CCC (P2.2)
CCCD
CCDL
CCDH
CCD (P2.3)
T2CCC
T2CCL
T2CCH
T2CCA
13.1
CCA Registers
The Compare/Capture Array has five Special Function Registers: T2CCA, T2CCC, T2CCL,
T2CCH and T2CCF. The T2CCF register contains the interrupt flags for each CCA channel. The
CCA interrupt is a logic OR of the bits in T2CCF. The flags are set by hardware when a compare/capture event occurs on the relevant channel and must be cleared by software. The
T2CCF bits will only generate an interrupt when the ECC bit (IE2.1) is set and the CIENx bit in
the associated channel’s CCCx register is set.
The T2CCC, T2CCL and T2CCH register locations are not true SFRs. These locations represent
access points to the contents of the array. Writes/reads to/from the T2CCC, T2CCL and T2CCH
locations will access the control, data low and data high bytes of the CCA channel currently
selected by the index in T2CCA. Channels currently not indexed by T2CCA are not accessible.
When writing to T2CCH, the value is stored in a shadow register. When T2CCL is written, the
16-bit value formed by the contents of T2CCL and the T2CCH shadow is written into the array.
Therefore, T2CCH must be written prior to writing T2CCL. All four channels use the same
T2CCH shadow register. If the value of T2CCH remains constant for multiple writes, there is no
need to update T2CCH between T2CCL writes. Every write to T2CCL will use the last value of
T2CCH for the upper data byte. It is not possible to write to the data register of a channel configured for capture mode.
The configuration bits for each channel are stored in the CCCx registers accessible through
T2CCC. See Table 13-5 on page 73 for a description of the CCCx register.
69
3706A–MICRO–9/09
Table 13-1.
T2CCA – Timer/Counter 2 Compare/Capture Address
T2CCA Address = 0D1H
Reset Value = xxxx xx00B
Not Bit Addressable
—
—
—
—
—
—
T2CCA.1
T2CCA.0
Bit
7
6
5
4
3
2
1
0
Symbol
Function
Compare/Capture Address. Selects which CCA channel is currently accessible through the T2CCH, T2CCL and T2CCC
registers. Only one channel may be accessed at a time.
T2CCA
[1-0]
T2CCA1
T2CCA0
Channel
0
0
A – T2CCH, T2CCL and T2CCC access data and control for Channel A
0
1
B – T2CCH, T2CCL and T2CCC access data and control for Channel B
1
0
C – T2CCH, T2CCL and T2CCC access data and control for Channel C
1
1
D – T2CCH, T2CCL and T2CCC access data and control for Channel D
Table 13-2.
T2CCH – Timer/Counter 2 Compare/Capture Data High
T2CCH Address = 0D2H
Reset Value = 0000 0000B
Not Bit Addressable
T2CCD.15
T2CCD.14
T2CCD.13
T2CCD.12
T2CCD.11
T2CCD.10
T2CCD.9
T2CCD.8
Bit
7
6
5
4
3
2
1
0
Symbol
Function
T2CCD
[15-8]
Compare/Capture Data (High Byte). Reads from T2CCH will return the high byte from the CCA channel currently
selected by T2CCA. The high byte of the selected CCA channel will be updated with the contents of T2CCH when
T2CCL is written. When writing multiple channels with the same high byte, T2CCH need not be updated between writes
to T2CCL.
Note:
All writes/reads to/from T2CCH will access channel X as currently selected by T2CCA.The data registers for the remaining
unselected channels are not accessible.
Table 13-3.
T2CCL – Timer/Counter 2 Compare/Capture Data Low
T2CCC Address = 0D3H
Reset Value = 0000 0000B
Not Bit Addressable
T2CCD.7
T2CCD.6
T2CCD.5
T2CCD.4
T2CCD.3
T2CCD.2
T2CCD.1
T2CCD.0
Bit
7
6
5
4
3
2
1
0
Symbol
Function
T2CCD
[7-0]
Compare/Capture Data (Low Byte). Reads from T2CCL will return the low byte from the CCA channel currently selected
by T2CCA. Writes to T2CCL will update the selected CCA channel with the 16-bit contents of T2CCH and T2CCL.
Note:
70
All writes/reads to/from T2CCL will access channel X as currently selected by T2CCA.The data registers for the remaining
unselected channels are not accessible.
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Table 13-4.
T2CCF – Timer/Counter 2 Compare/Capture Flags
T2CCF Address = 0D5H
Reset Value = XXXX 0000B
Not Bit Addressable
–
–
–
–
CCFD
CCFC
CCFB
CCFA
Bit
7
6
5
4
3
2
1
0
Symbol
Function
CCFD
Channel D Compare/Capture Interrupt Flag. Set by a compare/capture event on channel D. Must be cleared by software.
CCFD will generate an interrupt when CIEND = 1 and ECC = 1.
CCFC
Channel C Compare/Capture Interrupt Flag. Set by a compare/capture event on channel C. Must be cleared by
software. CCFC will generate an interrupt when CIENC = 1 and ECC = 1.
CCFB
Channel B Compare/Capture Interrupt Flag. Set by a compare/capture event on channel B. Must be cleared by software.
CCFB will generate an interrupt when CIENB = 1 and ECC = 1.
CCFA
Channel A Compare/Capture Interrupt Flag. Set by a compare/capture event on channel A. Must be cleared by software.
CCFA will generate an interrupt when CIENA = 1 and ECC = 1.
13.2
Input Capture Mode
The Compare/Capture Array provides a variety of capture modes suitable for time-stamping
events or performing measurements of pulse width, frequency, slope, etc. CCA channels are
configured for capture mode by clearing the CCMx bit in the associated CCCx register to 0.
Each time a capture event occurs, the contents of Timer 2 (TH2 and TL2) are transferred to the
16-bit data register of the corresponding channel, and the channel’s interrupt flag CCFx is set in
T2CCF. Optionally, the capture event may also clear Timer 2 to 0000H by setting the CTCx bit in
CCCx. The capture event is defined by the CxM2-0 bits in CCCx and may be either externally or
internally generated. A diagram of a CCA channel in capture mode is shown in Figure 13-2.
Figure 13-2. CCA Capture Mode Diagram
00H
00H
TL2
TH2
(P2.x) CCx
CIENx
CTCx
“0”
Timer 0 Overflow
Timer 1 Overflow
Comparator A
Comparator B
0
1
2
3
4
5
6
7
CCxL
Interrupt
CCxH
CCFx
CCCx
CxM2-0
T2CCL
T2CCH
T2CCC
Each CCA channel has an associated external capture input pin: CCA (P2.0), CCB (P2.1), CCC
(P2.2) and CCD (P2.3). External capture events are always edge-triggered and can be selected
71
3706A–MICRO–9/09
to occur at a negative edge, positive edge, or both (toggle). Capture inputs are sampled every
clock cycle and a new value must be held for at least 2 clock cycles to be correctly sampled by
the device. The maximum achievable capture rate will be determined by how fast the software
can retrieve the captured data. There is no protection against capture events overrunning the
data register.
Capture events may also be triggered internally by the overflows of Timer 0 or Timer 1, or by an
event from the dual analog comparators. Any comparator event which can generate a comparator interrupt may also be used as a capture event. However, Timer 2 should not be selected as
the comparator clock source when using the comparator as the capture trigger.
When the DAC output is enabled on P2.2 and P2.3, channels C and D cannot use their external
pin capture modes. However, those channels may still use the timer or comparator triggers to
capture data. The same applies for all four channels when Port 2 is used for the external memory interface.
13.2.1
72
Timer 2 Operation for Capture Mode
Capture channels are intended to work with Timer 2 in capture mode CP/RL2 = 1. Captures can
still occur when Timer 2 operates in other modes; however, the full 16-bit count range may not
be available. The TF2 flag can be used to determine if the timer overflowed before the capture
occurred. If the timer is operating in dual-slope mode (CP/RL2 = 0, T2CM1-0 = 1xB), the count
direction (Up = 0 and Down = 1) at the time of the event will be captured into the channel’s
CDIRx bit in CCCx. CTCx must be cleared to 0 for all channels if Timer 2 is operating in Baud
Rate mode or errors may occur in the serial communication.
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Table 13-5.
T2CCC – Timer/Counter 2 Compare/Capture Control
T2CCC Address = 0D4H
Reset Value = 00X0 0000B
Not Bit Addressable
CIENx
CDIRx
–
CTCx
CCMx
CxM2
CxM1
CxM0
Bit
7
6
5
4
3
2
1
0
Symbol
Function
CIENx
Channel X Interrupt Enable. When set, channel X’s interrupt flag, CCFx in T2CCF, will generate an interrupt when ECC
= 1. Clear to disable interrupts from channel X.
CDIRx
Channel X Capture Direction. In dual-slope modes, a compare/capture event on channel X will store the current count
direction into CDIRx. Up-counting = 0 and down-counting = 1. Modifying this bit has no effect.
CTCx
Clear Timer on Compare/Capture of Channel X. When set, the Timer 2 registers TL2 and TH2 will be cleared by a
compare/capture event on channel X. When cleared, Timer 2 is unaffected by channel X events.
CCMx
Channel X Compare/Capture Mode. When CCMx = 1, channel X operates in compare mode. When CCMx = 0, channel
X operates in capture mode.
CxM[2-0]
Channel X Mode. Selects the output/input events for compare/capture channel X.
Notes:
CxM2
CxM1
CxM0
Capture Event (CCMx = 0)
0
0
0
Disabled
0
0
1
Trigger on negative edge of CCx pin
0
1
0
Trigger on positive edge of CCx pin
0
1
1
Trigger on either edge of CCx pin
1
0
0
Trigger on Timer 0 overflow
1
0
1
Trigger on Timer 1 overflow
1
1
0
Trigger on Analog Comparator A Event(2)
1
1
1
Trigger on Analog Comparator B Event(3)
CxM2
CxM1
CxM0
Compare Action (CCMx = 1)
0
0
0
Output disabled (interrupt only)
0
0
1
Set CCx pin on compare match
0
1
0
Clear CCx pin on compare match
0
1
1
Toggle CCx pin on compare match
1
0
0
Inverting Pulse Width Modulation(4)
1
0
1
Non-Inverting Pulse Width Modulation(4)
1
1
0
Reserved
1
1
1
Reserved
1. All writes/reads to/from T2CCC will access channel X as currently selected by T2CCA.The control registers for the remaining unselected channels are not accessible.
2. Analog Comparator A events are determined by the CMA2-0 bits in ACSRA. See Table 19-1 on page 130.
3. Analog Comparator B events are determined by the CMB2-0 bits in ACSRB. See Table 19-2 on page 131.
4. Asymmetrical versus Symmetrical PWM is determined by the Timer 2 Count Mode. See Section 13.4 on page 76.
73
3706A–MICRO–9/09
13.3
Output Compare Mode
The Compare/Capture Array provides a variety of compare modes suitable for event timing or
waveform generation. CCA channels are configured for compare mode by setting the CCMx bit
in the associated CCCx register to 1. A compare event occurs when the 16-bit contents of a
channel’s data register match the contents of Timer 2 (TH2 and TL2). The compare event also
sets the channel’s interrupt flag CCFx in T2CCF and may optionally clear Timer 2 to 0000H if the
CTCx bit in CCCx is set. A diagram of a CCA channel in compare mode is shown in Figure 13-3.
Figure 13-3. CCA Compare Mode Diagram
00H
00H
TL2
TH2
CCx (P2.x)
CTCx
=
CxM2-0
CCFx
CCxL
CCxH
CCCx
T2CCL
shadow
T2CCC
Interrupt
CIENx
T2CCH
13.3.1
Waveform Generation
Each CCA channel has an associated external compare output pin: CCA (P2.0), CCB (P2.1),
CCC (P2.2) and CCD (P2.3). The CxM2-0 bits in CCCx determine what action is taken when a
compare event occurs. The output pin may be set to 1, cleared to 0 or toggled. Output actions
take place even if the interrupt is disabled; however, the associated I/O pin must be set to the
desired output mode before the compare event occurs. The state of the compare outputs are initialized to 1 by reset. Channels C and D cannot use their output pin when the DAC is enabled.
These channels may still be used to generate interrupts or to clear the timebase. The same
applies to all four channels when Port 2 is used for the external memory interface.
Multiple compare events per channel can occur within a single time period, provided that the
software has time to update the compare value before the timer reaches the next compare point.
In this case other interrupts should be disabled or the CCA interrupt given a higher priority in
order to ensure that the interrupt is serviced in time.
A wide range of waveform generation configurations are possible using the various operating
modes of Timer 2 and the CCA. Some example configurations are detailed below. Pulse width
modulation is a special case of output compare. See Section 13.4 on page 76 for more details of
PWM operation.
74
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
13.3.1.1
Normal Mode
The simplest waveform mode is when CP/RL2 = 0 and T2CM1-0 = 01B. In this mode the frequency of the output is determined by the TOP value stored in RCAP2L and RCAP2H and
output edges occur at fractions of the timer period. Figure 13-4 shows an example of outputting
two waveforms of the same frequency but different phase by using the toggle on match action.
More complex waveforms are achieved by changing the TOP value and the compare values
more frequently.
Figure 13-4. Normal Mode Waveform Example
CP/RL2 = 0, T2CM1-0 = 01B, DCEN = 0
{RCAP2H,RCA2L}
{CCAH,CCAL}
{CCBH,CCBL}
CCA
CCB
13.3.1.2
Clear-Timer-on-Compare Mode
Clear-Timer-on-Compare (CTC) mode occurs when the CTCx bit of a compare channel is set to
one. CTC Mode works best when Timer 2 is in capture mode (CP/RL2 = 1) to allow the full range
of compare values. In CTC Mode the compare value defines the interval between output events
because the timer is cleared after every compare match. Figure 13-5 shows an example waveform using the toggle on match action in CTC Mode.
Figure 13-5. CTC Mode Waveform Example
CP/RL2 = 1
{CCAH,CCAL}
CCA
13.3.1.3
Dual-Slope Mode
The dual-slope mode occurs when CP/RL2 = 0 and T2CM1-0 = 1xB. In this mode the frequency
of the output is determined by the TOP value stored in RCAP2L and RCAP2H and output edges
occur at fractions of the timer period on both the up and down count. Figure 13-6 shows an
example of outputting two symmetrical waveforms using the toggle on match action. More complex waveforms are achieved by changing the TOP value and the compare values more
frequently.
75
3706A–MICRO–9/09
Figure 13-6. Dual-Slope Waveform Example
CP/RL2 = 0, T2CM1-0 = 10B, DCEN = 0
{RCAP2H,RCA2L}
{CCAH,CCAL}
{CCBH,CCBL}
CCA
CCB
13.3.2
13.4
Timer 2 Operation for Compare Mode
Compare channels will work with any Timer 2 operating mode. The full 16-bit compare range
may not be available in all modes. In order for a compare output action to take place, the compare values must be within the counting range of Timer 2. CTCx must be cleared to 0 for all
channels if Timer 2 is operating in Baud Rate mode or errors may occur in the serial
communication.
Pulse Width Modulation Mode
In Pulse Width Modulation (PWM) Mode, a compare channel can output a square wave with programmable frequency and duty cycle. Setting CCMx = 1 and CxM 2-0 = 10xB enables PWM
Mode. PWM Mode is similar to Output Compare Mode except that the compare value is doublebuffered. A diagram of a CCA channel in PWM Mode is shown in Figure 13-7. The PWM polarity
is selectable between inverting and non-inverting modes. PWM is intended for use with Timer 2
in Auto-Reload Mode (CP/RL2 = 0, DCEN = 0) using count modes 1, 2 or 3. The PWM can operate in asymmetric (edge-aligned) or symmetric (center-aligned) mode depending on the T2CM
selection. The CCA PWM has variable precision from 2 to 16 bits. A trade-off between frequency
and precision is made by changing the TOP value of the timer. The CCA PWM always uses the
greatest precision allowable for the selected output frequency, as compared to Timer 0 and 1
whose PWMs are fixed at 8-bit precision regardless of frequency.
Figure 13-7. CCA PWM Mode Diagram
TL2
TH2
CCx (P2.x)
=
CxM2-0
shadow
CCxL
shadow
CCxH
CCFx
CCCx
Interrupt
CIENx
T2CCL
shadow
T2CCC
T2CCH
76
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
13.4.1
Asymmetrical PWM
For Asymmetrical PWM, Timer 2 should be configured for Auto-Reload mode and Count Mode 1
(CP/RL2 = 0, DCEN = 0, T2CM1-0 = 01B). Asymmetrical PWM uses single slope operation as
shown in Figure 13-8. The timer counts up from BOTTOM to TOP and then restarts from BOTTOM. In non-inverting mode, the output CCx is set on the compare match between Timer 2
(TL2, TH2) and the channel data register (CCxL, CCxH), and cleared at BOTTOM. In inverting
mode, the output CCx is cleared on the compare match between Timer 2 and the data register,
and set at BOTTOM. The resulting asymmetrical output waveform is left-edge aligned.
The TOP value in RCAP2L and RCAP2H is double buffered such that the output frequency is
only updated at the TOP to BOTTOM overflow. The channel data register (CCxL, CCxH) is also
double-buffered such that the duty cycle is only updated at the TOP to BOTTOM overflow to prevent glitches. The output frequency and duty cycle for asymmetrical PWM are given by the
following equations:
Oscillator Frequency
1
f OUT = ---------------------------------------------------------------- × --------------------{RCAP2H, RCAP2L} + 1 TPS + 1
{CCxH, CCxL}
Duty Cycle = 100% × ---------------------------------------------------------------{RCAP2H, RCAP2L} + 1
{RCAP2H, RCAP2L} – {CCxH, CCxL} + 1
Duty Cycle = 100% × ------------------------------------------------------------------------------------------------------------{RCAP2H, RCAP2L} + 1
Inverting:
Non-Inverting:
The extreme compare values represent special cases when generating a PWM waveform. If the
compare value is set equal to (or greater than) TOP, the output will remain low or high for noninverting and inverting modes, respectively. If the compare value is set to BOTTOM (0000H), the
output will remain high or low for non-inverting and inverting modes, respectively.
Figure 13-8. Asymmetrical (Edge-Aligned) PWM
CP/RL2 = 0, T2CM1-0 = 01B, DCEN = 0
{RCAP2H,RCA2L}
{CCxH,CCxL}
Inverted
CCx
Non-inverted
13.4.2
Symmetrical PWM
For Symmetrical PWM, Timer 2 should be configured for Auto-Reload mode and Count Mode 2
or 3 (CP/RL2 = 0, DCEN = 0, T2CM1-0 = 1xB). Symmetrical PWM uses dual-slope operation as
shown in Figure 13-9. The timer counts up from MIN to TOP and then counts down from TOP to
MIN. The timer is equal to TOP for exactly one clock cycle. In non-inverting mode, the output
CCx is cleared on the up-count compare match between Timer 2 (TL2, TH2) and the channel
data register (CCxL, CCxH), and set at the down-count compare match. In inverting mode, the
output CCx is set on the up-count compare match between Timer 2 and the data register, and
cleared at the down-count compare match. The resulting symmetrical PWM output waveform is
77
3706A–MICRO–9/09
center-aligned around the timer equal to TOP point. Symmetrical PWM may be used to generate
non-overlapping waveforms.
The TOP value in RCAP2L and RCAP2H is double buffered such that the output frequency is
only updated at the underflow. The channel data register (CCxL, CCxH) is also double-buffered
to prevent glitches. The output frequency and duty cycle for symmetrical PWM are given by the
following equations:
Oscillator Frequency
1
f OUT = ----------------------------------------------------------------- × --------------------2 × {RCAP2H, RCAP2L} TPS + 1
{CCxH, CCxL}
Duty Cycle = 100% × -----------------------------------------------------{RCAP2H, RCAP2L}
{RCAP2H, RCAP2L} – {CCxH, CCxL}
Duty Cycle = 100% × --------------------------------------------------------------------------------------------------{RCAP2H, RCAP2L}
Non-Inverting:
Inverting:
The extreme compare values represent special cases when generating a PWM waveform. If the
compare value is set equal to (or greater than) TOP, the output will remain high or low for noninverting and inverting modes, respectively. If the compare value is set to MIN (0000H), the output will remain low or high for non-inverting and inverting modes, respectively.
Figure 13-9. Non-overlapping Waveforms Using Symmetrical PWM
CP/RL2 = 0, T2CM1-0 = 10B, DCEN = 0
{RCAP2H,RCA2L}
{CCAH,CCAL}
{CCBH,CCBL}
(Inverted) CCA
(Non-Inverted) CCB
13.4.2.1
Phase and Frequency Correct PWM
When T2CM1-0 = 10B the Symmetrical PWM operates in phase and frequency correct mode. In
this mode the compare value double buffer is only updated when the timer equals MIN (underflow). This guarantees that the resulting waveform is always symmetrical around the TOP value
as shown in Figure 13-10 because the up and down count compare values are identical. The
TF2 interrupt flag is only set at underflow.
13.4.2.2
Phase Correct PWM
When T2CM1-0 = 11B the Symmetrical PWM operates in phase correct mode. In this mode the
compare value double buffer is updated when the timer equals MIN (underflow) and TOP (overflow). The resulting waveform may not be completely symmetrical around the TOP value as
shown in Figure 13-11 because the up and down count compare values may not be identical.
However, this allows the pulses to be weighted toward one edge or another. The TF2 interrupt
flag is set at both underflow and overflow.
78
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Figure 13-10. Phase and Frequency Correct Symmetrical (Center-Aligned) PWM
CP/RL2 = 0, T2CM1-0 = 10B, DCEN = 0
Duty Cycle Updated
{RCAP2H,RCA2L}
{CCxH,CCxL}
Inverted
CCx
Non-Inverted
Figure 13-11. Phase Correct Symmetrical (Center-Aligned) PWM
CP/RL2 = 0, T2CM1-0 = 11B, DCEN = 0
Duty Cycle Updated
{RCAP2H,RCA2L}
{CCxH,CCxL}
Inverted
CCx
Non-Inverted
13.4.3
Multi-Phasic PWM
The PWM channels may be configured to provide multi-phasic alternating outputs by the PHS2-0
bits in T2MOD. The AT89LP6440 provides 1 out of 2, 1 out of 3, 1 out of 4 and 2 out of 4 phase
modes (See Table 13-6). In Multi-phasic mode the PWM outputs on CCA, CCB, CCC and CCD
are connected to a one-hot shift register that selectively enables and disables the outputs (See
Figure 13-12). Compare points on disabled channels are blocked from toggling the output as if
the compare value was set equal to TOP. The PHSD bit in T2MOD controls the direction of the
shift register. Example waveforms are shown in Figure 13-14 on page 81. In order to use multiphasic PWM, the associated channels must be configured for PWM operation. Non-PWM channels are not affected by multi-phasic operation. However, their locations in the shift register are
maintained such that some periods in the PWM outputs may not have any pulses as shown in
Figure 13-13.
The PHS2-0 bits may only be modified when the timer is not operational (TR2 = 0). Updates to
PHSD are allowed at any time. Note that channels C and D in 1:2 phase mode and channel D in
1:3 phase mode operate normally.
79
3706A–MICRO–9/09
Table 13-6.
Summary of Multi-Phasic Modes
Behavior
PHS2-0
Mode
PHSD = 0
PHSD = 1
000
Off
001
1:2
A →B →A →B
B →A →B →A
010
1:3
A →B →C →A →B →C
C →B →A →C →B →A
011
1:4
100
2:4
A →B →C →D →A →B →C →D
A →B →A →B
C →D →C →D
D →C →B →A →D →C →B →A
B →A →B →A
D →C →D →C
Normal Operation (all channels active at all times)
Figure 13-12. Multi-Phasic PWM Output Stage
PHS = 010B
1
EN
1
EN
EN
0
EN
1
EN
CCC
CCB
0
EN
0
EN
0
EN
CCC
CCD
1
EN
PHSD
PHSD
CCB
0
CCA
EN
CCB
0
EN
1
EN
CCC
CCD
CCD
CCC
PHSD
EN
1
PHSD
CCB
0
PHS = 100B
CCA
EN
PHSD
1
PHSD
EN
PHSD
PHSD
1
PHS = 011B
CCA
PHSD
CCA
PHSD
PHS = 001B
0
CCD
EN
Figure 13-13. Three-Phase Mode with Channel B Disabled
PHS = 010B, CCB disabled
CCA
CCB
CCC
PHSD
80
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Figure 13-14. Multi-Phasic PWM Modes
PHS = 000B
CCA
CCB
CCC
CCD
PHS = 001B
CCA
CCB
CCC
CCD
PHS = 010B
CCA
CCB
CCC
CCD
PHSD
PHS = 011B
CCA
CCB
CCC
CCD
PHSD
PHS = 100B
CCA
CCB
CCC
CCD
14. External Interrupts
The INT0 (P3.2) and INT1 (P3.3) pins of the AT89LP6440 may be used as external interrupt
sources. The external interrupts can be programmed to be level-activated or transition-activated
by setting or clearing bit IT1 or IT0 in Register TCON. If ITx = 0, external interrupt x is triggered
by a detected low at the INTx pin. If ITx = 1, external interrupt x is edge-triggered. In this mode if
successive samples of the INTx pin show a high in one cycle and a low in the next cycle, interrupt request flag IEx in TCON is set. Flag bit IEx then requests the interrupt. Since the external
interrupt pins are sampled once each clock cycle, an input high or low should hold for at least 2
oscillator periods to ensure sampling. If the external interrupt is transition-activated, the external
source has to hold the request pin high for at least two clock cycles, and then hold it low for at
least two clock cycles to ensure that the transition is seen so that interrupt request flag IEx will
be set. IEx will be automatically cleared by the CPU when the service routine is called if generated in edge-triggered mode. If the external interrupt is level-activated, the external source has
to hold the request active until the requested interrupt is actually generated. Then the external
source must deactivate the request before the interrupt service routine is completed, or else
81
3706A–MICRO–9/09
another interrupt will be generated. Both INT0 and INT1 may wake up the device from the
Power-down state.
15. General-purpose Interrupts
The General-purpose Interrupt (GPI) function provides 8 configurable external interrupts on
Port 1. Each port pin can detect high/low levels or positive/negative edges. The GPIEN register
select which bits of Port 1 are enabled to generate an interrupt. The GPMOD and GPLS registers determine the mode for each individual pin. GPMOD selects between level-sensitive and
edge-triggered mode. GPLS selects between high/low in level mode and positive/negative in
edge mode. A block diagram is shown in Figure 15-1. The pins of Port 1 are sampled every
clock cycle. In level-sensitive mode, a valid level must appear in two successive samples before
generating the interrupt. In edge-triggered mode, a transition will be detected if the value
changes from one sample to the next. When an interrupt condition on a pin is detected, and that
pin is enabled, the appropriate flag in the GPIF register is set. The flags in GPIF must be cleared
by software. Any GPI interrupt may wake up the device from the Power-down state.
Figure 15-1. GPI Block Diagram
GPLS
1
7
GPMOD
1
7
GPIEN
GPIF
7
7
D Q
(P1.7) GPI7
0
1
0
6
1
6
6
6
D Q
(P1.6) GPI6
0
1
0
5
1
5
5
5
D Q
(P1.5) GPI5
0
1
0
4
1
4
4
4
D Q
(P1.4) GPI4
Interrupt
0
1
0
3
1
3
3
3
D Q
(P1.3) GPI3
0
1
0
2
1
2
2
2
D Q
(P1.2) GPI2
0
1
0
1
1
1
1
1
D Q
(P1.1) GPI1
0
1
0
0
1
0
0
0
D Q
(P1.0) GPI0
0
0
CLK
82
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
.
Table 15-1.
GPMOD – General-purpose Interrupt Mode Register
GPMOD = 9AH
Reset Value = 0000 0000B
Not Bit Addressable
GPMOD7
GPMOD6
GPMOD5
GPMOD4
GPMOD3
GPMOD2
GPMOD1
GPMOD0
7
6
5
4
3
2
1
0
Bit
GPMOD.x
0 = level-sensitive interrupt for P1.x
1 = edge-triggered interrupt for P1.x
Table 15-2.
GPLS – General-purpose Interrupt Level Select Register
GPLS = 9BH
Reset Value = 0000 0000B
Not Bit Addressable
GPLS7
GPLS6
GPLS5
GPLS4
GPLS3
GPLS2
GPLS1
GPLS0
7
6
5
4
3
2
1
0
Bit
GPMOD.x
0 = detect low level or negative edge on P1.x
1 = detect high level or positive edge on P1.x
.
Table 15-3.
GPIEN – General-purpose Interrupt Enable Register
GPIEN = 9CH
Reset Value = 0000 0000B
Not Bit Addressable
GPIEN7
GPIEN6
GPIEN5
GPIEN4
GPIEN3
GPIEN2
GPIEN1
GPIEN0
7
6
5
4
3
2
1
0
Bit
GPIEN.x
0 = interrupt for P1.x disabled
1 = interrupt for P1.x enabled
.
Table 15-4.
GPIF – General-purpose Interrupt Flag Register
GPIF = 9DH
Reset Value = 0000 0000B
Not Bit Addressable
Bit
GPIF7
GPIF6
GPIF5
GPIF4
GPIF3
GPIF2
GPIF1
GPIF0
7
6
5
4
3
2
1
0
GPIF.x
0 = interrupt on P1.x inactive
1 = interrupt on P1.x active. Must be cleared by software.
83
3706A–MICRO–9/09
16. Serial Interface (UART)
The serial interface on the AT89LP6440 implements a Universal Asynchronous Receiver/Transmitter (UART). The UART has the following features:
• Full-duplex Operation
• 8 or 9 Data Bits
• Framing Error Detection
• Multiprocessor Communication Mode with Automatic Address Recognition
• Baud Rate Generator Using Timer 1 or Timer 2
• Interrupt on Receive Buffer Full or Transmission Complete
The serial interface is full-duplex, which means it can transmit and receive simultaneously. It is
also receive-buffered, which means it can begin receiving a second byte before a previously
received byte has been read from the receive register. (However, if the first byte still has not
been read when reception of the second byte is complete, one of the bytes will be lost.) The
serial port receive and transmit registers are both accessed at the Special Function Register
SBUF. Writing to SBUF loads the transmit register, and reading SBUF accesses a physically
separate receive register. The serial port can operate in the following four modes.
• Mode 0: Serial data enters and exits through RXD. TXD outputs the shift clock. Eight data
bits are transmitted/received, with the LSB first. The baud rate is programmable to 1/2 or 1/4
the oscillator frequency, or variable based on Time 1.
• Mode 1: 10 bits are transmitted (through TXD) or received (through RXD): a start bit (0),
8 data bits (LSB first), and a stop bit (1). On receive, the stop bit goes into RB8 in the Special
Function Register SCON. The baud rate is variable based on Timer 1 or Timer 2.
• Mode 2: 11 bits are transmitted (through TXD) or received (through RXD): a start bit (0),
8 data bits (LSB first), a programmable 9th data bit, and a stop bit (1). On transmit, the 9th
data bit (TB8 in SCON) can be assigned the value of “0” or “1”. For example, the parity bit
(P, in the PSW) can be moved into TB8. On receive, the 9th data bit goes into RB8 in the
Special Function Register SCON, while the stop bit is ignored. The baud rate is
programmable to either 1/16 or 1/32 the oscillator frequency.
• Mode 3: 11 bits are transmitted (through TXD) or received (through RXD): a start bit (0),
8 data bits (LSB first), a programmable 9th data bit, and a stop bit (1). In fact, Mode 3 is the
same as Mode 2 in all respects except the baud rate, which is variable based on Timer 1 or
Timer 3 in Mode 3.
In all four modes, transmission is initiated by any instruction that uses SBUF as a destination
register. Reception is initiated in Mode 0 by the condition RI = 0 and REN = 1. Reception is initiated in the other modes by the incoming start bit if REN = 1.
16.1
Multiprocessor Communications
Modes 2 and 3 have a special provision for multiprocessor communications. In these modes,
9 data bits are received, followed by a stop bit. The 9th bit goes into RB8. Then comes a stop bit.
The port can be programmed such that when the stop bit is received, the serial port interrupt is
activated only if RB8 = 1. This feature is enabled by setting bit SM2 in SCON.
The following example shows how to use the serial interrupt for multiprocessor communications.
When the master processor must transmit a block of data to one of several slaves, it first sends
out an address byte that identifies the target slave. An address byte differs from a data byte in
that the 9th bit is “1” in an address byte and “0” in a data byte. With SM2 = 1, no slave is
interrupted by a data byte. An address byte, however, interrupts all slaves. Each slave can
examine the received byte and see if it is being addressed. The addressed slave clears its SM2
84
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
bit and prepares to receive the data bytes that follows. The slaves that are not addressed set
their SM2 bits and ignore the data bytes. See “Automatic Address Recognition” on page 96.
The SM2 bit can be used to check the validity of the stop bit in Mode 1. In a Mode 1 reception, if
SM2 = 1, the receive interrupt is not activated unless a valid stop bit is received.
Table 16-1.
SCON – Serial Port Control Register
SCON Address = 98H
Reset Value = 0000 0000B
Bit Addressable
Bit
SM0/FE
SM1
SM2
REN
TB8
RB8
T1
RI
7
6
5
4
3
2
1
0
(SMOD0 = 0/1)(1)
Symbol
Function
FE
Framing error bit. This bit is set by the receiver when an invalid stop bit is detected. The FE bit is not cleared by valid
frames and must be cleared by software. The SMOD0 bit must be set to enable access to the FE bit. FE will be set
regardless of the state of SMOD0.
SM0
Serial Port Mode Bit 0, (SMOD0 must = 0 to access bit SM0)
Serial Port Mode Bit 1
SM1
SM0
SM1
Mode
Description
Baud Rate(2)
0
0
0
shift register
fosc/2 or fosc/4 or Timer 1
0
1
1
8-bit UART
variable (Timer 1 or Timer 2)
1
0
2
9-bit UART
fosc/32 or fosc/16
1
1
3
9-bit UART
variable (Timer 1 or Timer 2)
SM2
Enables the Automatic Address Recognition feature in Modes 2 or 3. If SM2 = 1 then Rl will not be set unless the received
9th data bit (RB8) is 1, indicating an address, and the received byte is a Given or Broadcast Address. In Mode 1, if SM2 =
1 then Rl will not be activated unless a valid stop bit was received, and the received byte is a Given or Broadcast Address.
In Mode 0, SM2 determines the idle state of the shift clock such that the clock is the inverse of SM2, i.e. when SM2 = 0
the clock idles high and when SM2 = 1 the clock idles low.
REN
Enables serial reception. Set by software to enable reception. Clear by software to disable reception.
TB8
The 9th data bit that will be transmitted in Modes 2 and 3. Set or clear by software as desired. In Mode 0, setting TB8
enables Timer 1 as the shift clock generator.
RB8
In Modes 2 and 3, the 9th data bit that was received. In Mode 1, if SM2 = 0, RB8 is the stop bit that was received. In Mode
0, RB8 is not used.
TI
Transmit interrupt flag. Set by hardware at the end of the 8th bit time in Mode 0, or at the beginning of the stop bit in the
other modes, in any serial transmission. Must be cleared by software.
RI
Receive interrupt flag. Set by hardware at the end of the 8th bit time in Mode 0, or halfway through the stop bit time in the
other modes, in any serial reception (except see SM2). Must be cleared by software.
Notes:
1. SMOD0 is located at PCON.6.
2. fosc = oscillator frequency. The baud rate depends on SMOD1 (PCON.7).
85
3706A–MICRO–9/09
16.2
Baud Rates
The baud rate in Mode 0 depends on the value of the SMOD1 bit in Special Function Register
PCON.7. If SMOD1 = 0 (the value on reset) and TB8 = 0, the baud rate is 1/4 of the oscillator
frequency. If SMOD1 = 1 and TB8 = 0, the baud rate is 1/2 of the oscillator frequency, as shown
in the following equation:
Mode 0 Baud Rate
TB8 = 0
SMOD1
2
= -------------------- × Oscillator Frequency
4
The baud rate in Mode 2 also depends on the value of the SMOD1 bit. If SMOD1 = 0, the baud
rate is 1/32 of the oscillator frequency. If SMOD1 = 1, the baud rate is 1/16 of the oscillator frequency, as shown in the following equation:
SMOD1
2
Mode 2 Baud Rate = -------------------- × Oscillator Frequency
32
16.2.1
Using Timer 1 to Generate Baud Rates
Setting TB8 = 1 in Mode 0 enables Timer 1 as the baud rate generator. When Timer 1 is the
baud rate generator for Mode 0, the baud rates are determined by the Timer 1 overflow rate and
the value of SMOD1 according to the following equation:
Mode 0 Baud Rate
TB8 = 1
SMOD1
2
= -------------------- × (Timer 1 Overflow Rate)
4
The Timer 1 overflow rate normally determines the baud rates in Modes 1 and 3. When Timer 1
is the baud rate generator, the baud rates are determined by the Timer 1 overflow rate and the
value of SMOD1 according to the following equation:
Modes 1, 3
SMOD1
2
= -------------------- × (Timer 1 Overflow Rate)
32
Baud Rate
The Timer 1 interrupt should be disabled in this application. The Timer itself can be configured
for either timer or counter operation in any of its 3 running modes. In the most typical applications, it is configured for timer operation in auto-reload mode (high nibble of TMOD = 0010B). In
this case, the baud rate is given by the following formula:
Modes 1, 3
SMOD1
2
Oscillator Frequency
1
= -------------------- × ------------------------------------------------------- × --------------------32
[
256
–
(
TH1
)
]
TPS
+1
Baud Rate
Programmers can achieve very low baud rates with Timer 1 by configuring the Timer to run as a
16-bit auto-reload timer (high nibble of TMOD = 0001B). In this case, the baud rate is given by
the following formula.
Modes 1, 3
SMOD1
2
Oscillator Frequency
1
= -------------------- × --------------------------------------------------------- × --------------------32
[ 65536 – ( RH1,RL1 ) ] TPS + 1
Baud Rate
Table 16-2 lists commonly used baud rates and how they can be obtained from Timer 1.
86
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Table 16-2.
Commonly Used Baud Rates Generated by Timer 1 (TPS = 0000B)
Timer 1
16.2.2
Baud Rate
fOSC (MHz)
SMOD1
C/T
Mode
Reload Value
Mode 0: 1 MHz
4
0
X
X
X
Mode 2: 750K
12
1
X
X
X
62.5K
12
1
0
2
F4H
38.4K
11.059
0
0
2
F7H
19.2K
11.059
1
0
2
DCH
9.6K
11.059
0
0
2
DCH
4.8K
11.059
0
0
2
B8H
2.4K
11.059
0
0
2
70H
1.2K
11.059
0
0
1
FEE0H
137.5
11.986
0
0
1
F55CH
110
6
0
0
1
F958H
110
12
0
0
1
F304H
Using Timer 2 to Generate Baud Rates
Timer 2 is selected as the baud rate generator by setting TCLK and/or RCLK in T2CON. Under
these conditions, the baud rates for transmit and receive can be simultaneously different by
using Timer 1 for transmit and Timer 2 for receive, or vice versa. The baud rate generator mode
is similar to the auto-reload mode, in that a rollover causes the Timer 2 registers to be reloaded
with the 16-bit value in registers RCAP2H and RCAP2L, which are preset by software. In this
case, the baud rates in Modes 1 and 3 are determined by Timer 2’s overflow rate according to
the following equation:.
Modes 1 and 3
Baud Rate
1
Oscillator Frequency
1
= ------ × --------------------------------------------------------------------------------- × --------------------16 [ 65536 – ( RCAP2H,RCAP2L ) ] TPS + 1
Table 16-3 lists commonly used baud rates and how they can be obtained from Timer 2.
Table 16-3.
Commonly Used Baud Rates Generated by Timer 2 (TPS = 0000B)
Timer 2
Baud Rate
fOSC (MHz)
CP/RL2
C/T2
TCLK or RCLK
Reload Value
62.5K
12
0
0
1
FFF4H
19.2K
11.059
0
0
1
FFDCH
9.6K
11.059
0
0
1
FFB8H
4.8K
11.059
0
0
1
FF70H
2.4K
11.059
0
0
1
FEE0H
1.2K
11.059
0
0
1
FDC0H
137.5
11.986
0
0
1
EAB8H
110
6
0
0
1
F2AFH
110
12
0
0
1
E55EH
87
3706A–MICRO–9/09
16.3
More About Mode 0
In Mode 0, the UART is configured as a two wire half-duplex synchronous serial interface. Serial
data enters and exits through RXD. TXD outputs the shift clock. Eight data bits are transmitted/received, with the LSB first. Figure 16-1 on page 89 shows a simplified functional diagram of
the serial port in Mode 0 and associated timing. The baud rate is programmable to 1/2 or 1/4 the
oscillator frequency by setting/clearing the SMOD1 bit. However, changing SMOD1 has an
effect on the relationship between the clock and data as described below. The baud rate can
also be generated by Timer 1 by setting TB8. Table 16-4 lists the baud rate options for Mode 0.
Table 16-4.
Mode 0 Baud Rates
TB8
SMOD1
Baud Rate
0
0
fSYS/4
0
1
fSYS/2
1
0
(Timer 1 Overflow) / 4
1
1
(Timer 1 Overflow) / 2
Transmission is initiated by any instruction that uses SBUF as a destination register. The “write
to SBUF” signal also loads a “1” into the 9th position of the transmit shift register and tells the TX
Control Block to begin a transmission. The internal timing is such that one full bit slot may elapse
between “write to SBUF” and activation of SEND.
SEND transfers the output of the shift register to the alternate output function line of P3.0, and
also transfers Shift Clock to the alternate output function line of P3.1. As data bits shift out to the
right, “0”s come in from the left. When the MSB of the data byte is at the output position of the
shift register, the “1” that was initially loaded into the 9th position is just to the left of the MSB,
and all positions to the left of that contain “0”s. This condition flags the TX Control block to do
one last shift, then deactivate SEND and set TI.
Reception is initiated by the condition REN = 1 and R1 = 0. At the next clock cycle, the RX Control unit writes the bits 11111110 to the receive shift register and activates RECEIVE in the next
clock phase. RECEIVE enables Shift Clock to the alternate output function line of P3.1. As data
bits come in from the right, “1”s shift out to the left. When the “0” that was initially loaded into the
right-most position arrives at the left-most position in the shift register, it flags the RX Control
block to do one last shift and load SBUF. Then RECEIVE is cleared and RI is set.
The relationship between the shift clock and data is determined by the combination of the SM2
and SMOD1 bits as listed in Table 16-5 and shown in Figure 16-2. The SM2 bit determines the
idle state of the clock when not currently transmitting/receiving. The SMOD1 bit determines if the
output data is stable for both edges of the clock, or just one.
Table 16-5.
88
Mode 0 Clock and Data Modes
SM2
SMOD1
Clock Idle
Data Changed
Data Sampled
0
0
High
While clock is high
Positive edge of clock
0
1
High
Negative edge of clock
Positive edge of clock
1
0
Low
While clock is low
Negative edge of clock
1
1
Low
Negative edge of clock
Positive edge of clock
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Figure 16-1. Serial Port Mode 0
INTERNAL BUS
TIMER 1
OVERFLOW
1
“1“
f osc
0
TB8
÷2
÷2
0
1
SMOD1
SM2
INTERNAL BUS
WRITE TO SBUF
SEND
SHIFT
RXD (DATA OUT)
TXD (SHIFT CLOCK)
TI
WRITE TO SCON (CLEAR RI)
RI
RECEIVE
SHIFT
RXD (DATA IN)
TXD (SHIFT CLOCK)
89
3706A–MICRO–9/09
Figure 16-2. Mode 0 Waveforms
SMOD1 = 0
SM2 = 0
TXD
RXD (TX)
0
0
RXD (RX)
SMOD1 = 1
SM2 = 0
1
0
2
1
0
RXD (RX)
4
3
5
4
6
5
7
6
7
2
1
3
2
4
3
5
4
6
5
7
6
7
TXD
RXD (TX)
0
1
0
RXD (RX)
SMOD1 = 1
SM2 = 1
1
3
TXD
RXD (TX)
SMOD1 = 0
SM2 = 1
2
2
1
3
2
4
3
5
4
6
5
7
6
7
TXD
0
RXD (TX)
1
0
RXD (RX)
2
1
3
2
4
3
5
4
6
5
6
7
7
Mode 0 may be used as a hardware accelerator for software emulation of serial interfaces such
as a half-duplex Serial Peripheral Interface (SPI) in mode (0,0) or (1,1) or a Two-Wire Interface
(TWI) in master mode. An example of Mode 0 emulating a TWI master device is shown in Figure
16-3. In this example, the start, stop, and acknowledge are handled in software while the byte
transmission is done in hardware. Falling/rising edges on TXD are created by setting/clearing
SM2. Rising/falling edges on RXD are forced by setting/clearing the P3.0 register bit. SM2 and
P3.0 must be 1 while the byte is being transferred.
Figure 16-3. UART Mode 0 TWI Emulation (SMOD1 = 1)
(SCL) TXD
(SDA) RXD
0
1
2
3
4
5
6
7
ACK
SM2
P3.0
Sample ACK
Write to SBUF
TI
Mode 0 transfers data LSB first whereas SPI or TWI are generally MSB first. Emulation of these
interfaces may require bit reversal of the transferred data bytes. The following code example
reverses the bits in the accumulator:
EX:
MOV
REVRS: RLC
XCH
RRC
XCH
DJNZ
90
R7, #8
A
A, R6
A
A, R6
R7, REVRS
; C << msb(ACC)
; msb(ACC) >> B
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
16.4
More About Mode 1
Ten bits are transmitted (through TXD), or received (through RXD): a start bit (0), 8 data bits
(LSB first), and a stop bit (1). On receive, the stop bit goes into RB8 in SCON. In the
AT89LP6440, the baud rate is determined either by the Timer 1 overflow rate, the TImer 2 overflow rate, or both. In this case one timer is for transmit and the other is for receive. Figure 16-4
shows a simplified functional diagram of the serial port in Mode 1 and associated timings for
transmit and receive.
Transmission is initiated by any instruction that uses SBUF as a destination register. The “write
to SBUF” signal also loads a “1” into the 9th bit position of the transmit shift register and flags the
TX Control unit that a transmission is requested. Transmission actually commences at S1P1 of
the machine cycle following the next rollover in the divide-by-16 counter. Thus, the bit times are
synchronized to the divide-by-16 counter, not to the “write to SBUF” signal.
The transmission begins when SEND is activated, which puts the start bit at TXD. One bit time
later, DATA is activated, which enables the output bit of the transmit shift register to TXD. The
first shift pulse occurs one bit time after that.
As data bits shift out to the right, “0”s are clocked in from the left. When the MSB of the data byte
is at the output position of the shift register, the “1” that was initially loaded into the 9th position is
just to the left of the MSB, and all positions to the left of that contain “0”s. This condition flags the
TX Control unit to do one last shift, then deactivate SEND and set TI. This occurs at the tenth
divide-by-16 rollover after “write to SBUF.”
Reception is initiated by a 1-to-0 transition detected at RXD. For this purpose, RXD is sampled
at a rate of 16 times the established baud rate. When a transition is detected, the divide-by-16
counter is immediately reset, and 1FFH is written into the input shift register. Resetting the
divide-by-16 counter aligns its roll-overs with the boundaries of the incoming bit times.
The 16 states of the counter divide each bit time into 16ths. At the 7th, 8th, and 9th counter
states of each bit time, the bit detector samples the value of RXD. The value accepted is the
value that was seen in at least 2 of the 3 samples. This is done to reject noise. In order to reject
false bits, if the value accepted during the first bit time is not 0, the receive circuits are reset and
the unit continues looking for another 1-to-0 transition. If the start bit is valid, it is shifted into the
input shift register, and reception of the rest of the frame proceeds.
As data bits come in from the right, “1”s shift out to the left. When the start bit arrives at the leftmost position in the shift register, (which is a 9-bit register in Mode 1), it flags the RX Control
block to do one last shift, load SBUF and RB8, and set RI. The signal to load SBUF and RB8
and to set RI is generated if, and only if, the following conditions are met at the time the final shift
pulse is generated.
RI = 0 and
Either SM2 = 0, or the received stop bit = 1
If either of these two conditions is not met, the received frame is irretrievably lost. If both conditions are met, the stop bit goes into RB8, the 8 data bits go into SBUF, and RI is activated. At
this time, whether or not the above conditions are met, the unit continues looking for a 1-to-0
transition in RXD.
91
3706A–MICRO–9/09
Figure 16-4. Serial Port Mode 1
TIMER 2
OVERFLOW
TIMER 1
OVERFLOW
INTERNAL BUS
“1”
WRITE
TO
SBUF
÷2
“0”
S
D Q
CL
“1”
SBUF
SMOD1
TXD
ZERO DETECTOR
“0”
“1”
SHIFT DATA
START
TX CONTROL
TCLK
÷16
SEND
TI
SERIAL
PORT
INTERRUPT
“1”
“0”
RX CLOCK
RCLK
÷16
SAMPLE
1-TO-0
TRANSITION
DETECTOR
RX CLOCK RI
START
RX CONTROL
LOAD
SBUF
SHIFT
1FFH
BIT
DETECTOR
INPUT SHIFT REG.
(9 BITS)
RXD
SHIFT
LOAD
SBUF
SBUF
READ
SBUF
INTERNAL BUS
TRANSMIT
TX
CLOCK
WRITE TO SBUF
SEND
DATA
SHIFT
D0
TXD
TI
RX
CLOCK
RECEIVE
D1
D2
D3
D4
D5
D6
D7
STOP BIT
START BIT
RXD
÷16 RESET
START BIT
D0
D1
D2
D3
D4
D5
D6
D7
STOP BIT
BIT DETECTOR SAMPLE TIMES
SHIFT
RI
92
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
16.5
More About Modes 2 and 3
Eleven bits are transmitted (through TXD), or received (through RXD): a start bit (0), 8 data bits
(LSB first), a programmable 9th data bit, and a stop bit (1). On transmit, the 9th data bit (TB8)
can be assigned the value of “0” or “1”. On receive, the 9th data bit goes into RB8 in SCON. The
baud rate is programmable to either 1/16 or 1/32 of the oscillator frequency in Mode 2. Mode 3
may have a variable baud rate generated from either Timer 1 or Timer 2, depending on the state
of RCLK and TCLK.
Figures 16-5 and 16-6 show a functional diagram of the serial port in Modes 2 and 3. The
receive portion is exactly the same as in Mode 1. The transmit portion differs from Mode 1 only
in the 9th bit of the transmit shift register.
Transmission is initiated by any instruction that uses SBUF as a destination register. The “write
to SBUF” signal also loads TB8 into the 9th bit position of the transmit shift register and flags the
TX Control unit that a transmission is requested. Transmission commences at S1P1 of the
machine cycle following the next rollover in the divide-by-16 counter. Thus, the bit times are synchronized to the divide-by-16 counter, not to the “write to SBUF” signal.
The transmission begins when SEND is activated, which puts the start bit at TXD. One bit time
later, DATA is activated, which enables the output bit of the transmit shift register to TXD. The
first shift pulse occurs one bit time after that. The first shift clocks a “1” (the stop bit) into the 9th
bit position of the shift register. Thereafter, only “0”s are clocked in. Thus, as data bits shift out to
the right, “0”s are clocked in from the left. When TB8 is at the output position of the shift register,
then the stop bit is just to the left of TB8, and all positions to the left of that contain “0”s. This condition flags the TX Control unit to do one last shift, then deactivate SEND and set TI. This occurs
at the 11th divide-by-16 rollover after “write to SBUF.”
Reception is initiated by a 1-to-0 transition detected at RXD. For this purpose, RXD is sampled
at a rate of 16 times the established baud rate. When a transition is detected, the divide-by-16
counter is immediately reset, and 1FFH is written to the input shift register.
At the 7th, 8th and 9th counter states of each bit time, the bit detector samples the value of RXD.
The value accepted is the value that was seen in at least 2 of the 3 samples. If the value
accepted during the first bit time is not 0, the receive circuits are reset and the unit continues
looking for another 1-to-0 transition. If the start bit proves valid, it is shifted into the input shift
register, and reception of the rest of the frame proceeds.
As data bits come in from the right, “1”s shift out to the left. When the start bit arrives at the leftmost position in the shift register (which in Modes 2 and 3 is a 9-bit register), it flags the RX Control block to do one last shift, load SBUF and RB8, and set RI. The signal to load SBUF and RB8
and to set RI is generated if, and only if, the following conditions are met at the time the final shift
pulse is generated:
RI = 0, and
Either SM2 = 0 or the received 9th data bit = 1
If either of these conditions is not met, the received frame is irretrievably lost, and RI is not set. If
both conditions are met, the received 9th data bit goes into RB8, and the first 8 data bits go into
SBUF. One bit time later, whether the above conditions were met or not, the unit continues looking for a 1-to-0 transition at the RXD input.
Note that the value of the received stop bit is irrelevant to SBUF, RB8, or RI.
93
3706A–MICRO–9/09
Figure 16-5. Serial Port Mode 2
INTERNAL BUS
CPU CLOCK
SMOD1 1
SMOD1 0
INTERNAL BUS
94
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Figure 16-6. Serial Port Mode 3
TIMER 2
OVERFLOW
TIMER 1
OVERFLOW
INTERNAL BUS
TB8
WRITE
TO
SBUF
÷2
“0”
S
D Q
CL
“1”
SMOD1
SBUF
TXD
ZERO DETECTOR
“0”
“1”
TCLK
÷16
“0”
SHIFT DATA
START STOP BIT
TX CONTROL
RX CLOCK
SEND
TI
SERIAL
PORT
INTERRUPT
“1”
RCLK
÷16
SAMPLE
1-TO-0
TRANSITION
DETECTOR
RX CLOCK RI
START
RX CONTROL
LOAD
SBUF
SHIFT
1FFH
BIT
DETECTOR
INPUT SHIFT REG.
(9 BITS)
RXD
SHIFT
LOAD
SBUF
SBUF
READ
SBUF
INTERNAL BUS
TRANSMIT
TX
CLOCK
WRITE TO SBUF
SEND
DATA
SHIFT
D0
TXD
TI
D1
D2
D3
D4
D5
D6
D7
TB8
START BIT
STOP BIT
RECEIVE
STOP BIT GEN
RX
CLOCK
÷16 RESET
RXD
START BIT
BIT DETECTOR SAMPLE TIMES
D0
D1
D2
D3
D4
D5
D6
D7
RB8
STOP
BIT
SHIFT
RI
95
3706A–MICRO–9/09
16.6
Framing Error Detection
In addition to all of its usual modes, the UART can perform framing error detection by looking for
missing stop bits, and automatic address recognition. When used for framing error detect, the
UART looks for missing stop bits in the communication. A missing bit will set the FE bit in the
SCON register. The FE bit shares the SCON.7 bit with SM0 and the function of SCON.7 is determined by PCON.6 (SMOD0). If SMOD0 is set then SCON.7 functions as FE. SCON.7 functions
as SM0 when SMOD0 is cleared. When used as FE, SCON.7 can only be cleared by software.
The FE bit will be set by a framing error regardless of the state of SMOD0.
16.7
Automatic Address Recognition
Automatic Address Recognition is a feature which allows the UART to recognize certain
addresses in the serial bit stream by using hardware to make the comparisons. This feature
saves a great deal of software overhead by eliminating the need for the software to examine
every serial address which passes by the serial port. This feature is enabled by setting the SM2
bit in SCON for Modes 1, 2 or 3. In the 9-bit UART modes, Mode 2 and Mode 3, the Receive
Interrupt flag (RI) will be automatically set when the received byte contains either the “Given”
address or the “Broadcast” address. The 9-bit mode requires that the 9th information bit be a “1”
to indicate that the received information is an address and not data.
In Mode 1 (8-bit) the RI flag will be set if SM2 is enabled and the information received has a valid
stop bit following the 8th address bits and the information is either a Given or Broadcast
address.
Automatic Address Recognition is not available during Mode 0.
Using the Automatic Address Recognition feature allows a master to selectively communicate
with one or more slaves by invoking the given slave address or addresses. All of the slaves may
be contacted by using the Broadcast address. Two special Function Registers are used to
define the slave’s address, SADDR, and the address mask, SADEN. SADEN is used to define
which bits in the SADDR are to be used and which bits are “don’t care”. The SADEN mask can
be logically ANDed with the SADDR to create the “Given” address which the master will use for
addressing each of the slaves. Use of the Given address allows multiple slaves to be recognized
while excluding others. The following examples show the versatility of this scheme:
Slave 0
SADDR = 1100 0000
SADEN = 1111 1101
Given = 1100 00X0
Slave 1
SADDR = 1100 0000
SADEN = 1111 1110
Given = 1100 000X
In the previous example, SADDR is the same and the SADEN data is used to differentiate
between the two slaves. Slave 0 requires a “0” in bit 0 and it ignores bit 1. Slave 1 requires a “0”
in bit 1 and bit 0 is ignored. A unique address for slave 0 would be 1100 0010 since slave 1
requires a “0” in bit 1. A unique address for slave 1 would be 1100 0001 since a “1” in bit 0 will
exclude slave 0. Both slaves can be selected at the same time by an address which has bit 0 = 0
(for slave 0) and bit 1 = 0 (for slave 1). Thus, both could be addressed with 1100 0000.
96
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
In a more complex system, the following could be used to select slaves 1 and 2 while excluding
slave 0:
Slave 0
SADDR = 1100 0000
SADEN = 1111 1001
Given = 1100 0XX0
Slave 1
SADDR = 1110 0000
SADEN = 1111 1010
Given = 1110 0X0X
Slave 2
SADDR = 1110 0000
SADEN = 1111 1100
Given = 1110 00XX
In the above example, the differentiation among the 3 slaves is in the lower 3 address bits. Slave
0 requires that bit 0 = 0 and it can be uniquely addressed by 1110 0110. Slave 1 requires that
bit 1 = 0 and it can be uniquely addressed by 1110 and 0101. Slave 2 requires that bit 2 = 0 and
its unique address is 1110 0011. To select Slaves 0 and 1 and exclude Slave 2, use address
1110 0100, since it is necessary to make bit 2 = 1 to exclude slave 2.
The Broadcast Address for each slave is created by taking the logic OR of SADDR and SADEN.
Zeros in this result are trended as don’t cares. In most cases, interpreting the don’t cares as
ones, the broadcast address will be FF hexadecimal.
Upon reset SADDR (SFR address 0A9H) and SADEN (SFR address 0B9H) are loaded with
“0”s. This produces a given address of all “don’t cares” as well as a Broadcast address of all
“don’t cares”. This effectively disables the Automatic Addressing mode and allows the microcontroller to use standard 80C51-type UART drivers which do not make use of this feature.
17. Enhanced Serial Peripheral Interface
The Serial Peripheral Interface (SPI) allows high-speed synchronous data transfer between the
AT89LP6440 and peripheral devices or between multiple AT89LP6440 devices, including multiple masters and slaves on a single bus. The SPI includes the following features:
• Full-duplex, 3-wire or 4-wire Synchronous Data Transfer
• Master or Slave Operation
• Maximum Bit Frequency = fOSC/4
• LSB First or MSB First Data Transfer
• Four Programmable Bit Rates or Timer 1-based Baud Generation (Master Mode)
• End of Transmission Interrupt Flag
• Write Collision Flag Protection
• Double-buffered Receive and Transmit
• Transmit Buffer Empty Interrupt Flag
• Mode Fault (Master Collision) Detection and Interrupt
• Wake up from Idle Mode
A block diagram of the SPI is shown below in Figure 17-1.
97
3706A–MICRO–9/09
Figure 17-1. SPI Block Diagram
S
Oscillator
M
M
T1 OVF
0
MSB
TSCK
LSB
Read Data Buffer
Write Data Buffer
SPI Clock (Master)
SPR0
M
SPI Status Register
DORD
SPR0
SPR1
CPHA
CPOL
MSTR
DORD
SPI Control Register
8
SPI Interrupt
Request
SPE
8
TSCK
MSTR
SPE
ENH
SSIG
DISSO
TXE
MODF
SPE
MSTR
SS
P1.4
SPI Control
WCOL
SCK
1.7
S
Clock
Logic
Select
SPR1
S
8-bit Shift Register
Divider
÷4/÷8/÷32/÷64
SPIF
MOSI
P1.5
Pin Control Logic
1
MISO
P1.6
8
Internal
Data Bus
The interconnection between master and slave CPUs with SPI is shown in Figure 17-2. The four
pins in the interface are Master-In/Slave-Out (MISO), Master-Out/Slave-In (MOSI), Shift Clock
(SCK), and Slave Select (SS). The SCK pin is the clock output in master mode, but is the clock
input in slave mode. The MSTR bit in SPCR determines the directions of MISO and MOSI. Also
notice that MOSI connects to MOSI and MISO to MISO. By default SS/P1.4 is an input to both
master and slave devices.
In slave mode, SS must be driven low to select an individual device as a slave. When SS is held
low, the SPI is activated, and MISO becomes an output if configured so by the user. All other
pins are inputs. When SS is driven high, all pins are inputs, and the SPI is passive, which means
that it will not receive incoming data. Note that the SPI logic will be reset once the SS pin is
driven high. The SS pin is useful for packet/byte synchronization to keep the slave bit counter
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.The slave may ignore SS by setting its SSIG bit in SPSR. When SSIG = 1, the slave is
always enabled and operates in 3-wire mode. However, the slave output on MISO may still be
disabled by setting DISSO = 1.
98
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Figure 17-2. SPI Master-Slave Interconnection
Master
MSB
LSB
Slave
MSB
MISO
LSB
MISO
8-Bit Shift Register
8-Bit Shift Register
MOSI
MOSI
SSIG
SS
MODF
DISSO
VCC
SS
SSIG
GPIO
Clock
Generator
SCK
SCK
When the SPI is configured as a Master (MSTR in SPCR is set), the operation of the SS pin
depends on the setting of the Slave Select Ignore bit, SSIG. If SSIG = 1, the SS pin is a general
purpose output pin which does not affect the SPI system. Typically, the pin will be driving the SS
pin of an SPI Slave. If SSIG = 0, SS 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 SSIG = 0,
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 MODF Flag in SPSR is set, and if the SPI interrupt is enabled, the interrupt routine
will be executed.
Thus, when interrupt-driven SPI transmission is used in Master mode, and there exists a possibility that SS may be 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.
17.1
Master Operation
An SPI master device initiates all data transfers on the SPI bus. The AT89LP6440 is configured
for master operation by setting MSTR = 1 in SPCR. Writing to the SPI data register (SPDR)
while in master mode loads the transmit buffer. If the SPI shift register is empty, the byte in the
transmit buffer is moved to the shift register; the transmit buffer empty flag, TXE, is set; and a
transmission begins. The transfer may start after an initial delay, while the clock generator waits
for the next full bit slot of the specified baud rate. The master shifts the data out serially on the
MOSI line while providing the serial shift clock on SCK. When the transfer finishes, the SPIF flag
is set to “1” and an interrupt request is generated, if enabled. The data received from the
addressed SPI slave device is also transferred from the shift register to the receive buffer.
Therefore, the SPIF bit flags both the transmit-complete and receive-data-ready conditions. The
received data is accessed by reading SPDR.
While the TXE flag is set, the transmit buffer is empty. TXE can be cleared by software or by
writing to SPDR. Writing to SPDR will clear TXE and load the transmit buffer. The user may load
the buffer while the shift register is busy, i.e. before the current transfer completes. When the
current transfer completes, the queued byte in the transmit buffer is moved to the shift register
and the next transfer commences. TXE will generate an interrupt if the SPI interrupt is enabled
99
3706A–MICRO–9/09
and if the ENH bit in SPSR is set. For multi-byte transfers, TXE may be used to remove any
dead time between byte transmissions.
The SPI master can operate in two modes: multi-master mode and single-master mode. By
default, multi-master mode is active when SSIG = 0. In this mode, the SS input is used to disable a master device when another master is accessing the bus. When SS is driven low, the
master device becomes a slave by clearing its MSTR bit and a Mode Fault is generated by setting the MODF bit in SPSR. MODF will generate an interrupt if enabled. The MSTR bit must be
set in software before the device may become a master again. Single-master mode is enabled
by setting SSIG = 1. In this mode SS is ignored and the master is always active. SS may be
used as a general purpose I/O in this mode.
17.2
Slave Operation
When the AT89LP6440 is not configured for master operation, MSTR = 0, it will operate as an
SPI slave. In slave mode, bytes are shifted in through MOSI and out through MISO by a master
device controlling the serial clock on SCK. When a byte has been transferred, the SPIF flag is
set to “1” and an interrupt request is generated, if enabled. The data received from the
addressed master device is also transferred from the shift register to the receive buffer. The
received data is accessed by reading SPDR. A slave device cannot initiate transfers. Data to be
transferred to the master device must be preloaded by writing to SPDR. Writes to SPDR are
double-buffered. The transmit buffer is loaded first and if the shift register is empty, the contents
of the buffer will be transferred to the shift register.
While the TXE flag is set, the transmit buffer is empty. TXE can be cleared by software or by
writing to SPDR. Writing to SPDR will clear TXE and load the transmit buffer. The user may load
the buffer while the shift register is busy, i.e. before the current transfer completes. When the
current transfer completes, the queued byte in the transmit buffer is moved to the shift register
and waits for the master to initiate another transfer. TXE will generate an interrupt if the SPI
interrupt is enabled and if the ENH bit in SPSR is set.
The SPI slave can operate in two modes: 4-wire mode and 3-wire mode. By default, 4-wire
mode is active when SSIG = 0. In this mode, the SS input is used to enable/disable the slave
device when addressed by a master. When SS is driven low, the slave device is enabled and will
shift out data on MISO in response to the serial clock on SCK. While SS is high, the SPI slave
will remain sleeping with MISO inactive. Three-wire mode is enabled by setting SSIG = 1. In this
mode SS is ignored and the slave is always active. SS may be used as a general purpose I/O in
this mode.
The Disable Slave Output bit, DISSO in SPSR, may be used to disable the MISO line of a slave
device. DISSO can allow several slave devices to share MISO while operating in 3-wire mode. In
this case some protocol other than SS may be used to determine which slave is enabled.
17.3
Pin Configuration
When the SPI is enabled (SPE = 1), the data direction of the MOSI, MISO, SCK, and SS pins is
automatically overridden according to the MSTR bit as shown in Table 17-1. The user need not
reconfigure the pins when switching from master to slave or vice-versa. For more details on port
configuration, refer to “Port Configuration” on page 44.
100
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
.
Table 17-1.
Pin
SPI Pin Configuration and Behavior when SPE = 1
Mode
Master (MSTR = 1)
Slave (MSTR = 0)
Quasi-bidirectional
Output
Input (Internal Pull-up)
Push-Pull Output
Output
Input (Tristate)
Input-Only
No output (Tristated)
Input (Tristate)
Open-Drain Output
Output
Input (External Pull-up)
SCK
(1)
Input (Internal Pull-up)
Quasi-bidirectional
Output
Push-Pull Output
Output(2)
Input (Tristate)
Input-Only
No output (Tristated)
Input (Tristate)
MOSI
(1)
Open-Drain Output
Output
Input (External Pull-up)
Quasi-bidirectional
Input (Internal Pull-up)
Output (SS = 0)
Internal Pull-up (SS = 1 or DISSO = 1)
Push-Pull Output
Input (Tristate)
Output (SS = 0)
Tristated (SS = 1 or DISSO = 1)
Input-Only
Input (Tristate)
No output (Tristated)
Open-Drain Output
Input (External Pull-up)
Output (SS = 0)
External Pull-up (SS = 1 or DISSO = 1)
MISO
Notes:
1. In these modes MOSI is active only during transfers. MOSI will be pulled high between transfers to allow other masters to control the line.
2. In Push-Pull mode MOSI is active only during transfers, otherwise it is tristated to prevent line
contention. A weak external pull-up may be required to prevent MOSI from floating.
Table 17-2.
SPCR – SPI Control Register
SPCR Address = E9H
Reset Value = 0000 0000B
Not Bit Addressable
Bit
TSCK
SPE
DORD
MSTR
CPOL
CPHA
SPR1
SPR0
7
6
5
4
3
2
1
0
Symbol
Function
TSCK
SCK Clock Mode. When TSCK = 0, the SCK baud rate is based on the system clock, divided by the SPR1-0 ratio.When
TSCK = 1, the SCK baud rate is based on the Timer 1 overflow rate, divided by the SPR1-0 ratio.
SPE
SPI enable. SPI = 1 enables the SPI channel and connects SS, MOSI, MISO and SCK to pins P1.4, P1.5, P1.6, and P1.7.
SPI = 0 disables the SPI channel.
DORD
Data order. DORD = 1 selects LSB first data transmission. DORD = 0 selects MSB first data transmission.
MSTR
Master/slave select. MSTR = 1 selects Master SPI mode. MSTR = 0 selects slave SPI mode.
CPOL
Clock polarity. When CPOL = 1, SCK is high when idle. When CPOL = 0, SCK of the master device is low when not
transmitting. Please refer to figure on SPI clock phase and polarity control.
CPHA
Clock phase. The CPHA bit together with the CPOL bit controls the clock and data relationship between master and
slave. Please refer to figure on SPI clock phase and polarity control.
101
3706A–MICRO–9/09
Symbol
Function
SPI clock rate select. These two bits control the SCK rate of the device configured as master. SPR1 and SPR0 have no
effect on the slave. The relationship between SCK and the oscillator frequency, FOSC., is as follows:
SPR0
SPR1
Notes:
SPR1
SPR0
0
SCK (TSCK = 0)
SCK (TSCK = 1)
0
fOSC/4
fT1OVF/4
0
1
fOSC/8
fT1OVF/8
1
0
fOSC/32
fT1OVF/32
1
1
fOSC/64
fT1OVF/64
1. Set up the clock mode before enabling the SPI: set all bits needed in SPCR except the SPE bit, then set SPE.
2. Enable the master SPI prior to the slave device.
3. Slave echoes master on the next Tx if not loaded with new data.
Table 17-3.
SPDR – SPI Data Register
SPDR Address = EAH
Reset Value = 00H (after cold reset)
unchanged (after warm reset)
Not Bit Addressable
Bit
Table 17-4.
SPD7
SPD6
SPD5
SPD4
SPD3
SPD2
SPD1
SPD0
7
6
5
4
3
2
1
0
SPSR – SPI Status Register
SPSR Address = E8H
Reset Value = 0000 X000B
Not Bit Addressable
Bit
SPIF
WCOL
MODF
TXE
–
SSIG
DISSO
ENH
7
6
5
4
3
2
1
0
Symbol
Function
SPIF
SPI Transfer Complete Interrupt Flag. When a serial transfer is complete, the SPIF bit is set by hardware and an interrupt
is generated if ESP = 1. The SPIF bit may be cleared by software or by reading the SPI status register followed by
reading/writing the SPI data register.
WCOL
Write Collision Flag. The WCOL bit is set by hardware if SPDR is written while the transmit buffer is full. The ongoing
transfer is not affected. WCOL may be cleared by software or by reading the SPI status register followed by
reading/writing the SPI data register.
MODF
Mode Fault Flag. MODF is set by hardware when a master mode collision is detected (MSTR = 1, SSIG = 0 and SS = 0)
and an interrupt is generated if ESP = 1. MODF must be cleared by software.
TXE
Transmit Buffer Empty Flag. Set by hardware when the transmit buffer is loaded into the shift register, allowing a new byte
to be loaded. TXE must be cleared by software. When ENH = 1 and ESP = 1, TXE will generate an interrupt.
102
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
SSIG
Slave Select Ignore. If SSIG = 0, the SPI will only operate in slave mode if SS (P1.4) is pulled low. When SSIG = 1, the
SPI ignores SS in slave mode and is active whenever SPE (SPCR.6) is set. When MSTR = 1 and SSIG = 0, SS is
monitored for master mode collisions. Setting SSIG = 1 will ignore collisions on SS. P1.4 may be used as a regular I/O
pin when SSIG = 1.
DISSO
Disable slave output bit. When set, this bit causes the MISO pin to be tristated so that more than one slave device can
share the same interface without multiple SS lines. Normally, the first byte in a transmission could be the slave address
and only the selected slave should clear its DISSO bit.
ENH
TX Buffer Interrupt Enable. When ENH = 1, TXE will generate an SPI interrupt if ESP = 1. When ENH = 0, TXE does not
generate an interrupt.
17.4
Serial Clock Timing
The CPHA, CPOL and SPR bits in SPCR control the shape and rate of SCK. The two SPR bits
provide four possible clock rates when the SPI is in master mode. In slave mode, the SPI will
operate at the rate of the incoming SCK as long as it does not exceed the maximum bit rate.
There are also four possible combinations of SCK phase and polarity with respect to the serial
data. CPHA and CPOL determine which format is used for transmission. The SPI data transfer
formats are shown in Figures 17-3 and 17-4. To prevent glitches on SCK from disrupting the
interface, CPHA, CPOL, and SPR should not be modified while the interface is enabled, and the
master device should be enabled before the slave device(s).
Figure 17-3. SPI Transfer Format with CPHA = 0
Note:
*Not defined but normally MSB of character just received.
Figure 17-4. SPI Transfer Format with CPHA = 1
SCK CYCLE #
(FOR REFERENCE)
1
2
3
4
5
6
7
8
SCK (CPOL = 0)
SCK (CPOL = 1)
MOSI
(FROM MASTER)
MISO
(FROM SLAVE)
*
MSB
6
5
4
3
2
1
MSB
6
5
4
3
2
1
LSB
LSB
SS (TO SLAVE)
Note:
*Not defined but normally LSB of previously transmitted character.
103
3706A–MICRO–9/09
18. Two-Wire Serial Interface
The Two-Wire Interface (TWI) is a bi-directional 2-wire serial communication standard. It is
designed primarily for simple but efficient integrated circuit (IC) control. The system is comprised
of two lines, SCL (Serial Clock) and SDA (Serial Data) that carry information between the ICs
connected to them. 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. The
serial data transfer is limited to 400Kbit/s in standard mode. Various communication configurations can be designed using this bus. Figure 18-1 shows a typical 2-wire bus configuration. Any
of the devices connected to the bus can be master or slave.
The Two-Wire Interface on the AT89LP provides the following features:
• Simple Yet Powerful and Flexible Communication Interface, only two Bus Lines Needed
• Both Master and Slave Operation Supported
• Device can Operate as Transmitter or Receiver
• 7-bit Address Space Allows up to 128 Different Slave Addresses
• Multi-master Arbitration Support
• Up to 400 kHz Data Transfer Speed
• Fully Programmable Slave Address with General Call Support
Figure 18-1. Two-Wire Bus Configuration
VCC
Device 1
Device 2
Device 3
........
Device n
R1
R2
SDA
SCL
As depicted in Figure 18-1, 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 tristate their outputs, allowing the pull-up resistors to pull the line
high. Note that all AT89LP 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.
104
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
18.1
18.1.1
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 18-2. Data Validity
SDA
SCL
Data Stable
Data Stable
Data Change
18.1.2
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.
Figure 18-3. START, REPEATED START, and STOP Conditions
SDA
SCL
START
18.1.3
STOP START
REPEATED START
STOP
Address Packet Format
All address packets transmitted on the TWI bus are nine bits long, consisting of seven address
bits, one READ/WRITE control bit and an acknowledge bit. If the READ/WRITE bit is set, a read
operation is to be performed, otherwise a write operation should be 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 Mas105
3706A–MICRO–9/09
ter’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 18-4. Address Packet Format
Addr MSB
Addr LSB
R/W
ACK
7
8
9
SDA
SCL
1
2
START
18.1.4
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 18-5. Data Packet Format
Data MSB
Data LSB
ACK
8
9
Aggregate
SDA
SDA from
Transmitter
SDA from
Receiver
SCL from
Master
1
SLA+R/W
106
2
7
Data Byte
STOP, REPEATED
START, or Next
Data Byte
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
18.1.5
Combining Address and Data Packets Into a Transmission
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 18-6 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 18-6. 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
18.2
2
SLA+R/W
2
7
Data Byte
STOP
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.
107
3706A–MICRO–9/09
Figure 18-7. SCL Synchronization between Multiple Masters
TA low
TA high
SCL from
Master A
SCL from
Master B
SCL bus
Line
TBhigh
TBlow
Masters Start
Counting Low Period
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.
Figure 18-8. 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.
108
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
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.
18.3
Overview of the TWI Module
The TWI module is comprised of several submodules, as shown in Figure 18-9. All registers
drawn in a thick line are accessible through the AT89LP data bus.
Figure 18-9. 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)
Address Match Unit
Address Register
(TWAR)
Address Comparator
Bit Rate Generator
Prescaler
Bit Rate Register
(TWBR)
Ack
Control Unit
Status Register
(TWSR)
Control Register
(TWCR)
TWI Unit
SCL
State Machine and
Status Control
18.3.1
SCL and SDA Pins
These pins interface the AT89LP 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.
18.3.2
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). Slave operation does not depend on the
Bit Rate setting, 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
109
3706A–MICRO–9/09
average TWI bus clock period. The SCL frequency is generated according to the following
equation:
System Clock
SCL frequency = ---------------------------------------------16 × ( TWBR + 1 )
18.3.3
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.
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.
18.3.4
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 (GC) 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.
18.3.5
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 (TWIF) 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 TWIF 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 TWIF 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.
110
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
18.4
Register Overview
Table 18-1.
TWCR – Two-Wire Control Register
TWCR Address = AAH
Reset Value = X000 00XXB
Not Bit Addressable
Bit
–
TWEN
STA
STO
TWIF
AA
–
–
7
6
5
4
3
2
1
0
Symbol
Function
TWEN
Two-wire Serial Interface Enable. Set to enable the TWI. Clear to disable the TWI.
STA
Start Flag. Set to send a START condition on the bus. Must be cleared by software.
STO
Stop Flag. Set to send a STOP condition on the bus. Cleared automatically by hardware when the STOP occurs.
TWIF
Two-wire Interface Interrupt Flag. Set by hardware when the TWI requests an interrupt. TWIF must be cleared by
software. While TWIF is set, the SCL low period is stretched. Note that clearing this flag starts the operation of the TWI,
so all accesses to the other TWI registers (TWAR, TWSR and TWDR) must be complete before clearing this flag.
AA
Assert Acknowledge Flag. Clear in master and slave receiver modes, to force a not acknowledge (high level on SDA).
Clear to disable SLA or GCA recognition. Set to recognize SLA or GCA (if GC set) for entering slave receiver or
transmitter modes. Set in master and slave receiver modes, to force an acknowledge (low level on SDA). This bit has no
effect when in master transmitter mode. By clearing AA to zero, the device can be virtually disconnected from the Twowire Serial Bus temporarily. Address recognition can then be resumed by setting the AA bit to one again.
Table 18-2.
TWSR – Two-Wire Status Register
TWSR Address = ABH
Reset Value = 1111 1000B
Not Bit Addressable
TWS7
TWS6
TWS5
TWS4
TWS3
0
0
0
7
6
5
4
3
2
1
0
Bit
Symbol
Function
TWS7-0
Two-wire Interface Status. The current status code of the TWI logic and serial bus. See Table 18-6 through Table 18-10
for a description of the status codes. Note that the three least significant bits always read as zero. The Status code is
valid only while TWIF remains set.
Table 18-3.
TWAR – Two-Wire Address Register
TWAR Address = ACH
Reset Value = 1111 1110B
Not Bit Addressable
Bit
TWA6
TWA5
TWA4
TWA3
TWA2
TWA1
TWA0
GC
7
6
5
4
3
2
1
0
Symbol
Function
TWA6-0
Two-wire Interface Slave Address. The TWI will only respond to slave addresses that match this 7-bit address.
GC
General Call Enable. Set to enable General Call address (00h) recognition. Clear to disable General Call address
recognition.
111
3706A–MICRO–9/09
Table 18-4.
TWDR – Two-Wire Data Register
TWDR Address = ADH
Reset Value = 1111 1111B
Not Bit Addressable
TWD7
TWD6
TWD5
TWD4
TWD3
TWD2
TWD1
TWD0
7
6
5
4
3
2
1
0
Bit
Symbol
Function
TWD7-0
Two-wire Interface Data. Writes to TWDR queue the next address or data byte for transmission. Reads from TWDR
return the last address or data byte present on the bus. Writes/reads to/from TWDR must occur only while TWIF is set.
Writes to TWDR while TWIF = 0 are ignored. Reads from TWDR while TWIF = 0 may return random data.
Table 18-5.
TWBR – Two-Wire Bit Rate Register
TWBR Address = AEH
Reset Value = 0000 0000B
Not Bit Addressable
Bit
TWB7
TWB6
TWB5
TWB4
TWB3
TWB2
TWB1
TWB0
7
6
5
4
3
2
1
0
Symbol
Function
TWB7-0
Two-wire Interface Serial Bit Rate. TWBR is an 8-bit down counter that selects the division factor (÷1–256) for the bit rate
generator. The bit rate generator is a frequency divider which generates the SCL clock frequency from the system clock
in Master mode.
18.5
Using the TWI
The AT89LP 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 interruptbased, the application software is free to carry on other operations during a TWI byte transfer.
Note that the TWI Interrupt Enable (TWE) bit in IE2 together with the Global Interrupt Enable bit
in EA allow the application to decide whether or not assertion of the TWIF flag should generate
an interrupt request. If the TWE bit is cleared, the application must poll the TWIF flag in order to
detect actions on the TWI bus.
When the TWIF 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 18-10 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.
112
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Application
Action
Figure 18-10. 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
TWIF is written to zero and
STA is written to zero.
START
2. TWIF set.
Status code indicates
START condition sent
SLA+W
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
TWIF is written to zero.
A
4. TWIF 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
TWIF is written to zero.
A
6. TWIF set.
Status code indicates
data sent, ACK received
STOP
Indicates
TWIF 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
TWIF bit is cleared in the value written. The TWI will not start any operation as long as
the TWIF bit in TWCR is set. Immediately after the application has cleared TWIF, the
TWI will initiate transmission of the START condition.
2. When the START condition has been transmitted, the TWIF 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 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 TWIF bit is cleared in the value written. The TWI will not start any operation as long as the TWIF bit in TWCR is set.
Immediately after the application has cleared TWIF, the TWI will initiate transmission of
the address packet.
4. When the address packet has been transmitted, the TWIF 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 TWIF bit is cleared in the value written. The TWI will not start any operation as
113
3706A–MICRO–9/09
long as the TWIF bit in TWCR is set. Immediately after the application has cleared
TWIF, the TWI will initiate transmission of the data packet.
6. When the data packet has been transmitted, the TWIF 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 TWIF bit is cleared in the value written. The TWI will not start any
operation as long as the TWIF bit in TWCR is set. Immediately after the application has
cleared TWIF, the TWI will initiate transmission of the STOP condition. Note that TWIF
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 TWIF flag is
set. The SCL line is pulled low until TWIF is cleared.
• When the TWIF 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 TWIF bit should be cleared. The TWI
will then commence executing whatever operation was specified by the TWCR setting.
18.6
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
114
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
SLA: Slave Address
In Figure 18-11 to Figure 18-14, circles are used to indicate that the TWIF flag is set. The numbers in the circles show the status code held in TWSR. At these points, actions must be taken by
the application to continue or complete the TWI transfer. The TWI transfer is suspended until the
TWIF flag is cleared by software.
When the TWIF 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 18-6 to Table 18-9.
18.6.1
Master Transmitter Mode
In the Master Transmitter mode, a number of data bytes are transmitted to a Slave Receiver. 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.
A START condition is sent by writing the following value to TWCR:
TWCR
–
TWEN
STA
STO
TWIF
AA
–
–
Value
X
1
1
0
0
X
X
X
TWEN must be set to enable the Two-wire Serial Interface, STA must be written to one to transmit a START condition and TWIF must be cleared. 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 TWIF flag is set by hardware, and the status code in TWSR will be
08h (see Table 18-6). In order to enter MT mode, SLA+W must be transmitted. This is done by
writing SLA+W to TWDR. Thereafter the TWIF bit should be cleared to continue the transfer.
When SLA+W has been transmitted and an acknowledgment bit has been received, TWIF is set
again and a number of status codes in TWSR are possible. Possible status codes in Master
mode are 18h, 20h, or 38h. The appropriate action to be taken for each of these status codes is
detailed in Table 18-6.
After 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 TWIF is high. If not,
the access will be discarded and the previous value will be transmitted. After updating TWDR,
the TWIF bit should be cleared to continue the transfer. 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
–
TWEN
STA
STO
TWIF
AA
–
–
Value
X
1
0
1
0
X
X
X
A REPEATED START condition is generated by writing the following value to TWCR:
TWCR
–
TWEN
STA
STO
TWIF
AA
–
–
Value
X
1
1
0
0
X
X
X
After a repeated START condition (status 10h) 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.
115
3706A–MICRO–9/09
.
Table 18-6.
Status Codes for Master Transmitter Mode
Application Software Response
Status
Code
(TWSR)
Status of the Two-wire
Serial Bus and Two-wire
Serial Interface Hardware
0x08
A START condition has
been transmitted
10h
A repeated START
condition has been
transmitted
18h
20h
28h
30h
38h
116
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
To TWCR
To/from TWDR
STA
STO
TWIF
AA
Load SLA+W
0
0
1
X
SLA+W will be transmitted;
ACK or NOT ACK will be received
Load SLA+W
0
0
1
X
SLA+W will be transmitted;
ACK or NOT ACK will be received
Load SLA+R
0
0
1
X
SLA+R will be transmitted;
Logic will switch to Master Receiver mode
Load data byte
0
0
1
X
Data byte will be transmitted and ACK or NOT
ACK will be received
No action
1
0
1
X
Repeated START will be transmitted
No action
0
1
1
X
STOP condition will be transmitted andSTO flag
will be reset
No action
1
1
1
X
STOP condition followed by a START condition
will be transmitted and STO flag will be reset
Load data byte
0
0
1
X
Data byte will be transmitted and ACK or NOT
ACK will be received
No action
1
0
1
X
Repeated START will be transmitted
No action
0
1
1
X
STOP condition will be transmitted and STO
flag will be reset
No action
1
1
1
X
STOP condition followed by a START condition
will be transmitted and STO flag will be reset
Load data byte
0
0
1
X
Data byte will be transmitted and ACK or NOT
ACK will be received
No action
1
0
1
X
Repeated START will be transmitted
No action
0
1
1
X
STOP condition will be transmitted and STO
flag will be reset
No action
1
1
1
X
STOP condition followed by a START condition
will be transmitted and STO flag will be reset
Load data byte
0
0
1
X
Data byte will be transmitted and ACK or NOT
ACK will be received
No actio
1
0
1
X
Repeated START will be transmitted
No actio
0
1
1
X
STOP condition will be transmitted and STO
flag will be reset
No actio
1
1
1
X
STOP condition followed by a START condition
will be transmitted and STO flag will be reset
No action
0
0
1
X
Two-wire Serial Bus will be released and not
addressed slave mode entered
No action
1
0
1
X
A START condition will be transmitted when the
bus becomes free
Arbitration lost in SLA+W
or data bytes
Next Action Taken by TWI Hardware
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Figure 18-11. Format and States in Master Transmitter Mode
MT
Successfull
transmission
to a slave
receiver
S
SLA
08h
W
A
DATA
18h
A
P
28h
Next transfer
started with a
repeated start
condition
RS
SLA
W
10h
Not acknowledge
received after the
slave address
A
R
P
20h
MR
Not acknowledge
received after a data
byte
A
P
30h
Arbitration lost in slave
address or data byte
A or A
Other master
continues
38h
Arbitration lost and
addressed as slave
A
68h
From master to slave
From slave to master
18.6.2
A or A
Other master
continues
38h
Other master
continues
78h
DATA
To corresponding
states in slave mode
B0h
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
Master Receiver Mode
In the Master Receiver mode, a number of data bytes are received from a slave transmitter. 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.
117
3706A–MICRO–9/09
TWEN must be written to one to enable the Two-wire Serial Interface, STA must be written to
one to transmit a START condition and TWIF must be cleared. The TWI will then test the Twowire Serial Bus and generate a START condition as soon as the bus becomes free. After a
START condition has been transmitted, the TWIF flag is set by hardware, and the status code in
TWSR will be 08h (see Table 18-7). In order to enter MR mode, SLA+R must be transmitted.
This is done by writing SLA+R to TWDR. Thereafter the TWIF bit should be cleared to continue
the transfer.
When SLA+R has been transmitted and an acknowledgment bit has been received, TWIF is set
again and a number of status codes in TWSR are possible. Possible status codes in Master
mode are 38h, 40h or 48h. The appropriate action to be taken for each of these status codes is
detailed in Table 18-7. Received data can be read from the TWDR Register when the TWIF 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.
Table 18-7.
Status Codes for Master Receiver Mode
Application Software Response
Status
Code
(TWSR)
Status of the Two-wire
Serial Bus and Two-wire
Serial Interface Hardware
08h
A START condition has
been transmitted
10h
A repeated START
condition has been
transmitted
38h
40h
48h
50h
58h
118
To TWCR
To/from TWDR
STA
STO
TWIF
AA
Load SLA+R
0
0
1
X
SLA+R will be transmitted; ACK or NOT ACK
will be received
Load SLA+R
0
0
1
X
SLA+R will be transmitted; ACK or NOT ACK
will be received
Load SLA+W
0
0
1
X
SLA+W will be transmitted; Logic will switch to
Master Transmitter mode
No action
0
0
1
X
Two-wire Serial Bus will be released and not
addressed Slave mode will be entered
No action
1
0
1
X
A START condition will be transmitted when the
bus becomes free
No action
0
0
1
0
Data byte will be received and NOT ACK will be
returned
No action
0
0
1
1
Data byte will be received and ACK will be
returned
No action
1
0
1
X
Repeated START will be transmitted
No action
0
1
1
X
STOP condition will be transmitted and STO
flag will be reset
No action
1
1
1
X
STOP condition followed by a START condition
will be transmitted and STO flag will be reset
Read data byte
0
0
1
0
Data byte will be received and NOT ACK will be
returned
Read data byte
0
0
1
1
Data byte will be received and ACK will be
returned
Read data byte
1
0
1
X
Repeated START will be transmitted
Read data byte
0
1
1
X
STOP condition will be transmitted and STO
flag will be reset
Read data byte
1
1
1
X
STOP condition followed by a START condition
will be transmitted and STO flag will be reset
Arbitration lost in SLA+R or
NOT ACK bit
SLA+R has been
transmitted; ACK has been
received
SLA+R has been
transmitted; NOT ACK has
been received
Data byte has been
received; ACK has been
returned
Data byte has been
received; NOT ACK has
been returned
Next Action Taken by TWI Hardware
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Figure 18-12. Format and States in Master Receiver Mode
MR
Successfull
reception
from a slave
receiver
S
SLA
R
A
DATA
A
40h
08h
DATA
50h
A
P
58h
Next transfer
started with a
repeated start
condition
RS
SLA
R
10h
Not acknowledge
received after the
slave address
A
W
P
48h
MT
Arbitration lost in slave
address or data byte
A or A
Other master
continues
A
38h
Arbitration lost and
addressed as slave
A
68h
From master to slave
18.6.3
38h
Other master
continues
78h
To corresponding
states in slave mode
B0h
DATA
From slave to master
Other master
continues
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
Slave Receiver Mode
In the Slave Receiver mode, a number of data bytes are received from a master transmitter. To
initiate the Slave Receiver mode, TWAR and TWCR must be initialized as follows:
TWAR
TWA6
TWA5
Value
TWA4
TWA3
TWA2
TWA1
TWA0
Device’s own Slave Address
GC
X
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 (00h),
otherwise it will ignore the general call address.:
TWCR
–
TWEN
STA
STO
TWIF
AA
–
–
Value
X
1
0
0
0
1
X
X
119
3706A–MICRO–9/09
TWEN must be written to one to enable the TWI. The AA bit must be written to one to enable the
acknowledgment of the device’s own slave address or the general call address. STA and STO
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 TWIF 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 188. The Slave Receiver mode may also be entered if arbitration is lost while the TWI is in the Master mode (see states 68h and 78h).
If the AA 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 AA 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 AA. This implies that the AA bit may be used to temporarily isolate the TWI from the Twowire Serial Bus.
.
Table 18-8.
Status Codes for Slave Receiver Mode
Application Software Response
Status
Code
(TWSR)
60h
68h
70h
78h
80h
120
Status of the Two-wire
Serial Bus and Two-wire
Serial Interface Hardware
To TWCR
STA
STO
TWIF
AA
No action
X
0
1
0
Data byte will be received and NOT ACK will be
returned
No action
X
0
1
1
Data byte will be received and ACK will be
returned
No action
X
0
1
0
Data byte will be received and NOT ACK will be
returned
No action
X
0
1
1
Data byte will be received and ACK will be
returned
General call address has
been received; ACK has
been returned
No action
X
0
1
0
Data byte will be received and NOT ACK will be
returned
No action
X
0
1
1
Data byte will be received and ACK will be
returned
Arbitration lost in SLA+R/W
as master; General call
address has been received;
ACK has been returned
No action
X
0
1
0
Data byte will be received and NOT ACK will be
returned
No action
X
0
1
1
Data byte will be received and ACK will be
returned
Read data byte
X
0
1
0
Data byte will be received and NOT ACK will be
returned
Read data byte
X
0
1
1
Data byte will be received and ACK will be
returned
Own SLA+W has been
received; ACK has been
returned
Arbitration lost in SLA+R/W
as master; own SLA+W has
been received; ACK has
been returned
Previously addressed with
own SLA+W; data has been
received; ACK has been
returned
To/from TWDR
Next Action Taken by TWI Hardware
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Table 18-8.
88h
90h
98h
A0h
Status Codes for Slave Receiver Mode
Previously addressed with
own SLA+W; data has been
received; NOT ACK has
been returned
Previously addressed with
general call; data has been
received; ACK has been
returned
Previously addressed with
general call; data has been
received; NOT ACK has
been returned
A STOP condition or
repeated START condition
has been received while still
addressed as slave
Read data byte
0
0
1
0
Switched to the not addressed Slave mode; no
recognition of own SLA or GCA
Read data byte
0
0
1
1
Switched to the not addressed Slave mode;
own SLA will be recognized; GCA will be
recognized if GC = “1”
0
Switched to the not addressed Slave mode; no
recognition of own SLA or GCA; a START
condition will be transmitted when the bus
becomes free
Read data byte
1
0
1
Read data byte
1
0
1
1
Switched to the not addressed Slave mode;
own SLA will be recognized; GCA will be
recognized if GC = “1”; a START condition will
be transmitted when the bus becomes free
Read data byte
X
0
1
0
Data byte will be received and NOT ACK will be
returned
Read data byte
X
0
1
1
Data byte will be received and ACK will be
returned
Read data byte
0
0
1
0
Switched to the not addressed Slave mode; no
recognition of own SLA or GCA
Read data byte
0
0
1
1
Switched to the not addressed Slave mode;
own SLA will be recognized; GCA will be
recognized if GC = “1”
0
Switched to the not addressed Slave mode; no
recognition of own SLA or GCA; a START
condition will be transmitted when the bus
becomes free
Read data byte
1
0
1
Read data byte
1
0
1
1
Switched to the not addressed Slave mode;
own SLA will be recognized; GCA will be
recognized if GC = “1”; a START condition will
be transmitted when the bus becomes free
No Action
0
0
1
0
Switched to the not addressed Slave mode; no
recognition of own SLA or GCA
No Action
0
0
1
1
Switched to the not addressed Slave mode;
own SLA will be recognized; GCA will be
recognized if GC = “1”
0
Switched to the not addressed Slave mode; no
recognition of own SLA or GCA; a START
condition will be transmitted when the bus
becomes free
1
Switched to the not addressed Slave mode;
own SLA will be recognized; GCA will be
recognized if GC = “1”; a START condition will
be transmitted when the bus becomes free
No Action
No Action
1
1
0
0
1
1
121
3706A–MICRO–9/09
Figure 18-13. Format and States in Slave Receiver Mode
Reception of the own
slave address and one or
more data bytes. All are
acknowledged
S
SLA
W
A
DATA
60h
A
DATA
80h
Last data byte received
is not acknowledged
A
P or S
80h
A0h
A
P or S
88h
Arbitration lost as master
and addressed as slave
A
68h
Reception of the general call
address and one or more data
bytes
General Call
A
DATA
70h
A
DATA
90h
Last data byte received is
not acknowledged
A
P or S
90h
A0h
A
P or S
98h
Arbitration lost as master and
addressed as slave by general call
A
78h
From master to slave
From slave to master
122
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
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
18.6.4
Slave Transmitter Mode
In the Slave Transmitter mode, a number of data bytes are transmitted to a master receiver. To
initiate the Slave Transmitter mode, upper 7 bits of TWAR must be initialized with 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 (00h), otherwise it will ignore the general call
address. TWEN must be written to one to enable the TWI. The AA bit must be written to one to
enable the acknowledgment of the device’s own slave address or the general call address. STA
and STO 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 18-9.
The Slave Transmitter mode may also be entered if arbitration is lost while the TWI is in the
Master mode (see state B0h).
If the AA bit is written to zero during a transfer, the TWI will transmit the last byte of the transfer.
State C0h or state C8h 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 “1s” as
serial data. State C8h is entered if the master demands additional data bytes (by transmitting
ACK), even though the slave has transmitted the last byte (AA zero and expecting NACK from
the master). While AA 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 AA. This implies that the AA bit may be used to temporarily isolate the TWI from the Twowire Serial Bus.
Figure 18-14. Format and States in Slave Transmitter Mode
Reception of the own
slave address and one or
more data bytes
S
SLA
R
A
DATA
A8h
Arbitration lost as master
and addressed as slave
A
DATA
B8h
A
P or S
C0h
A
B0h
Last data byte transmitted.
Switched to not addressed
slave (TWEA = '0')
A
All 1's
P or S
C8h
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
123
3706A–MICRO–9/09
.
Table 18-9.
Status Codes for Slave Transmitter Mode
Application Software Response
Status
Code
(TWSR)
A8h
B0h
B8h
C0h
C8h
Status of the Two-wire
Serial Bus and Two-wire
Serial Interface Hardware
To TWCR
To/from TWDR
STA
STO
TWIF
AA
Own SLA+R has been
received; ACK has been
returned
Load data byte
X
0
1
0
Last data byte will be transmitted and NOT
ACK should be received
Load data byte
X
0
1
1
Data byte will be transmitted and ACK should
be received
Arbitration lost in SLA+R/W
as master; own SLA+R has
been received; ACK has
been returned
Load data byte
X
0
1
0
Last data byte will be transmitted and NOT
ACK should be received
Load data byte
X
0
1
1
Data byte will be transmitted and ACK should
be received
Data byte in TWDR has
been transmitted; ACK has
been received
Load data byte
X
0
1
0
Last data byte will be transmitted and NOT
ACK should be received
Load data byte
X
0
1
1
Data byte will be transmitted and ACK should
be received
No action
0
0
1
0
Switched to the not addressed Slave mode; no
recognition of own SLA or GCA
No action
0
0
1
1
Switched to the not addressed Slave mode;
own SLA will be recognized; GCA will be
recognized if GC = “1”
0
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
Data byte in TWDR has
been transmitted; NOT ACK
has been received
Last data byte in TWDR has
been transmitted (AA = “0”);
ACK has been received
No action
0
1
No action
1
0
1
1
Switched to the not addressed Slave mode;
own SLA will be recognized; GCA will be
recognized if GC = “1”; a START condition will
be transmitted when the bus becomes free
No action
0
0
1
0
Switched to the not addressed Slave mode; no
recognition of own SLA or GCA
No action
0
0
1
1
Switched to the not addressed Slave mode;
own SLA will be recognized; GCA will be
recognized if GC = “1”
0
Switched to the not addressed Slave mode; no
recognition of own SLA or GCA; a START
condition will be transmitted when the bus
becomes free
1
Switched to the not addressed Slave mode;
own SLA will be recognized; GCA will be
recognized if GC = “1”; a START condition will
be transmitted when the bus becomes free
No action
No action
124
1
Next Action Taken by TWI Hardware
1
1
0
0
1
1
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
18.6.5
Miscellaneous States
There are two status codes that do not correspond to a defined TWI state, see Table 18-10.
Status F8h indicates that no relevant information is available because the TWIF flag is not set.
This occurs between other states, and when the TWI is not involved in a serial transfer.
Status 00h 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, TWIF is set. To recover from a bus error, the
STO flag must set and TWIF must be cleared. This causes the TWI to enter the not addressed
Slave mode and to clear the STO flag (no other bits in TWCR are affected). The SDA and SCL
lines are released, and no STOP condition is transmitted.
Table 18-10. Miscellaneous States
Application Software Response
Status
Code
(TWSR)
Status of the Two-wire
Serial Bus and Two-wire
Serial Interface hardware
To TWCR
To/from TWDR
F8h
No relevant state
information available;
TWIF = “0”
No action
00h
Bus error due to an illegal
START or STOP condition
No action
18.6.6
STA
STO
TWIF
AA
No 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 STO is cleared.
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 multi-master 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.
125
3706A–MICRO–9/09
Figure 18-15. Combining Several TWI Modes to Access a Serial EEPROM
Master Transmitter
S
SLA+W
A
ADDRESS
Master Receiver
A
S = START
Rs
SLA+R
A
DATA
Rs = REPEATED START
Transmitted from master to slave
A
P
P = STOP
Transmitted from slave to master
19. Dual Analog Comparators
The AT89LP6440 provides two analog comparators. The analog comparators have the following
features:
• Internal 3-level Voltage Reference (1.125V, 1.25V, 1.375V)
• Four Shared Analog Input Channels
– Configure as Multiple Input Window Comparator
• Selectable Interrupt Conditions
– High- or Low-level
– Rising- or Falling-edge
– Output Toggle
• Hardware Debouncing Modes
Figure 19-1. Dual Comparator Block Diagram
11
(P2.7) AIN3
10
01
00
B
11
CSB1
CSB0
(P2.6) AIN2
10
01
00
CMB2
CMB1
CSA0
CSA1
11
(P2.4) AIN0
Interrupt
CMA2
CMA1
00
01
10
00
01
10
CFB
CMB0
RFB0
RFB1
RFA1
RFA0
(P2.5) AIN1
CMPB (P4.7)
CFA
EC
CMA0
A
CMPA (P4.6)
11
VAREF+Δ
VAREF
VAREF-Δ
A block diagram of the dual analog comparators with relevant connections is shown in Figure
19-1. Input options allow the comparators to function in a number of different configurations as
shown in Figure 19-4. Comparator operation is such that the output is a logic “1” when the positive input is greater than the negative input. Otherwise the output is a zero. Setting the CENA
(ACSRA.3) and CENB (ACSRB.3) bits enable Comparator A and B respectively. The user must
126
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
also set the CONA (ACSRA.5) or CONB (ACSRB.5) bits to connect the comparator inputs
before using a comparator. When a comparator is first enabled, the comparator output and interrupt flag are not guaranteed to be stable for 10 µs. The corresponding comparator interrupt
should not be enabled during that time, and the comparator interrupt flag must be cleared before
the interrupt is enabled in order to prevent an immediate interrupt service. Before enabling the
comparators, the analog inputs should be tristated by putting P2.4, P2.5, P2.6 and P2.7 into
input-only mode. See “Port Analog Functions” on page 47.
Each comparator may be configured to cause an interrupt under a variety of output value conditions by setting the CMx2-0 bits in ACSRx. The comparator interrupt flags CFx in ACSRx are set
whenever the comparator outputs match the conditions specified by CMx2-0. The flags may be
polled by software or may be used to generate an interrupt and must be cleared by software.
Both comparators share a common interrupt vector. If both comparators are enabled, the user
needs to read the flags after entering the interrupt service routine to determine which comparator caused the interrupt.
The CAC1-0 and CBC1-0 bits in AREF control when the comparator interrupts sample the comparator outputs. Normally the outputs are sampled every clock system; however, the outputs
may also be sampled whenever Timer 0, Timer 1 or Timer 2 overflows. These settings allow the
comparators to be sampled at a specific time or to reduce the number of comparator events
seen by the system when using level sensitive modes. The comparators will continue to function
during Idle mode. If this is not the desired behavior, the comparators should be disabled before
entering Idle. The comparators are always disabled during Power-down mode.
19.1
Analog Input Muxes
The positive input terminal of each comparator may be connected to any of the four analog input
pins by changing the CSA1-0 or CSB1-0 bits in ACSRA and ACSRB. When changing the analog
input pins, the comparator must be disconnected from its inputs by clearing the CONA or CONB
bits. The connection is restored by setting the bits again after the muxes have been modified.
CLR
EC
ANL
...
ORL
ACSRA, #0DFh ; Clear CONA to disconnect COMP A
; Modify CSA or RFA bits
ACSRA, #020h ; Set CONA to connect COMP A
ANL
ACSRA, #0EFh ; Clear any spurious interrupt
SETB
EC
; Disable comparator interrupts
; Re-enable comparator interrupts
The corresponding comparator interrupt should not be enabled while the inputs are being
changed, and the comparator interrupt flag must be cleared before the interrupt is re-enabled in
order to prevent an unintentional interrupt request.
The equivalent model for the analog input circuitry is illustrated in Figure 20-3. An analog source
applied to AINn is subjected to the pin capacitance and input leakage of that pin, regardless of
whether that channel is selected as input to the comparator. When the channel is selected, the
source must drive the input capacitance of the comparator through the series resistance (combined resistance in the input path).
127
3706A–MICRO–9/09
Figure 19-2. Equivalent Analog Input Model
RMUX =
10 kΩ
RIN =
10 kΩ
AINn
CPIN =
10 pF
19.2
CCMP <
0.3 pF
Internal Reference Voltage
The negative input terminal of each comparator may be connected to an internal voltage reference by changing the RFB1-0 or RFA1-0 bits in AREF. The internal reference voltage, VAREF, is
set to 1.25 V ±5%. The voltage reference also provides two additional voltage levels approximately 125 mV above and below VAREF. These levels may be used to configure the comparators
as an internally referenced window comparator with up to four input channels. Changing the reference input must follow the same routine used for changing the positive input as described in
“Analog Input Muxes” above.
19.3
Comparator Interrupt Debouncing
The comparator output is normally sampled every clock cycle. The conditions on the analog
inputs may be such that the comparator output will toggle excessively. This is especially true if
applying slow moving analog inputs. Three debouncing modes are provided to filter out this
noise for edge-triggered interrupts. In debouncing mode, the comparator uses Timer 1 to modulate its sampling time when CxC1-0 = 00B. When a relevant transition occurs, the comparator
waits until two Timer 1 overflows have occurred before resampling the output. If the new sample
agrees with the expected value, CFx is set. Otherwise, the event is ignored. The filter may be
tuned by adjusting the time-out period of Timer 1. Because Timer 1 is free running, the
debouncer must wait for two overflows to guarantee that the sampling delay is at least 1 time-out
period. Therefore, after the initial edge event, the interrupt may occur between 1 and 2 time-out
periods later. See Figure 19-3. When the comparator clock is provided by one of the timer overflows, i.e. CxC1-0 != 00B, any change in the comparator output must be valid after 4 samples to
be accepted as an edge event.
Figure 19-3. Negative Edge with Debouncing Example
Comparator Out
Timer 1 Overflow
CFx
Start
128
Compare
(rejected)
Start
Compare
(accepted)
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Figure 19-4. Dual Comparator Configuration Examples
a. dual independent comparators with external references
+
AIN0
-
AIN1
A
AIN3
+
AIN2
-
CMPA
B
f. 2-channel window
comparator with
external reference
CMPB
-
AIN2
CSA = 00 RFA = 00
CSB = 11 RFB = 00
+
B
CMPB
AIN0
b. 3-channel comparator with external reference
AIN3
AIN0
+
AIN2
AIN1
AIN3
-
+
A
CMPA
-
AIN1
CSA = 00/10/11 RFA = 00
A
CMPA
CSA = CSB = 00/11
RFA = RFB = 00
c. 4-channel comparator with internal reference
AIN0
+
AIN1
VAREF
AIN2
AIN3
-
A
CMPA
g. 4-channel window
comparator with
internal reference
CSA = 00/01/10/11 RFA = 10
d. 2-channel comparator with internal reference &
comparator with external reference
VAREF+Δ
AIN0
AIN0
+
AIN1
VAREF
-
A
CMPA
AIN3
AIN2
CSA = 00/01 RFA = 10
+
-
B
CMPB
+
+
AIN1
-
A
CMPA
CSB = 11 RFB = 00
CSA = 00/10 RFA = 00
AIN3
VAREF
CMPB
AIN2
e. 2-channel comparator with external reference &
comparator with internal reference
AIN0
B
AIN1
+
AIN3
VAREF-Δ
AIN2
-
-
A
CMPA
CSA = CSB = 00/01/10/11
RFA = 01 RFB = 11
+
-
B
CMPB
CSB = 11 RFB = 10
129
3706A–MICRO–9/09
Table 19-1.
ACSRA – Analog Comparator A Control & Status Register
ACSRA = 97H
Reset Value = 0000 0000B
Not Bit Addressable
Bit
CSA1
CSA0
CONA
CFA
CENA
CMA2
CMA1
CMA0
7
6
5
4
3
2
1
0
Symbol
Function
CSA [1-0]
Comparator A Positive Input Channel Select(1)
CSA1
CSA0
A+ Channel
0
0
AIN0 (P2.4)
0
1
AIN1 (P2.5)
1
0
AIN2 (P2.6)
1
1
AIN3 (P2.7)
CONA
Comparator A Input Connect. When CONA = 1 the analog input pins are connected to the comparator. When CONA = 0
the analog input pins are disconnected from the comparator. CONA must be cleared to 0 before changing CSA[1-0] or
RFA[1-0].
CFA
Comparator A Interrupt Flag. Set when the comparator output meets the conditions specified by the CMA [2-0] bits and
CENA is set. The flag must be cleared by software. The interrupt may be enabled/disabled by setting/clearing bit 6 of IE.
CENA
Comparator A Enable. Set this bit to enable the comparator. Clearing this bit will force the comparator output low and
prevent further events from setting CFA. When CENA = 1 the analog input pins, P2.4—P2.7, have their digital inputs
disabled if they are configured in input-only mode.
CMA [2-0]
Comparator A Interrupt Mode
Notes:
CMA2
CMA1
CMA0
Interrupt Mode
0
0
0
Negative (Low) level
0
0
1
Positive edge
0
1
0
Toggle with debouncing(2)
0
1
1
Positive edge with debouncing(2)
1
0
0
Negative edge
1
0
1
Toggle
1
1
0
Negative edge with debouncing(2)
1
1
1
Positive (High) level
1. CONA must be cleared to 0 before changing CSA[1-0].
2. Debouncing modes require the use of Timer 1 to generate the sampling delay.
130
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Table 19-2.
ACSRB – Analog Comparator B Control & Status Register
ACSRB = 9FH
Reset Value = 1100 0000B
Not Bit Addressable
Bit
CSB1
CSB0
CONB
CFB
CENB
CMB2
CMB1
CMB0
7
6
5
4
3
2
1
0
Symbol
Function
CSB [1-0]
Comparator B Positive Input Channel Select(1)
CSB1
CSB0
B+ Channel
0
0
AIN0 (P2.4)
0
1
AIN1 (P2.5)
1
0
AIN2 (P2.6)
1
1
AIN3 (P2.7)
CONB
Comparator B Input Connect. When CONB = 1 the analog input pins are connected to the comparator. When CONB = 0
the analog input pins are disconnected from the comparator. CONB must be cleared to 0 before changing CSB[1-0] or
RFB[1-0].
CFB
Comparator B Interrupt Flag. Set when the comparator output meets the conditions specified by the CMB [2-0] bits and
CENB is set. The flag must be cleared by software. The interrupt may be enabled/disabled by setting/clearing bit 6 of IE.
CENB
Comparator B Enable. Set this bit to enable the comparator. Clearing this bit will force the comparator output low and
prevent further events from setting CFB. When CENB = 1 the analog input pins, P2.4—P2.7, have their digital inputs
disabled if they are configured in input-only mode.
CMB [2-0]
Comparator B Interrupt Mode
Notes:
CMB2
CMB1
CMB0
Interrupt Mode
0
0
0
Negative (Low) level
0
0
1
Positive edge
0
1
0
Toggle with debouncing(2)
0
1
1
Positive edge with debouncing(2)
1
0
0
Negative edge
1
0
1
Toggle
1
1
0
Negative edge with debouncing(2)
1
1
1
Positive (High) level
1. CONB must be cleared to 0 before changing CSB[1-0].
2. Debouncing modes require the use of Timer 1 to generate the sampling delay.
131
3706A–MICRO–9/09
Table 19-3.
AREF – Analog Comparator Reference Control Register
AREF = AFH
Reset Value = 0000 0000B
Not Bit Addressable
Bit
CBC1
CBC0
RFB1
RFB0
CAC1
CAC0
RFA1
RFA0
7
6
5
4
3
2
1
0
Symbol
Function
CSC [1-0]
Comparator B Clock Select
RFB [1-0]
CAC [1-0]
RFA [1-0]
Notes:
CBC1
CBC0
Clock Source
0
0
System Clock
0
0
Timer 0 Overflow
0
1
Timer 1 Overflow
0
1
Timer 2 Overflow
Comparator B Negative Input Channel Select(1)
CRF1
RFB0
B- Channel
0
0
AIN2 (P2.6)
0
0
Internal VAREF-Δ (~1.125V)
0
1
Internal VAREF (~1.25V)
0
1
Internal VAREF+Δ (~1.375V)
Comparator A Clock Select
CAC1
CAC0
Clock Source
0
0
System Clock
0
0
Timer 0 Overflow
0
1
Timer 1 Overflow
0
1
Timer 2 Overflow
Comparator A Negative Input Channel Select(2)
RFA1
RFA0
A- Channel
0
0
AIN1 (P2.5)
0
0
Internal VAREF-Δ (~1.125V)
0
1
Internal VAREF (~1.25V)
0
1
Internal VAREF+Δ (~1.375V)
1. CONB (ACSRB.5) must be cleared to 0 before changing RFB[1-0].
2. CONA (ACSRA.5) must be cleared to 0 before changing RFA[1-0].
132
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
20. Digital-to-Analog/Analog-to-Digital Converter
The AT89LP6440 includes a 10-bit Data Converter (DADC) with the following features:
• Digital-to-Analog (DAC) or Analog-to-Digital (ADC) Mode
• 10-bit Resolution
• 6.5 µs Conversion Time
• 8 Multiplexed Single-ended Channels or 4 Differential Channels
• Selectable 1.0V±10% Internal Reference Voltage
• Optional Left-Adjust of Conversion Results
• Single Conversion or Timer-triggered Mode
• Interrupt on Conversion Complete
The AT89LP6440 features a 10-bit successive approximation data converter that functions in
either Analog-to-Digital (ADC) or Digital-to-Analog (DAC) mode. A block diagram of the converter is shown in Figure 20-1. An 8-channel Analog Multiplexer connects eight single-ended or
four differential voltage inputs from the pins of Port 0 to a sample-and-hold circuit that in turn
provides an input to the successive approximation block. The Sample-and-Hold circuit ensures
that the input voltage to the ADC is held at a constant level during conversion. The SAR block
digitizes the analog voltage into a 10-bit value accessible through a data register. The SAR
block also operates in reverse to generate an analog voltage on Port 2 from a 10-bit digital
value. The DADC has separate analog supply pins, AVDD and AGND, which must not differ more
than ± 0.3V from the standard supplies VDD and GND.
ADC results are available in the DADL and DADH register pair. The ADC result scale is determined by the reference voltage (VREF) generated either internally from a 1.0V reference or
externally from AVDD/2. The ADC results are always represented in signed 2’s complement
form, with single-ended voltage channels referring to the level above or below AVDD/2. The 10bit results may be right or left adjusted within the 16-bit register. The sign is extended through
the 6 MSBs of right-adjusted results and the 6 LSBs of left-adjusted results are zeroed. If only 8bit precision is required, the user should select left-adjusted by setting LADJ in DADC and read
only the DADH register. Example results are listed in Table 20-1.
The conversion formulas are as follows:
(Singled-Ended)
ADC
511 ×
(Differential)
ADC
511 ×
V IN
AV DD ⁄ 2
V REF
V IN+
V IN-
V REF
Conversion results can be converted into unsigned binary by adding 02h to DADH in rightadjusted mode or 80h to DADH in left-adjusted mode. When using the external reference
(AVDD/2) in single-ended mode this is equivalent to:
(Unsigned Singled-Ended)
ADC
1023 ×
V IN
AV DD
To convert the unsigned binary value back to 2’s complement, subtract 02h from DADH in rightadjusted mode or 80h from DADH in left-adjusted mode. Note that the DADH/DADL registers
cannot be directly manipulated as they are read-only in ADC mode and write-only in DAC mode.
133
3706A–MICRO–9/09
Table 20-1.
Example ADC Conversion Codes
Right Adjust
Left Adjust
Single-Ended Mode (VIN)
Differential Mode (VIN+ – VIN-)
0
0
AVDD/2
0
0100h
4000h
AVDD/2 + 1/2 x VREF
1/2 x VREF
01FFh
7FC0h
AVDD/2 + 511/512 x VREF
511/512 x VREF
FF00h
C000h
AVDD/2 – 1/2 x VREF
–1/2 x VREF
FE01h
8040h
AVDD/2 – 511/512 x VREF
–511/512 x VREF
Figure 20-1. DADC Block Diagram
INTERRUPT
FLAG
Timer
Overflows
8-BIT DATA BUS
15
ACK0
ACK1
ACK2
ADIF
LADJ
DAC
ADCE
GO
TRG0
ACS0
TRG1
ACS1
DIFF
ACS2
ACON
IREF
AVDD
0
ADC DATA REGISTER
(DADH/DADL)
ADC CTRL & STATUS
REGISTER (DADC)
ADC INPUT SELECT
REGISTER (DADI)
R
AVDD/2
R
CHANNEL SELECTION
PRESCALER
TRIGGER
SELECT
AGND
VREF
INTERNAL
1.0V
REFERENCE
START
ADC7
ADC6
DA+
ADC5
SAMPLE &
HOLD
ADC4
POS.
INPUT
MUX
ADC3
VIN+
+
10-BIT SAR
-
ADC2
ADC1
DA-
ADC0
NEG.
INPUT
MUX
VIN-
AVDD/2
134
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
20.1
ADC Operation
The ADC converts an analog input voltage to a 10-bit signed digital value through successive
approximation. When DIFF (DADI.3) is zero, the ADC operates in single-ended mode and the
input voltage is the difference between the voltage at the input pin and AVDD/2. In differential
mode (DIFF = 1) the input voltage is the difference between the positive and negative input pins.
The minimum value represents zero difference and the maximum values represent a difference
of positive or negative VREF minus 1 LSB.
The analog input channel is selected by writing to the ACS bits in DADI. Any of the eight Port 0
input pins can be selected as single-ended inputs to the ADC. Four pairs of Port 0 pins can be
selected as differential inputs. The ACON bit (DADI.7) must be set to one to connect the input
pins to the ADC. Prior to changing ACS, ACON must be cleared to zero. This ensures that
crosstalk between channels is limited. ACON must be set back to one after ACS is updated.
ACON and ACS should not be changed while a conversion is in progress. ADC input channels
must have their port pins configured for input-only mode.
The ADC is enabled by setting the ADCE bit in DADC. Some settling time is required for the reference circuits to stabilize after the ADC is enabled. The ADC does not consume power when
ADCE is cleared, so it is recommended to switch off the ADC before entering power saving
modes.
A timing diagram of an ADC conversion is shown in Figure 20-2. The conversion requires 13
ADC clock cycles to complete. The analog input is sampled during the third cycle of the conversion and is held constant for the remainder of the conversion. At the end of the conversion, the
interrupt flag, ADIF, is set and the result is written to the data registers. An additional 1 ADC
clock cycle and up to 2 system clock cycles may be required to synchronize ADIF with the rest of
the system. The results in DADH/DADL remain valid until the next conversion completes. DADH
and DADL are read-only registers during ADC mode.
Figure 20-2. ADC Timing Diagram
One Conversion
Cycle Number
1
2
3
4
5
6
7
8
Next Conversion
9
10
11
12
13
1
2
3
ADC Clock
GO/BSY
ADIF
DADH
MSB of Result
DADL
LSB of Result
Sample & Hold
Initialize Circuitry
Conversion
Complete
Initialize
The equivalent model for the analog input circuitry is illustrated in Figure 20-3. 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 to 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). To achieve 10-bit resolution the S/H capacitor must be charged to within 1/2 LSB of the
expected value within the 1 ADC clock period sample time. High impedance sources may
require a reduction in the ADC clock frequency to achieve full resolution.
135
3706A–MICRO–9/09
Figure 20-3. Equivalent Analog Input Model
RIN =
10 kΩ
RMUX =
10 kΩ
ADCn
CPIN =
10 pF
20.2
CS/H =
2 pF
DAC Operation
The DAC converts a 10-bit signed digital value to an analog output voltage through successive
approximation. The DAC always operates in differential mode, outputting a voltage differential
between its positive (P2.2) and negative (P2.3) outputs with a common mode of AVDD/2. The
minimum value represents zero difference and the maximum values represent a difference of
positive or negative VREF minus 1 LSB.
The DAC is enabled by setting the ADCE and DAC bits in DADC. Some settling time is required
for the reference circuits to stabilize after the DAC is enabled. The DAC does not have multiple
output channels and the DIFF, ACON and ACS bits have no effect in DAC mode. P2.2 and P2.3
are automatically forced to input-only mode while the DAC is enabled.
A timing diagram of a DAC conversion is shown in Figure 20-4. The conversion requires 11 ADC
clock cycles to complete. Construction of the analog output starts in the second cycle of the conversion and the DAC will allow the new value to propagate to the outputs during cycle 7, after the
5 MSBs are complete. At the end of the conversion, the interrupt flag is set. An additional 1 ADC
clock cycle and up to 2 system clock cycles may be required to synchronize ADIF with the rest of
the system. The DADL and DADH registers hold the value to be output and are write-only during
DAC mode. An internal buffer samples DADH/DADL at the start of the conversion and holds the
value constant for the remainder of the conversion. One system clock cycle is required to transfer the contents of DADH/DADL into the buffer at the start of the conversion and therefore the
ADC clock frequency must always be equal to or less than the system clock frequency during
DAC mode to ensure that the buffer is updated before the second cycle.
Figure 20-4. DAC Timing Diagram
One Conversion
Cycle Number
1
2
3
4
5
6
7
Next Conversion
8
9
10
11
1
2
3
ADC Clock
GO/BSY
ADIF
DADH
MSB of Output
DADL
LSB of Output
Begin Output
Initialize Circuitry
Conversion
Complete
Initialize
The equivalent model for the analog output circuitry is illustrated in Figure 20-5. The series output resistance of the DAC must drive the pin capacitance and any external load on the pin.
136
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Figure 20-5. Equivalent Analog Output Model
ROUT =
100 kΩ
DAn
CPIN =
10 pF
VOUT
AVDD/2
20.3
Clock Selection
The DADC requires a clock of 2 MHz or less to achieve full resolution. By default the DADC will
use an internal 2 MHz clock generated from the 8 MHz internal oscillator. The internal oscillator
will be enabled even if it is not supplying the system clock. This may result in higher power consumption. Conversely, the DADC clock can be generated directly from the system clock using a
7-bit prescaler. The prescaler output is controlled by the ACK bits in DADC as shown in Figure
20-6.
In ADC mode, there are no requirements on the clock frequency with respect to the system
clock. The ADC prescaler selection is independent of the system clock divider and the ADC may
operate at both higher or lower frequencies than the CPU. However, in DAC mode the ADC
clock frequency must not be higher than the CPU clock, including any clock division from the
system clock.
Figure 20-6. DADC Clock Selection
CK/128
CK/64
CK/32
CK/16
CK/8
÷4
CK/4
INTERNAL
8MHz OSC
CK/2
7-BIT ADC PRESCALER
CK
ACK0
ACK1
ACK2
ADC CLOCK SOURCE
20.4
Starting a Conversion
Setting the GO/BSY bit (DADC.6) when ADCE = 1 starts a single conversion in both ADC and
DAC modes. The bit remains set while the conversion is in progress and is cleared by hardware
when the conversion completes. The ADC channel should not be changed while a conversion is
in progress.
Alternatively, a conversion can be started automatically by various timer sources. Conversion
trigger sources are selected by the TRG bits in DADI. A conversion is started every time the
selected timer overflows, allowing for conversions to occur at fixed intervals. The GO/BSY bit will
137
3706A–MICRO–9/09
be set by hardware while the conversion is in progress. Note that the timer overflow rate must be
slower than the conversion time.
20.5
Noise Considerations
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:
• Connect the AVDD pin to the digital VDD supply voltage via an LC network as shown in Figure
20-7. Values should be chosen to fit the system. Note this example has a cutoff frequency of
~160kHz and works best for devices operate at 1 MHz or above.
• Keep analog signal paths as short as possible. Make sure to run analog signals tracks over
an analog ground plane, and keep them well away from high-speed digital tracks.
• Place the CPU in Idle during a conversion.
• If any Port 0 pins are used as digital outputs, it is essential that these do not switch while a
conversion is in progress.
P0.2 (ADC2)
P0.3 (ADC3)
38
P0.1 (ADC1)
39
AVDD
VDD
10
μH
P0.0 (ADC0)
Figure 20-7. Example ADC Power Connections (TQFP Package)
37
36
35
34
Analog Ground
Plane
100
nF
33 P0.4 (ADC4)
32 P0.5 (ADC5)
31 P0.6 (ADC6)
30 P0.7 (ADC7)
29
28
138
GND
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Table 20-2.
DADC – DADC Control Register
DADC = D9H
Reset Value = 0000 0000B
Not Bit Addressable
ADIF
GO/BSY
DAC
ADCE
LADJ
ACK2
ACK1
ACK0
7
6
5
4
3
2
1
0
Bit
Symbol
Function
ADIF
ADC Interrupt Flag. Set by hardware when a conversion completes. Cleared by hardware when calling the interrupt
service routine.
GO/BSY
Conversion Start/Busy Flag. In software triggered mode, writing a 1 to this bit starts a conversion. The bit remains high
while the conversion is in progress and is cleared by hardware when the conversion completes. In hardware triggered
mode, this bit is set and cleared by hardware to flag when the DADC is busy.
DAC
Digital-to-Analog Conversion Enable. Set to configure the DADC in Digital-to-Analog (DAC) mode. Clear to configure the
DADC in Analog-to-Digital (ADC) mode.
ADCE
DADC Enable. Set to enable the DADC. Clear to disable the DADC.
LADJ
Left Adjust Enable. When cleared, the ADC results are right adjusted and the MSBs are sign extended. When set, the
ADC results are left adjusted and the LSBs are zeroed.
ACK [2-0]
DADC Clock Select
Table 20-3.
ACK3
ACK1
ACK0
Clock Source
0
0
0
Internal RC Oscillator/4 (2MHz)
0
0
1
fsys/2
0
1
0
fsys/4
0
1
1
fsys/8
1
0
0
fsys/16
1
0
1
fsys/32
1
1
0
fsys/64
1
1
1
fsys/128
DADL – DADC Data Low Register
DADL = DCH
Reset Value = 0000 0000B
Not Bit Addressable
ADC.7
ADC.6
ADC.5
ADC.4
ADC.3
ADC.2
ADC.1
ADC.0
7
6
5
4
3
2
1
0
Bit
Table 20-4.
DADH – DADC Data High Register
DADH = DDH
Reset Value = 0000 0000B
Not Bit Addressable
Bit
ADC.15
ADC.14
ADC.13
ADC.12
ADC.11
ADC.10
ADC.9
ADC.8
7
6
5
4
3
2
1
0
139
3706A–MICRO–9/09
Table 20-5.
DADI – DADC Input Control Register
DADI = DAH
Reset Value = 0000 0000B
Not Bit Addressable
ACON
IREF
TRG1
TRG0
DIFF
ACS2
ACS1
ACS0
7
6
5
4
3
2
1
0
Bit
Symbol
Function
ACON
Analog Input Connect. When cleared, the analog inputs are disconnected from the ADC. When set, the analog inputs
selected by ACS2-0 are connected to the ADC. ACON must be zero when changing the input channel multiplexor (ACS20).
IREF
Internal Reference Enable. When set, the DADC uses the internal voltage reference. When cleared the DADC uses
AVDD for its reference.
DIFF
Differential Mode Enable. Set to configure the ADC in differential mode. Clear to configure the ADC in single-ended
mode.
TRG[1-0]
Trigger Select.
ACS [2-0]
140
TRG1
TRG0
Trigger
0
0
Software (GO bit)
0
1
Timer 0 Overflow
1
0
Timer 1 Overflow
1
1
Timer 2 Overflow
DADC Channel Select
DIFF
ACS2
ACS1
ACS0
V+
V-
0
0
0
0
P0.0
AVDD/2
0
0
0
1
P0.1
AVDD/2
0
0
1
0
P0.2
AVDD/2
0
0
1
1
P0.3
AVDD/2
0
1
0
0
P0.4
AVDD/2
0
1
0
1
P0.5
AVDD/2
0
1
1
0
P0.6
AVDD/2
0
1
1
1
P0.7
AVDD/2
1
0
0
0
P0.0
P0.1
1
0
0
1
P0.2
P0.3
1
0
1
0
P0.4
P0.5
1
0
1
1
P0.6
P0.7
1
1
0
0
reserved
1
1
0
1
reserved
1
1
1
0
reserved
1
1
1
1
reserved
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
21. Programmable Watchdog Timer
The programmable Watchdog Timer (WDT) protects the system from incorrect execution by triggering a system reset when it times out after the software has failed to feed the timer prior to the
timer overflow. By Default the WDT counts CPU clock cycles. The prescaler bits, PS0, PS1 and
PS2 in SFR WDTCON are used to set the period of the Watchdog Timer from 16K to 2048K
clock cycles. The Timer Prescaler can also be used to lengthen the time-out period (see Table
6-2 on page 32) The WDT is disabled by Reset and during Power-down mode. When the WDT
times out without being serviced, an internal RST pulse is generated to reset the CPU. See
Table 21-1 for the available WDT period selections.
Table 21-1.
Watchdog Timer Time-out Period Selection
WDT Prescaler Bits
Note:
PS2
PS1
PS0
Period(1)
(Clock Cycles)
0
0
0
16K
0
0
1
32K
0
1
0
64K
0
1
1
128K
1
0
0
256K
1
0
1
512K
1
1
0
1024K
1
1
1
2048K
1. The WDT time-out period is dependent on the system clock frequency.
( PS + 14 )
2
Time-out Period = ------------------------------------------------------- × ( TPS + 1 )
Oscillator Frequency
The Watchdog Timer consists of a 14-bit timer with 7-bit programmable prescaler. Writing the
sequence 1EH/E1H to the WDTRST register enables the timer. When the WDT is enabled, the
WDTEN bit in WDTCON will be set to “1”. To prevent the WDT from generating a reset when if
overflows, the watchdog feed sequence must be written to WDTRST before the end of the timeout period. To feed the watchdog, two write instructions must be sequentially executed successfully. Between the two write instructions, SFR reads are allowed, but writes are not allowed. The
instructions should move 1EH to the WDTRST register and then 1EH to the WDTRST register.
An incorrect feed or enable sequence will cause an immediate watchdog reset. The program
sequence to feed or enable the watchdog timer is as follows:
MOV WDTRST, #01Eh
MOV WDTRST, #0E1h
141
3706A–MICRO–9/09
21.1
Software Reset
A Software Reset of the AT89LP6440 is accomplished by writing the software reset sequence
5AH/A5H to the WDTRST SFR. The WDT does not need to be enabled to generate the software
reset. A normal software reset will set the SWRST flag in WDTCON. However, if at any time an
incorrect sequence is written to WDTRST (i.e. anything other than 1EH/E1H or 5AH/A5H), a
software reset will immediately be generated and both the SWRST and WDTOVF flags will be
set. In this manner an intentional software reset may be distinguished from a software error-generated reset. The program sequence to generate a software reset is as follows:
MOV WDTRST, #05Ah
MOV WDTRST, #0A5h
Table 21-2.
WDTCON – Watchdog Control Register
WDTCON Address = A7H
Reset Value = 0000 X000B
Not Bit Addressable
PS2
PS1
PS0
WDIDLE
–
SWRST
WDTOVF
WDTEN
7
6
5
4
3
2
1
0
Bit
Symbol
Function
PS2
PS1
PS0
Prescaler bits for the watchdog timer (WDT). When all three bits are cleared to 0, the watchdog timer has a nominal
period of 16K clock cycles. When all three bits are set to 1, the nominal period is 2048K clock cycles.
WDIDLE
Disable/enable the Watchdog Timer in IDLE mode. When WDIDLE = 0, WDT continues to count in IDLE mode. When
WDIDLE = 1, WDT freezes while the device is in IDLE mode.
SWRST
Software Reset Flag. Set when a software reset is generated by writing the sequence 5AH/A5H to WDTRST. Also set
when an incorrect sequence is written to WDTRST. Must be cleared by software.
WDTOVF
Watchdog Overflow Flag. Set when a WDT rest is generated by the WDT timer overflow. Also set when an incorrect
sequence is written to WDTRST. Must be cleared by software.
WDTEN
Watchdog Enable Flag. This bit is READ-ONLY and reflects the status of the WDT (whether it is running or not). The
WDT is disabled after any reset and must be re-enabled by writing 1EH/E1H to WDTRST
Table 21-3.
WDTRST – Watchdog Reset Register
WDTCON Address = A6H
(Write-Only)
Not Bit Addressable
Bit
–
–
–
–
–
–
–
–
7
6
5
4
3
2
1
0
The WDT is enabled by writing the sequence 1EH/E1H to the WDTRST SFR. The current status may be checked by reading
the WDTEN bit in WDTCON. To prevent the WDT from resetting the device, the same sequence 1EH/E1H must be written to
WDTRST before the time-out interval expires. A software reset is generated by writing the sequence 5AH/A5H to WDTRST.
142
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
22. Instruction Set Summary
The AT89LP6440 is fully binary compatible with the 8051 instruction set. The difference
between the AT89LP6440 and the standard 8051 is the number of cycles required to execute an
instruction. Instructions in the AT89LP6440 may take 1 to 9 clock cycles to complete. The execution times of most instructions may be computed using Table 22-1.
Table 22-1.
Instruction Execution Times and Exceptions
Generic Instruction Types
Cycle Count Formula
Most arithmetic, logical, bit and transfer instructions
# bytes
Branches and Calls
# bytes + 1
Single Byte Indirect (i.e. ADD A, @Ri, etc.)
2
4/5(4)
RET, RETI
MOVC
3
MOVX
2/4(2)
MUL
2
DIV
4
MAC
9
INC DPTR
2
Clock Cycles
Arithmetic
Bytes
8051
AT89LP
Hex Code
ADD A, Rn
1
12
1
28-2F
ADD A, direct
2
12
2
25
ADD A, @Ri
1
12
2
26-27
ADD A, #data
2
12
2
24
ADDC A, Rn
1
12
1
38-3F
ADDC A, direct
2
12
2
35
ADDC A, @Ri
1
12
2
36-37
ADDC A, #data
2
12
2
34
SUBB A, Rn
1
12
1
98-9F
SUBB A, direct
2
12
2
95
SUBB A, @Ri
1
12
2
96-97
SUBB A, #data
2
12
2
94
INC Rn
1
12
1
08-0F
INC direct
2
12
2
05
INC @Ri
1
12
2
06-07
INC A
2
12
2
04
DEC Rn
1
12
1
18-1F
DEC direct
2
12
2
15
DEC @Ri
1
12
2
16-17
DEC A
2
12
2
14
INC DPTR
1
24
2
A3
143
3706A–MICRO–9/09
Table 22-1.
Instruction Execution Times and Exceptions (Continued)
INC /DPTR(1)
2
–
3
A5 A3
MUL AB
1
48
2
A4
DIV AB
1
48
4
84
DA A
1
12
1
D4
MAC AB(1)
2
–
9
A5 A4
CLR M(1)
2
–
2
A5 E4
ASR M(1)
2
–
2
A5 03
LSL M(1)
2
–
2
A5 23
Clock Cycles
Bit Operations
Bytes
8051
AT89LP
Hex Code
CLR C
1
12
1
C3
CLR bit
2
12
2
C2
SETB C
1
12
1
D3
SETB bit
2
12
2
D2
CPL C
1
12
1
B3
CPL bit
2
12
2
B2
ANL C, bit
2
24
2
82
ANL C, bit
2
24
2
B0
ORL C, bit
2
24
2
72
ORL C, /bit
2
24
2
A0
MOV C, bit
2
12
2
A2
MOV bit, C
2
24
2
92
Clock Cycles
144
Logical
Bytes
8051
AT89LP
Hex Code
CLR A
1
12
1
E4
CPL A
1
12
1
F4
ANL A, Rn
1
12
1
58-5F
ANL A, direct
2
12
2
55
ANL A, @Ri
1
12
2
56-57
ANL A, #data
2
12
2
54
ANL direct, A
2
12
2
52
ANL direct, #data
3
24
3
53
ORL A, Rn
1
12
1
48-4F
ORL A, direct
2
12
2
45
ORL A, @Ri
1
12
2
46-47
ORL A, #data
2
12
2
44
ORL direct, A
2
12
2
42
ORL direct, #data
3
24
3
43
XRL A, Rn
1
12
1
68-6F
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Table 22-1.
Instruction Execution Times and Exceptions (Continued)
XRL A, direct
2
12
2
65
XRL A, @Ri
1
12
2
66-67
XRL A, #data
2
12
2
64
XRL direct, A
2
12
2
62
XRL direct, #data
3
24
3
63
RL A
1
12
1
23
RLC A
1
12
1
33
RR A
1
12
1
03
RRC A
1
12
1
13
SWAP A
1
12
1
C4
Clock Cycles
Data Transfer
Bytes
8051
AT89LP
Hex Code
MOV A, Rn
1
12
1
E8-EF
MOV A, direct
2
12
2
E5
MOV A, @Ri
1
12
2
E6-E7
MOV A, #data
2
12
2
74
MOV Rn, A
1
12
1
F8-FF
MOV Rn, direct
2
24
2
A8-AF
MOV Rn, #data
2
12
2
78-7F
MOV direct, A
2
12
2
F5
MOV direct, Rn
2
24
2
88-8F
MOV direct, direct
3
24
3
85
MOV direct, @Ri
2
24
2
86-87
MOV direct, #data
3
24
3
75
MOV @Ri, A
1
12
1
F6-F7
MOV @Ri, direct
2
24
2
A6-A7
MOV @Ri, #data
2
12
2
76-77
MOV DPTR, #data16
3
24
3
90
MOV /DPTR, #data16(1)
4
–
4
A5 90
MOVC A, @A+DPTR
1
24
3
93
MOVC A, @A+/DPTR(1)
2
–
4
A5 93
MOVC A, @A+PC
1
24
3
83
MOVX A, @Ri
1
24
2
E2-E3
MOVX A, @DPTR
1
24
2/4(2)
E0
MOVX A, @/DPTR(1)
2
–
3/5(2)
A5 E0
MOVX @Ri, A
1
24
2
F2-F3
MOVX @DPTR, A
1
24
2/4(2)
F0
MOVX @/DPTR, A(1)
2
–
3/5(2)
A5 F0
PUSH direct
2
24
2/3(4)
C0
145
3706A–MICRO–9/09
Table 22-1.
Instruction Execution Times and Exceptions (Continued)
POP direct
2
24
2/3(4)
D0
XCH A, Rn
1
12
1
C8-CF
XCH A, direct
2
12
2
C5
XCH A, @Ri
1
12
2
C6-C7
XCHD A, @Ri
1
12
2
D6-D7
Clock Cycles
Branching
Bytes
8051
AT89LP
Hex Code
JC rel
2
24
3
40
JNC rel
2
24
3
50
JB bit, rel
3
24
4
20
JNB bit, rel
3
24
4
30
JBC bit, rel
3
24
4
10
JZ rel
2
24
3
60
JNZ rel
2
24
3
70
SJMP rel
2
24
3
80
ACALL addr11
2
24
3/5(4)
11,31,51,71,91,
B1,D1,F1
LCALL addr16
3
24
4/6(4)
12
24
4/5
(4)
22
(4)
32
RET
1
RETI
1
24
4/5
AJMP addr11
2
24
3
01,21,41,61,81,
A1,C1,E1
LJMP addr16
3
24
4
02
JMP @A+DPTR
1
24
2
73
JMP @A+PC(1)
2
–
3
A5 73
CJNE A, direct, rel
3
24
4
B5
CJNE A, #data, rel
3
24
4
B4
CJNE Rn, #data, rel
3
24
4
B8-BF
CJNE @Ri, #data, rel
3
24
4
B6-B7
CJNE A, @R0, rel(1)
3
–
4
A5 B6
CJNE A, @R1, rel(1)
3
–
4
A5 B7
DJNZ Rn, rel
2
24
3
D8-DF
DJNZ direct, rel
3
24
4
D5
NOP
1
12
1
00
BREAK(1)(3)
2
–
2
A5 00
Notes:
1. This escaped instruction is an extension to the instruction set. See Section 22.1 on page 147.
2. MOVX @DPTR instructions take 2 clock cycles when accessing ERAM and 4 clock cycles
when accessing FDATA, XDATA or CODE. (3 and 5 cycles for MOVX @/DPTR).
3. The BREAK instruction acts as a 2 cycle NOP.
4. Instructions accessing the stack require additional cycles when using the extended stack.
146
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
22.1
Instruction Set Extensions
The following instructions are extensions to the standard 8051 instruction set that provide
enhanced capabilities not found in standard 8051 devices. All extended instructions start with an
A5H escape code. For this reason random A5H reserved codes should not be placed in the
instruction stream even though other devices may have treated these as NOPs.
Other AT89LP devices may not support all of these instructions.
22.1.1
ASR M
Function: Shift MAC Accumulator Right Arithmetically
Description: The forty bits in the M register are shifted one bit to the right. Bit 39 retains its value to preserve the sign of the
value. No flags are affected.
Example: The M register holds the value 0C5B1A29384H . The following instruction,
ASR M
leaves the M register holding the value 0E2D8D149C2H.
Bytes: 2
Cycles: 2
Encoding:
A5
0
0
0
0
0
0
1
1
Operation: ASR
(Mn) ←(Mn + 1) n = 0 - 38
(M39) ←(M39)
22.1.2
BREAK
Function: Software Breakpoint (Halt execution)
Description: BREAK transfers control from normal execution to the On-Chip Debug (OCD) handler if OCD is enabled. The PC
is left pointing to the following instruction. If OCD is disabled, BREAK acts as a double NOP. No flags are
affected.
Example: If On-Chip Debugging is allowed, the following instruction,
BREAK
will halt instruction execution prior to the immediately following instruction. If debugging is not allowed, the
BREAK is treated as a double NOP.
Bytes: 2
Cycles: 2
Encoding:
A5
0
0
0
0
0
0
0
0
Operation: BREAK
(PC) ←(PC) + 2
147
3706A–MICRO–9/09
22.1.3
CJNE A, @Ri, rel
Function: Compare and Jump if Not Equal
Description: CJNE compares the magnitudes of the Accumulator and indirect RAM location and branches if their values are
not equal. The branch destination is computed by adding the signed relative-displacement in the last instruction
byte to the PC, after incrementing the PC to the start of the next instruction. The carry flag is set if the unsigned
integer value of ACC is less than the unsigned integer value of the indirect location; otherwise, the carry is
cleared. Neither operand is affected.
Example: The Accumulator contains 34H. Register 0 contains 78H and 78H contains 56H. The first instruction in the
sequence,
CJNE A, @R0, NOT_EQ
;
. . . . . . ...... ...... ; ACC = @R0.
NOT_EQ:
JC
REQ_LOW .. ;IF ACC< @R0.
;
. . . . . . ...... ...... ;ACC > @R0.
sets the carry flag and branches to the instruction at label NOT_EQ. By testing the carry flag, the second
instruction determines whether ACC is greater or less than the location pointed to by R0.
Bytes: 2
Cycles: 9
Encoding:
A5
1
0
1
1
0
1
1
i
rel. address
Operation: CJNE
(PC) ← (PC) + 3
IF (A) ≠ ((Ri))
THEN
(PC) ← (PC) + relative offset
IF (A) < ((Ri))
THEN
(C) ← 1
ELSE
(C) ← 0
22.1.4
CLR M
Function: Clear MAC Accumulator
Description: CLR M clears the 40-bit M register. No flags are affected.
Example: The M registercontains 123456789AH. The following instruction,
CLR M
leaves the M register set to 0000000000H.
Bytes: 2
Cycles: 2
Encoding:
A5
1
1
1
0
0
1
0
0
Operation: JMP
(M) ←0
148
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
22.1.5
INC /DPTR
Function: Increment Alternate Data Pointer
Description: INC /DPTR increments the unselected 16-bit data pointer by 1. A 16-bit increment (modulo 216 ) is performed,
and an overflow of the low-order byte of the data pointer from 0FFH to 00H increments the high-order byte. No
flags are affected.
Example: Registers DP1H and DP1L contain 12H and 0FEH, respectively, and DPS = 0. The following instruction
sequence,
INC
INC
INC
/DPTR
/DPTR
/DPTR
changes DP1H and DP1L to 13H and 01H.
Bytes: 2
Cycles: 3
Encoding:
A5
1
0
1
0
0
0
1
1
Operation: INC
IF (DPS) = 0
THEN
(DPTR1) ← (DPTR1) + 1
ELSE
(DPTR0) ← (DPTR0) + 1
22.1.6
JMP @A+PC
Function: Jump indirect relative to PC
Description: JMP @A+PC adds the eight-bit unsigned contents of the Accumulator to the program counter, which is first
incremented by two. This is the address for subsequent instruction fetches. Sixteen-bit addition is performed
(modulo 216): a carry-out from the low-order eight bits propagates through the higher-order bits. The
Accumulator is not altered. No flags are affected.
Example: An even number from 0 to 6 is in the Accumulator. The following sequence of instructions branches to one of four
AJMP instructions in a jump table starting at JMP_TBL.
JMP
JMP_TBL:
AJMP
AJMP
AJMP
AJMP
@A + PC
LABEL0
LABEL1
LABEL2
LABEL3
If the Accumulator equals 04H when starting this sequence, execution jumps to label LABEL2. Because AJMP is
a 2-byte instruction, the jump instructions start at every other address.
Bytes: 2
Cycles: 3
Encoding:
A5
0
1
1
1
0
0
1
1
Operation: JMP
(PC) ←(A) + (PC) + 2
149
3706A–MICRO–9/09
22.1.7
LSL M
Function: Shift MAC Accumulator Left Logically
Description: The forty bits in the M register are shifted one bit to the left. Bit 0 is cleared. No flags are affected.
Example: The M register holds the value 0C5B1A29384H. The following instruction,
LSL M
leaves the M register holding the value 8B63452708H.
Bytes: 2
Cycles: 2
Encoding:
A5
0
0
1
0
0
0
1
1
Operation: LSL
(Mn+1) ←(Mn) n = 0 - 38
(M0) ← 0
22.1.8
MOVC A, @A+/DPTR
Function: Move code byte relative to Alternate Data Pointer
Description: The MOVC instructions load the Accumulator with a code byte or constant from program memory. The address
of the byte fetched is the sum of the original unsigned 8-bit Accumulator contents and the contents of the
unselected Data Pointer. The base register is not altered. Sixteen-bit addition is performed so a carry-out from
the low-order eight bits may propagate through higher-order bits. No flags are affected.
Example: A value between 0 and 3 is in the Accumulator. The following instructions will translate the value in the
Accumulator to one of four values defined by the DB (define byte) directive.
MOV
/DPTR, #TABLE
MOVC A, @A+PC
RET
TABLE:
DB
DB
DB
DB
66H
77H
88H
99H
If the subroutine is called with the Accumulator equal to 01H, it returns with 77H in the Accumulator.
Bytes: 2
Cycles: 4
Encoding:
A5
1
0
0
1
0
0
1
1
Operation: MOVC
IF (DPS) = 0
THEN
(A) ←( (A) + (DPTR1) )
ELSE
(A) ←( (A) + (DPTR0) )
150
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
22.1.9
MAC AB
Function: Multiply and Accumulate
Description: MAC AB multiplies the signed 16-bit integers in the register pairs {AX, A} and {BX, B} and adds the 32-bit product
to the 40-bit M register. The low-order bytes of the 16-bit operands are stored in A and B, and the high-order
bytes in AX and BX respectively. The four operand registers are unaffected by the operation. If the addition of the
product to the accumulated sum in M results in a two's complement overflow, the overflow flag is set; otherwise it
is not cleared. The carry flag is set if the result is negative and cleared if positive.
Example: Originally the Accumulator holds the value 80 (50H). Register B holds the value 160 (0A0H). The instruction,
MAC AB
will give the product 12, 800 (3200H), so B is changed to 32H (00110010B) and the Accumulator is cleared. The
overflow flag is set, carry is cleared.
Bytes: 2
Cycles: 9
Encoding:
A5
1
0
1
0
0
1
0
0
Operation: MAC
(M39-0) ← (M) + { (AX), (A) } X { (BX), (B) }
22.1.10
MOV /DPTR, #data16
Function: Load Alternate Data Pointer with a 16-bit constant
Description: MOV /DPTR, #data16 loads the unselected Data Pointer with the 16-bit constant indicated. The third byte is the
high-order byte, while the fourth byte holds the lower-order byte. No flags are affected.
Example: When DPS = 0, the instruction sequence,
MOV DPTR, # 1234H
MOV /DPTR, # 5678H
loads the value 1234H into the first Data Pointer: DPH0 holds 12H and DPL0 holds 34H; and loads the value
5678H into the second Data Pointer: DPH1 hold 56H and DPL1 holds 78H.
Bytes: 2
Cycles: 3
Encoding:
A5
90
immed. data 15-8
immed. data 7-0
Operation: MOV
IF (DPS) = 0
THEN
(DP1H) ← #data15-8
(DP1L) ← #data7-0
ELSE
(DP0H) ← #data15-8
(DP0L) ← #data7-0
151
3706A–MICRO–9/09
22.1.11
MOVX A, @/DPTR
Function: Move External using Alternate Data Pointer
Description: The MOVX instruction transfesr data from external data memory to the Accumulator. The unselected Data
Pointer generates a 16-bit address which targets EDATA, FDATA or XDATA.
Example: DPS = 0, DPTR0 contains 0123H and DPTR1 contains 4567H. The following instruction sequence,
MOVX
MOVX
A, @DPTR
@/DPTR, A
copies the data from address 0123H to 4567H.
Bytes: 2
Cycles: 3 (EDATA)
5 (FDATA or XDATA)
Encoding:
A5
1
1
1
0
0
0
0
0
Operation: MOVX
IF (DPS) = 0
(A) ← ((DPTR1))
ELSE
(A) ← ((DPTR0))
22.1.12
MOVX @/DPTR, A
Function: Move External using Alternate Data Pointer
Description: The MOVX instruction transfesr data from the Accumulator to external data memory. The unselected Data
Pointer generates a 16-bit address which targets EDATA, FDATA or XDATA.
Example: DPS = 0, DPTR0 contains 0123H and DPTR1 contains 4567H. The following instruction sequence,
MOVX
MOVX
A, @DPTR
@/DPTR, A
copies the data from address 0123H to 4567H.
Bytes: 2
Cycles: 3 (EDATA)
5 (FDATA or XDATA)
Encoding:
A5
1
1
1
1
0
0
0
0
Operation: MOVX
IF (DPS) = 0
THEN
((DPTR1)) ← (A)
ELSE
((DPTR0)) ← (A)
152
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
23. Register Index
Table 23-1.
Special Function Register Cross Reference
Name
Address
Description Index
ACC
E0H
ACSRA
97H
Table 19-1 on page 130
ACSRB
9FH
Table 19-2 on page 131
AREF
AFH
Table 19-3 on page 132
AUXR
8EH
Table 3-4 on page 17
AX
E1H
Section 5.1 on page 23
B
F0H
BX
F7H
Section 5.1 on page 23
CLKREG
8FH
Table 6-2 on page 32
DADC
D9H
Table 20-2 on page 139
DADH
DDH
Table 20-4 on page 139
DADI
DAH
Table 20-3 on page 139
DADL
DCH
Table 20-3 on page 139
DPCF (AUXR1)
A2H
Table 5-5 on page 27
DPH0
83H
Section 5.2 on page 24
DPH1
85H
Section 5.2 on page 24
DPL0
82H
Section 5.2 on page 24
DPL1
83H
Section 5.2 on page 24
DSPR
E2H
Table 5-1 on page 25
FIRD
E3H
Section 5.2.2.3 on page 28
GPIEN
9CH
Table 15-3 on page 83
GPIF
9DH
Table 15-4 on page 83
GPLS
9BH
Table 15-2 on page 83
GPMOD
9AH
Table 15-1 on page 83
IE
A8H
Table 9-2 on page 41
IE2
B4H
Table 9-3 on page 42
IP
B8H
Table 9-4 on page 42
IP2
B5H
Table 9-5 on page 42
IPH
B7H
Table 9-6 on page 43
IPH2
B6H
Table 9-7 on page 43
MACH
E5H
Section 5.1 on page 23
MACL
E4H
Section 5.1 on page 23
MEMCON
96H
Table 3-3 on page 16
P0
80H
Table 10-3 on page 44
P0M0
BAH
Table 10-2 and Table 10-3 on page 44
P0M1
BBH
Table 10-2 and Table 10-3 on page 44
153
3706A–MICRO–9/09
Table 23-1.
154
Special Function Register Cross Reference
P1
90H
Table 10-3 on page 44
P1M0
C2H
Table 10-2 and Table 10-3 on page 44
P1M1
C3H
Table 10-2 and Table 10-3 on page 44
P2
A0H
Table 10-3 on page 44
P2M0
C4H
Table 10-2 and Table 10-3 on page 44
P2M1
C5H
Table 10-2 and Table 10-3 on page 44
P3
B0H
Table 10-3 on page 44
P3M0
C6H
Table 10-2 and Table 10-3 on page 44
P3M1
C7H
Table 10-2 and Table 10-3 on page 44
P4
C0H
Table 10-3 on page 44
P4M0
BEH
Table 10-2 and Table 10-3 on page 44
P4M1
BFH
Table 10-2 and Table 10-3 on page 44
PAGE
86H
Table 3-2 on page 13
PCON
87H
Table 8-1 on page 36
PSW
D0H
RCAP2H
CBH
Section 12.1 on page 60
RCAP2L
CAH
Section 12.1 on page 60
RH0
94H
Table 11-1 on page 50
RH1
95H
Table 11-1 on page 50
RL0
92H
Table 11-1 on page 50
RL1
93H
Table 11-1 on page 50
SADDR
A9H
Section 16.7 on page 96
SADEN
B9H
Section 16.7 on page 96
SBUF
99H
Section 16.3 on page 88
SCON
98H
Table 16-1 on page 85
SP
81H
Section 3.4 on page 19
SPCR
E9H
Table 17-2 on page 101
SPDR
EAH
Table 17-3 on page 102
SPSR
E8H
Table 17-4 on page 102
SPX
9EH
Section 3.4 on page 19
T2CCA
D1H
Table 13-1 on page 70
T2CCC
D4H
Table 13-5 on page 73
T2CCF
D5H
Table 13-4 on page 71
T2CCH
D3H
Table 13-2 on page 70
T2CCL
D2H
Table 13-3 on page 70
T2CON
C8H
Table 12-3 on page 60
T2MOD
C9H
Table 12-4 on page 60
TCON
88H
Table 11-2 on page 53
TCONB
91H
Table 11-4 on page 55
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Table 23-1.
Special Function Register Cross Reference
TH0
8CH
Table 11-1 on page 50
TH1
8DH
Table 11-1 on page 50
TH2
CDH
Section 12.1 on page 60
TL0
8AH
Table 11-1 on page 50
TL1
8BH
Table 11-1 on page 50
TL2
CCH
Section 12.1 on page 60
TMOD
89H
Table 11-3 on page 54
TWAR
ACH
Table 18-3 on page 111
TWBR
AEH
Table 18-5 on page 112
TWCR
AAH
Table 18-1 on page 111
TWDR
ADH
Table 18-4 on page 112
TWSR
ABH
Table 18-2 on page 111
WDTCON
A7H
Table 21-2 on page 142
WDTRST
A6H
Table 21-3 on page 142
24. On-Chip Debug System
The AT89LP6440 On-Chip Debug (OCD) System uses a two-wire serial interface to control program flow; read, modify, and write the system state; and program the nonvolatile memory. The
OCD System has the following features:
• Complete program flow control
• Read-Modify-Write access to all internal SFRs and data memories
• Four hardware program address breakpoints, plus four program/data address breakpoints
• Unlimited program software breakpoints using BREAK instruction
• Break on change in program memory flow
• Break on stack overflow/underflow
• Break on Watchdog overflow
• Break on reset
• Non-intrusive operation
• Programming of nonvolatile memory
24.1
Physical Interface
The On-Chip Debug System uses a two-wire synchronous serial interface to establish communication between the target device and the controlling emulator system. The OCD interface is
enabled by clearing the OCD Enable Fuse. The OCD device connections are shown in Figure
24-1. When OCD is enabled, the RST port pin is configured as an input for the Debug Clock
(DCL). Either the XTAL1, XTAL2 or P4.3 pin is configured as a bi-directional data line for the
Debug Data (DDA) depending on the clock source selected. If the Internal RC Oscillator is
selected, XTAL1 is configured as DDA (A).If the External Clock is selected, XTAL2 is configured
as DDA (B). If the Crystal Oscillator is selected, P4.3 is configured as DDA (C).
When designing a system where On-Chip Debug will be used, the following observations must
be considered for correct operation:
155
3706A–MICRO–9/09
• P4.2/RST cannot be connected directly to VDD and any external capacitors connected to RST
must be removed.
• All external reset sources must be removed.
• If P4.3 needs to be debugged in systems using the crystal oscillator, the external clock option
should be selected. The quartz crystal and any capacitors on XTAL1 or XTAL2 must be
removed and an external clock signal must be driven on XTAL1. Some emulator systems may
provide a user-configurable clock for this purpose.
Figure 24-1. AT89LP6440 On-Chip Debug Connections
VDD
DCL
P4.2/RST
DDA
XTAL1
VDD
DCL
P4.2/RST
CLK
XTAL1
A
B
GND
XTAL2
DDA
GND
CLK = Internal RC
CLK = External Clock
VDD
DCL
P4.2/RST
XTAL2
XTAL1
C
P4.3
DDA
GND
CLK = Crystal Oscillator
24.2
Software Breakpoints
The AT89LP6440 microcontroller includes a BREAK instruction for implementing program memory breakpoints in software. A software breakpoint can be inserted manually by placing the
BREAK instruction in the program code. Some emulator systems may allow for automatic insertion/deletion of software breakpoints. The Flash memory must be re-programmed each time a
software breakpoint is changed. Frequent insertions/deletions of software breakpoints will
reduce the endurance of the nonvolatile memory. Devices used for debugging purposes should
not be shipped to end customers. The BREAK instruction is treated as a two-cycle NOP when
OCD is disabled.
24.3
Limitations of On-Chip Debug
The AT89LP6440 is a fully-featured microcontroller that multiplexes several functions on its limited I/O pins. Some device functionality must be sacrificed to provide resources for On-Chip
Debugging. The On-Chip Debug System has the following limitations:
• The Debug Clock pin (DCL) is physically located on that same pin as Port Pin P4.2 and the
External Reset (RST). Therefore, neither P4.2 nor an external reset source may be emulated
when OCD is enabled.
156
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
• When using the Internal RC Oscillator during debug, DDA is located on the XTAL1/P4.0 pin.
The P4.0 I/O function cannot be emulated in this mode.
• When using the External Clock during debug, DDA is located on the XTAL2/P4.1 pin and the
system clock drives XTAL1/P4.0. The P4.1 I/O and CLKOUT functions cannot be emulated in
this mode.
• When using the Crystal Oscillator during debug, DDA is located on the P4.3 pin and the
crystal connects to XTAL1/P4.0 and XTAL2/P4.1. The P4.3 I/O function cannot be emulated
in this mode.
25. Programming the Flash Memory
The Atmel AT89LP6440 microcontroller features 64K bytes of on-chip In-System Programmable
Flash program memory and 8K bytes of nonvolatile Flash data memory. In-System Programming allows programming and reprogramming of the microcontroller positioned inside the end
system. Using a simple 4-wire SPI interface, the programmer communicates serially with the
AT89LP6440 microcontroller, reprogramming all nonvolatile memories on the chip. In-System
Programming eliminates the need for physical removal of the chips from the system. This will
save time and money, both during development in the lab, and when updating the software or
parameters in the field. The programming interface of the AT89LP6440 includes the following
features:
• Four-wire serial SPI Programming Interface or 11-pin Parallel Interface
• Active-low Reset Entry into Programming
• Slave Select allows multiple devices on same interface
• User Signature Array
• Flexible Page Programming
• Row Erase Capability
• Page Write with Auto-Erase Commands
• Programming Status Register
For more detailed information on In-System Programming, refer to the Application Note entitled
“AT89LP In-System Programming Specification”.
25.1
Physical Interface
The AT89LP6440 provides a standard programming command set with two physical interfaces:
a bit-serial and a byte-parallel interface. Normal Flash programming utilizes the Serial Peripheral
Interface (SPI) pins of an AT89LP6440 microcontroller. The SPI is a full-duplex synchronous
serial interface consisting of four wires: Serial Clock (SCK), Master-In/Slave-out (MISO), Masterout/Slave-in (MOSI), and an active-low Slave Select (SS). When programming an AT89LP6440
device, the programmer always operates as the SPI master, and the target system always operates as the SPI slave. To enter or remain in Programming mode the device’s reset line (RST)
must be held active (low). With the addition of VDD and GND, an AT89LP6440 microcontroller
can be programmed with a minimum of seven connections as shown in Figure 25-1.
157
3706A–MICRO–9/09
Figure 25-1. In-System Programming Device Connections
AT89LP6440
Serial Clock
P1.7/SCK
Serial Out
P1.6/MISO
Serial In
P1.5/MOSI
SS
VDD
P1.4/SS
P4.2/RST
RST
GND
The Parallel interface is a special mode of the serial interface, i.e. the serial interface is used to
enable the parallel interface. After enabling the interface serially over P1.7/SCK and P1.5/MOSI,
P1.5 is reconfigured as an active-low output enable (OE) for data on Port 0. When OE = 1, command, address and write data bytes are input on Port 0 and sampled at the rising edge of SCK.
When OE = 0, read data bytes are output on Port 0 and should be sampled on the falling edge of
SCK. The P1.7/SCK, P1.4/SS and P4.2/RST pins continue to function in the same manner. With
the addition of VDD and GND, the parallel interface requires a minimum of fourteen connections
as shown in Figure 25-2. Note that a connection to P1.6/MISO is not required for using the parallel interface.
Figure 25-2. Parallel Programming Device Connections
AT89LP6440
Clock
P1.7/SCK
RST
P4.2/RST
OE
P1.5/MOSI
SS
P1.4/SS
VDD
P0.7-0
8
Data In/Out
GND
The Programming Interface is the only means of externally programming the AT89LP6440
microcontroller. The Interface can be used to program the device both in-system and in a standalone serial programmer. The Interface does not require any clock other than SCK and is not
limited by the system clock frequency. During Programming the system clock source of the target device can operate normally.
When designing a system where In-System Programming will be used, the following observations must be considered for correct operation:
158
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
• The ISP interface uses the SPI clock mode 0 (CPOL = 0, CPHA = 0) exclusively with a
maximum frequency of 5 MHz.
• The AT89LP6440 will enter programming mode only when its reset line (RST) is active (low).
To simplify this operation, it is recommended that the target reset can be controlled by the InSystem programmer. To avoid problems, the In-System programmer should be able to keep
the entire target system reset for the duration of the programming cycle. The target system
should never attempt to drive the four SPI lines while reset is active.
• The RST input may be disabled to gain an extra I/O pin. In these cases the RST pin will
always function as a reset during power up. To enter programming the RST pin must be
driven low prior to the end of Power-On Reset (POR). After POR has completed the device
will remain in ISP mode until RST is brought high. Once the initial ISP session has ended, the
power to the target device must be cycled OFF and ON to enter another session.
• The SS pin should not be left floating during reset if ISP is enabled.
• The ISP Enable Fuse must be set to allow programming during any reset period. If the ISP
Fuse is disabled, ISP may only be entered at POR.
• For standalone programmers, RST may be tied directly to GND to ensure correct entry into
Programming mode regardless of the device settings.
25.2
Memory Organization
The AT89LP6440 offers 64K bytes of In-System Programmable (ISP) nonvolatile Flash code
memory and 8K bytes of nonvolatile Flash data memory. In addition, the device contains a 256byte User Signature Array and a 128-byte read-only Atmel Signature Array. The memory organization is shown in Table 25-1 and Figure 25-3. The memory is divided into pages of 128 bytes
each. A single read or write command may only access half a page (64 bytes) in the memory;
however, write with auto-erase commands will erase an entire 128-byte page even though they
can only write one half page. Each memory type resides in its own address space and is
accessed by commands specific to that memory. However, all memory types share the same
page size.
User configuration fuses are mapped as a row in the memory, with each byte representing one
fuse. From a programming standpoint, fuses are treated the same as normal code bytes except
they are not affected by Chip Erase. Fuses can be enabled at any time by writing 00h to the
appropriate locations in the fuse row. However, to disable a fuse, i.e. set it to FFh, the entire
fuse row must be erased and then reprogrammed. The programmer should read the state of all
the fuses into a temporary location, modify those fuses which need to be disabled, then issue a
Fuse Write with Auto-Erase command using the temporary data. Lock bits are treated in a similar manner to fuses except they may only be erased (unlocked) by Chip Erase.
Table 25-1.
AT89LP6440 Memory Organization
Memory
Capacity
Page Size
# Pages
Address Range
CODE
65536 bytes
128 bytes
512
0000H – FFFFH
DATA
8192 bytes
128 bytes
64
1000H – 3FFFH
User Signature
256 bytes
128 bytes
2
0000H – 00FFH
Atmel Signature
128 bytes
128 bytes
1
0000H – 007FH
159
3706A–MICRO–9/09
Figure 25-3. AT89LP6440 Memory Organization
00
3F
Page Buffer
User Fuse Row
Page 0 Low
User Signature Array
Page 1 Low
Page 0 Low
Page 1 High
Page 0 High
Atmel Signature Array
Page 0 Low
Page 0 High
Page 63 Low
Page 63 High
3FFF
Page 0 Low
Page 0 High
1000
Page 511 Low
Page 510 Low
Page 511 High
Page 510 High
FFFF
Page 1 Low
Page 0 Low
Page 1 High
Page 1 High
Data Memory
Code Memory
00
25.3
3F 40
0000
7F
Command Format
Programming commands consist of an opcode byte, two address bytes, and zero or more data
bytes. In addition, all command packets must start with a two-byte preamble of AAH and 55H.
The preamble increases the noise immunity of the programming interface by making it more difficult to issue unintentional commands. Figure 25-4 on page 161 shows a simplified flow chart of
a command sequence.
A sample command packet is shown in Figure 25-5 on page 161. The SS pin defines the packet
frame. SS must be brought low before the first byte in a command is sent and brought back high
after the final byte in the command has been sent. The command is not complete until SS
returns high. Command bytes are issued serially on MOSI. Data output bytes are received serially on MISO. Packets of variable length are supported by returning SS high when the final
required byte has been transmitted. In some cases command bytes have a don’t care value.
Don’t care bytes in the middle of a packet must be transmitted. Don’t care bytes at the end of a
packet may be ignored.
Page oriented instructions always include a full 16-bit address. The higher order bits select the
page and the lower order bits select the byte within that page. The AT89LP6440 allocates 6 bits
for byte address, 1 bit for low/high half page selection and 9 bits for page address. The half page
to be accessed is always fixed by the page address and half select as transmitted. The byte
address specifies the starting address for the first data byte. After each data byte has been
transmitted, the byte address is incremented to point to the next data byte. This allows a page
command to linearly sweep the bytes within a page. If the byte address is incremented past the
last byte in the half page, the byte address will roll over to the first byte in the same half page.
While loading bytes into the page buffer, overwriting previously loaded bytes will result in data
corruption.
160
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
For a summary of available commands, see Table 25-2 on page 162.
Figure 25-4. Command Sequence Flow Chart
Input Preamble 1
(AAh)
Input Preamble 2
(55h)
Input Opcode
Input Address
High Byte
Input Address
Low Byte
Input/Output
Data
Address +1
Figure 25-5. ISP Command Packet (Serial)
SS
SCK
MOSI
MISO
7 6 5 4 3 2 1 0
7 6 5 4 3 2 1 0
7 6 5 4 3 2 1 0
Preamble 1
Preamble 2
Opcode
Address High
Address Low
7 6 5 4 3 2 1 0
Data In
X
X
X
X
X
7 6 5 4 3 2 1 0
Data Out
Figure 25-6. ISP Command Packet (Parallel)
SS
SCK
WRITE
OE
P0
AAh
55h
Opcode
Address High
Address Low
Data In
READ
OE
P0
AAh
55h
Opcode
Address High
Address Low
Data Out
161
3706A–MICRO–9/09
Table 25-2.
Programming Command Summary
Command
Opcode
Addr High
Addr Low
Data 0
Data n
Program Enable(1)
1010 1100
0101 0011
–
–
–
Parallel Enable(2)
1010 1100
0011 0101
–
–
–
Chip Erase
1000 1010
–
–
–
–
Read Status
0110 0000
xxxx xxxx
xxxx xxxx
Status Out
Load Page Buffer(3)
0101 0001
xxxx xxxx
00bb bbbb
Data In 0 ... Data In n
Write Code Page(3)
0101 0000
aaaa aaaa
asbb bbbb
Data In 0 ... Data In n
Write Code Page with Auto-Erase(3)
0111 0000
aaaa aaaa
asbb bbbb
Data In 0 ... Data In n
Read Code Page(3)
0011 0000
aaaa aaaa
asbb bbbb
Data Out 0 ... Data Out n
Write Data Page(3)
1101 0000
000a aaaa
asbb bbbb
Data In 0 ... Data In n
Write Data Page with Auto-Erase(3)
1101 0010
000a aaaa
asbb bbbb
Data In 0 ... Data In n
Read Data Page(3)
1011 0000
000a aaaa
asbb bbbb
Data Out 0 ... Data Out n
Write User Fuses(3)(4)(5)
1110 0001
0000 0000
00bb bbbb
Data In 0 ... Data In n
Write User Fuses with Auto-Erase(3)(4)(5)
1111 0001
0000 0000
00bb bbbb
Data In 0 ... Data In n
Read User Fuses(3)(4)(5)
0110 0001
0000 0000
00bb bbbb
Data Out 0 ... Data Out n
Write Lock Bits(3)(4)(6)
1110 0100
0000 0000
00bb bbbb
Data In 0 ... Data In n
Read Lock Bits(3)(4)(6)
0110 0100
0000 0000
00bb bbbb
Data Out 0 ... Data Out n
Write User Signature Page(3)
0101 0010
0000 0000
asbb bbbb
Data In 0 ... Data In n
Write User Signature Page with Auto-Erase(3)
0111 0010
0000 0000
asbb bbbb
Data In 0 ... Data In n
Read User Signature Page(3)
0011 0010
0000 0000
asbb bbbb
Data Out 0 ... Data Out n
Read Atmel Signature Page(3)(7)
0011 1000
0000 0000
0sbb bbbb
Data Out 0 ... Data Out n
Notes:
1. Program Enable must be the first command issued after entering into programming mode.
2. Parallel Enable switches the interface from serial to parallel format until RST returns high.
3. Any number of Data bytes from 1 to 64 may be written/read. The internal address is incremented between each byte.
4. Each byte address selects one fuse or lock bit. Data bytes must be 00h or FFh.
5. See Table 25-5 on page 164 for Fuse definitions.
6. See Table 25-4 on page 163 for Lock Bit definitions.
7. Atmel Signature Bytes:
Address:
AT89LP6440:
0000H
0001H
0002H
1EH
64H
FFH
8. Symbol Key:
162
a:
Page Address Bit
s:
Half Page Select Bit
b:
Byte Address Bit
x:
Don’t Care Bit
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
25.4
Status Register
The current state of the memory may be accessed by reading the status register. The status register is shown in Table 25-3.
Table 25-3.
Status Register
Bit
–
–
–
–
LOAD
SUCCESS
WRTINH
BUSY
7
6
5
4
3
2
1
0
Symbol
Function
LOAD
Load flag. Cleared low by the load page buffer command and set high by the next memory write. This flag signals that
the page buffer was previously loaded with data by the load page buffer command.
SUCCESS
Success flag. Cleared low at the start of a programming cycle and will only be set high if the programming cycle
completes without interruption from the brownout detector.
WRTINH
Write Inhibit flag. Cleared low by the brownout detector (BOD) whenever programming is inhibited due to VDD falling
below the minimum required programming voltage. If a BOD episode occurs during programming, the SUCCESS flag
will remain low after the cycle is complete.
BUSY
Busy flag. Cleared low whenever the memory is busy programming or if write is currently inhibited.
25.5
DATA Polling
The AT89LP6440 implements DATA polling to indicate the end of a programming cycle. While
the device is busy, any attempted read of the last byte written will return the data byte with the
MSB complemented. Once the programming cycle has completed, the true value will be accessible. During Erase the data is assumed to be FFH and DATA polling will return 7FH. When
writing multiple bytes in a page, the DATA value will be the last data byte loaded before programming begins, not the written byte with the highest physical address within the page.
25.6
Flash Security
The AT89LP6440 provides two Lock Bits for Flash Code and Data Memory security. Lock bits
can be left unprogrammed (FFh) or programmed (00h) to obtain the protection levels listed in
Table 25-4. Lock bits can only be erased (set to FFh) by Chip Erase. Lock bit mode 2 disables
programming of all memory spaces, including the User Signature Array and User Configuration
Fuses. User fuses must be programmed before enabling Lock bit mode 2 or 3. Lock bit mode 3
implements mode 2 and also blocks reads from the code and data memories; however, reads of
the User Signature Array, Atmel Signature Array, and User Configuration Fuses are still allowed.
The Lock Bits will not disable FDATA or IAP programming initiated by the application software.
Table 25-4.
Lock Bit Protection Modes
Program Lock Bits (by address)
Mode
00h
01h
Protection Mode
1
FFh
FFh
No program lock features
2
00h
FFh
Further programming of the Flash is disabled
3
00h
00h
Further programming of the Flash is disabled and verify (read) is also
disabled; OCD is disabled
163
3706A–MICRO–9/09
25.7
User Configuration Fuses
The AT89LP6440 includes 11 user fuses for configuration of the device. Each fuse is accessed
at a separate address in the User Fuse Row as listed in Table 25-5. Fuses are cleared by programming 00h to their locations. Programming FFh to a fuse location will cause that fuse to
maintain its previous state. To set a fuse (set to FFh) the fuse row must be erased and then
reprogrammed using the Fuse Write with Auto-erase command. The default state for all fuses is
FFh.
Table 25-5.
Address
User Configuration Fuse Definitions
Fuse Name
Description
Selects source for the system clock:
00 – 01h
Clock Source – CS[0:1](2)
CS1
CS0
Selected Source
00h
00h
High Speed Crystal Oscillator (XTAL)
00h
FFh
Low Speed Crystal Oscillator (XTAL)
FFh
00h
External Clock on XTAL1 (XCLK)
FFh
FFh
Internal RC Oscillator (IRC)
Selects time-out delay for the POR/BOD/PWD wake-up period:
02 – 03h
SUT1
SUT0
Selected Time-out
00h
00h
1 ms (XTAL); 16 µs (XCLK/IRC)
00h
FFh
2 ms (XTAL); 512 µs (XCLK/IRC)
FFh
00h
4 ms (XTAL); 1 ms (XCLK/IRC)
FFh
FFh
16 ms (XTAL); 4 ms (XCLK/IRC)
Start-up Time – SUT[0:1]
04h
Reset Pin Enable(3)
FFh: RST pin functions as reset
00h: RST pin functions as general purpose I/O
05h
Brown-Out Detector Enable
FFh: Brown-out Detector Enabled
00h: Brown-out Detector Disabled
06h
On-Chip Debug Enable
FFh: On-Chip Debug Disabled
00h: On-Chip Debug Enabled
07h
ISP Enable(3)
FFh: In-System Programming Enabled
00h: In-System Programming Disabled (Enabled at POR only)
08H
User Signature Programming
FFh: Programming of User Signature Disabled
00h: Programming of User Signature Enabled
09H
Tristate Ports
FFh: I/O Ports start in input-only mode (tristated) after reset
00h: I/O Ports start in quasi-bidirectional mode after reset
0AH
OCD Interface Select
FFh: Fast two-wire interface
00h: Do not use
0BH
In-Application Programming
FFh: In-Application Programming Disabled
00h: In-Application Programming Enabled
Notes:
1. The default state for all fuses is FFh.
2. Changes to these fuses will only take effect after a device POR.
3. Changes to these fuses will only take effect after the ISP session terminates by bringing RST high.
164
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
25.8
User Signature and Analog Configuration
The User Signature Array contains 256 bytes of non-volatile memory in two 128-byte pages. The
first page of the User Signature Array (0000H–007FH) is available for serial numbers, firmware
revision information, date codes or other user parameters. The User Signature Array may only
be written by an external device when the User Signature Programming Fuse is enabled. When
the fuse is enabled, Chip Erase will also erase the first page of the array. When the fuse is disabled, the array is not affected by write or erase commands. Programming of the Signature
Array can also be disabled by the Lock Bits. However, reading the signature is always allowed
and the array should not be used to store security sensitive information. The User Signature
Array may be modified during execution through the In-Application Programming interface,
regardless of the state of the User Signature Programming fuse or Lock Bits, provided that the
IAP Fuse is enabled. Note that the address of the User Signature Array, as seen by the IAP
interface, equals the User Signature address plus 256 (0100H–01FFH instead of 0000H–
00FFH).
The second page of the User Signature Array (0080H–00FFH) contains analog configuration
parameters for the AT89LP6440. Each byte represents a parameter as listed in Table 25-6 and
is preset in the factory. The parameters are read at POR and the device is configured accordingly. The second page of the array is not affected by Chip Erase. Other bytes in this page may
be used as additional signature space; however, care should be taken to preserve the parameter values when modifying other bytes.
Table 25-6.
Address
0080H
25.9
Analog Configuration Definitions
Parameter Name
Description
RC Oscillator Calibration Byte
The RC Calibration Byte controls the frequency of the internal RC oscillator. The
frequency is inversely proportional to the calibration value such that higher values
result in lower frequencies. A copy of the factory-set calibration value is stored at
location 0008H of the Atmel Signature.
Programming Interface Timing
This section details general system timing sequences and constraints for entering or exiting InSystem Programming as well as parameters related to the Serial Peripheral Interface during
ISP. The general timing parameters for the following waveform figures are listed in section “Timing Parameters” on page 169.
25.9.1
Power-up Sequence
Execute this sequence to enter programming mode immediately after power-up. In the RST pin
is disabled or if the ISP Fuse is disabled, this is the only method to enter programming (see
“External Reset” on page 34).
1. Apply power between VDD and GND pins. RST should remain low.
2. Wait at least tPWRUP. and drive SS high.
3. Wait at least tSUT for the internal Power-on Reset to complete. The value of tSUT will
depend on the current settings of the device.
4. Start programming session.
165
3706A–MICRO–9/09
Figure 25-7. Serial Programming Power-up Sequence
VDD
tPWRUP
RST
tPOR + tSUT
SS
tZSS
SCK
25.9.2
MISO
HIGH Z
MOSI
HIGH Z
Power-down Sequence
Execute this sequence to power-down the device after programming.
1. Drive SCK low.
2. Wait at least tSSD and bring SS high.
3. Tristate MOSI.
4. Wait at least tSSZ and then tristate SS and SCK.
5. Wait no more than tPWRDN and power off VDD.
Figure 25-8. Serial Programming Power-down Sequence
VDD
tPWRDN
RST
SS
SCK
25.9.3
tSSD
tSSZ
MISO
HIGH Z
MOSI
HIGH Z
ISP Start Sequence
Execute this sequence to exit CPU execution mode and enter ISP mode when the device has
passed Power-On Reset and is already operational.
1. Drive RST low.
2. Drive SS high.
3. Wait tRLZ + tSTL.
4. Start programming session.
166
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Figure 25-9. In-System Programming (ISP) Start Sequence
tRLZ
VDD
XTAL1
RST
tSTL
SS
tZSS
tSSE
SCK
25.9.4
MISO
HIGH Z
MOSI
HIGH Z
ISP Exit Sequence
Execute this sequence to exit ISP mode and resume CPU execution mode.
1. Drive SCK low.
1. Wait at least tSSD and drive SS high.
2. Tristate MOSI.
3. Wait at least tSSZ and bring RST high.
4. Tristate SCK.
5. Wait tRHZ and tristate SS.
Figure 25-10. In-System Programming (ISP) Exit Sequence
VDD
XTAL1
RST
tSSZ
SS
SCK
Note:
25.9.5
tRHZ
tSSD
MISO
HIGH Z
MOSI
HIGH Z
The waveforms on this page are not to scale.
Serial Peripheral Interface
The Serial Peripheral Interface (SPI) is a byte-oriented full-duplex synchronous serial communication channel. During In-System Programming, the programmer always acts as the SPI master
and the target device always acts as the SPI slave. The target device receives serial data on
MOSI and outputs serial data on MISO. The Programming Interface implements a standard
SPI Port with a fixed data order and For In-System Programming, bytes are transferred MSB
first as shown in Figure 25-11. The SCK phase and polarity follow SPI clock mode 0 (CPOL = 0,
167
3706A–MICRO–9/09
CPHA = 0) where bits are sampled on the rising edge of SCK and output on the falling edge of
SCK. For more detailed timing information see Figure 25-12.
Figure 25-11. ISP Byte Sequence
SCK
MOSI
7
6
5
4
3
2
1
0
MISO
7
6
5
4
3
2
1
0
Data Sampled
Figure 25-12. Serial Programming Interface Timing
SS
tSCK
tSSE
tSHSL
SCK
tSR
tSSD
tSF
tSLSH
tSOV
tSOE
tSOX
tSOH
MISO
tSIS
tSIH
MOSI
Figure 25-13. Parallel Programming Interface Timing
SS
tSCK
tSSE
tSHSL
SCK
tSR
tSF
tSSD
tSLSH
OE
tPIS
tPIH
tPOE
tPOV
tPOH
tPOX
P0
168
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
25.9.6
Timing Parameters
The timing parameters for Figure 25-7, Figure 25-8, Figure 25-9, Figure 25-10, Figure 25-12 and
Figure 25-13 are shown in Table .
Table 25-7.
Symbol
Parameter
Min
Max
Units
60
ns
tCLCL
System Clock Cycle Time
0
tPWRUP
Power On to SS High Time
10
tPOR
Note:
Programming Interface Timing Parameters
Power-on Reset Time
µs
100
µs
1
µs
2 tCLCL
ns
tPWRDN
SS Tristate to Power Off
tRLZ
RST Low to I/O Tristate
tCLCL
tSTL
RST Low Settling Time
100
tRHZ
RST High to SS Tristate
0
tSCK
Serial Clock Cycle Time
200(1)
ns
tSHSL
Clock High Time
75
ns
tSLSH
Clock Low Time
50
ns
ns
2 tCLCL
ns
tSR
Rise Time
25
ns
tSF
Fall Time
25
ns
tSIS
Serial Input Setup Time
10
ns
tSIH
Serial Input Hold Time
10
ns
tSOH
Serial Output Hold Time
10
ns
tSOV
Serial Output Valid Time
35
ns
tPIS
Parallel Input Setup Time
10
ns
tPIH
Parallel Input Hold Time
10
ns
tPOH
Parallel Output Hold Time
10
ns
tPOV
Parallel Output Valid Time
35
ns
tSOE
Serial Output Enable Time
10
ns
tSOX
Serial Output Disable Time
25
ns
tPOE
Parallel Output Enable Time
10
ns
tPOX
Parallel Output Disable Time
25
ns
tSSE
SS Enable Lead Time
tSLSH
ns
tSSD
SS Disable Lag Time
tSLSH
ns
tZSS
SCK Setup to SS Low
25
ns
tSSZ
SCK Hold after SS High
25
ns
tWR
Write Cycle Time
2.5
ms
tAWR
Write Cycle with Auto-Erase Time
5
ms
7.5
ms
tERS
Chip Erase Cycle Time
1. tSCK is independent of tCLCL.
169
3706A–MICRO–9/09
26. Electrical Characteristics
26.1
Absolute Maximum Ratings*
Operating Temperature ................................... -40°C to +85°C
*NOTICE:
Storage Temperature ..................................... -65°C to +150°C
Voltage on Any Pin with Respect to Ground......-0.7V to +3.6V
Maximum Operating Voltage ............................................ 3.6V
DC Output Current...................................................... 15.0 mA
26.2
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 any
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.
DC Characteristics
TA = -40°C to 85°C, VDD = 2.4V to 3.6V (unless otherwise noted)
Symbol
Parameter
VIL
Input Low-voltage
VIH
Input High-voltage
Condition
(1)
VOL
Output Low-voltage
VOH
Output High-voltage
With Weak Pull-ups Enabled
Min
Max
Units
-0.5
0.2 VDD - 0.1
V
0.2 VDD + 0.9
VDD + 0.5
V
0.5
V
IOL = 10 mA, VDD = 2.7V, TA = 85°C
IOH = -80 µA, VDD = 3V ± 10%
2.4
V
IOH = -30 µA
0.75 VDD
V
IOH = -12 µA
0.9 VDD
V
VOH1
Output High-voltage
With Strong Pull-ups Enabled
IOH = -10 mA, TA = 85°C
0.9 VDD
IOH = -5 mA, TA = 85°C
0.75 VDD
IIL
Logic 0 Input Current
VIN = 0.45V
-50
µA
ITL
Logic 1 to 0 Transition Current
VIN = 2V, VDD = 3V ± 10%
-750
µA
ILI
Input Leakage Current
0 < VIN < VDD
±10
µA
RRST
Reset Pull-up Resistor
150
kΩ
CIO
Pin Capacitance
Test Freq. = 1 MHz, TA = 25°C
10
pF
Active Mode, 12 MHz, VDD = 3.6V
7
mA
Idle Mode, 12 MHz, VDD = 3.6V
P1.0 & P1.1 = 0V or VDD
3
mA
VDD = 3.6V, P1.0 & P1.1 = 0V or VDD
5
µA
VDD = 3V, P1.0 & P1.1 = 0V or VDD
2
µA
Power Supply Current
ICC
Power-down Mode(2)
Notes:
50
1. Under steady state (non-transient) conditions, IOL must be externally limited as follows:
Maximum IOL per port pin: 10 mA
Maximum total IOL for all output pins: 15 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 conditions.
2. Minimum VDD for Power-down is 2V.
3. All characteristics contained in this datasheet are based on simulation and characterization of other microcontrollers manufactured in the same process technology. These values are preliminary values representing design targets, and will be
updated after characterization of actual silicon.
170
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
26.3
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 quasi-bidirectional (with internal pull-ups). A square wave generator with rail-to-rail output is used as an
external clock source for consumption versus frequency measurements.
26.3.1
Supply Current (Internal Oscillator)
Figure 26-1. Active Supply Current vs. Vcc (8MHz Internal Oscillator)
Active Supply Current vs. Vcc
8MHz Internal Oscillator
9
85C
8
-40C
Icc (mA)
7
25C
6
5
4
3
2
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
Vcc (V)
Figure 26-2. Idle Supply Current vs. Vcc (8MHz Internal Oscillator)
Idle Supply Current vs. Vcc
8MHz Internal Oscillator
Icc (mA)
3.0
85C
2.5
-40C
2.0
25C
1.5
1.0
0.5
0.0
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
Vcc (V)
171
3706A–MICRO–9/09
26.3.2
Supply Current (External Clock)
Figure 26-3. Active Supply Current vs. Frequency
Active Supply Current vs. Frequency
External Clock Source
20
5.5V
18
5.0V
16
Icc (mA)
14
4.5V
12
3.6V
10
8
3.0V
6
2.4V
4
2
0
0
5
10
15
20
25
Frequency (MHz)
Figure 26-4. Idle Supply Current vs. Frequency
Idle Supply Current vs. Frequency
External Clock Source
7
5.5V
6
5.0V
Icc (mA)
5
4.5V
4
3.6V
3
3.0V
2
2.4V
1
0
0
5
10
15
20
25
Frequency (MHz)
Note:
172
All characteristics contained in this datasheet are based on simulation and characterization of
other microcontrollers manufactured in the same process technology. These values are preliminary values representing design targets, and will be updated after characterization of actual
silicon.
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
26.4
Clock Characteristics
The values shown in this table are valid for TA = -40°C to 85°C and VDD = 2.4 to 3.6V, unless otherwise noted.
Figure 26-5. External Clock Drive Waveform
Table 26-1.
External Clock Parameters
VDD = 2.0V to 3.6V
VDD = 2.4V to 3.6V
Min
Min
Max
Units
20
MHz
Max
Symbol
Parameter
1/tCLCL
Oscillator Frequency
0
tCLCL
Clock Period
50
ns
tCHCX
External Clock High Time
12
ns
tCLCX
External Clock Low Time
12
ns
tCLCH
External Clock Rise Time
5
ns
tCHCL
External Clock Fall Time
5
ns
Min
Max
Units
Low Speed Oscillator
10
100
kHz
High Speed Oscillator
0.5
24
MHz
TA = 25°C; VDD = 3.0V
7.92
8.08
MHz
VDD = 2.4 to 3.6V
7.60
8.40
MHz
Table 26-2.
Clock Characteristics
Symbol
Parameter
fXTAL
Crystal Oscillator Frequency
fRC
Internal Oscillator Frequency
26.5
Condition
Reset Characteristics
The values shown in this table are valid for TA = -40°C to 85°C and VDD = 2.4 to 3.6V, unless otherwise noted.
Table 26-3.
Reset Characteristics
Symbol
Parameter
RRST
Condition
Min
Max
Units
Reset Pull-up Resistor
50
150
kΩ
VPOR
Power-On Reset Threshold
1.3
1.6
V
VBOD
Brown-Out Detector Threshold
1.9
2.2
V
VBH
Brown-Out Detector Hysteresis
200
300
mV
tPOR
Power-On Reset Delay
135
150
µs
tWDTRST
Watchdog Reset Pulse Width
16tCLCL
ns
173
3706A–MICRO–9/09
26.6
External Data Memory Characteristics
The values shown in this table are valid for TA = -40°C to 85°C and VDD = 2.4 to 3.6V, unless otherwise noted. Under operating conditions, load capacitance for Port 0 and ALE = 100 pF; load capacitance for all other outputs = 80 pF. Parameters
refer to Figure 26-6 and Figure 26-7.
Table 26-4.
External Data Memory Characteristics
Variable Oscillator
Symbol
Parameter
1/tCLCL
Oscillator Frequency
tLHLL
ALE Pulse Width(3)
tAVLL
Min
Max
Units
0
24
MHz
tCLCL - d
Address Valid to ALE Low
ns
(1)
ns
(2)
0.5tCLCL - d
ns
tCLCL - d
ns
tCLCL - d
ns
0.5tCLCL - d
tLLAX
Address Hold after ALE Low
tRLRH
RD Pulse Width(4)
(4)
tWLWH
WR Pulse Width
tRLDV
RD Low to Valid Data In
tRHDX
Data Hold after RD
tRHDZ
Data Float after RD
tCLCL - d
tLLDV
ALE Low to Valid Data In
2tCLCL - d
tCLCL - d
ns
0
ns
ns
ns
(1)
tAVDV
Address to Valid Data In
tLLWL
ALE Low to RD or WR Low
tCLCL - d
tAVWL
Address to RD or WR Low
1.5tCLCL - d(1)
ns
tQVWX
Data Valid to WR Transition
0.5tCLCL - d(1)
ns
(1)
ns
(2)
ns
tQVWH
Data Valid to WR High
tWHQX
Data Hold after WR
tRLAZ
RD Low to Address Float
tWHAX
tWHLH
Notes:
ns
tCLCL + d
ns
1.5tCLCL - d
0.5tCLCL - d
-0.5tCLCL + d(1)
Address Hold after RD or WR High
RD or WR High to ALE High
2.5tCLCL - d
(5)
(2)
0.5tCLCL - d
tCLCL - d
ns
ns
tCLCL + d
ns
1. This assumes 50% clock duty cycle. The half period depends on the clock high value tCHCX (high duty cycle).
2. This assumes 50% clock duty cycle. The half period depends on the clock low value tCLCX (low duty cycle).
3. Parameter tLHLL applies only when ALES = 1.
4. The strobe pulse width may be lengthened by 1, 2 or 3 additional tCLCL using wait states.
5. Parameter tWHLH applies only when ALES = 0, or when two MOVX instructions occur in succession.
174
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Figure 26-6. External Data Memory Read Cycle
tLHLL
ALE
tLLDV
tLLWL
tWHLH
tRLRH
RD
tRLAZ
tAVLL
PORT 0
tRLDV
tLLAX
tRHDX
DATA IN
A0 - A7
tAVWL
tAVDV
PORT 2
P2
tRHDZ
tWHAX
A8 - A15 FROM DPH OR P2.0 - P2.7
P2
Figure 26-7. External Data Memory Write Cycle
tLHLL
ALE
tLLWL
tWLWH
tWHLH
WR
tQVWX
tAVLL
PORT 0
tWHQX
tLLAX
A0 - A7
DATA OUT
tQVWH
tAVWL
PORT 2
26.7
P2
tWHAX
A8 - A15 FROM DPH OR P2.0 - P2.7
P2
Serial Peripheral Interface Timing
The values shown in these tables are valid for TA = -40°C to 85°C and VDD = 2.4 to 3.6V, unless otherwise noted.
Table 26-5.
SPI Master Characteristics
Symbol
Parameter
Min
tCLCL
Oscillator Period
41.6
ns
tSCK
Serial Clock Cycle Time
4tCLCL
ns
tSHSL
Clock High Time
tSCK/2 - 25
ns
tSLSH
Clock Low Time
tSCK/2 - 25
ns
tSR
Rise Time
25
ns
tSF
Fall Time
25
ns
tSIS
Serial Input Setup Time
10
Max
Units
ns
175
3706A–MICRO–9/09
Table 26-5.
SPI Master Characteristics
Symbol
Parameter
tSIH
Serial Input Hold Time
tSOH
Serial Output Hold Time
10
ns
tSOV
Serial Output Valid Time
35
ns
Max
Units
Table 26-6.
Min
Max
10
Units
ns
SPI Slave Characteristics
Symbol
Parameter
Min
tCLCL
Oscillator Period
41.6
ns
tSCK
Serial Clock Cycle Time
4tCLCL
ns
tSHSL
Clock High Time
1.5 tCLCL - 25
ns
tSLSH
Clock Low Time
1.5 tCLCL - 25
ns
tSR
Rise Time
25
ns
tSF
Fall Time
25
ns
tSIS
Serial Input Setup Time
10
ns
tSIH
Serial Input Hold Time
10
ns
tSOH
Serial Output Hold Time
10
ns
tSOV
Serial Output Valid Time
35
ns
tSOE
Output Enable Time
10
ns
tSOX
Output Disable Time
25
ns
tSSE
Slave Enable Lead Time
10
ns
tSSD
Slave Disable Lag Time
0
ns
Figure 26-8. SPI Master Timing (CPHA = 0)
SS
tSCK
SCK
(CPOL = 0)
SCK
(CPOL = 1)
tSR
tSHSL
tSLSH
tSLSH
tSHSL
tSF
tSIS
tSIH
MISO
tSOH
tSOV
MOSI
176
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Figure 26-9. SPI Slave Timing (CPHA = 0)
SS
tSR
tSCK
tSSE
SCK
(CPOL = 0)
SCK
(CPOL= 1)
tSHSL
tSLSH
tSLSH
tSHSL
tSOV
tSOE
tSSD
tSF
tSOX
tSOH
MISO
tSIS
tSIH
MOSI
Figure 26-10. SPI Master Timing (CPHA = 1)
SS
tSCK
SCK
(CPOL = 0)
SCK
(CPOL = 1)
tSF
tSHSL
tSLSH
tSLSH
tSHSL
tSR
tSIS
tSIH
MISO
tSOV
tSOH
MOSI
Figure 26-11. SPI Slave Timing (CPHA = 1)
SS
tSCK
tSSE
SCK
(CPOL = 0)
SCK
(CPOL = 1)
tSR
tSF
tSHSL
tSLSH
tSLSH
tSHSL
tSOE
tSOV
tSOH
tSSD
tSOX
MISO
tSIS
tSIH
MOSI
177
3706A–MICRO–9/09
26.8
Two-wire Serial Interface Characteristics
Table 26-7 describes the requirements for devices connected to the Two-wire Serial Bus. The AT89LP6440 Two-wire
Serial Interface meets or exceeds these requirements under the noted conditions. The values shown in this table are valid
for TA = -40°C to 85°C and VDD = 2.4 to 3.6V, unless otherwise noted.
Timing symbols refer to Figure 26-12.
Table 26-7.
Two-wire Serial Bus Requirements
Symbol
Parameter
VIL
VIH
Vhys
(1)
Min
Max
Units
Input Low-voltage
-0.5
0.3 VDD
V
Input High-voltage
0.7 VDD
VDD + 0.5
V
–
V
0.4
V
20 + 0.1Cb(3)(2)
300
ns
(3)(2)
250
ns
(2)
ns
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
0.05 VDD
3 mA sink current
10 pF < Cb < 400 pF
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
(2)
0
(3)
20 + 0.1Cb
0
0.1VDD < Vi < 0.9VDD
50
-10
10
µA
–
10
pF
fCK(4) > 16fSCL
0
400
kHz
fSCL ≤100 kHz
V DD – 0.4V
---------------------------3mA
1000ns
------------------Cb
Ω
fSCL > 100 kHz
V DD – 0.4V
---------------------------3mA
300ns
---------------Cb
Ω
fSCL ≤100 kHz
4.0
–
µs
fSCL > 100 kHz
0.6
–
µs
fSCL ≤100 kHz
4.7
–
µs
fSCL > 100 kHz
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
Value of Pull-up resistor
tHD;STA
Notes:
Condition
1. In AT89LP6440, this parameter is characterized and not 100% tested.
2. Required only for fSCL > 100 kHz.
178
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
3. Cb = capacitance of one bus line in pF.
4. fCK = CPU clock frequency
Figure 26-12. Two-wire Serial Bus Timing
tHIGH
tof
tr
tLOW
tLOW
SCL
tSU;STA
tHD;STA
tHD;DAT
tSU;DAT
tSU;STO
SDA
tBUF
26.9
Serial Port Timing: Shift Register Mode
The values in this table are valid for VDD = 2.4V to 3.6V and Load Capacitance = 80 pF.
SMOD1 = 0
Min
SMOD1 = 1
Symbol
Parameter
Max
Min
Max
Units
tXLXL
Serial Port Clock Cycle Time
4tCLCL -15
2tCLCL -15
µs
tQVXH
Output Data Setup to Clock Rising Edge
3tCLCL -15
tCLCL -15
ns
tXHQX
Output Data Hold after Clock Rising Edge
tCLCL -15
tCLCL -15
ns
tXHDX
Input Data Hold after Clock Rising Edge
0
0
ns
tXHDV
Input Data Valid to Clock Rising Edge
15
15
ns
Figure 26-13. Shift Register Mode Timing Waveform
SMOD1 = 0
Clock
Write to SBUF
Output Data
0
1
2
3
4
5
6
7
Clear RI
Input Data
Valid
Valid
Valid
Valid
Valid
Valid
Valid
Valid
SMOD1 = 1
Clock
Write to SBUF
Output Data
0
1
2
3
4
5
6
7
Clear RI
Input Data
Valid
Valid
Valid
Valid
Valid
Valid
Valid
Valid
179
3706A–MICRO–9/09
26.10 Dual Analog Comparator Characteristics
The values shown in this table are valid for TA = -40°C to 85°C and VDD = 2.4 to 3.6V, unless otherwise noted.
Table 26-8.
Dual Analog Comparator Characteristics
Symbol
Parameter
VCM
Common Mode Input Voltage
VOS
Input Offset Voltage
VAREF
Analog Reference Voltage
VΔREF
Reference Delta Voltage
tCMP
Comparator Propagation Delay
tAREF
Reference Settling Time
Condition
Min
Max
Units
GND
VDD
V
20
mV
1.2
1.3
V
70
170
mV
200
ns
VDD = 3.6V
VIN+ – VIN- = 20mV; VDD = 2.4V
3
µs
26.11 DADC Characteristics
The values shown in these tables are valid for TA = -40°C to 85°C and VDD = 2.4 to 3.6V, unless otherwise noted.
Table 26-9.
Symbol
ADC Characteristics
Parameter
Condition
Min
Resolution
Typical
Max
Units
10
Bits
Absolute Accuracy (including
INL, DNL, quantization error,
gain and offset error)
TBD
LSB
Integral Non-Linearity (INL)
TBD
LSB
DIfferential Non-Linearity
(DNL)
TBD
LSB
Gain Error
TBD
LSB
Offset Error
TBD
LSB
tACK
Clock Period
tADC
Conversion Time
AVDD
Analog Supply Voltage
VREF
Reference Voltage
VIN
Single-Ended Input Voltage
VCMI
Differential Input Common
Mode Voltage
VDI
Differential Input Voltage
RIN
Analog Input Resistance
TBD
Ω
RMUX
Analog Mux Resistance
TBD
Ω
CS/H
Sample & Hold Capacitance
TBD
pF
180
500
ns
13tACK
14tACK +
2tCLCL
ns
VDD - 0.3
VDD + 0.3
V
External Reference
AVDD/2 - 0.2
AVDD/2
AVDD/2 + 0.2
V
Internal Reference
0.9
1.0
1.1
V
AVDD/2 - VREF
AVDD/2 + VREF
V
GND
AVDD
V
0
±VREF
V
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
.
Table 26-10. DAC Characteristics
Symbol
Parameter
Condition
Min
Resolution
tACK
Clock Period
tDAC
Conversion Time
AVDD
Analog Supply Voltage
VREF
Reference Voltage
VIN
Single-Ended Input Voltage
VCMO
Differential Output
Common Mode Voltage
VDO
Differential Output Voltage
ROUT
Analog Output Resistance
Typical
Max
10
tACK ≥ tCLCL
Units
Bits
500
ns
11tACK
12tACK +
2tCLCL
ns
VDD - 0.3
VDD + 0.3
V
External Reference
AVDD/2 - 0.2
AVDD/2
AVDD/2 + 0.2
V
Internal Reference
0.9
1.0
1.1
V
AVDD/2 + VREF
V
AVDD/2 + 0.2
V
0
±VREF
V
50
100
kΩ
AVDD/2 - VREF
AVDD/2 - 0.2
AVDD/2
26.12 Test Conditions
26.12.1
Note:
26.12.2
Note:
AC Testing Input/Output Waveform(1)
1. AC Inputs during testing are driven at VDD - 0.5V for a logic “1” and 0.45V for a logic “0”. Timing measurements are made at
VIH min. for a logic “1” and VIL max. for a logic “0”.
Float Waveform(1)
1. For timing purposes, a port pin is no longer floating when a 100 mV change from load voltage occurs. A port pin begins to
float when 100 mV change from the loaded VOH/VOL level occurs.
181
3706A–MICRO–9/09
26.12.3
ICC Test Condition, Active Mode, All Other Pins are Disconnected
VDD
VDD
ICC
RST
XTAL2
(NC)
CLOCK SIGNAL
26.12.4
VDD
XTAL1
GND
ICC Test Condition, Idle Mode, All Other Pins are Disconnected
VDD
VDD
ICC
RST
XTAL2
(NC)
CLOCK SIGNAL
26.12.5
VDD
XTAL1
GND
Clock Signal Waveform for ICC Tests in Active and Idle Modes, tCLCH = tCHCL = 5 ns
VCC - 0.5V
0.45V
0.7 VCC
tCHCX
0.2 VCC - 0.1V
tCHCL
tCLCH
tCHCX
tCLCL
26.12.6
ICC Test Condition, Power-down Mode, All Other Pins are Disconnected, VDD = 2V to 3.6V
VDD
VDD
ICC
RST
(NC)
VDD
XTAL2
XTAL1
GND
182
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
27. Ordering Information
27.1
Green Package Option (Pb/Halide-free)
Speed
(MHz)
20
Power
Supply
2.4V to 3.6V
Ordering Code
Package
AT89LP6440-20AU
AT89LP6440-20PU
AT89LP6440-20JU
AT89LP6440-20MU
44A
40P6
44J
44M1
Operation Range
Industrial
(-40° C to 85° C)
Package Types
44A
44-lead, Thin Plastic Quad Flat Package (TQFP)
40P6
40-lead, 0.600” Wide, Plastic Dual Inline Package (PDIP)
44J
44-lead, Plastic J-leaded Chip Carrier (PLCC)
44M1
44-pad, 7 x 7 x 1.0 mm Body, Plastic Very Thin Quad Flat No Lead Package (VQFN/MLF)
183
3706A–MICRO–9/09
28. Packaging Information
28.1
44A – TQFP
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 ACB.
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
11.75
12.00
12.25
D1
9.90
10.00
10.10
E
11.75
12.00
12.25
E1
9.90
10.00
10.10
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
184
2325 Orchard Parkway
San Jose, CA 95131
TITLE
44A, 44-lead, 10 x 10 mm Body Size, 1.0 mm Body Thickness,
0.8 mm Lead Pitch, Thin Profile Plastic Quad Flat Package (TQFP)
DRAWING NO.
REV.
44A
B
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
28.2
40P6 – PDIP
D
PIN
1
E1
A
SEATING PLANE
A1
L
B
B1
e
E
0º ~ 15º
C
COMMON DIMENSIONS
(Unit of Measure = mm)
REF
MIN
NOM
MAX
A
–
–
4.826
A1
0.381
–
–
D
52.070
–
52.578
E
15.240
–
15.875
E1
13.462
–
13.970
B
0.356
–
0.559
B1
1.041
–
1.651
L
3.048
–
3.556
C
0.203
–
0.381
eB
15.494
–
17.526
SYMBOL
eB
Notes:
1. This package conforms to JEDEC reference MS-011, Variation AC.
2. Dimensions D and E1 do not include mold Flash or Protrusion.
Mold Flash or Protrusion shall not exceed 0.25 mm (0.010").
e
NOTE
Note 2
Note 2
2.540 TYP
09/28/01
R
2325 Orchard Parkway
San Jose, CA 95131
TITLE
40P6, 40-lead (0.600"/15.24 mm Wide) Plastic Dual
Inline Package (PDIP)
DRAWING NO.
40P6
REV.
B
185
3706A–MICRO–9/09
28.3
44J – PLCC
1.14(0.045) X 45˚
PIN NO. 1
1.14(0.045) X 45˚
0.318(0.0125)
0.191(0.0075)
IDENTIFIER
E1
D2/E2
B1
E
B
e
A2
D1
A1
D
A
0.51(0.020)MAX
45˚ MAX (3X)
COMMON DIMENSIONS
(Unit of Measure = mm)
Notes:
1. This package conforms to JEDEC reference MS-018, Variation AC.
2. Dimensions D1 and E1 do not include mold protrusion.
Allowable protrusion is .010"(0.254 mm) per side. Dimension D1
and E1 include mold mismatch and are measured at the extreme
material condition at the upper or lower parting line.
3. Lead coplanarity is 0.004" (0.102 mm) maximum.
SYMBOL
MIN
NOM
MAX
A
4.191
–
4.572
A1
2.286
–
3.048
A2
0.508
–
–
D
17.399
–
17.653
D1
16.510
–
16.662
E
17.399
–
17.653
E1
16.510
–
16.662
D2/E2
14.986
–
16.002
B
0.660
–
0.813
B1
0.330
–
0.533
e
NOTE
Note 2
Note 2
1.270 TYP
10/04/01
R
186
2325 Orchard Parkway
San Jose, CA 95131
TITLE
44J, 44-lead, Plastic J-leaded Chip Carrier (PLCC)
DRAWING NO.
REV.
44J
B
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
28.4
44M1 – VQFN/MLF
D
Marked Pin# 1 ID
E
SEATING PLANE
A1
TOP VIEW
A3
A
K
L
Pin #1 Corner
D2
1
2
3
Option A
SIDE VIEW
Pin #1
Triangle
E2
Option B
K
Option C
b
e
Pin #1
Chamfer
(C 0.30)
Pin #1
Notch
(0.20 R)
BOTTOM VIEW
COMMON DIMENSIONS
(Unit of Measure = mm)
SYMBOL
MIN
NOM
MAX
A
0.80
0.90
1.00
A1
–
0.02
0.05
A3
0.20 REF
b
0.18
0.23
0.30
D
6.90
7.00
7.10
D2
5.00
5.20
5.40
E
6.90
7.00
7.10
E2
5.00
5.20
5.40
e
Note: JEDEC Standard MO-220, Fig. 1 (SAW Singulation) VKKD-3.
NOTE
0.50 BSC
L
0.59
0.64
0.69
K
0.20
0.26
0.41
9/26/08
Package Drawing Contact:
[email protected]
TITLE
44M1, 44-pad, 7 x 7 x 1.0 mm Body, Lead
Pitch 0.50 mm, 5.20 mm Exposed Pad, Thermally
Enhanced Plastic Very Thin Quad Flat No
Lead Package (VQFN)
GPC
ZWS
DRAWING NO.
REV.
44M1
H
187
3706A–MICRO–9/09
29. Revision History
188
Revision No.
History
Revision A – September 2009
•
Initial Release
AT89LP6440 - Preliminary
3706A–MICRO–9/09
AT89LP6440 - Preliminary
Table of Contents
Features ..................................................................................................... 1
1
2
3
Pin Configurations ................................................................................... 2
1.1
40P6: 40-lead PDIP ...........................................................................................2
1.2
44A: 44-lead TQFP (Top View) .........................................................................2
1.3
44J: 44-lead PLCC ............................................................................................3
1.4
44M1: 44-pad VQFN/MLF .................................................................................3
1.5
Pin Description ..................................................................................................4
Overview ................................................................................................... 6
2.1
Block Diagram ...................................................................................................7
2.2
Comparison to Standard 8051 ...........................................................................8
Memory Organization ............................................................................ 10
3.1
Program Memory .............................................................................................10
3.2
Internal Data Memory ......................................................................................11
3.3
External Data Memory .....................................................................................12
3.4
Extended Stack ...............................................................................................19
3.5
In-Application Programming (IAP) ...................................................................20
4
Special Function Registers ................................................................... 21
5
Enhanced CPU ....................................................................................... 22
6
7
5.1
Multiply–Accumulate Unit (MAC) .....................................................................23
5.2
Enhanced Dual Data Pointers .........................................................................24
5.3
Instruction Set Extensions ...............................................................................29
System Clock ......................................................................................... 30
6.1
Crystal Oscillator .............................................................................................30
6.2
External Clock Source .....................................................................................31
6.3
Internal RC Oscillator ......................................................................................31
6.4
System Clock Out ............................................................................................31
6.5
System Clock Divider ......................................................................................31
Reset ....................................................................................................... 32
7.1
Power-on Reset ...............................................................................................32
7.2
Brown-out Reset ..............................................................................................34
7.3
External Reset .................................................................................................34
7.4
Watchdog Reset ..............................................................................................35
i
3706A–MICRO–9/09
7.5
8
9
Software Reset ................................................................................................35
Power Saving Modes ............................................................................. 35
8.1
Idle Mode .........................................................................................................35
8.2
Power-down Mode ...........................................................................................36
8.3
Reducing Power Consumption ........................................................................37
Interrupts ................................................................................................ 38
9.1
Interrupt Response Time .................................................................................40
10 I/O Ports .................................................................................................. 44
10.1
Port Configuration ............................................................................................44
10.2
Port Analog Functions .....................................................................................47
10.3
Port Read-Modify-Write ...................................................................................47
10.4
Port Alternate Functions ..................................................................................48
11 Enhanced Timer 0 and Timer 1 with PWM ........................................... 50
11.1
Mode 0 – Variable Width Timer/Counter .........................................................51
11.2
Mode 1 – 16-bit Auto-Reload Timer/Counter ...................................................51
11.3
Mode 2 – 8-bit Auto-Reload Timer/Counter .....................................................52
11.4
Mode 3 – 8-bit Split Timer ...............................................................................52
11.5
Pulse Width Modulation ...................................................................................55
12 Enhanced Timer 2 .................................................................................. 59
12.1
Timer 2 Registers ............................................................................................60
12.2
Capture Mode ..................................................................................................61
12.3
Auto-Reload Mode ...........................................................................................62
12.4
Baud Rate Generator ......................................................................................66
12.5
Frequency Generator (Programmable Clock Out) ...........................................67
13 Compare/Capture Array ........................................................................ 68
13.1
CCA Registers .................................................................................................69
13.2
Input Capture Mode .........................................................................................71
13.3
Output Compare Mode ....................................................................................74
13.4
Pulse Width Modulation Mode .........................................................................76
14 External Interrupts ................................................................................. 81
15 General-purpose Interrupts .................................................................. 82
3706A–MICRO–9/09
16 Serial Interface (UART) .......................................................................... 84
16.1
Multiprocessor Communications .....................................................................84
16.2
Baud Rates ......................................................................................................86
16.3
More About Mode 0 .........................................................................................88
16.4
More About Mode 1 .........................................................................................91
16.5
More About Modes 2 and 3 .............................................................................93
16.6
Framing Error Detection ..................................................................................96
16.7
Automatic Address Recognition ......................................................................96
17 Enhanced Serial Peripheral Interface .................................................. 97
17.1
Master Operation .............................................................................................99
17.2
Slave Operation .............................................................................................100
17.3
Pin Configuration ...........................................................................................100
17.4
Serial Clock Timing ........................................................................................103
18 Two-Wire Serial Interface .................................................................... 104
18.1
Data Transfer and Frame Format ..................................................................105
18.2
Multi-master Bus Systems, Arbitration and Synchronization .........................107
18.3
Overview of the TWI Module .........................................................................109
18.4
Register Overview .........................................................................................111
18.5
Using the TWI ................................................................................................112
18.6
Transmission Modes .....................................................................................114
19 Dual Analog Comparators ................................................................... 126
19.1
Analog Input Muxes .......................................................................................127
19.2
Internal Reference Voltage ............................................................................128
19.3
Comparator Interrupt Debouncing .................................................................128
20 Digital-to-Analog/Analog-to-Digital Converter .................................. 133
3706A–MICRO–9/09
20.1
ADC Operation ..............................................................................................135
20.2
DAC Operation ..............................................................................................136
20.3
Clock Selection ..............................................................................................137
20.4
Starting a Conversion ....................................................................................137
20.5
Noise Considerations ....................................................................................138
21 Programmable Watchdog Timer ......................................................... 141
21.1
Software Reset ..............................................................................................142
22 Instruction Set Summary .................................................................... 143
22.1
Instruction Set Extensions .............................................................................147
23 Register Index ...................................................................................... 153
24 On-Chip Debug System ....................................................................... 155
24.1
Physical Interface ..........................................................................................155
24.2
Software Breakpoints ....................................................................................156
24.3
Limitations of On-Chip Debug .......................................................................156
25 Programming the Flash Memory ........................................................ 157
25.1
Physical Interface ..........................................................................................157
25.2
Memory Organization ....................................................................................159
25.3
Command Format ..........................................................................................160
25.4
Status Register ..............................................................................................163
25.5
DATA Polling .................................................................................................163
25.6
Flash Security ................................................................................................163
25.7
User Configuration Fuses ..............................................................................164
25.8
User Signature and Analog Configuration .....................................................165
25.9
Programming Interface Timing ......................................................................165
26 Electrical Characteristics .................................................................... 170
26.1
Absolute Maximum Ratings* .........................................................................170
26.2
DC Characteristics .........................................................................................170
26.3
Typical Characteristics ..................................................................................171
26.4
Clock Characteristics .....................................................................................173
26.5
Reset Characteristics ....................................................................................173
26.6
External Data Memory Characteristics ..........................................................174
26.7
Serial Peripheral Interface Timing .................................................................175
26.8
Two-wire Serial Interface Characteristics ......................................................178
26.9
Serial Port Timing: Shift Register Mode ........................................................179
26.10
Dual Analog Comparator Characteristics ......................................................180
26.11
DADC Characteristics ....................................................................................180
26.12
Test Conditions ..............................................................................................181
3706A–MICRO–9/09
27 Ordering Information ........................................................................... 183
27.1
Green Package Option (Pb/Halide-free) ........................................................183
28 Packaging Information ........................................................................ 184
28.1
44A – TQFP ...................................................................................................184
28.2
40P6 – PDIP ..................................................................................................185
28.3
44J – PLCC ...................................................................................................186
28.4
44M1 – VQFN/MLF .......................................................................................187
29 Revision History ................................................................................... 188
3706A–MICRO–9/09
Headquarters
International
Atmel Corporation
2325 Orchard Parkway
San Jose, CA 95131
USA
Tel: 1(408) 441-0311
Fax: 1(408) 487-2600
Atmel Asia
Unit 1-5 & 16, 19/F
BEA Tower, Millennium City 5
418 Kwun Tong Road
Kwun Tong, Kowloon
Hong Kong
Tel: (852) 2245-6100
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.
© 2009 Atmel Corporation. All rights reserved. Atmel ®, Atmel logo and combinations thereof, and others are registered trademarks or trademarks of Atmel Corporation or its subsidiaries. Other terms and product names may be trademarks of others.
3706A–MICRO–9/09