AVR128DB28/32/48/64
AVR® DB Family
Introduction
The AVR128DB28/32/48/64 microcontrollers of the AVR® DB family of microcontrollers are using the AVR® CPU with
hardware multiplier running at clock speeds up to 24 MHz. They come with 128 KB of Flash, 16 KB of SRAM, and
512 bytes of EEPROM. The microcontrollers are available in 28-, 32-, 48- and 64- pin packages. The AVR® DB
family uses the latest technologies from Microchip with a flexible and low-power architecture, including Event System,
accurate analog subsystems, and advanced digital peripherals.
AVR® DB Family Overview
The figure below shows the AVR DB devices, laying out pin count variants and memory sizes:
•
•
Vertical migration is possible without code modification, as these devices are fully pin and feature compatible.
Horizontal migration to the left reduces the pin count and therefore the available features.
Figure 1. AVR® DB Family Overview
Devices described in this data sheet
Devices described in other data sheets
Flash
128 KB
AVR128DB28
AVR128DB32
AVR128DB48
AVR128DB64
64 KB
AVR64DB28
AVR64DB32
AVR64DB48
AVR64DB64
32 KB
AVR32DB28
AVR32DB32
AVR32DB48
28
32
48
64
Pins
The name of a device in the AVR DB family is decoded as follows:
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 1
AVR128DB28/32/48/64
Figure 2. AVR® DB Device Designations
Memory Overview
The following table shows the memory overview of the entire family, but the further documentation describes only the
AVR128DB28/32/48/64 devices.
Table 1. Memory Overview
AVR32DB28
AVR32DB32
AVR32DB48
AVR64DB28
AVR64DB32
AVR64DB48
AVR64DB64
AVR128DB28
AVR128DB32
AVR128DB48
AVR128DB64
Flash memory
32 KB
64 KB
128 KB
SRAM
4 KB
8 KB
16 KB
EEPROM
512B
512B
512B
User row
32B
32B
32B
Devices
Peripheral Overview
The following table shows the peripheral overview of the entire AVR DB family, but the further documentation
describes only the AVR128DB28/32/48/64 devices.
Table 2. Peripheral Overview
Feature
AVR32DB28
AVR64DB28
AVR128DB28
AVR32DB32
AVR64DB32
AVR128DB32
AVR32DB48
AVR64DB48
AVR128DB48
AVR64DB64
AVR128DB64
Pins
28
32
48
64
Max. frequency (MHz)
24
24
24
24
16-bit Timer/Counter type A (TCA)
1
1
2
2
16-bit Timer/Counter type B (TCB)
3
3
4
5
12-bit Timer/Counter type D (TCD)
1
1
1
1
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 2
AVR128DB28/32/48/64
...........continued
Feature
AVR32DB28
AVR64DB28
AVR128DB28
AVR32DB32
AVR64DB32
AVR128DB32
AVR32DB48
AVR64DB48
AVR128DB48
AVR64DB64
AVR128DB64
Pins
28
32
48
64
Real-Time Counter (RTC)
1
1
1
1
USART
3
3
5
6
SPI
2
2
2
2
TWI/I2C
1(1)
2(1)
2(1)
2(1)
12-bit differential ADC (channels)
1 (9)
1 (13)
1 (18)
1 (22)
10-bit DAC (outputs)
1 (1)
1 (1)
1 (1)
1 (1)
Analog Comparator (AC)
3
3
3
3
Zero-Cross Detector (ZCD)
1
1
2
3
Peripheral Touch Controller (PTC)
-
-
-
-
Op amp (OP)
2
2
3
3
Configurable Custom Logic Look-up
Table (CCL LUT)
4
4
6
6
Watchdog Timer (WDT)
1
1
1
1
Event System channels (EVSYS)
8
8
10
10
22/21(2)
26/25(2)
41/40(2)
55/54(2)
PA[7:0], PC[3:0],
PD[7:1],
PF[6,1,0]
PA[7:0], PC[3:0],
PD[7:1], PF[6:0]
PA[7:0], PB[5:0],
PC[7:0], PD[7:0],
PE[3:0], PF[6:0]
PA[7:0], PB[7:0],
PC[7:0], PD[7:0],
PE[7:0], PF[6:0],
PG[7:0]
External Interrupts
22
26
41
55
CRCSCAN
1
1
1
1
Unified Program and Debug Interface
(UPDI)
1
1
1
1
General Purpose
I/O(2)
PORT
Notes:
1. The TWI/I2C can operate simultaneously as master and slave on different pins.
2. PF6/RESET pin is input only.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 3
AVR128DB28/32/48/64
Features
•
•
•
•
AVR® CPU
– Running at up to 24 MHz
– Single-cycle I/O register access
– Two-level interrupt controller
– Two-cycle hardware multiplier
– Supply voltage range: 1.8V to 5.5V
Memories
– 128 KB In-System self-programmable Flash memory
– 512B EEPROM
– 16 KB SRAM
– 32B of user row in nonvolatile memory that can keep data during chip-erase and be programmed while the
device is locked
– Write/erase endurance
• Flash: 10,000 cycles
• EEPROM: 100,000 cycles
– Data retention: 40 Years at 55°C
System
– Power-on Reset (POR) circuit
– Brown-out Detector (BOD) with user-programmable levels
– Voltage Level Monitor (VLM) with interrupt at a programmable level above the BOD level
– Clock failure detection
– Clock options
• High-precision internal oscillator with selectable frequency up to 24 MHz (OSCHF)
– Auto-tuning for improved internal oscillator accuracy
• Internal PLL up to 48 MHz for high-frequency operation of Timer/Counter type D (PLL)
• Internal ultra-low power 32.768 kHz oscillator (OSC32K)
• External 32.768 kHz crystal oscillator (XOSC32K)
• External clock input
• External high-frequency crystal oscillator (XOSCHF) with clock failure detection
– Single pin Unified Program and Debug Interface (UPDI)
– Three sleep modes
• Idle with all peripherals running for immediate wake-up
• Standby
– Configurable operation of selected peripherals
– SleepWalking peripherals
• Power-Down with full data retention
Peripherals
– Up to two 16-bit Timer/Counters type A (TCA) with three compare channels for PWM and waveform
generation
– Up to five 16-bit Timer/Counters type B (TCB) with input capture for capture and signal measurements
– One 12-bit PWM Timer/Counter type D (TCD) optimized for power control
– One 16-bit Real-Time Counter (RTC) that can run from external crystal or internal oscillator
– Up to six USARTs
• Operation modes: RS-485, LIN slave, master SPI, and IrDA
• Fractional baud rate generator, auto-baud, and start-of-frame detection
– Two SPIs with master/slave operation modes
– Up to two Two-Wire Interface (TWI) with dual address match
• Independent master and slave operation Dual mode)
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 4
AVR128DB28/32/48/64
•
•
• Phillips I2C compatible
• Standard mode (Sm, 100 kHz)
• Fast mode (Fm, 400 kHz)
• Fast mode plus (Fm+, 1 MHz)(1)
– Event System for CPU-independent and predictable inter-peripherals signaling
– Configurable Custom Logic (CCL) with up to six programmable Look-up Tables (LUTs)
– One 12-bit 130 ksps differential Analog-to-Digital Converter (ADC)
– Three Analog Comparators (ACs) with window compare functions
– One 10-bit Digital-to-Analog Converter (DAC)
– Up to three Zero-Cross Detectors (ZCDs)
– Analog Signal Conditioning (OPAMP) peripheral with up to three op amps, each with an internal resistor
ladder that allows for many useful configurations with no external components
– Multiple voltage references (VREF)
• 1.024V
• 2.048V
• 2.500V
• 4.096V
• External Voltage Reference (VREFA)
• Supply Voltage (VDD)
– Automated Cyclic Redundancy Check (CRC) Flash program memory scan
– Watchdog Timer (WDT) with Window mode, and separate on-chip oscillator
– External interrupt on all general purpose pins
I/O and Packages
– Multi-Voltage I/O (MVIO) on I/O port C
– Selectable input voltage threshold
– Up to 55/54 programmable I/O pins
– 28-pin SSOP, SOIC and SPDIP
– 32-pin VQFN 5x5 mm and TQFP 7x7 mm
– 48-pin VQFN 5x5 mm and TQFP 7x7 mm
– 64-pin VQFN 9x9 mm and TQFP 10x10 mm
Temperature Ranges
– Industrial: -40°C to 85°C
– Extended: -40°C to 125°C
Note:
1. I2C Fm+ is only supported for supply voltage VDD above 2.7 VDC.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 5
AVR128DB28/32/48/64
Table of Contents
Introduction.....................................................................................................................................................1
AVR® DB Family Overview............................................................................................................................ 1
1.
2.
Memory Overview........................................................................................................................ 2
Peripheral Overview..................................................................................................................... 2
Features......................................................................................................................................................... 4
1.
Block Diagram.......................................................................................................................................13
2.
Pinout.................................................................................................................................................... 14
2.1.
2.2.
2.3.
2.4.
3.
I/O Multiplexing and Considerations..................................................................................................... 18
3.1.
4.
Numerical Notation.....................................................................................................................29
Memory Size and Type...............................................................................................................29
Frequency and Time...................................................................................................................29
Registers and Bits...................................................................................................................... 30
ADC Parameter Definitions........................................................................................................ 31
AVR® CPU............................................................................................................................................ 34
7.1.
7.2.
7.3.
7.4.
7.5.
7.6.
8.
Power Domains.......................................................................................................................... 27
Voltage Regulator.......................................................................................................................27
Power-Up................................................................................................................................... 28
Conventions.......................................................................................................................................... 29
6.1.
6.2.
6.3.
6.4.
6.5.
7.
General Guidelines.....................................................................................................................21
Connection for Power Supply.....................................................................................................21
Connection for RESET............................................................................................................... 23
Connection for UPDI Programming............................................................................................23
Connecting External Crystal Oscillators..................................................................................... 24
Connection for External Voltage Reference............................................................................... 25
Power Supply........................................................................................................................................ 27
5.1.
5.2.
5.3.
6.
I/O Multiplexing...........................................................................................................................18
Hardware Guidelines.............................................................................................................................21
4.1.
4.2.
4.3.
4.4.
4.5.
4.6.
5.
28-pin SSOP, SOIC and SPDIP................................................................................................. 14
32-pin VQFN and TQFP.............................................................................................................15
48-pin VQFN and TQFP.............................................................................................................16
64-pin VQFN and TQFP.............................................................................................................17
Features..................................................................................................................................... 34
Overview.................................................................................................................................... 34
Architecture................................................................................................................................ 34
Functional Description................................................................................................................36
Register Summary......................................................................................................................40
Register Description................................................................................................................... 40
Memories.............................................................................................................................................. 45
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 6
AVR128DB28/32/48/64
8.1.
8.2.
8.3.
8.4.
8.5.
8.6.
8.7.
8.8.
8.9.
8.10.
9.
Overview.................................................................................................................................... 45
Memory Map.............................................................................................................................. 45
In-System Reprogrammable Flash Program Memory................................................................45
SRAM Data Memory.................................................................................................................. 46
EEPROM Data Memory............................................................................................................. 46
SIGROW - Signature Row..........................................................................................................47
USERROW - User Row..............................................................................................................51
FUSE - Configuration and User Fuses.......................................................................................51
LOCK - Memory Sections Access Protection.............................................................................59
I/O Memory.................................................................................................................................62
GPR - General Purpose Registers........................................................................................................65
9.1.
9.2.
Register Summary......................................................................................................................66
Register Description................................................................................................................... 66
10. Peripherals and Architecture.................................................................................................................68
10.1. Peripheral Address Map.............................................................................................................68
10.2. Interrupt Vector Mapping............................................................................................................ 70
10.3. SYSCFG - System Configuration............................................................................................... 72
11. NVMCTRL - Nonvolatile Memory Controller......................................................................................... 75
11.1.
11.2.
11.3.
11.4.
11.5.
Features..................................................................................................................................... 75
Overview.................................................................................................................................... 75
Functional Description................................................................................................................76
Register Summary......................................................................................................................84
Register Description................................................................................................................... 84
12. CLKCTRL - Clock Controller................................................................................................................. 92
12.1.
12.2.
12.3.
12.4.
12.5.
Features..................................................................................................................................... 92
Overview.................................................................................................................................... 92
Functional Description................................................................................................................94
Register Summary....................................................................................................................100
Register Description................................................................................................................. 100
13. SLPCTRL - Sleep Controller............................................................................................................... 115
13.1.
13.2.
13.3.
13.4.
13.5.
Features................................................................................................................................... 115
Overview...................................................................................................................................115
Functional Description.............................................................................................................. 115
Register Summary....................................................................................................................120
Register Description................................................................................................................. 120
14. RSTCTRL - Reset Controller.............................................................................................................. 123
14.1.
14.2.
14.3.
14.4.
14.5.
Features................................................................................................................................... 123
Overview.................................................................................................................................. 123
Functional Description..............................................................................................................124
Register Summary....................................................................................................................127
Register Description................................................................................................................. 127
15. CPUINT - CPU Interrupt Controller..................................................................................................... 130
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 7
AVR128DB28/32/48/64
15.1.
15.2.
15.3.
15.4.
15.5.
Features................................................................................................................................... 130
Overview.................................................................................................................................. 130
Functional Description..............................................................................................................131
Register Summary ...................................................................................................................136
Register Description................................................................................................................. 136
16. EVSYS - Event System.......................................................................................................................141
16.1.
16.2.
16.3.
16.4.
16.5.
Features................................................................................................................................... 141
Overview.................................................................................................................................. 141
Functional Description..............................................................................................................142
Register Summary....................................................................................................................148
Register Description................................................................................................................. 148
17. PORTMUX - Port Multiplexer.............................................................................................................. 155
17.1. Overview.................................................................................................................................. 155
17.2. Register Summary....................................................................................................................156
17.3. Register Description................................................................................................................. 156
18. PORT - I/O Pin Configuration..............................................................................................................170
18.1.
18.2.
18.3.
18.4.
18.5.
18.6.
18.7.
Features................................................................................................................................... 170
Overview.................................................................................................................................. 170
Functional Description..............................................................................................................172
Register Summary - PORTx.....................................................................................................176
Register Description - PORTx.................................................................................................. 176
Register Summary - VPORTx.................................................................................................. 193
Register Description - VPORTx................................................................................................193
19. MVIO - Multi-Voltage I/O..................................................................................................................... 198
19.1.
19.2.
19.3.
19.4.
19.5.
Features................................................................................................................................... 198
Overview.................................................................................................................................. 198
Functional Description..............................................................................................................199
Register Summary....................................................................................................................202
Register Description................................................................................................................. 202
20. BOD - Brown-out Detector.................................................................................................................. 206
20.1.
20.2.
20.3.
20.4.
20.5.
Features................................................................................................................................... 206
Overview.................................................................................................................................. 206
Functional Description..............................................................................................................207
Register Summary....................................................................................................................209
Register Description................................................................................................................. 209
21. VREF - Voltage Reference..................................................................................................................216
21.1.
21.2.
21.3.
21.4.
21.5.
Features................................................................................................................................... 216
Overview.................................................................................................................................. 216
Functional Description..............................................................................................................216
Register Summary....................................................................................................................217
Register Description................................................................................................................. 217
22. WDT - Watchdog Timer ......................................................................................................................221
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 8
AVR128DB28/32/48/64
22.1.
22.2.
22.3.
22.4.
22.5.
Features................................................................................................................................... 221
Overview.................................................................................................................................. 221
Functional Description..............................................................................................................221
Register Summary....................................................................................................................225
Register Description................................................................................................................. 225
23. TCA - 16-bit Timer/Counter Type A.....................................................................................................229
23.1.
23.2.
23.3.
23.4.
23.5.
23.6.
23.7.
Features................................................................................................................................... 229
Overview.................................................................................................................................. 229
Functional Description..............................................................................................................232
Register Summary - Normal Mode...........................................................................................242
Register Description - Normal Mode........................................................................................ 242
Register Summary - Split Mode............................................................................................... 261
Register Description - Split Mode.............................................................................................261
24. TCB - 16-bit Timer/Counter Type B.....................................................................................................277
24.1.
24.2.
24.3.
24.4.
24.5.
Features................................................................................................................................... 277
Overview.................................................................................................................................. 277
Functional Description..............................................................................................................279
Register Summary....................................................................................................................289
Register Description................................................................................................................. 289
25. TCD - 12-Bit Timer/Counter Type D.................................................................................................... 300
25.1.
25.2.
25.3.
25.4.
25.5.
Features................................................................................................................................... 300
Overview.................................................................................................................................. 300
Functional Description..............................................................................................................302
Register Summary....................................................................................................................325
Register Description................................................................................................................. 325
26. RTC - Real-Time Counter................................................................................................................... 350
26.1. Features................................................................................................................................... 350
26.2. Overview.................................................................................................................................. 350
26.3. Clocks.......................................................................................................................................351
26.4. RTC Functional Description..................................................................................................... 351
26.5. PIT Functional Description....................................................................................................... 352
26.6. Crystal Error Correction............................................................................................................353
26.7. Events...................................................................................................................................... 353
26.8. Interrupts.................................................................................................................................. 354
26.9. Sleep Mode Operation............................................................................................................. 355
26.10. Synchronization........................................................................................................................355
26.11. Debug Operation...................................................................................................................... 355
26.12. Register Summary................................................................................................................... 356
26.13. Register Description.................................................................................................................356
27. USART - Universal Synchronous and Asynchronous Receiver and Transmitter................................373
27.1.
27.2.
27.3.
27.4.
Features................................................................................................................................... 373
Overview.................................................................................................................................. 373
Functional Description..............................................................................................................374
Register Summary....................................................................................................................389
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 9
AVR128DB28/32/48/64
27.5. Register Description................................................................................................................. 389
28. SPI - Serial Peripheral Interface..........................................................................................................407
28.1.
28.2.
28.3.
28.4.
28.5.
Features................................................................................................................................... 407
Overview.................................................................................................................................. 407
Functional Description..............................................................................................................408
Register Summary....................................................................................................................415
Register Description................................................................................................................. 415
29. TWI - Two-Wire Interface.................................................................................................................... 422
29.1.
29.2.
29.3.
29.4.
29.5.
Features................................................................................................................................... 422
Overview.................................................................................................................................. 422
Functional Description..............................................................................................................423
Register Summary....................................................................................................................434
Register Description................................................................................................................. 434
30. CRCSCAN - Cyclic Redundancy Check Memory Scan...................................................................... 452
30.1.
30.2.
30.3.
30.4.
30.5.
Features................................................................................................................................... 452
Overview.................................................................................................................................. 452
Functional Description..............................................................................................................452
Register Summary....................................................................................................................455
Register Description................................................................................................................. 455
31. CCL – Configurable Custom Logic......................................................................................................459
31.1.
31.2.
31.3.
31.4.
31.5.
Features................................................................................................................................... 459
Overview.................................................................................................................................. 459
Functional Description..............................................................................................................461
Register Summary....................................................................................................................469
Register Description................................................................................................................. 469
32. AC - Analog Comparator.....................................................................................................................482
32.1.
32.2.
32.3.
32.4.
32.5.
Features................................................................................................................................... 482
Overview.................................................................................................................................. 482
Functional Description..............................................................................................................483
Register Summary ...................................................................................................................487
Register Description................................................................................................................. 487
33. ADC - Analog-to-Digital Converter...................................................................................................... 494
33.1.
33.2.
33.3.
33.4.
33.5.
Features................................................................................................................................... 494
Overview.................................................................................................................................. 494
Functional Description..............................................................................................................495
Register Summary....................................................................................................................506
Register Description................................................................................................................. 506
34. DAC - Digital-to-Analog Converter...................................................................................................... 524
34.1.
34.2.
34.3.
34.4.
Features................................................................................................................................... 524
Overview.................................................................................................................................. 524
Functional Description..............................................................................................................524
Register Summary....................................................................................................................526
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 10
AVR128DB28/32/48/64
34.5. Register Description................................................................................................................. 526
35. OPAMP - Analog Signal Conditioning................................................................................................. 529
35.1.
35.2.
35.3.
35.4.
35.5.
Features................................................................................................................................... 529
Overview.................................................................................................................................. 529
Functional Description..............................................................................................................530
Register Summary....................................................................................................................542
Register Description................................................................................................................. 542
36. ZCD - Zero-Cross Detector................................................................................................................. 553
36.1.
36.2.
36.3.
36.4.
36.5.
Features................................................................................................................................... 553
Overview.................................................................................................................................. 553
Functional Description..............................................................................................................554
Register Summary....................................................................................................................561
Register Description................................................................................................................. 561
37. UPDI - Unified Program and Debug Interface.....................................................................................565
37.1.
37.2.
37.3.
37.4.
37.5.
Features................................................................................................................................... 565
Overview.................................................................................................................................. 565
Functional Description..............................................................................................................567
Register Summary....................................................................................................................586
Register Description................................................................................................................. 586
38. Instruction Set Summary.....................................................................................................................597
39. Electrical Characteristics.....................................................................................................................598
39.1.
39.2.
39.3.
39.4.
39.5.
Disclaimer.................................................................................................................................598
Absolute Maximum Ratings .....................................................................................................598
Standard Operating Conditions ............................................................................................... 598
DC Characteristics................................................................................................................... 599
AC Characteristics....................................................................................................................606
40. Typical Characteristics........................................................................................................................ 619
40.1. OPAMP.....................................................................................................................................619
41. Ordering Information .......................................................................................................................... 628
42. Package Drawings.............................................................................................................................. 630
42.1. Online Package Drawings........................................................................................................ 630
42.2. 28-Pin SPDIP........................................................................................................................... 631
42.3. 28-Pin SOIC............................................................................................................................. 632
42.4. 28-Pin SSOP............................................................................................................................ 635
42.5. 32-Pin VQFN............................................................................................................................ 638
42.6. 32-Pin TQFP............................................................................................................................ 641
42.7. 48-Pin VQFN............................................................................................................................ 644
42.8. 48-Pin TQFP............................................................................................................................ 647
42.9. 64-Pin VQFN............................................................................................................................ 650
42.10. 64-Pin TQFP............................................................................................................................ 653
43. Data Sheet Revision History............................................................................................................... 656
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 11
AVR128DB28/32/48/64
43.1. Rev. A - 08/2020.......................................................................................................................656
The Microchip Website...............................................................................................................................657
Product Change Notification Service..........................................................................................................657
Customer Support...................................................................................................................................... 657
Product Identification System.....................................................................................................................658
Microchip Devices Code Protection Feature.............................................................................................. 658
Legal Notice............................................................................................................................................... 658
Trademarks................................................................................................................................................ 659
Quality Management System..................................................................................................................... 659
Worldwide Sales and Service.....................................................................................................................660
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 12
AVR128DB28/32/48/64
Block Diagram
1.
Block Diagram
UPDI
UPDI
M
M
SRAM
CPU
OCD
CRC
Legend:
M = Master
S = Slave
M
BUS Matrix
S
S
S
OPn-INP
OPn-INN
OPn-OUT
AINPn
AINNn
OUT
PORTx
OPAMP
PORTMUX
GPR
ACn
AINn
ADCn
ZCIN
OUT
ZCDn
OUT
DACn
VREFA
VREF
WOn
TCAn
WO
TCBn
WOx
TCDn
RxD
TxD
XCK
XDIR
USARTn
MISO
MOSI
SCK
SS
SPIn
SDA (Master)
SCL (Master)
SDA (Slave)
SCL (Slave)
TWIn
Flash
S
E
V
E
N
T
R
O
U
T
I
N
G
N
E
T
W
O
R
K
D
A
T
A
B
U
S
CPUINT
System
Management
EEPROM
I
N
/
O
U
T
D
A
T
A
B
U
S
NVMCTRL
Pxn
VDDIO2
MVIO
PCn
Detectors/
Power Control
POR
VREG
BOD
VLM
VDD
RESET
RSTCTRL
CLKCTRL
Clock Generation
SLPCTRL
PLL
XTALHF2
XOSCHF
XTALHF1
WDT
OSCHF
RTC
OSC32K
CLKOUT
EXTCLK
XTAL32K2
XOSC32K
© 2020 Microchip Technology Inc.
XTAL32K1
EVSYS
CCL
Preliminary Datasheet
EVOUTx
LUTn-OUT
LUTn-INn
DS40002247A-page 13
AVR128DB28/32/48/64
Pinout
2.
Pinout
2.1
28-pin SSOP, SOIC and SPDIP
PA7
1
28
PA6
PC0
2
27
PA5
PC1
3
26
PA4
PC2
4
25
PA3
PC3
5
24
PA2
VDDIO2
6
23
PA1 (XTALHF2)
PD1
7
22
PA0 (XTALHF1)
PD2
8
21
GND
PD3
9
20
VDD
PD4
10
19
UPDI
PD5
11
18
PF6
PD6
12
17
PF1 (XTAL32K2)
PD7
13
16
PF0 (XTAL32K1)
AVDD
14
15
GND
Note: For the AVR® DBFamily of devices, AVDD is internally connected to VDD (not separate power domains).
Power
Functionality
Power Supply
Programming/Debug
Ground
Clock/Crystal
Pin on VDD Power Domain
Digital Function Only
Pin on AVDD Power Domain
Analog Function
Pin on VDDIO2 Power Domain
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 14
AVR128DB28/32/48/64
Pinout
PA2
PA1 (XTALHF2)
PA0 (XTALHF1)
GND
VDD
UPDI
PF6
PF5
32
31
30
29
28
27
26
25
32-pin VQFN and TQFP
21
PF1 (XTAL32K2)
PA7
5
20
PF0 (XTAL32K1)
PC0
6
19
GND
PC1
7
18
AVDD
PC2
8
17
PD7
PD6
16
4
15
PA6
PD5
PF2
14
22
PD4
3
13
PA5
PD3
PF3
12
23
PD2
2
11
PA4
PD1
PF4
10
24
VDDIO2
1
9
PA3
PC3
2.2
Note: For the AVR® DBFamily of devices, AVDD is internally connected to VDD (not separate power domains).
Power
Functionality
Power Supply
Programming/Debug
Ground
Clock/Crystal
Pin on VDD Power Domain
Digital Function Only
Pin on AVDD Power Domain
Analog Function
Pin on VDDIO2 Power Domain
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 15
AVR128DB28/32/48/64
Pinout
PA4
PA3
PA2
PA1 (XTALHF2)
PA0 (XTALHF1)
GND
VDD
UPDI
PF6
PF5
PF4
PF3
48
47
46
45
44
43
42
41
40
39
38
37
48-pin VQFN and TQFP
PE1
PB3
7
30
PE0
PB4
8
29
GND
PB5
9
28
AVDD
PC0
10
27
PD7
PC1
11
26
PD6
PC2
12
25
PD5
24
31
PD4
6
23
PB2
PD3
PE2
22
32
PD2
5
21
PB1
PD1
PE3
20
33
PD0
4
19
PB0
PC7
PF0 (XTAL32K1)
18
34
PC6
3
17
PA7
PC5
PF1 (XTAL32K2)
16
35
PC4
2
15
PA6
GND
PF2
14
36
VDDIO2
1
13
PA5
PC3
2.3
Note: For the AVR® DBFamily of devices, AVDD is internally connected to VDD (not separate power domains).
Power
Functionality
Power Supply
Programming/Debug
Ground
Clock/Crystal
Pin on VDD Power Domain
Digital Function Only
Pin on AVDD Power Domain
Analog Function
Pin on VDDIO2 Power Domain
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 16
AVR128DB28/32/48/64
Pinout
PA2
PA1 (XTALHF2)
PA0 (XTALHF1)
PG7
PG6
PG5
PG4
GND
VDD
PG3
PG2
PG1
PG0
UPDI
PF6
PF5
64
63
62
61
60
59
58
57
56
55
54
53
52
51
50
49
64-pin VQFN and TQFP
PA3
1
48
PF4
PA4
2
47
PF3
PA5
3
46
PF2
PA6
4
45
PF1 (XTAL32K2)
PA7
5
44
PF0 (XTAL32K1)
15
34
AVDD
PC0
16
33
PD7
32
PB7
31
GND
PD6
35
PD5
14
30
PB6
PD4
PE0
29
36
PD3
13
28
PB5
PD2
PE1
27
PE2
37
PD1
38
12
26
11
PB4
25
PB3
PD0
PE3
PC7
39
24
10
PC6
PB2
23
PE4
PC5
40
22
9
PC4
PB1
21
PE5
GND
41
20
8
19
PB0
PC3
PE6
VDDIO2
PE7
42
18
43
7
PC2
6
17
VDD
GND
PC1
2.4
Note: For the AVR® DBFamily of devices, AVDD is internally connected to VDD (not separate power domains).
Power
Functionality
Power Supply
Programming/Debug
Ground
Clock/Crystal
Pin on VDD Power Domain
Digital Function Only
Pin on AVDD Power Domain
Analog Function
Pin on VDDIO2 Power Domain
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 17
AVR128DB28/32/48/64
44
30
22
PA0
XTALHF1
EXTCLK
63
45
31
23
PA1
XTALHF2
0, RxD
64
46
32
24
PA2
TWI
Fm+
0, XCK
0, SDA(MS)
0, WO2
0, WO
1
47
1
25
PA3
TWI
Fm+
0, XDIR
0, SCL(MS)
0, WO3
1, WO
2
48
2
26
PA4
0, TxD(3)
0, MOSI
0, WO4
WOA
3
1
3
27
PA5
0, RxD(3)
0, MISO
0, WO5
WOB
PA6
0, XCK(3)
0, SCK
PA7
0, XDIR(3)
5
2
3
4
5
28
1
6
CLKOUT
0, OUT
1, OUT
2, OUT
0, OUT
1, OUT
2, OUT
0, WO0
CCL
EVSYS
TCD0
TCBn
62
4
0, TxD
TCAn
TWIn(4)
SPIn
ZCDn
OPAMP
DAC0
ACn
ADC0
Special
Pin name(1,2)
SSOP28/SPDIP28
TQFP32
SOIC28/
TQFP48
VQFN32/
I/O Multiplexing
VQFN48/
3.1
TQFP64
I/O Multiplexing and Considerations
VQFN64/
3.
USARTn
I/O Multiplexing and Considerations
LUT0, IN0
0, WO1
LUT0, IN1
EVOUTA
LUT0, OUT
LUT0, OUT(3)
WOC
0, SS
LUT0, IN2
WOD
EVOUTA(3)
VDD
7
GND
8
4
PB0
3, TxD
0, WO0(3)
1, WO0
LUT4, IN0
9
5
PB1
3, RxD
0, WO1(3)
1, WO1
LUT4, IN1
10
6
PB2
TWI
3, XCK
1, SDA(MS)(3)
0, WO2(3)
1, WO2
11
7
PB3
TWI
3, XDIR
1, SCL(MS)(3)
0, WO3(3)
1, WO3
12
8
PB4
3, TxD(3)
1, MOSI(3)
0, WO4(3)
1, WO4
2, WO(3)
WOA(3)
13
9
PB5
3, RxD(3)
1, MISO(3)
0, WO5(3)
1, WO5
3, WO
WOB(3)
EVOUTB
LUT4, OUT
14
PB6
3, XCK(3)
1, SCK(3)
1, SDA(S)(3)
WOC(3)
15
PB7
3, XDIR(3)
1, SS(3)
1, SCL(S)(3)
WOD(3)
16
10
6
2
PC0
1, TxD
1, MOSI
0, WO0(3)
2, WO
17
11
7
3
PC1
1, RxD
1, MISO
0, WO1(3)
3, WO(3)
18
12
8
4
PC2
TWI
Fm+
1, XCK
1, SCK
0, SDA(MS)(3)
0, WO2(3)
19
13
9
5
PC3
TWI
Fm+
1, XDIR
1, SS
0, SCL(MS)(3)
0, WO3(3)
20
14
10
6
VDDIO2
21
15
GND
22
16
PC4
1, TxD(3)
1, MOSI(3)
0, WO4(3)
1, WO0(3)
23
17
PC5
1, RxD(3)
1, MISO(3)
0, WO5(3)
1, WO1(3)
24
18
PC6
1, XCK(3)
1, SCK(3)
0, SDA(S)(3)
25
19
PC7
1, XDIR(3)
1, SS(3)
0, SCL(S)(3)
26
20
PD0
0, OUT(3)
1, OUT(3)
2, OUT(3)
0, OUT(3)
1, OUT(3)
2, OUT(3)
AIN0
© 2020 Microchip Technology Inc.
Preliminary Datasheet
LUT4, OUT(3)
EVOUTB(3)
LUT1, IN0
LUT1, IN1
EVOUTC
LUT1, IN2
LUT1, OUT
1, WO2(3)
LUT1, OUT(3)
EVOUTC(3)
0, WO0(3)
0, AINN1
1, AINN1
2, AINN1
LUT4, IN2
LUT2, IN0
DS40002247A-page 18
AVR128DB28/32/48/64
I/O Multiplexing and Considerations
21
11
7
PD1
AIN1
28
22
12
8
PD2
AIN2
0, AINP0
1, AINP0
2, AINP0
OP0, OUT
0, WO2(3)
29
23
13
9
PD3
AIN3
0, AINN0
1, AINP1
OP0, INN
0, WO3(3)
30
24
14
10
PD4
AIN4
1, AINP2
2, AINP1
OP1, INP
0, WO4(3)
31
25
15
11
PD5
AIN5
1, AINN0
OP1, OUT
0, WO5(3)
32
26
16
12
PD6
AIN6
0, AINP3
1, AINP3
2, AINP3
33
27
17
13
PD7
AIN7
0, AINN2
1, AINN2
2, AINN0/AINN2
34
28
18
14
AVDD
35
29
19
15
AGND
36
30
PE0
AIN8
0, AINP1
37
31
PE1
AIN9
2, AINP2
38
32
PE2
AIN10
0, AINP2
39
33
0, ZCIN
EVSYS
TCD0
CCL
LUT2, IN2
LUT2, OUT
LUT2, OUT(3)
EVOUTD(3)
OP1, INN
4, TxD
0, MOSI(3)
0, WO0(3)
OP2, INP
4, RxD
0, MISO(3)
0, WO1(3)
OP2, OUT
4, XCK
0, SCK(3)
0, WO2(3)
4, XDIR
0, SS(3)
0, WO3(3)
PE3
AIN11
PE4
AIN12
4, TxD(3)
0, WO4(3)
1, WO0(3)
41
PE5
AIN13
4, RxD(3)
0, WO5(3)
1, WO1(3)
42
PE6
AIN14
4, XCK(3)
1, WO2(3)
AIN15
4, XDIR(3)
PE7
LUT2, IN1
EVOUTD
OUT
40
43
TCBn
TCAn
TWIn(4)
SPIn
0, WO1(3)
27
VREFA
OP0, INP
USARTn
ZCDn
OPAMP
DAC0
ACn
ADC0
Special
Pin name(1,2)
SSOP28/SPDIP28
TQFP32
SOIC28/
TQFP48
VQFN32/
VQFN48/
TQFP64
VQFN64/
...........continued
OP2, INN
1, ZCIN
2, ZCIN
EVOUTE
EVOUTE(3)
44
34
20
16
PF0
XTAL32K1
AIN16(5)
2, TxD
0, WO0(3)
WOA(3)
LUT3, IN0
45
35
21
17
PF1
XTAL32K2
AIN17(5)
2, RxD
0, WO1(3)
WOB(3)
LUT3, IN1
2, XCK
1, SDA(MS)
0, WO2(3)
WOC(3)
1, SCL(MS)
0, WO3(3)
WOD(3)
46
36
22
PF2
TWI
Fm+
AIN18(5)
47
37
23
PF3
TWI
Fm+
AIN19(5)
2, XDIR
48
38
24
PF4
AIN20(5)
2, TxD(3)
0, WO4(3)
0, WO(3)
25
PF5
AIN21(5)
2, RxD(3)
0, WO5(3)
1, WO(3)
RESET
UPDI
49
39
50
40
26
18
PF6(6)
51
41
27
19
UPDI
EVOUTF
LUT3, IN2
LUT3, OUT
52
PG0
5, TxD
0, WO0(3)
1, WO0(3)
LUT5, IN0
53
PG1
5, RxD
0, WO1(3)
1, WO1(3)
LUT5, IN1
54
PG2
5, XCK
0, WO2(3)
1, WO2(3)
55
PG3
5, XDIR
0, WO3(3)
1, WO3(3)
56
42
28
20
VDD
57
43
29
21
GND
EVOUTG
4, WO
LUT5, OUT
58
PG4
5, TxD(3)
0, MOSI(3)
0, WO4(3)
1, WO4(3)
WOA(3)
59
PG5
5, RxD(3)
0, MISO(3)
0, WO5(3)
1, WO5(3)
WOB(3)
60
PG6
5, XCK(3)
0, SCK(3)
WOC(3)
61
PG7
5, XDIR(3)
0, SS(3)
WOD(3)
© 2020 Microchip Technology Inc.
Preliminary Datasheet
LUT5, IN2
LUT5, OUT(3)
EVOUTG(3)
DS40002247A-page 19
AVR128DB28/32/48/64
I/O Multiplexing and Considerations
Notes:
1. Pin names are of type Pxn, with x being the PORT instance (A, B, C, ...) and n the pin number. Notation for
signals is PORTx_PINn. All pins can be used as event inputs.
2. All pins can be used for external interrupt, where pins Px2 and Px6 of each port have full asynchronous
detection.
3. Alternative pin positions.
4. TWI pins are marked MS if they can be used as TWI Master or Slave pins, and S if they can only be used as
TWI Slave pins.
5. AIN16 - AIN21 cannot be used as a negative ADC input for differential measurements.
6. Input only.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 20
AVR128DB28/32/48/64
Hardware Guidelines
4.
Hardware Guidelines
This section contains guidelines for designing or reviewing electrical schematics using AVR 8-bit microcontrollers.
The information presented here is a brief overview of the most common topics. More detailed information can be
found in application notes, listed in this section where applicable.
This section covers the following topics:
•
•
•
•
•
•
4.1
General guidelines
Connection for power supply
Connection for RESET
Connection for UPDI (Unified Program and Debug Interface)
Connection for external crystal oscillators
Connection for VREF (external voltage reference)
General Guidelines
Unused pins must be soldered to their respective soldering pads. The soldering pads must not be connected to the
circuit.
The PORT pins are in their default state after Reset. Follow the recommendations in the PORT section to reduce
power consumption.
All values are given as typical values and serve only as a starting point for circuit design.
Refer to the following application notes for further information:
•
•
4.1.1
AVR040 - EMC Design Considerations
AVR042 - AVR Hardware Design Considerations
Special Consideration for Packages with Center Pad
Flat packages often come with an exposed pad located on the bottom, often referred to as the center pad or the
thermal pad. This pad is not electrically connected to the internal circuit of the chip, but it is mechanically bonded to
the internal substrate and serves as a thermal heat sink as well as providing added mechanical stability. This pad
must be connected to GND since the ground plane is the best heat sink (largest copper area) of the printed circuit
board (PCB).
4.2
Connection for Power Supply
The basics and details regarding the design of the power supply itself lie beyond the scope of these guidelines. For
more detailed information about this subject, see the application notes mentioned at the beginning of this section.
A decoupling capacitor must be placed close to the microcontroller for each supply pin pair (VDD, AVDD, or other
power supply pin and its corresponding GND pin). If the decoupling capacitor is placed too far from the
microcontroller, a high-current loop might form that will result in increased noise and increased radiated emission.
Each supply pin pair (power input pin and ground pin) must have separate decoupling capacitors.
It is recommended to place the decoupling capacitor on the same side of the PCB as the microcontroller. If space
does not allow it, the decoupling capacitor may be placed on the other side through a via, but make sure the distance
to the supply pin is kept as short as possible.
If the board is experiencing high-frequency noise (upward of tens of MHz), add a second ceramic type capacitor in
parallel to the decoupling capacitor described above. Place this second capacitor next to the primary decoupling
capacitor.
On the board layout from the power supply circuit, run the power and return traces to the decoupling capacitors first,
and then to the device pins. This ensures that the decoupling capacitors are first in the power chain. Equally
important is to keep the trace length between the capacitor and the power pins to a minimum, thereby reducing PCB
trace inductance.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 21
AVR128DB28/32/48/64
Hardware Guidelines
As mentioned at the beginning of this section, all values used in examples are typical values. The actual design may
require other values.
4.2.1
Digital Power Supply
For larger pin count package types, there are several VDD and corresponding GND pins. All the VDD pins in the
microcontroller are internally connected. The same voltage must be applied to each of the VDD pins.
The following figure shows the recommendation for connecting a power supply to the VDD pin(s) of the device.
Figure 4-1. Recommended VDD Connection Circuit Schematic
VDD
VDD
C2
Typical values (recommended):
C1: 1 µF (primary decoupling capacitor)
C2: 10-100 nF (HF decoupling capacitor)
C1
GND
4.2.2
Analog Power Supply
These devices have a separate analog supply voltage pin, AVDD. This separate voltage supply pin is provided to
make the analog circuits less exposed to the digital noise originating from the switching of the digital circuits.
The following figure shows the recommendation for connecting a power supply to the AVDD pin of the device.
Figure 4-2. Recommended AVDD Connection Circuit Schematic
VDD
Typical values (recommended):
C1: 1 µF (primary decoupling capacitor)
C2: 10-100 nF (HF decoupling capacitor)
AVDD
C2
C1
GND
4.2.3
Multi-Voltage I/O
This additional Multi-Voltage I/O (MVIO) power supply input pin and corresponding grounding pin must be treated the
same way as any other power supply pin pair: By connecting a separate decoupling capacitor at the shortest possible
trace distance from pins. If there is more than one MVIO power supply pin, each supply pin and its corresponding
ground pin must have a decoupling capacitor.
The following figure shows the recommendation for connecting a power supply to the VDDIO2 pin(s) of the device.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 22
AVR128DB28/32/48/64
Hardware Guidelines
Figure 4-3. Recommended VDDIO2 Connection Circuit Schematic
VDDIO2
Typical values (recommended):
C1: 1 µF (primary decoupling capacitor)
C2: 10-100 nF (HF decoupling capacitor)
VDDIO2
C1
C2
GND
4.3
Connection for RESET
The RESET pin on the device is active-low, and setting the pin low externally will result in a Reset of the device.
AVR devices feature an internal pull-up resistor on the RESET pin, and an external pull-up resistor is usually not
required.
The following figure shows the recommendation for connecting an external Reset switch to the device.
Figure 4-4. Recommended External Reset Circuit Schematic
RESET
R1
SW1
Typical values (Recommended):
C1: 1 µF (Filtering capacitor)
R1: 330Ω (Switch series resistance)
C1
GND
A resistor in series with the switch can safely discharge the filtering capacitor. This prevents a current surge when
shorting the filtering capacitor, as this may cause a noise spike that can harm the system.
4.4
Connection for UPDI Programming
The standard connection for UPDI programming is a 100-mil 6-pin 2x3 header. Even though three pins are sufficient
for programming most AVR devices, it is recommended to use a 2x3 header since most programming tools are
delivered with 100-mil 6-pin 2x3 connectors.
The following figure shows the recommendation for connecting a UPDI connector to the device.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 23
AVR128DB28/32/48/64
Hardware Guidelines
Figure 4-5. Recommended UPDI Programming Circuit Schematic
VDD
UPDI
Typical values (recommended):
C1: 1 µF (primary decoupling capacitor)
C2: 10 nF-100 nF (HF decoupling capacitor)
NC = Not Connected
VDD
UPDI 1
NC 3
NC 5
2 VDD
4 NC
6 GND
C2
C1
GND
100-mil 6-pin
2x3 connector
The decoupling capacitor between VDD and GND must be placed as close to the pin pair as possible. The
decoupling capacitor must be included even if the UPDI connector is not included in the circuit.
4.5
Connecting External Crystal Oscillators
The use of external oscillators and the design of oscillator circuits are not trivial. This because there are many
variables: VDD, operating temperature range, crystal type and manufacture, loading capacitors, circuit layout, and
PCB material. Presented here are some typical guidelines to help with the basic oscillator circuit design.
•
•
•
•
•
•
•
Even the best performing oscillator circuits and high-quality crystals will not perform well if the layout and
materials used during the assembly are not carefully considered
The crystal circuit must be placed on the same side of the board as the device. Place the crystal circuit as close
to the respective oscillator pins as possible and avoid long traces. This will reduce parasitic capacitance and
increase immunity against noise and crosstalk. The load capacitors must be placed next to the crystal itself, on
the same side of the board. Any kind of sockets must be avoided.
Place a grounded copper area around the crystal circuit to isolate it from surrounding circuits. If the circuit board
has two sides, the copper area on the bottom layer must be a solid area covering the crystal circuit. The copper
area on the top layer must surround the crystal circuit and tie to the bottom layer area using via(s).
Do not run any signal traces or power traces inside the grounded copper area. Avoid routing digital lines,
especially clock lines, close to the crystal lines.
If using a two-sided PCB, avoid any traces beneath the crystal. For a multilayer PCB, avoid routing signals
below the crystal lines.
Dust and humidity will increase parasitic capacitance and reduce signal isolation. A protective coating is
recommended.
Successful oscillator design requires good specifications of operating conditions, a component selection phase
with initial testing, and testing in actual operating conditions to ensure that the oscillator performs as desired
For more detailed information about oscillators and oscillator circuit design, read the following application notes:
• AN2648 - Selecting and Testing 32 KHz Crystal Oscillators for AVR® Microcontrollers
• AN949 - Making Your Oscillator Work
4.5.1
Connection for XTAL32K (External 32.768 kHz crystal oscillator)
Ultra-low power 32.768 kHz oscillators typically dissipate significantly below 1 μW, and the current flowing in the
circuit is, therefore, extremely small. The crystal frequency is highly dependent on the capacitive load.
The following figure shows how to connect an external 32.768 kHz crystal oscillator.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 24
AVR128DB28/32/48/64
Hardware Guidelines
Figure 4-6. Recommended External 32.768 kHz Oscillator Connection Circuit Schematic
XTAL32K1
C1
32.768 kHz
Crystal Oscillator
XTAL32K2
C2
4.5.2
Connection for XTALHF (external HF crystal oscillator)
The following figure shows how to connect an external high-frequency crystal oscillator.
Figure 4-7. Recommended External High-Frequency Oscillator Connection Circuit Schematic
XTALHF1
C1
High Frequency
Crystal Oscillator
XTALHF2
C2
4.6
Connection for External Voltage Reference
If the design includes the use of an external voltage reference, the general recommendation is to use a suitable
capacitor connected in parallel with the reference. The value of the capacitor depends on the nature of the reference
and the type of electrical noise that needs to be filtered out.
Additional filtering components may be needed. This depends on the type of external voltage reference used.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 25
AVR128DB28/32/48/64
Hardware Guidelines
Figure 4-8. Recommended External Voltage Reference Connection
+
Voltage
Reference
C1
-
© 2020 Microchip Technology Inc.
VREFA
GND
Preliminary Datasheet
DS40002247A-page 26
AVR128DB28/32/48/64
Power Supply
5.
Power Supply
5.1
Power Domains
Figure 5-1. Power Domain Overview
ADC
GND
PD[7:0]
VDDIO2
GND
VDD
GND
AVDD
AVDD
VDDIO2
Voltage Regulator
MVIO
ADC
PC[7:0]
VDDCORE
VDDIO
PA[7:2]
AC
PE[7:0]
DAC
OPAMP
Digital Logic
PB[7:0]
(CPU, peripherals)
PF[6]
PF[5:2]
OSCHF
PG[7:0]
PF[1:0]
XOSC32K
OSC32K
XOSCHF
PA[1:0]
Note: For the AVR® DBFamily of devices, AVDD is internally connected to VDD (not separate power domains).
The AVR DB family of devices has several power domains with the following power supply pins:
VDD
Powers I/O lines, XOSCHF and the internal voltage regulator
AVDD
Powers I/O lines, XOSC32K (external 32.768 kHz oscillator) and the analog peripherals
VDDIO2
Powers I/O lines, optionally at a different voltage from VDD
The same voltage must be applied to all VDD and AVDD pins. This common voltage is referred to as VDD in the data
sheet.
The ground pins, GND, are common to VDD, AVDD and VDDIO2.
A subset of the device I/O pins can be powered by VDDIO2. This power domain is independent of VDD. Refer to the
Multi-Voltage I/O section for further information.
For recommendations on layout and decoupling, refer to the Hardware Guidelines section.
5.2
Voltage Regulator
The device has an internal voltage regulator that powers the VDDCORE domain. This domain has most of the digital
logic and the internal oscillators. The voltage regulator balances power consumption when the CPU is active or in a
sleep mode. Refer to the Sleep Controller (SLPCTRL) section for further information.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 27
AVR128DB28/32/48/64
Power Supply
5.3
Power-Up
The AVDD voltage must ramp up closely with the VDD voltage during power-up to ensure proper operation.
If the device is configured in single-supply mode, the VDDIO2 voltage must also rise closely to VDD. In Dual-supply
mode, the VDDIO2 voltage can ramp up or down at any time without affecting the proper operation. Refer to the MultiVoltage I/O (MVIO) section for further information.
The Power-on Reset (POR) and the Brown-out Detector (BOD) monitors VDD and will keep the system in reset if the
voltage level is below the respective voltage thresholds. Refer to the Reset Controller (RSTCTRL) and Brown-out
Detector (BOD) sections for further information.
Refer to the Electrical Characteristics section for further information on voltage thresholds.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 28
AVR128DB28/32/48/64
Conventions
6.
Conventions
6.1
Numerical Notation
Table 6-1. Numerical Notation
6.2
Symbol
Description
165
Decimal number
0b0101
Binary number
‘0101’
Binary numbers are given without prefix if unambiguous
0x3B24
Hexadecimal number
X
Represents an unknown or do not care value
Z
Represents a high-impedance (floating) state for either a
signal or a bus
Memory Size and Type
Table 6-2. Memory Size and Bit Rate
6.3
Symbol
Description
KB
kilobyte (210B = 1024B)
MB
megabyte (220B = 1024 KB)
GB
gigabyte (230B = 1024 MB)
b
bit (binary ‘0’ or ‘1’)
B
byte (8 bits)
1 kbit/s
1,000 bit/s rate
1 Mbit/s
1,000,000 bit/s rate
1 Gbit/s
1,000,000,000 bit/s rate
word
16-bit
Frequency and Time
Table 6-3. Frequency and Time
Symbol
Description
kHz
1 kHz = 103 Hz = 1,000 Hz
MHz
1 MHz = 106 Hz = 1,000,000 Hz
GHz
1 GHz = 109 Hz = 1,000,000,000 Hz
ms
1 ms = 10-3s = 0.001s
µs
1 µs = 10-6s = 0.000001s
ns
1 ns = 10-9s = 0.000000001s
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 29
AVR128DB28/32/48/64
Conventions
6.4
Registers and Bits
Table 6-4. Register and Bit Mnemonics
Symbol
Description
R/W
Read/Write accessible register bit. The user can read from and write to this bit.
R
Read-only accessible register bit. The user can only read this bit. Writes will be ignored.
W
Write-only accessible register bit. The user can only write this bit. Reading this bit will return an
undefined value.
BITFIELD
Bitfield names are shown in uppercase. Example: INTMODE.
BITFIELD[n:m]
A set of bits from bit n down to m. Example: PINA[3:0] = {PINA3, PINA2, PINA1, PINA0}.
Reserved
Reserved bits, bit fields, and bit field values are unused and reserved for future use. For
compatibility with future devices, always write reserved bits to ‘0’ when the register is written.
Reserved bits will always return zero when read.
PERIPHERALn
If several instances of the peripheral exist, the peripheral name is followed by a single number to
identify one instance. Example: USARTn is the collection of all instances of the USART module,
while USART3 is one specific instance of the USART module.
PERIPHERALx
If several instances of the peripheral exist, the peripheral name is followed by a single capital
letter (A-Z) to identify one instance. Example: PORTx is the collection of all instances of the
PORT module, while PORTB is one specific instance of the PORT module.
Reset
Value of a register after a Power-on Reset. This is also the value of registers in a peripheral after
performing a software Reset of the peripheral, except for the Debug Control registers.
SET/CLR/TGL
Registers with SET/CLR/TGL suffix allow the user to clear and set bits in a register without doing
a read-modify-write operation.
Each SET/CLR/TGL register is paired with the register it is affecting. Both registers in a register
pair return the same value when read.
Example: In the PORT peripheral, the OUT and OUTSET registers form such a register pair. The
contents of OUT will be modified by a write to OUTSET. Reading OUT and OUTSET will return
the same value.
Writing a ‘1’ to a bit in the CLR register will clear the corresponding bit in both registers.
Writing a ‘1’ to a bit in the SET register will set the corresponding bit in both registers.
Writing a ‘1’ to a bit in the TGL register will toggle the corresponding bit in both registers.
6.4.1
Addressing Registers from Header Files
In order to address registers in the supplied C header files, the following rules apply:
1.
2.
3.
4.
A register is identified by ., e.g., CPU.SREG, USART2.CTRLA,
or PORTB.DIR.
The peripheral name is given in the “Peripheral Address Map” in the “Peripherals and Architecture” section.
is obtained by substituting any n or x in the peripheral name with the correct
instance identifier.
When assigning a predefined value to a peripheral register, the value is constructed following the rule:
___gc
is , but remove any instance identifier.
can be found in the “Name” column in the tables in the Register Description sections
describing the bit fields of the peripheral registers.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 30
AVR128DB28/32/48/64
Conventions
Example 6-1. Register Assignments
// EVSYS channel 0 is driven by TCB3 OVF event
EVSYS.CHANNEL0 = EVSYS_CHANNEL0_TCB3_OVF_gc;
// USART0 RXMODE uses Double Transmission Speed
USART0.CTRLB = USART_RXMODE_CLK2X_gc;
Note: For peripherals with different register sets in different modes, and
must be followed by a mode name, for example:
// TCA0 in Normal Mode (SINGLE) uses waveform generator in frequency mode
TCA0.SINGLE.CTRL=TCA_SINGLE_WGMODE_FRQ_gc;
6.5
ADC Parameter Definitions
An ideal n-bit single-ended ADC converts a voltage linearly between GND and VREF in 2n steps (LSb). The lowest
code is read as ‘0’, and the highest code is read as ‘2n-1’. Several parameters describe the deviation from the ideal
behavior:
Offset Error
The deviation of the first transition (0x000 to 0x001) compared to the ideal transition (at 0.5
LSb). Ideal value: 0 LSb.
Figure 6-1. Offset Error
Output Code
Ideal ADC
Actual ADC
Offset
Error
Gain Error
VREF Input Voltage
After adjusting for offset, the gain error is found as the deviation of the last transition (e.g.,
0x3FE to 0x3FF for a 10-bit ADC) compared to the ideal transition (at 1.5 LSb below
maximum). Ideal value: 0 LSb.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 31
AVR128DB28/32/48/64
Conventions
Figure 6-2. Gain Error
Gain
Error
Output Code
Ideal ADC
Actual ADC
VREF
Integral
Nonlinearity (INL)
Input Voltage
After adjusting for offset and gain error, the INL is the maximum deviation of an actual
transition compared to an ideal transition for any code. Ideal value: 0 LSb.
Figure 6-3. Integral Nonlinearity
Output Code
INL
Ideal ADC
Actual ADC
VREF
Differential
Nonlinearity (DNL)
Input Voltage
The maximum deviation of the actual code width (the interval between two adjacent
transitions) from the ideal code width (1 LSb). Ideal value: 0 LSb.
Figure 6-4. Differential Nonlinearity
Output Code
0x3FF
1 LSb
DNL
0x000
0
© 2020 Microchip Technology Inc.
VREF
Preliminary Datasheet
Input Voltage
DS40002247A-page 32
AVR128DB28/32/48/64
Conventions
Quantization Error
Due to the quantization of the input voltage into a finite number of codes, a range of input
voltages (1 LSb wide) will code to the same value. Always ±0.5 LSb.
Absolute Accuracy The maximum deviation of an actual (unadjusted) transition compared to an ideal transition
for any code. This is the compound effect of all errors mentioned before. Ideal value: ±0.5
LSb.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 33
AVR128DB28/32/48/64
AVR® CPU
7.
AVR® CPU
7.1
Features
•
•
•
•
•
•
•
•
7.2
8-bit, High-Performance AVR RISC CPU:
– 135 instructions
– Hardware multiplier
32 8-bit Registers Directly Connected to the ALU
Stack in RAM
Stack Pointer Accessible in I/O Memory Space
Direct Addressing of up to 64 KB of Unified Memory
Efficient Support for 8-, 16-, and 32-bit Arithmetic
Configuration Change Protection for System-Critical Features
Native On-Chip Debugging (OCD) Support:
– Two hardware breakpoints
– Change of flow, interrupt, and software breakpoints
– Run-time read-out of Stack Pointer (SP) register, Program Counter (PC), and Status Register (SREG)
– Register file read- and writable in Stopped mode
Overview
The AVR CPU can access memories, perform calculations, control peripherals, execute instructions from the
program memory, and handling interrupts.
7.3
Architecture
To maximize performance and parallelism, the AVR CPU uses a Harvard architecture with separate buses for
program and data. Instructions in the program memory are executed with a single-level pipeline. While one
instruction is being executed, the next instruction is pre-fetched from the program memory. This enables instructions
to be executed on every clock cycle.
Refer to the Instruction Set Summary section for a summary of all AVR instructions.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 34
AVR128DB28/32/48/64
AVR® CPU
Figure 7-1. AVR® CPU Architecture
Register file
R31 (ZH)
R29 (YH)
R27 (XH)
R25
R23
R21
R19
R17
R15
R13
R11
R9
R7
R5
R3
R1
R30 (ZL)
R28 (YL)
R26 (XL)
R24
R22
R20
R18
R16
R14
R12
R10
R8
R6
R4
R2
R0
Program
Counter
Flash Program
Memory
Instruction
Register
Instruction
Decode
Data Memory
Stack
Pointer
Status
Register
© 2020 Microchip Technology Inc.
ALU
Preliminary Datasheet
DS40002247A-page 35
AVR128DB28/32/48/64
AVR® CPU
7.3.1
Arithmetic Logic Unit (ALU)
The Arithmetic Logic Unit (ALU) supports arithmetic and logic operations between working registers, or between a
constant and a working register. Also, single-register operations can be executed.
The ALU operates in a direct connection with all the 32 general purpose working registers in the register file.
Arithmetic operations between working registers or between a working register and an immediate operand are
executed in a single clock cycle, and the result is stored in the register file. After an arithmetic or logic operation, the
Status Register (CPU.SREG) is updated to reflect information about the result of the operation.
ALU operations are divided into three main categories – arithmetic, logical, and bit functions. Both 8- and 16-bit
arithmetic are supported, and the instruction set allows for efficient implementation of the 32-bit arithmetic. The
hardware multiplier supports signed and unsigned multiplication and fractional formats.
7.3.1.1
Hardware Multiplier
The multiplier is capable of multiplying two 8-bit numbers into a 16-bit result. The hardware multiplier supports
different variations of signed and unsigned integer and fractional numbers:
•
•
•
•
Multiplication of signed/unsigned integers
Multiplication of signed/unsigned fractional numbers
Multiplication of a signed integer with an unsigned integer
Multiplication of a signed fractional number with an unsigned fractional number
A multiplication takes two CPU clock cycles.
7.4
7.4.1
Functional Description
Program Flow
After being reset, the CPU will execute instructions from the lowest address in the Flash program memory, 0x0000.
The CPU supports instructions that can change the program flow conditionally or unconditionally and are capable of
addressing the whole address space directly. Most AVR instructions use a 16-bit word format, and a limited number
use a 32-bit format.
During interrupts and subroutine calls, the return address PC is stored on the stack as a word pointer. The stack is
allocated in the general data SRAM, and consequently, the stack size is limited only by the total SRAM size and the
usage of the SRAM. After the Stack Pointer (SP) is reset, it points to the highest address in the internal SRAM. The
SP is read/write accessible in the I/O memory space, enabling easy implementation of multiple stacks or stack areas.
The data SRAM can easily be accessed through different LD*/ST* instructions supported by the AVR CPU. See the
Instruction Set Summary section for details.
7.4.2
Instruction Execution Timing
The AVR CPU is clocked by the CPU clock, CLK_CPU. No internal clock division is applied. The figure below shows
the parallel instruction fetches and executions enabled by the Harvard architecture and the fast-access register file
concept. This is a two-stage pipelining concept enabling up to 1 MIPS/MHz performance with high efficiency.
Figure 7-2. The Parallel Instruction Fetches and Executions
T1
T2
T3
T4
CLK_CPU
Fetch
Execute
© 2020 Microchip Technology Inc.
Instruction 1
Instruction 2
Instruction 3
Instruction 1
Instruction 2
Preliminary Datasheet
Instruction 4
Instruction 3
DS40002247A-page 36
AVR128DB28/32/48/64
AVR® CPU
7.4.3
Status Register
The Status Register (CPU.SREG) contains information about the result of the most recently executed arithmetic or
logic instructions. This information can be used for altering the program flow to perform conditional operations.
CPU.SREG is updated after all ALU operations, as specified in the Instruction Set Summary section. This will, in
many cases, remove the need for using the dedicated compare instructions, resulting in a faster and more compact
code. CPU.SREG is not automatically stored or restored when entering or returning from an Interrupt Service Routine
(ISR). Therefore, maintaining the Status Register between context switches must be handled by user-defined
software. CPU.SREG is accessible in the I/O memory space.
7.4.4
Stack and Stack Pointer
The stack is used for storing return addresses after interrupts and subroutine calls. Also, it can be used for storing
temporary data. The Stack Pointer (SP) always points to the top of the stack. The address pointed to by the SP is
stored in the Stack Pointer (CPU.SP) register. CPU.SP is implemented as two 8-bit registers that are accessible in
the I/O memory space.
Data are pushed and popped from the stack using the PUSH and POP instructions. The stack grows from higher to
lower memory locations. This means that when pushing data onto the stack, the SP decreases, and when popping
data off the stack, the SP increases. The SP is automatically set to the highest address of the internal SRAM after
being reset. If the stack is changed, it must be set to point within the SRAM address space (see the SRAM Data
Memory section in the Memories section for the SRAM start address), and it must be defined before any subroutine
calls are executed and before interrupts are enabled. See the table below for SP details.
Table 7-1. Stack Pointer Instructions
Instruction
PUSH
Stack Pointer
Description
Decremented by 1 Data are pushed onto the stack
CALL
ICALL
EICALL
Decremented by 2 A return address is pushed onto the stack with a subroutine call or interrupt
RCALL
POP
RET
RETI
Incremented by 1
Data are popped from the stack
Incremented by 2
A return address is popped from the stack with a return from subroutine or return
from interrupt
During interrupts or subroutine calls, the return address is automatically pushed on the stack as a word, and the SP is
decremented by two. The return address consists of two bytes, and the Least Significant Byte (LSB) is pushed on the
stack first (at the higher address). As an example, a byte pointer return address of 0x0006 is saved on the stack as
0x0003 (shifted one bit to the right), pointing to the fourth 16-bit instruction word in the program memory. The return
address is popped off the stack with RETI (when returning from interrupts) and RET (when returning from subroutine
calls), and the SP is incremented by two.
The SP is decremented by one when data are pushed on the stack with the PUSH instruction and incremented by one
when data are popped off the stack using the POP instruction.
To prevent corruption when updating the SP from software, a write to SPL will automatically disable interrupts for up
to four instructions or until the next I/O memory write, whichever comes first.
7.4.5
Register File
The register file consists of 32 8-bit general purpose working registers used by the CPU. The register file is located in
a separate address space from the data memory.
All CPU instructions that operate on working registers have direct and single-cycle access to the register file. Some
limitations apply to which working registers can be accessed by an instruction, like the constant arithmetic and logic
instructions SBCI, SUBI, CPI, ANDI ORI, and LDI. These instructions apply to the second half of the working
registers in the register file, R16 to R31. See the AVR Instruction Set Manual for further details.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 37
AVR128DB28/32/48/64
AVR® CPU
Figure 7-3. AVR® CPU General Purpose Working Registers
0 Addr.
7
0x00
R0
0x01
R1
0x02
R2
...
R13
R14
R15
R16
R17
0x0D
0x0E
0x0F
0x10
0x11
R26
R27
R28
R29
R30
R31
0x1A
0x1B
0x1C
0x1D
0x1E
0x1F
...
7.4.5.1
X-register Low Byte
X-register High Byte
Y-register Low Byte
Y-register High Byte
Z-register Low Byte
Z-register High Byte
The X-, Y-, and Z-Registers
Working registers R26...R31 have added functions besides their general purpose usage.
These registers can form 16-bit Address Pointers for indirect addressing of data memory. These three address
registers are called the X-register, Y-register, and Z-register. The Z-register can also be used as Address Pointer for
program memory.
Figure 7-4. The X-, Y-, and Z-Registers
Bit (individually)
7
X-register
15
Bit (individually)
7
Y-register
Bit (individually)
7
8
7
R29
0
7
15
7
R31
0
0
R28
0
YL
8
7
0
7
ZH
15
R26
XL
YH
Z-register
Bit (Z-register)
0
XH
Bit (X-register)
Bit (Y-register)
R27
0
R30
0
ZL
8
7
0
The lowest register address holds the Least Significant Byte (LSB), and the highest register address holds the Most
Significant Byte (MSB). These address registers can function as fixed displacement, automatic increment, and
automatic decrement, with different LD*/ST* instructions. See the Instruction Set Summary section for details.
7.4.5.2
Extended Pointers
To access program memory above 64 KB, the Address Pointer must be larger than 16 bits. This is done by
concatenating one of the address extension I/O registers (RAMPZ) with the internal Z-pointer. The RAMPZ register
then holds the Most Significant Byte (MSB) in a 24-bit address or Address Pointer.
This address extension register is available only on devices with more than 64 KB of program memory. For the
devices where extension pointers are required, only the number of bits required to address the whole program and
data memory space in the device are implemented.
7.4.5.2.1 Extended Program Memory Pointer
The RAMPZ register is concatenated with the Z-register to enable indirect addressing of the entire program memory.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 38
AVR128DB28/32/48/64
AVR® CPU
Figure 7-5. The Combined RAMPZ + Z Register
Bit (Individually)
7
0
7
RAMPZ
Bit (Z-pointer)
23
0
7
ZH
16
15
0
ZL
8
7
0
When reading (ELPM) above the first 64 KB of the program memory, RAMPZ is concatenated with the Z-register to
form the 24-bit address. The LPM instruction is not affected by the RAMPZ setting.
7.4.6
Configuration Change Protection (CCP)
System critical I/O register settings are protected from accidental modification, and Flash self-programming is
protected from accidental programming. This is handled globally by the Configuration Change Protection (CCP)
register.
Changes to the protected I/O registers or bits, or execution of protected instructions, are only possible after the CPU
writes a signature to the CCP register. The different signatures are listed in the description of the CCP (CPU.CCP)
register.
Once the correct signature is written by the CPU, interrupts will be ignored for the duration of the configuration
change enable period. Any interrupt request (including non-maskable interrupts) during the CCP period will set the
corresponding interrupt flag as normal, and the request is kept pending. After the CCP period is completed, any
pending interrupts are executed according to their level and priority.
There are two modes of CCP operation: One for protected I/O registers, and one for protected self-programming.
7.4.6.1
Sequence for Write Operation to Configuration Change Protected I/O Registers
To write to I/O registers protected by CCP, the following steps are required:
1.
2.
7.4.6.2
The software writes the signature that enables change of protected I/O registers to the CCP bit field in the
CPU.CCP register.
Within four instructions, the software must write the appropriate data to the protected I/O register.
The protected change is automatically disabled after CPU executes a write instruction.
Sequence for Execution of Self-Programming
To execute self-programming (the execution of writes to the NVM controller’s command register), the following steps
are required:
1.
2.
7.4.7
The software temporarily enables self-programming by writing the SPM signature to the CCP (CPU.CCP)
register.
Within four instructions, the software must execute the appropriate instruction or change to NVM Command
Register.
The protected change is automatically disabled after the CPU executes a write instruction.
On-Chip Debug Capabilities
The AVR CPU includes native On-Chip Debug (OCD) support. It includes some powerful debug capabilities to enable
profiling and detailed information about the CPU state. It is possible to alter the CPU state and resume code
execution. Also, normal debug capabilities like hardware Program Counter breakpoints, breakpoints on change of
flow instructions, breakpoints on interrupts, and software breakpoints (BREAK instruction) are present. Refer to the
Unified Program and Debug Interface section for details about OCD.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 39
AVR128DB28/32/48/64
AVR® CPU
7.5
Offset
0x00
...
0x03
0x04
0x05
...
0x0A
0x0B
0x0C
Register Summary
Name
7
6
5
4
3
2
1
0
V
N
Z
C
Reserved
CCP
7:0
CCP[7:0]
7:0
RAMPZ[7:0]
7:0
15:8
7:0
SP[7:0]
SP[15:8]
Reserved
RAMPZ
Reserved
0x0D
SP
0x0F
SREG
7.6
Bit Pos.
I
T
H
S
Register Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 40
AVR128DB28/32/48/64
AVR® CPU
7.6.1
Configuration Change Protection
Name:
Offset:
Reset:
Property:
Bit
7
CCP
0x04
0x00
-
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
CCP[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – CCP[7:0] Configuration Change Protection
Writing the correct signature to this bit field allows changing protected I/O registers or executing protected
instructions within the next four CPU instructions executed.
All interrupts are ignored during these cycles. After these cycles are completed, the interrupts will automatically be
handled again by the CPU, and any pending interrupts will be executed according to their level and priority.
When the protected I/O register signature is written, CCP[0] will read as ‘1’ as long as the CCP feature is enabled.
When the protected self-programming signature is written, CCP[1] will read as ‘1’ as long as the CCP feature is
enabled.
CCP[7:2] will always read as ‘0’.
Value
Name
Description
0x9D
SPM
Allow Self-Programming
0xD8
IOREG
Unlock protected I/O registers
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 41
AVR128DB28/32/48/64
AVR® CPU
7.6.2
Stack Pointer
Name:
Offset:
Reset:
Property:
SP
0x0D
Top of stack
-
The CPU.SP register holds the Stack Pointer (SP) that points to the top of the stack. After being reset, the SP points
to the highest internal SRAM address.
Only the number of bits required to address the available data memory, including external memory (up to 64 KB), is
implemented for each device. Unused bits will always read as ‘0’.
The CPU.SPL and CPU.SPH register pair represents the 16-bit value, CPU.SP. The low byte [7:0] (suffix L) is
accessible at the original offset. The high byte [15:8] (suffix H) can be accessed at offset + 0x01.
To prevent corruption when updating the SP from software, a write to CPU.SPL will automatically disable interrupts
for the next four instructions or until the next I/O memory write, whichever comes first.
Bit
15
14
13
12
11
10
9
8
R/W
R/W
R/W
R/W
3
2
1
0
R/W
R/W
R/W
R/W
SP[15:8]
Access
Reset
Bit
R/W
R/W
R/W
R/W
7
6
5
4
SP[7:0]
Access
Reset
R/W
R/W
R/W
R/W
Bits 15:8 – SP[15:8] Stack Pointer High Byte
These bits hold the MSB of the 16-bit register.
Bits 7:0 – SP[7:0] Stack Pointer Low Byte
These bits hold the LSB of the 16-bit register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 42
AVR128DB28/32/48/64
AVR® CPU
7.6.3
Status Register
Name:
Offset:
Reset:
Property:
SREG
0x0F
0x00
-
The Status Register contains information about the result of the most recently executed arithmetic or logic
instructions. For details about the bits in this register and how they are influenced by different instructions, see the
Instruction Set Summary section.
Bit
Access
Reset
7
I
R/W
0
6
T
R/W
0
5
H
R/W
0
4
S
R/W
0
3
V
R/W
0
2
N
R/W
0
1
Z
R/W
0
0
C
R/W
0
Bit 7 – I Global Interrupt Enable Bit
Writing a ‘1’ to this bit enables interrupts on the device.
Writing a ‘0’ to this bit disables interrupts on the device, independent of the individual interrupt enable settings of the
peripherals.
This bit is not cleared by hardware while entering an Interrupt Service Routine (ISR) or set when the RETI instruction
is executed.
This bit can be set and cleared by software with the SEI and CLI instructions.
Changing the I bit through the I/O register results in a one-cycle Wait state on the access.
Bit 6 – T Transfer Bit
The bit copy instructions, Bit Load (BLD) and Bit Store (BST), use the T bit as source or destination for the operated
bit.
Bit 5 – H Half Carry Flag
This flag is set when there is a half carry in arithmetic operations that support this, and is cleared otherwise. Half
carry is useful in BCD arithmetic.
Bit 4 – S Sign Flag
This flag is always an Exclusive Or (XOR) between the Negative flag (N) and the Two’s Complement Overflow flag
(V).
Bit 3 – V Two’s Complement Overflow Flag
This flag is set when there is an overflow in arithmetic operations that support this, and is cleared otherwise.
Bit 2 – N Negative Flag
This flag is set when there is a negative result in an arithmetic or logic operation, and is cleared otherwise.
Bit 1 – Z Zero Flag
This flag is set when there is a zero result in an arithmetic or logic operation, and is cleared otherwise.
Bit 0 – C Carry Flag
This flag is set when there is a carry in an arithmetic or logic operation, and is cleared otherwise.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 43
AVR128DB28/32/48/64
AVR® CPU
7.6.4
Extended Z-Pointer Register
Name:
Offset:
Reset:
Property:
RAMPZ
0x0B
0x00
-
This register is concatenated with the Z-register for indirect addressing (LD/LDD/ST/STD) of the whole data memory
space on devices with more than 64 KB of data memory. RAMPZ is concatenated with the Z-register when reading
the (ELPM) program memory locations above the first 64 KB and writing the (SPM) program memory locations above
the first 128 KB of the program memory.
This register is not available if the data memory and program memory in the device are less than 64 KB.
Bit
Access
Reset
7
6
5
R/W
0
R/W
0
R/W
0
4
3
RAMPZ[7:0]
R/W
R/W
0
0
2
1
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – RAMPZ[7:0] Extended Z-pointer Address Bits
These bits hold the MSB of the 24-bit address created by RAMPZ and the 16-bit Z-register. Only the number of bits
required to address the available data and program memory is implemented for each device. Unused bits will always
read as ‘0’.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 44
AVR128DB28/32/48/64
Memories
8.
Memories
8.1
Overview
The main memories of the AVR128DB28/32/48/64 devices are SRAM data memory space, EEPROM data memory
space, and Flash program memory space. In addition, the peripheral registers are located in the I/O memory space.
8.2
Memory Map
The figure below shows the memory map for the largest memory derivative in the AVR DB family. Refer to the
subsequent sections and the Peripheral Address Map table for further details.
Figure 8-1. Memory Map
Code space
Data space
0x00000
Flash code
128 KB
0x18000
I/O Memory
0x0000 - 0x103F
LOCK
0x1040 - 0x104F
FUSE
0x1050 - 0x105F
USERROW
0x1080 - 0x109F
SIGROW
0x1100 - 0x117F
EEPROM
512 Bytes
0x1400 - 0x15FF
(Reserved)
0x1600 - 0x3FFF
SRAM
16 KB
0x4000 - 0x7FFF
Mapped Flash
32 KB
Single-cycle
I/O registers
0x0000 - 0x003F
Extended
I/O registers
0x0040 - 0x103F
0x8000 - 0xFFFF
0x1FFFF
8.3
In-System Reprogrammable Flash Program Memory
The AVR128DB28/32/48/64 contains 128 KB on-chip in-system reprogrammable Flash memory for program storage.
Since all AVR instructions are 16 or 32 bits wide, the Flash is organized with 16-bit data width. For write protection,
the Flash program memory space can be divided into three sections: Boot Code section, Application Code section,
and Application Data section. Code placed in one section may be restricted from writing to addresses in other
sections. The Program Counter (PC) can address the whole program memory.
Refer to the Code Size (CODESIZE) and Boot Size (BOOTSIZE) descriptions and the Nonvolatile Memory Controller
section for further details.
The Program Counter can address the whole program memory. The procedure for writing Flash memory is described
in detail in the Nonvolatile Memory Controller (NVMCTRL) peripheral documentation.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 45
AVR128DB28/32/48/64
Memories
Each 32 KB section from Flash memory is mapped into the data memory space and is accessible with LD/ST
instructions. For LD/ST instructions, the Flash is mapped from address 0x8000 to 0xFFFF. The entire Flash memory
space can be accessed with the LPM/SPM instruction. For the LPM/SPM instruction, the Flash start address is
0x0000.
Table 8-1. Physical Properties of Flash Memory
AVR128DB64
AVR128DB48
Property
AVR128DB32
AVR128DB28
Size
128 KB
Page size
512B
Number of pages
8.4
256
Start address in data space
0x8000
Start address in code space
0x0
SRAM Data Memory
The primary task of the SRAM memory is to store application data. Also, the program stack is located at the end of
SRAM. It is not possible to execute from SRAM.
Table 8-2. Physical Properties of SRAM Memory
AVR128DB64
AVR128DB48
Property
AVR128DB32
AVR128DB28
8.5
Size
16 KB
Start address
0x4000
EEPROM Data Memory
The task of the EEPROM memory is to store nonvolatile application data. The EEPROM memory supports singleand multi-byte read and write. The EEPROM is controlled by the Nonvolatile Memory Controller (NVMCTRL).
Table 8-3. Physical Properties of EEPROM Memory
AVR128DB64
AVR128DB48
Property
AVR128DB32
AVR128DB28
Size
512B
Start address
© 2020 Microchip Technology Inc.
0x1400
Preliminary Datasheet
DS40002247A-page 46
AVR128DB28/32/48/64
Memories
8.6
SIGROW - Signature Row
The content of the Signature Row fuses (SIGROW) is pre-programmed and read-only. SIGROW contains information
such as device ID, serial number, and calibration values.
All the AVR128DB28/32/48/64 devices have a three-byte device ID that identifies the device. The device ID can be
read using the Unified Program and Debug Interface (UPDI), also when a device is locked. The device ID for the
AVR128DB28/32/48/64 devices consists of three signature bytes, which is given by the following table.
Table 8-4. Device ID
Device Name
Signature Byte Address and Value
0x00
0x01
0x02
AVR128DB28
0x1E
0x97
0x0E
AVR128DB32
0x1E
0x97
0x0D
AVR128DB48
0x1E
0x97
0x0C
AVR128DB64
0x1E
0x97
0x0B
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 47
AVR128DB28/32/48/64
Memories
8.6.1
Signature Row Summary
Offset
Name
Bit Pos.
0x00
0x01
0x02
0x03
DEVID0
DEVID1
DEVID2
Reserved
7:0
7:0
7:0
DEVID[7:0]
DEVID[7:0]
DEVID[7:0]
0x04
TEMPSENSE0
0x06
TEMPSENSE1
7:0
15:8
7:0
15:8
TEMPSENSE[7:0]
TEMPSENSE[15:8]
TEMPSENSE[7:0]
TEMPSENSE[15:8]
SERNUM0
7:0
SERNUM[7:0]
SERNUM15
7:0
SERNUM[7:0]
0x08
...
0x0F
0x10
...
0x1F
8.6.2
7
6
5
4
3
2
1
0
Reserved
Signature Row Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 48
AVR128DB28/32/48/64
Memories
8.6.2.1
Device ID n
Name:
Offset:
Reset:
Property:
DEVIDn
0x00 + n*0x01 [n=0..2]
[Signature byte n of device ID]
-
Each device has a device ID identifying the device and its properties such as memory sizes and pin count. This can
be used to identify a device and hence, the available features by software. The Device ID consists of three bytes:
SIGROW.DEVID[2:0].
Bit
7
6
5
4
3
2
1
0
R
x
R
x
R
x
R
x
DEVID[7:0]
Access
Reset
R
x
R
x
R
x
R
x
Bits 7:0 – DEVID[7:0] Byte n of the Device ID
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 49
AVR128DB28/32/48/64
Memories
8.6.2.2
Temperature Sensor Calibration n
Name:
Offset:
Reset:
Property:
TEMPSENSEn
0x04 + n*0x02 [n=0..1]
[Temperature sensor calibration value]
-
The Temperature Sensor Calibration value contains correction factors for temperature measurements from the onchip temperature sensor. The SIGROW.TEMPSENSE0 is a correction factor for the gain/slope (unsigned), and
SIGROW.TEMPSENSE1 is a correction factor for the offset (signed).
Bit
15
14
13
Access
Reset
R
x
R
x
R
x
Bit
7
6
5
Access
Reset
R
x
R
x
R
x
12
11
TEMPSENSE[15:8]
R
R
x
x
10
9
8
R
x
R
x
R
x
4
3
TEMPSENSE[7:0]
R
R
x
x
2
1
0
R
x
R
x
R
x
Bits 15:0 – TEMPSENSE[15:0] Temperature Sensor Calibration word n
Refer to the Analog-to-Digital Converter section for a description of how to use the value stored in this bit field.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 50
AVR128DB28/32/48/64
Memories
8.6.2.3
Serial Number Byte n
Name:
Offset:
Reset:
Property:
SERNUMn
0x10 + n*0x01 [n=0..15]
[Byte n of device serial number]
-
Each device has an individual serial number, representing a unique ID. This can be used to identify a specific device
in the field. The serial number consists of 16 bytes: SIGROW.SERNUM[15:0].
Bit
7
6
5
4
3
2
1
0
R
x
R
x
R
x
R
x
SERNUM[7:0]
Access
Reset
R
x
R
x
R
x
R
x
Bits 7:0 – SERNUM[7:0] Serial Number Byte n
8.7
USERROW - User Row
The AVR128DB28/32/48/64 devices have a special 32-byte memory section called the User Row (USERROW).
USERROW can be used for end-production data and is not affected by chip erase. It can be written by the Unified
Program and Debug Interface (UPDI) even if the part is locked, which enables storage of final configuration without
having access to any other memory. When the part is locked, the UPDI is not allowed to read the content of the
USERROW.
The CPU can write and read this memory as a normal flash. Refer to the System Memory Address Map for further
details.
8.8
FUSE - Configuration and User Fuses
Fuses are part of the nonvolatile memory and hold factory calibration and device configuration. The fuses can be
read by the CPU or the UPDI, but can only be programmed or cleared by the UPDI. The configuration values stored
in the fuses are written to their respective target registers at the end of the start-up sequence.
The fuses for peripheral configuration (FUSE) are pre-programmed but can be altered by the user. Altered values in
the configuration fuse will be effective only after a Reset.
Note: When writing the fuses, all reserved bits must be written to ‘0’.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 51
AVR128DB28/32/48/64
Memories
8.8.1
Fuse Summary
Offset
Name
Bit Pos.
0x00
0x01
0x02
0x03
...
0x04
0x05
0x06
0x07
0x08
WDTCFG
BODCFG
OSCCFG
7:0
7:0
7:0
8.8.2
7
6
5
WINDOW[3:0]
LVL[2:0]
4
3
2
1
0
PERIOD[3:0]
SAMPFREQ
ACTIVE[1:0]
SLEEP[1:0]
CLKSEL[3:0]
Reserved
SYSCFG0
SYSCFG1
CODESIZE
BOOTSIZE
7:0
7:0
7:0
7:0
CRCSRC[1:0]
CRCSEL
RSTPINCFG
MVSYSCFG[1:0]
CODESIZE[7:0]
BOOTSIZE[7:0]
EESAVE
SUT[2:0]
Fuse Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 52
AVR128DB28/32/48/64
Memories
8.8.2.1
Watchdog Timer Configuration
Name:
Offset:
Default:
Property:
WDTCFG
0x00
0x00
-
The default value given in this fuse description is the factory-programmed value, and should not be mistaken for the
Reset value.
Bit
7
6
5
4
3
2
WINDOW[3:0]
Access
Default
R
0
R
0
1
0
R
0
R
0
PERIOD[3:0]
R
0
R
0
R
0
R
0
Bits 7:4 – WINDOW[3:0] Watchdog Window Time-out Period
This value is loaded into the WINDOW bit field of the Watchdog Control A (WDT.CTRLA) register at the end of the
start-up sequence, after power-on or Reset.
Bits 3:0 – PERIOD[3:0] Watchdog Time-out Period
This value is loaded into the PERIOD bit field of the Watchdog Control A (WDT.CTRLA) register at the end of the
start-up sequence after power-on or Reset.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 53
AVR128DB28/32/48/64
Memories
8.8.2.2
Brown-Out Detector Configuration
Name:
Offset:
Default:
Property:
BODCFG
0x01
0x00
-
The bit values of this fuse register are written to the corresponding BOD configuration registers at the start-up.
The default value given in this fuse description is the factory-programmed value, and should not be mistaken for the
Reset value.
Bit
Access
Default
7
R
0
6
LVL[2:0]
R
0
5
R
0
4
SAMPFREQ
R
0
3
2
1
ACTIVE[1:0]
R
0
0
SLEEP[1:0]
R
0
R
0
R
0
Bits 7:5 – LVL[2:0] BOD Level
This value is loaded into the LVL bit field of the BOD Control B (BOD.CTRLB) register during Reset.
Value
Name
Description
0x0
BODLEVEL0
1.9V
0x1
BODLEVEL1
2.45V
0x2
BODLEVEL2
2.70V
0x3
BODLEVEL3
2.85V
Other
Reserved
Note: Values in the Description column are typical values. Refer to the Electrical Characteristics section for further
details.
Bit 4 – SAMPFREQ BOD Sample Frequency
This value is loaded into the Sample Frequency (SAMPFREQ) bit of the BOD Control A (BOD.CTRLA) register
during Reset. Refer to the Brown-out Detector section for further details.
Value
Name
Description
0
128HZ
The sample frequency is 128 Hz
1
32HZ
The sample frequency is 32 Hz
Bits 3:2 – ACTIVE[1:0] BOD Operation Mode in Active and Idle
This value is loaded into the ACTIVE bit field of the BOD Control A (BOD.CTRLA) register during Reset. Refer to the
Brown-out Detector section for further details.
Value
Name
Description
0x0
DISABLE
BOD disabled
0x1
ENABLE
BOD enabled in continuous mode
0x2
SAMPLE
BOD enabled in sampled mode
0x3
ENABLEWAIT BOD enabled in continuous mode. Execution is halted at wake-up until BOD is running
Bits 1:0 – SLEEP[1:0] BOD Operation Mode in Sleep
The value is loaded into the SLEEP bit field of the BOD Control A (BOD.CTRLA) register during Reset. Refer to the
Brown-out Detector section for further details.
Value
Name
Description
0x0
DISABLE
BOD disabled
0x1
ENABLE
BOD enabled in continuous mode
0x2
SAMPLE
BOD enabled in sampled mode
0x3
Reserved
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 54
AVR128DB28/32/48/64
Memories
8.8.2.3
Oscillator Configuration
Name:
Offset:
Default:
Property:
OSCCFG
0x02
0x00
-
The default value given in this fuse description is the factory-programmed value and should not be mistaken for the
Reset value.
Bit
7
6
5
4
3
2
1
0
R
0
R
0
CLKSEL[3:0]
Access
Default
R
0
R
0
Bits 3:0 – CLKSEL[3:0] Clock Select
This bit field controls the default oscillator for the device.
Value
Name
Description
0x0
OSCHF
Device running on internal high frequency oscillator
0x1
OSC32K
Device running on internal 32.768 kHz oscillator
Other
Reserved
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 55
AVR128DB28/32/48/64
Memories
8.8.2.4
System Configuration 0
Name:
Offset:
Default:
Property:
SYSCFG0
0x05
0xC0
-
The default value given in this fuse description is the factory-programmed value and should not be mistaken for the
Reset value.
Bit
7
6
CRCSRC[1:0]
Access
Default
R
1
R
1
5
CRCSEL
R
0
4
3
RSTPINCFG
R
0
2
1
0
EESAVE
R
0
Bits 7:6 – CRCSRC[1:0] CRC Source
This bit field controls which section of the Flash will be checked by the CRCSCAN peripheral after Reset. Refer to the
CRCSCAN section for more information about the functionality.
Value
Name
Description
0x0
FLASH
CRC of full Flash (boot, application code, and application data)
0x1
BOOT
CRC of the Boot section
0x2
BOOTAPP
CRC of the Application code and Boot sections
0x3
NOCRC
No CRC
Bit 5 – CRCSEL CRC Mode Selection
This bit controls the type of CRC performed by the CRCSCAN peripheral. Refer to the CRCSCAN section for more
information about the functionality.
Value
Name
Description
0
CRC16
CRC-16-CCITT
1
CRC32
CRC-32 (IEEE 802.3)
Bit 3 – RSTPINCFG Reset Pin Configuration at Start-Up
This bit controls the pin configuration for the Reset pin.
Value
Name
Description
0
INPUT
No external reset
1
RESET
External reset with pull-up enabled on PF6
Bit 0 – EESAVE EEPROM Saved During Chip Erase
This bit controls if the EEPROM will be erased or saved during a chip erase. If the device is locked, the EEPROM is
always erased by a chip erase regardless of this bit.
Value
Name
Description
0
DISABLE
EEPROM is erased during a chip erase
1
ENABLE
EEPROM is saved during a chip erase
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 56
AVR128DB28/32/48/64
Memories
8.8.2.5
System Configuration 1
Name:
Offset:
Default:
Property:
SYSCFG1
0x06
0x08
-
The default value given in this fuse description is the factory-programmed value and should not be mistaken for the
Reset value.
Bit
7
6
Access
Default
5
4
3
MVSYSCFG[1:0]
R
R
0
1
2
R
0
1
SUT[2:0]
R
0
0
R
0
Bits 4:3 – MVSYSCFG[1:0] MVIO System Configuration
This bit field controls the power supply mode.
Value
Name
Description
0x0
Reserved
0x1
DUAL
Device used in a dual supply configuration
0x2
SINGLE
Device used in a single supply configuration
0x3
Reserved
Bits 2:0 – SUT[2:0] Start-up Time
This bit field controls the start-up time between power-on and code execution.
Value
Description
0x0
0 ms
0x1
1 ms
0x2
2 ms
0x3
4 ms
0x4
8 ms
0x5
16 ms
0x6
32 ms
0x7
64 ms
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 57
AVR128DB28/32/48/64
Memories
8.8.2.6
Code Size
Name:
Offset:
Default:
Property:
CODESIZE
0x07
0x00
-
The default value given in this fuse description is the factory-programmed value, and should not be mistaken for the
Reset value.
Bit
7
6
5
Access
Default
R
0
R
0
R
0
4
3
CODESIZE[7:0]
R
R
0
0
2
1
0
R
0
R
0
R
0
Bits 7:0 – CODESIZE[7:0] Code section size configuration
This bit field defines the combined size of the Boot Code (BOOT) section and Application Code (APPCODE) section
in blocks of 512 bytes. For more details, refer to the Nonvolatile Memory Controller section.
Note: If FUSE.BOOTSIZE is 0x00, the entire Flash memory is set as the Boot Code section, and the value of
FUSE.CODESIZE is not used.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 58
AVR128DB28/32/48/64
Memories
8.8.2.7
Boot Size
Name:
Offset:
Default:
Property:
BOOTSIZE
0x08
0x00
-
The default value given in this fuse description is the factory-programmed value, and should not be mistaken for the
Reset value.
Bit
7
6
5
Access
Default
R
0
R
0
R
0
4
3
BOOTSIZE[7:0]
R
R
0
0
2
1
0
R
0
R
0
R
0
Bits 7:0 – BOOTSIZE[7:0] Boot Section Size
This bitfield controls the size of the boot section in blocks of 512 bytes. A value of 0x00 defines the entire Flash as
Boot Code section.
For more details, refer to the Nonvolatile Memory Controller section.
8.9
LOCK - Memory Sections Access Protection
The device can be locked so that the memories cannot be read using the Unified Program and Debug Interface
(UPDI). The locking protects both the Flash (all Boot Code, Application Code, and Application Data sections), SRAM,
and the EEPROM including the FUSE data. This prevents the reading of application data or code using the debugger
interface. Regular memory access from within the application is still enabled.
The device is locked by writing a non-valid key to the Lock Key (LOCK.KEY) register.
Table 8-5. Memory Access Unlocked (LOCK.KEY Valid Key)(1)
Memory Section
CPU Access
UPDI Access
Read
Write
Read
Write
Flash
Yes
Yes
Yes
Yes
SRAM
Yes
Yes
Yes
Yes
EEPROM
Yes
Yes
Yes
Yes
USERROW
Yes
Yes
Yes
Yes
SIGROW
Yes
No
Yes
No
FUSE
Yes
No
Yes
Yes
LOCK
Yes
No
Yes
Yes
Registers
Yes
Yes
Yes
Yes
Table 8-6. Memory Access Locked (LOCK.KEY Invalid Key)(1)
Memory Section
CPU Access
UPDI Access
Read
Write
Read
Write
Flash
Yes
Yes
No
No
SRAM
Yes
Yes
No
No
EEPROM
Yes
Yes
No
No
USERROW
Yes
Yes
No
Yes(2)
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 59
AVR128DB28/32/48/64
Memories
...........continued
Memory Section
CPU Access
UPDI Access
Read
Write
Read
Write
SIGROW
Yes
No
No
No
FUSE
Yes
No
No
No
LOCK
Yes
No
No
No
Registers
Yes
Yes
No
No
Notes:
1. Read operations marked No in the tables may appear to be successful, but the data is not valid. Hence, any
attempt of code validation through the UPDI will fail on these memory sections.
2. In the Locked mode, the USERROW can be written using the Fuse Write command, but the current
USERROW values cannot be read.
Important: The only way to unlock a device is to perform a CHIPERASE. No application data is retained.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 60
AVR128DB28/32/48/64
Memories
8.9.1
Offset
0x00
8.9.2
Lock Summary
Name
Bit Pos.
KEY
7:0
15:8
23:16
31:24
7
6
5
4
3
2
1
0
KEY[7:0]
KEY[15:8]
KEY[23:16]
KEY[31:24]
Lock Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 61
AVR128DB28/32/48/64
Memories
8.9.2.1
Lock Key
Name:
Offset:
Reset:
Property:
Bit
KEY
0x00
Initial factory value 0x5CC5C55C
-
31
30
29
28
27
26
25
24
R
x
R
x
R
x
R
x
19
18
17
16
R
x
R
x
R
x
R
x
11
10
9
8
R
x
R
x
R
x
R
x
3
2
1
0
R
x
R
x
R
x
R
x
KEY[31:24]
Access
Reset
R
x
R
x
R
x
R
x
Bit
23
22
21
20
KEY[23:16]
Access
Reset
R
x
R
x
R
x
R
x
Bit
15
14
13
12
KEY[15:8]
Access
Reset
R
x
R
x
R
x
R
x
Bit
7
6
5
4
KEY[7:0]
Access
Reset
R
x
R
x
R
x
R
x
Bits 31:0 – KEY[31:0] Lock Key
This bit field controls whether the device is locked or not.
8.10
Value
Name
Description
0x5CC5C55C
Other
UNLOCKED
LOCKED
Device unlocked
Device locked
I/O Memory
All AVR128DB28/32/48/64 devices I/O and peripheral registers are located in the I/O memory space. Refer to the
Peripheral Address Map table for further details.
For compatibility with future devices, if a register containing reserved bits is written, the reserved bits should be
written to ‘0’. Reserved I/O memory addresses should never be written.
8.10.1
Single-Cycle I/O Registers
The I/O memory ranging from 0x00 to 0x3F can be accessed by a single-cycle CPU instruction using the IN or OUT
instructions.
The peripherals available in the single-cycle I/O registers are as follows:
• VPORTx
– Refer to the I/O Configuration section for further details
• GPR
– Refer to the General Purpose Registers section for further details
• CPU
– Refer to the AVR CPU section for further details
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 62
AVR128DB28/32/48/64
Memories
The single-cycle I/O registers ranging from 0x00 to 0x1F (VPORTx and GPR) are also directly bit-accessible using
the SBI or CBI instruction. In these single-cycle I/O registers, single bits can be checked by using the SBIS or SBIC
instruction.
Refer to the Instruction Set Summary section for further details.
8.10.2
Extended I/O Registers
The I/O memory space ranging from 0x0040 to 0x103F can only be accessed by the LD/LDS/LDD or ST/STS/STD
instructions, transferring data between the 32 general purpose working registers (R0-R31) and the I/O memory
space.
Refer to the Peripheral Address Map table and the Instruction Set Summary section for further details.
8.10.3
Accessing 16-bit Registers
Most of the registers for the AVR128DB28/32/48/64 devices are 8-bit registers, but the devices also feature a few 16bit registers. As the AVR data bus has a width of 8 bits, accessing the 16-bit requires two read or write operations. All
the 16-bit registers of the AVR128DB28/32/48/64 devices are connected to the 8-bit bus through a temporary (TEMP)
register.
Figure 8-2. 16-Bit Register Write Operation
DATAH
TEMP
DATAL
Write Low Byte
A
V
R
D
A
T
A
B
U
S
DATAH
TEMP
DATAL
A
V
R
D
A
T
A
B
U
S
Write High Byte
For a 16-bit write operation, the low byte register (e.g., DATAL) of the 16-bit register must be written before the high
byte register (e.g., DATAH). Writing the low byte register will result in a write to the temporary (TEMP) register instead
of the low byte register, as shown on the left side of the 16-Bit Register Write Operation figure. When the high byte
register of the 16-bit register is written, TEMP will be copied into the low byte of the 16-bit register in the same clock
cycle, as shown on the right side of the 16-Bit Register Write Operation figure.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 63
AVR128DB28/32/48/64
Memories
Figure 8-3. 16-Bit Register Read Operation
DATAH
TEMP
DATAL
Read Low Byte
A
V
R
D
A
T
A
B
U
S
DATAH
TEMP
DATAL
A
V
R
D
A
T
A
B
U
S
Read High Byte
For a 16-bit read operation, the low byte register (e.g., DATAL) of the 16-bit register must be read before the high
byte register (e.g., DATAH). When the low byte register is read, the high byte register of the 16-bit register is copied
into the temporary (TEMP) register in the same clock cycle, as shown on the left side of the 16-Bit Register Read
Operation figure. Reading the high byte register will result in a read from TEMP instead of the high byte register, as
shown on right side of the 16-Bit Register Read Operation figure.
The described mechanism ensures that the low and high bytes of 16-bit registers are always accessed
simultaneously when reading or writing the registers.
Interrupts can corrupt the timed sequence if an interrupt is triggered during a 16-bit read/write operation and a 16-bit
register within the same peripheral is accessed in the interrupt service routine. To prevent this, interrupts should be
disabled when writing or reading 16-bit registers. Alternatively, the temporary register can be read before and
restored after the 16-bit access in the interrupt service routine.
8.10.4
Accessing 24-bit Registers
For 24-bit registers, the read and write access is done in the same way as described for 16-bit registers, except there
are two temporary registers for 24-bit registers. The Most Significant Byte must be written last when writing to the
register, and the Least Significant Byte must be read first when reading the register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 64
AVR128DB28/32/48/64
GPR - General Purpose Registers
9.
GPR - General Purpose Registers
The AVR128DB28/32/48/64 devices provide four General Purpose Registers. These registers can be used for storing
any information, and they are particularly useful for storing global variables and interrupt flags. No implicit or explicit
semantic applies to the bits in the General Purpose Registers. The interpretation of the bit values is completely
determined by software.
General Purpose Registers, which reside in the address range 0x001C - 0x001F, are directly bit-accessible using the
SBI, CBI, SBIS, and SBIC instructions.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 65
AVR128DB28/32/48/64
GPR - General Purpose Registers
9.1
Register Summary
Offset
Name
Bit Pos.
0x00
0x01
0x02
0x03
GPR0
GPR1
GPR2
GPR3
7:0
7:0
7:0
7:0
9.2
7
6
5
4
3
2
1
0
GPR[7:0]
GPR[7:0]
GPR[7:0]
GPR[7:0]
Register Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 66
AVR128DB28/32/48/64
GPR - General Purpose Registers
9.2.1
General Purpose Register n
Name:
Offset:
Reset:
Property:
GPRn
0x00 + n*0x01 [n=0..3]
0x00
-
These are General Purpose Registers that can be used to store data, such as global variables and flags, in the bit
accessible I/O memory space.
Bit
7
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
GPR[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – GPR[7:0] General Purpose Register Byte
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 67
AVR128DB28/32/48/64
Peripherals and Architecture
10.
Peripherals and Architecture
10.1
Peripheral Address Map
The address map shows the base address for each peripheral. For complete register description and summary for
each peripheral, refer to the respective peripheral sections.
Table 10-1. Peripheral Address Map
Base
Address
Name
Description
28pin
32pin
48pin
64pin
X
X
X
X
X
X
0x0000
VPORTA
Virtual Port A
0x0004
VPORTB
Virtual Port B
0x0008
VPORTC
Virtual Port C
X
X
X
X
0x000C
VPORTD
Virtual Port D
X
X
X
X
0x0010
VPORTE
Virtual Port E
X
X
0x0014
VPORTF
Virtual Port F
X
X
0x0018
VPORTG
Virtual Port G
0x001C
GPR
General Purpose Registers
X
X
X
X
0x0030
CPU
CPU
X
X
X
X
0x0040
RSTCTRL
Reset Controller
X
X
X
X
0x0050
SLPCTRL
Sleep Controller
X
X
X
X
0x0060
CLKCTRL
Clock Controller
X
X
X
X
0x00A0
BOD
Brown-Out Reset Detector
X
X
X
X
0x00B0
VREF
Voltage Reference
X
X
X
X
0x00C0
MVIO
MVIO Controller
X
X
X
X
0x0100
WDT
Watchdog Timer
X
X
X
X
0x0110
CPUINT
Interrupt Controller
X
X
X
X
0x0120
CRCSCAN
Cyclic Redundancy Check Memory Scan
X
X
X
X
0x0140
RTC
Real Time Counter
X
X
X
X
0x01C0
CCL
Configurable Custom Logic
X
X
X
X
0x0200
EVSYS
Event System
X
X
X
X
0x0400
PORTA
Port A Configuration
X
X
X
X
0x0420
PORTB
Port B Configuration
X
X
0x0440
PORTC
Port C Configuration
X
X
X
X
0x0460
PORTD
Port D Configuration
X
X
X
X
0x0480
PORTE
Port E Configuration
X
X
0x04A0
PORTF
Port F Configuration
X
X
0x04C0
PORTG
Port G Configuration
© 2020 Microchip Technology Inc.
X
X
X
Preliminary Datasheet
X
X
X
DS40002247A-page 68
AVR128DB28/32/48/64
Peripherals and Architecture
...........continued
Base
Address
Name
Description
28pin
32pin
48pin
64pin
0x05E0
PORTMUX
Port Multiplexer
X
X
X
X
0x0600
ADC0
Analog to Digital Converter 0
X
X
X
X
0x0680
AC0
Analog Comparator 0
X
X
X
X
0x0688
AC1
Analog Comparator 1
X
X
X
X
0x0690
AC2
Analog Comparator 2
X
X
X
X
0x06A0
DAC0
Digital to Analog converter 0
X
X
X
X
0x06C0
ZCD0
Zero Cross Detector 0
X
X
X
X
0x06C8
ZCD1
Zero Cross Detector 1
X
X
0x06D0
ZCD2
Zero Cross Detector 2
0x0700
OPAMP
Analog Signal Conditioning
X
X
X
X
0x0800
USART0
Universal Synchronous Asynchronous Receiver
and Transmitter 0
X
X
X
X
0x0820
USART1
Universal Synchronous Asynchronous Receiver
and Transmitter 1
X
X
X
X
0x0840
USART2
Universal Synchronous Asynchronous Receiver
and Transmitter 2
X
X
X
X
0x0860
USART3
Universal Synchronous Asynchronous Receiver
and Transmitter 3
X
X
0x0880
USART4
Universal Synchronous Asynchronous Receiver
and Transmitter 4
X
X
0x08A0
USART5
Universal Synchronous Asynchronous Receiver
and Transmitter 5
0x0900
TWI0
Two-Wire Interface 0
0x0920
TWI1
Two-Wire Interface 1
0x0940
SPI0
Serial Peripheral Interface 0
0x0960
SPI1
0x0A00
X
X
X
X
X
X
X
X
X
X
X
X
Serial Peripheral Interface 1
X
X
X
X
TCA0
Timer/Counter Type A 0
X
X
X
X
0x0A40
TCA1
Timer/Counter Type A 1
X
X
0x0B00
TCB0
Timer/Counter Type B 0
X
X
X
X
0x0B10
TCB1
Timer/Counter Type B 1
X
X
X
X
0x0B20
TCB2
Timer/Counter Type B 2
X
X
X
X
0x0B30
TCB3
Timer/Counter Type B 3
X
X
0x0B40
TCB4
Timer/Counter Type B 4
0x0B80
TCD0
Timer/Counter Type D 0
X
X
X
X
0x0F00
SYSCFG
System Configuration
X
X
X
X
0x1000
NVMCTRL
Non Volatile Memory Controller
X
X
X
X
© 2020 Microchip Technology Inc.
Preliminary Datasheet
X
X
DS40002247A-page 69
AVR128DB28/32/48/64
Peripherals and Architecture
Table 10-2. System Memory Address Map
Base Address
10.2
Name
Description
28-pin
32-pin
48-pin
64-pin
0x1040
LOCK
Lock bits
X
X
X
X
0x1050
FUSE
User
Configuration
X
X
X
X
0x1080
USERROW
User row
X
X
X
X
0x1100
SIGROW
Signature row
X
X
X
X
Interrupt Vector Mapping
Each of the interrupt vectors is connected to one peripheral instance, as shown in the table below. A peripheral can
have one or more interrupt sources. For more details on the available interrupt sources, see the Interrupt section in
the Functional Description of the respective peripheral.
An interrupt flag is set in the Interrupt Flags register of the peripheral (peripheral.INTFLAGS) when the interrupt
condition occurs, even if the interrupt is not enabled.
An interrupt is enabled or disabled by writing to the corresponding Interrupt Enable bit in the peripheral's Interrupt
Control register (peripheral.INTCTRL).
An interrupt request is generated when the corresponding interrupt is enabled, and the interrupt flag is set. Interrupts
must be enabled globally for interrupt request to be generated. The interrupt request remains active until the interrupt
flag is cleared. See the peripheral's INTFLAGS register for details on how to clear interrupt flags.
Table 10-3. Interrupt Vector Mapping
Vector
number
Program
Address
(word)
Peripheral
0
0x00
RESET
1
0x02
NMI
2
0x04
3
Description
28- 32- 48- 64pin pin pin pin
X
X
X
X
Non-Maskable Interrupt
X
X
X
X
BOD
Voltage Level Monitor Interrupt
X
X
X
X
0x06
CLKCTRL
External crystal oscillator/clock source failure
Interrupt (CFD)
X
X
X
X
4
0x08
MVIO
MVIO Interrupt
X
X
X
X
5
0x0A
RTC
Overflow or Compare Match Interrupt
X
X
X
X
6
0x0C
RTC
Periodic Interrupt (PIT)
X
X
X
X
7
0x0E
CCL
Configurable Custom Logic Interrupt
X
X
X
X
8
0x10
PORTA
PORTA External interrupt
X
X
X
X
9
0x12
TCA0
Normal: Overflow Interrupt
Split: Low Underflow Interrupt
X
X
X
X
10
0x14
TCA0
Normal: Unused
Split: High Underflow Interrupt
X
X
X
X
11
0x16
TCA0
Normal: Compare 0 Interrupt
Split: Low Compare 0 Interrupt
X
X
X
X
12
0x18
TCA0
Normal: Compare 1 Interrupt
Split: Low Compare 1 Interrupt
X
X
X
X
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 70
AVR128DB28/32/48/64
Peripherals and Architecture
...........continued
Vector
number
Program
Address
(word)
13
0x1A
TCA0
Normal: Compare 2 Interrupt
Split: Low Compare 2 Interrupt
X
X
X
X
14
0x1C
TCB0
Capture/Overflow Interrupt
X
X
X
X
15
0x1E
TCB1
Capture/Overflow Interrupt
X
X
X
X
16
0x20
TCD0
Overflow (OVF) Interrupt
X
X
X
X
17
0x22
TCD0
Trigger (TRIG) Interrupt
X
X
X
X
18
0x24
TWI0
Slave (TWIS) Interrupt
X
X
X
X
19
0x26
TWI0
Master (TWIM) Interrupt
X
X
X
X
20
0x28
SPI0
SPI0 Interrupt
X
X
X
X
21
0x2A
USART0
Receive Complete (RXC) Interrupt
X
X
X
X
22
0x2C
USART0
Data Register Empty (DRE) Interrupt
X
X
X
X
23
0x2E
USART0
Transmit Complete (TXC) Interrupt
X
X
X
X
24
0x30
PORTD
External Interrupt
X
X
X
X
25
0x32
AC0
Compare Interrupt
X
X
X
X
26
0x34
ADC0
Result Ready (RESRDY) Interrupt
X
X
X
X
27
0x36
ADC0
Window Compare (WCOMP) Interrupt
X
X
X
X
28
0x38
ZCD0
Zero Cross Interrupt
X
X
X
X
29
0x3A
AC1
Compare Interrupt
X
X
X
X
30
0x3C
PORTC
External Interrupt
X
X
X
X
31
0x3E
TCB2
Capture/Overflow Interrupt
X
X
X
X
32
0x40
USART1
Receive Complete (RXC) Interrupt
X
X
X
X
33
0x42
USART1
Data Register Empty (DRE) Interrupt
X
X
X
X
34
0x44
USART1
Transmit Complete (TXC) Interrupt
X
X
X
X
35
0x46
PORTF
External Interrupt
X
X
X
X
36
0x48
NVMCTRL
EEPROM Ready (EEREADY) Interrupt
X
X
X
X
37
0x4A
SPI1
SPI1 Interrupt
X
X
X
X
38
0x4C
USART2
Receive Complete (RXC) Interrupt
X
X
X
X
39
0x4E
USART2
Data Register Empty (DRE) Interrupt
X
X
X
X
40
0x50
USART2
Transmit Complete (TXC) Interrupt
X
X
X
X
41
0x52
AC2
Compare Interrupt
X
X
X
X
42
0x54
TWI1
Slave (TWIS) Interrupt
X
X
X
43
0x56
TWI1
Master (TWIM) Interrupt
X
X
X
44
0x58
TCB3
Capture Interrupt
X
X
45
0x5A
PORTB
External Interrupt
X
X
© 2020 Microchip Technology Inc.
Peripheral
Description
Preliminary Datasheet
28- 32- 48- 64pin pin pin pin
DS40002247A-page 71
AVR128DB28/32/48/64
Peripherals and Architecture
...........continued
10.3
Vector
number
Program
Address
(word)
Peripheral
Description
28- 32- 48- 64pin pin pin pin
46
0x5C
PORTE
External Interrupt
X
X
47
0x5E
TCA1
Normal: Overflow Interrupt
Split: Low Underflow Interrupt
X
X
48
0x60
TCA1
Normal: Unused Interrupt
Split: High Underflow Interrupt
X
X
49
0x62
TCA1
Normal: Compare 0 Interrupt
Split: Low Compare 0 Interrupt
X
X
50
0x64
TCA1
Normal: Compare 1 Interrupt
Split: Low Compare 1 Interrupt
X
X
51
0x66
TCA1
Normal: Compare 2 Interrupt
Split: Low Compare 2 Interrupt
X
X
52
0x68
ZCD1
Zero Cross Interrupt
X
X
53
0x6A
USART3
Receive Complete (RXD) Interrupt
X
X
54
0x6C
USART3
Data Register Empty (DRE) Interrupt
X
X
55
0x6E
USART3
Transmit Complete (TXD) Interrupt
X
X
56
0x70
USART4
Receive Complete (RXD) Interrupt
X
X
57
0x72
USART4
Data Register Empty (DRE) Interrupt
X
X
58
0x74
USART4
Transmit Complete (TXD) Interrupt
X
X
59
0x76
PORTG
External Interrupt
X
60
0x78
ZCD2
Zero Cross Interrupt
X
61
0x7A
TCB4
Capture/Overflow Interrupt
X
62
0x7C
USART5
Receive Complete (RXC) Interrupt
X
63
0x7E
USART5
Data Register Empty (DRE) Interrupt
X
64
0x80
USART5
Transmit Complete (TXD) Interrupt
X
SYSCFG - System Configuration
The system configuration contains the revision ID of the part. The revision ID is readable from the CPU, making it
useful for implementing application changes between part revisions.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 72
AVR128DB28/32/48/64
Peripherals and Architecture
10.3.1
Register Summary
Offset
Name
Bit Pos.
0x00
0x01
Reserved
REVID
7:0
10.3.2
7
6
5
4
MAJOR[3:0]
3
2
1
0
MINOR[3:0]
Register Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 73
AVR128DB28/32/48/64
Peripherals and Architecture
10.3.2.1 Device Revision ID Register
Name:
Offset:
Reset:
Property:
REVID
0x01
[revision ID]
-
This register is read only and give the device revision ID.
Bit
7
6
5
4
3
2
MAJOR[3:0]
Access
Reset
R
x
R
x
1
0
R
x
R
x
MINOR[3:0]
R
x
R
x
R
x
R
x
Bits 7:4 – MAJOR[3:0] Major revision
This bit field contains the major revision for the device. 0x00 = A, 0x01 = B, and so on.
Bits 3:0 – MINOR[3:0] Minor revision
This bit field contains the minor revision for the device. 0x00 = 0, 0x01 = 1, and so on.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 74
AVR128DB28/32/48/64
NVMCTRL - Nonvolatile Memory Controller
11.
NVMCTRL - Nonvolatile Memory Controller
11.1
Features
•
•
•
•
•
11.2
In-System Programmable
Self-Programming and Boot Loader Support
Configurable Memory Sections:
– Boot loader code section
– Application code section
– Application data section
Signature Row for Factory-Programmed Data:
– ID for each device type
– Serial number for each device
– Calibration bytes for factory-calibrated peripherals
User Row for Application Data:
– Can be read and written from software
– Can be written from the UPDI on a locked device
– Content is kept after chip erase
Overview
The NVM Controller (NVMCTRL) is the interface between the CPU and Nonvolatile Memories (Flash, EEPROM,
Signature Row, User Row and fuses). These are reprogrammable memory blocks that retain their values even when
they are not powered. The Flash is mainly used for program storage and can also be used for data storage. The
EEPROM is used for data storage and can be programmed while the CPU is running the program from the Flash.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 75
AVR128DB28/32/48/64
NVMCTRL - Nonvolatile Memory Controller
11.2.1
Block Diagram
Figure 11-1. NVMCTRL Block Diagram
Nonvolatile
Memory Block
Program Memory Bus
`
Flash
EEPROM
Data Memory Bus
Signature Row
User Row
Fuses
Register access
NVMCTRL
11.3
Functional Description
11.3.1
Memory Organization
11.3.1.1 Flash
The Flash is divided into a set of pages. A page is the smallest addressable unit when erasing the Flash. It is only
possible to erase an entire page or multiple pages at a time. Writes can be done per byte or word. One page consists
of 512 bytes.
The Flash can be divided into three sections, each consisting of a variable number of pages. These sections are:
Boot Loader Code (BOOT) Section
The Flash section with full write access. Boot loader software must be placed in this section if used.
Application Code (APPCODE) Section
The Flash section with limited write access. An executable application code is usually placed in this section.
Application Data (APPDATA) Section
The Flash section without write access. Parameters are usually placed in this section.
Inter-Section Write Protection
For security reasons, it is not possible to write to the section of Flash the code is currently executing from. Code
writing to the APPCODE section needs to be executed from the BOOT section, and code writing to the APPDATA
section needs to be executed from either the BOOT section or the APPCODE section.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 76
AVR128DB28/32/48/64
NVMCTRL - Nonvolatile Memory Controller
Table 11-1. Write Protection for Self-Programming
Program Execution Section
Section Being Addressed
BOOT
Programming
Allowed?
CPU Halted?
No
-
APPCODE
BOOT
Yes
APPDATA
Yes
Yes
EEPROM
No
BOOT
No
APPCODE
APPCODE
APPDATA
Yes
Yes
EEPROM
No
BOOT
APPCODE
APPDATA
No
APPDATA
-
EEPROM
Section Sizes
The sizes of these sections are set by the Boot Size (FUSE.BOOTSIZE) fuse and the Code Size (FUSE.CODESIZE)
fuse. The fuses select the section sizes in blocks of 512 bytes. The BOOT section stretches from FLASHSTART to
BOOTEND. The APPCODE section spreads from BOOTEND until APPEND. The remaining area is the APPDATA
section.
Figure 11-2. Flash Sections Sizes and Locations
FLASHSTART : 0x00000000
BOOT
BOOTEND: (BOOTSIZE*512) - 1
APPCODE
APPEND:
(CODESIZE*512) - 1
APPDATA
FLASHEND
If FUSE.BOOTSIZE is written to ‘0’, the entire Flash is regarded as the BOOT section. If FUSE.CODESIZE is written
to ‘0’ and FUSE.BOOTSIZE > 0, the APPCODE section runs from BOOTEND to the end of Flash (no APPDATA
section).
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 77
AVR128DB28/32/48/64
NVMCTRL - Nonvolatile Memory Controller
When FUSE.CODESIZE ≤ FUSE.BOOTSIZE, the APPCODE section is removed, and the APPDATA runs from
BOOTEND to the end of Flash.
Table 11-2. Setting Up Flash Sections
BOOTSIZE
CODESIZE
BOOT Section
APPCODE Section
APPDATA Section
0
-
0 to FLASHEND
-
-
>0
0
0 to BOOTEND
BOOTEND to
FLASHEND
-
>0
≤ BOOTSIZE
0 to BOOTEND
-
BOOTEND to
FLASHEND
>0
> BOOTSIZE
0 to BOOTEND
BOOTEND to APPEND
APPEND to
FLASHEND
If there is no boot loader software, it is recommended to use the BOOT section for application code.
Notes:
1. After Reset, the default vector table location is at the start of the APPCODE section. The peripheral interrupts
can be used in the code running in the BOOT section by relocating the interrupt vector table at the start of this
section. That is done by setting the IVSEL bit in the CPUINT.CTRLA register. Refer to the CPUINT section for
details.
2. If BOOTEND/APPEND, as resulted from BOOTSIZE/CODESIZE fuse setting, exceed the device FLASHEND,
the corresponding fuse setting is ignored, and the default value is used. Refer to “Fuse” in the Memories
section for default values.
Example 11-1. Size of Flash Sections Example
If FUSE.BOOTSIZE is written to 0x04 and FUSE.CODESIZE is written to 0x08, the first 4*512
bytes will be BOOT, the next 4*512 bytes will be APPCODE, and the remaining Flash will be
APPDATA.
Flash Protection
Additional to the inter-section write protection, the NVMCTRL provides a security mechanism to avoid unwanted
access to the Flash memory sections. Even if the CPU can never write to the BOOT section, a Boot Section Read
Protection (BOOTRP) bit in the Control B (NVMCTRL.CTRLB) register is provided to prevent the read and execution
of code from the BOOT section. This bit can be set only from the code executed in the BOOT section and has effect
only when leaving the BOOT section.
There are two other write protection (APPCODEWP and APPDATAWP) bits in the Control B (NVMCTRL.CTRLB)
register that can be set to prevent further updates of the respective Application Code and Application Data sections.
11.3.1.2 EEPROM
The EEPROM is a 512 bytes nonvolatile memory section that has byte granularity on erase/write. It can be erased in
blocks of 1/2/4/8/16/32 bytes, but writes are done only one byte at a time. It also has an option to do a byte erase and
write in one operation.
11.3.1.3 Signature Row
The Signature Row contains a device ID that identifies each microcontroller device type and a serial number for each
manufactured device. The serial number consists of the production lot number, wafer number, and wafer coordinates
for the device. The Signature Row cannot be written or erased, but it can be read by the CPU or through the UPDI
interface.
11.3.1.4 User Row
The User Row is 32 bytes. This section can be used to store various data, such as calibration/configuration data and
serial numbers. This section is not erased by a chip erase.
The User Row section can be read or written from the CPU. This section can be read from UPDI on a device
unlocked and can be written through UPDI even on a device locked.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 78
AVR128DB28/32/48/64
NVMCTRL - Nonvolatile Memory Controller
11.3.1.5 Fuses
The fuses contain device configuration values and are copied to their respective target registers at the end of the
start-up sequence.
The fuses can be read by the CPU or the UPDI, but can only be programmed or cleared by the UPDI.
11.3.2
Memory Access
For read/write operations, the Flash memory can be accessed either from the code space or from the CPU data
space. When the code space is used, the Flash is accessible through the LPM and SPM instructions.
Additionally, the Flash memory is byte accessible when accessed through the CPU data space. This means that it
shares the same address space and instructions as SRAM, EEPROM, and I/O registers and is accessible using
LD/ST instructions in assembly.
For the LPM and SPM instructions, address 0x0000 is the start of the Flash, but for LD and ST, it is 0x8000, as shown
in the Memory map section.
Addressing Flash Memory in Code Space
For read and write access to the Flash memory in the code space, the RAMPZ register concatenated with the Z
register to create the Address Pointer that is used for LPM/SPM access.
Figure 11-3. Flash Addressing for Self-Programming
Combined RAMPZ
and Z registers
RAMPZ
ZH
ZL
0
Low/High Byte select for (E)LPM
FPAGE
PAGE ADDRESS
WITHIN THE FLASH
FPAGE
FWORD
WORD ADDRESS
WITHIN A PAGE
PROGRAM MEMORY
PAGE
PAGE
INSTRUCTION WORD
00
FWORD
00
01
01
02
02
PAGEEND
FLASHEND
The Flash is word-accessed and organized in pages, so the Address Pointer can be treated as having two sections.
This is shown in Figure 11-3. The word address in the page (FWORD) is held by the least significant bits in the
Address Pointer, while the most significant bits in the Address Pointer hold the Flash page address (FPAGE).
Together, FWORD and FPAGE hold an absolute address to a word in the Flash.
The Flash is word-accessed for code space write operations, so the least significant bit (bit 0) in the Address Pointer
is ignored.
For Flash read operations, one byte is read at a time. For this, the least significant bit (bit 0) in the Address Pointer is
used to select the low byte or high byte in the word address. If this bit is ‘0’, the low byte is read, and if this bit is ‘1’,
the high byte is read.
Once a programming operation is initiated, the address is latched, and the Address Pointer can be updated and used
for other operations.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 79
AVR128DB28/32/48/64
NVMCTRL - Nonvolatile Memory Controller
Addressing Flash in CPU Data Space
The Flash area in data space has only 32 KB. For devices with Flash memory size greater than 32 KB, the Flash
memory is divided into blocks of 32 KB. Those blocks are mapped into data space using the FLMAP bit field of the
NVMCTRL.CTRLB register.
For read and write access to the Flash memory in the CPU data space, the LD/ST instructions are used to access
one byte at a time.
11.3.2.1 Read
Reading the Flash is done using Load Program Memory (LPM) instructions or load type (LD*) instructions with an
address according to the memory map. Reading the EEPROM and Signature Row is done using load type
instructions (LD*). Performing a read operation while a write or erase is in progress will result in a bus wait, and the
instruction will be suspended until the ongoing operation is complete.
11.3.2.2 Programming
The Flash programming is done by writing one byte or one word at a time. Writing from the CPU using store type
instructions (ST*) will write one byte at a time, while a write with the Store Program Memory (SPM) instruction will
write one word at a time.
The NVMCTRL command set supports multiple Flash erase operations. Up to 32 pages can be erased at the same
time. The duration of the erase operation is independent of the number of pages being erased.
The EEPROM erasing has byte granularity with the possibility of erasing up to 32 bytes in one operation. The
EEPROM is written one byte at a time, and it has an option to do the erase and write of one byte in the same
operation.
The User Row is erased/written as a normal Flash. When the erasing operation is used, the entire User Row is
erased at once. The User Row writing has byte granularity.
The Fuse programming is identical to the EEPROM programming, but it can be performed only via the UPDI
interface.
Table 11-3. Programming Granularity
Memory Section
Erase Granularity
Write Granularity
Flash array
Page
Word(1)
EEPROM array
Byte
Byte
User Row
Page(2)
Byte(3)
Fuses
Byte
Byte
Notes:
1. Byte granularity when writing to the CPU data space memory mapped section.
2. One page is 32 bytes.
3. Page granularity when programming from UPDI on a locked device.
11.3.2.3 Command Modes
Reading of the memory arrays is handled using the LD*/LPM(1) instructions.
The erase of the whole Flash (CHER) or the EEPROM (EECHER) is started by writing commands to the
NVMCTRL.CTRLA register. The other write/erase operations are just enabled by writing commands to the
NVMCTRL.CTRLA register and must be followed by writes using ST*/SPM(1) instructions to the memory arrays.
Note:
1. LPM/SPM cannot be used for EEPROM.
To write a command in the NVMCTRL.CTRLA register, the following sequence needs to be executed:
1. Confirm that any previous operation is completed by reading the Busy (EEBUSY and FBUSY) flags in the
NVMCTRL.STATUS register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 80
AVR128DB28/32/48/64
NVMCTRL - Nonvolatile Memory Controller
2.
3.
Write the appropriate key to the Configuration Change Protection (CPU.CCP) register to unlock the NVM
Control A (NVMCTRL.CTRLA) register.
Write the desired command value to the CMD bit field in the Control A (NVMCTRL.CTRLA) register within the
next four instructions.
To perform a write/erase operation in the NVM, the following steps are required:
1. Confirm that any previous operation is completed by reading the Busy (EEBUSY and FBUSY) flags in the
NVMCTRL.STATUS register.
2. Optional: If the Flash is accessed in the CPU data space, map the corresponding 32 KB Flash section into the
data space by writing the FLMAP bit field in the NVMCTRL.CTRLB register.
3. Write the desired command value to the NVMCTRL.CTRLA register as described before.
4. Write to the correct address in the data space/code space using the ST*/SPM instructions.
5.
6.
Optional: If multiple write operations are required, go to step 4.
Write a NOOP or NOCMD command to the NVMCTRL.CTRLA register to clear the current command.
11.3.2.3.1 Flash Write Mode
The Flash Write (FLWR) mode of the Flash controller enables writes to the Flash array to start a programming
operation. Several writes can be done while the FLWR mode is enabled in the NVMCTRL.CTRLA register. When the
FLWR mode is enabled, the ST* instructions write one byte at a time, while the SPM instruction writes one word at a
time.
Before a write is performed to an address, its content needs to be erased.
11.3.2.3.2 Flash Page Erase Mode
The Flash Page Erase (FLPER) mode will allow each write to the memory array to erase a page.
An erase operation to the Flash will halt the CPU.
11.3.2.3.3 Flash Multi-Page Erase Mode
The Multi-Page Erase (FLMPERn) mode will allow each write to the memory array to erase multiple pages. When
enabling FLMPERn, it is possible to select between erasing two, four, eight, 16, or 32 pages.
The LSbs of the page address are ignored when defining which Flash pages are erased. Using FLMPER4 as an
example, erasing any page in the 0x08 - 0x0B range will cause the erase of all pages in the range.
Table 11-4. Flash Multi-Page Erase
CMD
Pages Erased
Description
FLMPER2
2
Pages matching FPAGE[N:1] are erased. The value in FPAGE[0] is ignored.
FLMPER4
4
Pages matching FPAGE[N:2] are erased. The value in FPAGE[1:0] is ignored.
FLMPER8
8
Pages matching FPAGE[N:3] are erased. The value in FPAGE[2:0] is ignored.
FLMPER16
16
Pages matching FPAGE[N:4] are erased. The value in FPAGE[3:0] is ignored.
FLMPER32
32
Pages matching FPAGE[N:5] are erased. The value in FPAGE[4:0] is ignored.
Note: FPAGE is the page number when doing a Flash erase. Refer to Figure 11-3 for details.
11.3.2.3.4 EEPROM Write Mode
The EEPROM Write (EEWR) mode enables the EEPROM array for writing operations. Several writes can be done
while the EEWR mode is enabled in the NVMCTRL.CTRLA register. When the EEWR mode is enabled, writes with
the ST* instructions will be performed one byte at a time.
When writing the EEPROM, the CPU will continue executing code. If a new load/store operation is started before the
EEPROM erase/write is completed, the CPU will be halted.
Before a write is performed to an address, its content needs to be erased.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 81
AVR128DB28/32/48/64
NVMCTRL - Nonvolatile Memory Controller
11.3.2.3.5 EEPROM Erase/Write Mode
The EEPROM Erase/Write (EEERWR) mode enables the EEPROM array for the erase operation directly followed by
a write operation. Several erase/writes can be done while the EEERWR mode is enabled in the NVMCTRL.CTRLA
register. When the EEERWR mode is enabled, writes with the ST* instructions are performed one byte at a time.
When writing/erasing the EEPROM, the CPU will continue executing code.
If a new load or store instruction is started before the erase/write is completed, the CPU will be halted.
11.3.2.3.6 EEPROM Byte Erase Mode
The EEPROM Byte Erase (EEBER) mode will allow each write to the memory array to erase the selected byte. An
erased byte always reads back 0xFF, regardless of the value written to the EEPROM address.
When erasing the EEPROM, the CPU can continue to run from the Flash. If the CPU starts an erase or write
operation while the EEPROM is busy, the CPU will be halted until the previous operation is finished.
11.3.2.3.7 EEPROM Multi-Byte Erase Mode
The EEPROM Multi-Byte Erase (EEMBERn) mode allows erasing several bytes in one operation. When enabling the
EEMBERn mode, it is possible to select between erasing two, four, eight, 16, or 32 bytes in one operation.
The LSbs of the address are ignored when defining which EEPROM locations are erased. For example, while doing
an 8-byte erase, addressing any byte in the 0x18 - 0x1F range will result in erasing the entire range of bytes.
Table 11-5. EEPROM Multi-Byte Erase
Description(1)
CMD
Bytes Erased
EEMBER2
2
Addresses matching ADDR[N:1] are erased. The value in ADDR[0] is ignored.
EEMBER4
4
Addresses matching ADDR[N:2] are erased. The value in ADDR[1:0] is ignored.
EEMBER8
8
Addresses matching ADDR[N:3] are erased. The value in ADDR[2:0] is ignored.
EEMBER16
16
Addresses matching ADDR[N:4] are erased. The value in ADDR[3:0] is ignored.
EEMBER32
32
Addresses matching ADDR[N:5] are erased. The value in ADDR[4:0] is ignored.
Note: ADDR is the address written when doing an EEPROM erase.
When erasing the EEPROM, the CPU can continue to execute instructions from the Flash. If the CPU starts an erase
or write operation while the EEPROM is busy, the NVMCTRL module will give a wait on the bus, and the CPU will be
halted until the current operation is finished.
11.3.2.3.8 Chip Erase Command
The Chip Erase (CHER) command erases the Flash and the EEPROM. The EEPROM is unaltered if the EEPROM
Save During Chip Erase (EESAVE) fuse in FUSE.SYSCFG0 is set.
If the device is locked, the EEPROM is always erased by a chip erase regardless of the EESAVE bit. The read/write
protection (BOOTRP, APPCODEWP, APPDATAWP) bits in NVMCTRL.CTRLB do not prevent the operation. All Flash
and EEPROM bytes will read back 0xFF after this command.
This command can only be started from the UPDI.
11.3.2.3.9 EEPROM Erase Command
The EEPROM Erase (EECHER) command erases the EEPROM. All EEPROM bytes will read back 0xFF after the
operation. The CPU will be halted while the EEPROM is being erased.
11.3.3
Preventing Flash/EEPROM Corruption
A Flash/EEPROM write or erase can cause memory corruption if the supply voltage is too low for the CPU and the
Flash/EEPROM to operate properly. These issues are the same on-board level systems using Flash/EEPROM, and it
is recommended to use the internal or an external Brown-Out Detector (BOD) to ensure that the device is not
operating at too low voltage.
When the voltage is too low, a Flash/EEPROM corruption may be caused by two circumstances:
1. A regular write sequence to the Flash, which requires a minimum voltage to operate correctly.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 82
AVR128DB28/32/48/64
NVMCTRL - Nonvolatile Memory Controller
2.
The CPU itself can execute instructions incorrectly when the supply voltage is too low.
The chip erase does not clear fuses. If the BOD is enabled before starting the Chip Erase command, it is
automatically enabled at its previous configured level during the chip erase.
Refer to the Electrical Characteristics section for Maximum Frequency vs. VDD.
Attention: Flash/EEPROM corruption can be avoided by taking the following measures:
1. Keep the device in Reset during periods of insufficient power supply voltage. This can be done by
enabling the internal BOD.
2. The Voltage Level Monitor (VLM) in the BOD can be used to prevent starting a write to the
EEPROM close to the BOD level.
3. If the detection levels of the internal BOD do not match the required detection level, an external low
VDD Reset protection circuit can be used. If a Reset occurs while a write operation is ongoing, the
write operation will be aborted.
11.3.4
Interrupts
Table 11-6. Available Interrupt Vectors and Sources
Offset
Name
Vector Description
Conditions
0x00
EEREADY
NVM
The EEPROM is ready for new write/erase operations.
When an interrupt condition occurs, the corresponding interrupt flag is set in the Interrupt Flags
(NVMCTRL.INTFLAGS) register.
An interrupt source is enabled or disabled by writing to the corresponding bit in the Interrupt Control
(NVMCTRL.INTCTRL) register.
An interrupt request is generated when the corresponding interrupt source is enabled, and the interrupt flag is set.
The interrupt request remains active until the interrupt flag is cleared. See the NVMCTRL.INTFLAGS register for
details on how to clear interrupt flags.
11.3.5
Sleep Mode Operation
If there is no ongoing EEPROM write/erase operation, the NVMCTRL will enter sleep mode, when the system enters
sleep mode.
If an EEPROM write/erase operation is ongoing when the system enters a sleep mode, the NVM block, the
NVMCTRL and the peripheral clock will remain ON until the operation is finished and will be automatically turned off
once the operation is completed. This is valid for all sleep modes, including Power-Down.
The EEPROM Ready interrupt will wake up the device only from Idle sleep mode.
11.3.6
Configuration Change Protection
This peripheral has registers that are under Configuration Change Protection (CCP). To write to these registers, a
certain key must first be written to the CPU.CCP register, followed by a write access to the protected bits within four
CPU instructions.
Attempting to write to a protected register without following the appropriate CCP unlock sequence leaves the
protected register unchanged.
The following registers are under CCP:
Table 11-7. NVMCTRL - Registers under Configuration Change Protection
Register
Key
NVMCTRL.CTRLA
SPM
NVMCTRL.CTRLB
IOREG
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 83
AVR128DB28/32/48/64
NVMCTRL - Nonvolatile Memory Controller
11.4
Register Summary
Offset
Name
Bit Pos.
0x00
0x01
0x02
0x03
0x04
0x05
CTRLA
CTRLB
STATUS
INTCTRL
INTFLAGS
Reserved
7:0
7:0
7:0
7:0
7:0
0x06
DATA
0x08
ADDR
11.5
7
6
5
4
3
2
1
0
APPDATAWP
BOOTRP
EEBUSY
APPCODEWP
FBUSY
EEREADY
EEREADY
CMD[6:0]
FLMAPLOCK
7:0
15:8
7:0
15:8
23:16
FLMAP[1:0]
ERROR[2:0]
DATA[7:0]
DATA[15:8]
ADDR[7:0]
ADDR[15:8]
ADDR[23:16]
Register Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 84
AVR128DB28/32/48/64
NVMCTRL - Nonvolatile Memory Controller
11.5.1
Control A
Name:
Offset:
Reset:
Property:
Bit
7
Access
Reset
CTRLA
0x00
0x00
Configuration Change Protection
6
5
4
R/W
0
R/W
0
R/W
0
3
CMD[6:0]
R/W
0
2
1
0
R/W
0
R/W
0
R/W
0
Bits 6:0 – CMD[6:0] Command
Write this bit field to enable or issue a command. The Chip Erase and EEPROM Erase commands are started when
the command is written. The others enable an erase or write operation. The operation is started by doing a store
instruction to an address location.
A change from one command to another must always go through No command (NOCMD) or No operation (NOOP)
command to avoid the Command Collision error being set in the ERROR bit field from the NVMCTRL.STATUS
register.
Value
Name
Description
0x00
0x01
0x02
0x08
0x09
0x0A
0x0B
0x0C
0x0D
0x12
0x13
0x18
0x19
0x1A
0x1B
0x1C
0x1D
NOCMD
NOOP
FLWR
FLPER
FLMPER2
FLMPER4
FLMPER8
FLMPER16
FLMPER32
EEWR
EEERWR
EEBER
EEMBER2
EEMBER4
EEMBER8
EEMBER16
EEMBER32
0x20
CHER
0x30
Other
EECHER
-
No command
No operation
Flash Write Enable
Flash Page Erase Enable
Flash 2-page Erase Enable
Flash 4-page Erase Enable
Flash 8-page Erase Enable
Flash 16-page Erase Enable
Flash 32-page Erase Enable
EEPROM Write Enable
EEPROM Erase and Write Enable
EEPROM Byte Erase Enable
EEPROM 2-byte Erase Enable
EEPROM 4-byte Erase Enable
EEPROM 8-byte Erase Enable
EEPROM 16-byte Erase Enable
EEPROM 32-byte Erase Enable
Erase Flash and EEPROM. EEPROM is skipped if EESAVE fuse is set.
(UPDI access only.)
Erase EEPROM
Reserved
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 85
AVR128DB28/32/48/64
NVMCTRL - Nonvolatile Memory Controller
11.5.2
Control B
Name:
Offset:
Reset:
Property:
CTRLB
0x01
0x30
Configuration Change Protection
Bit
7
FLMAPLOCK
Access
R/W
Reset
0
6
5
4
3
FLMAP[1:0]
R/W
1
R/W
1
2
APPDATAWP
R/W
0
1
BOOTRP
R/W
0
0
APPCODEWP
R/W
0
Bit 7 – FLMAPLOCK Flash Mapping Lock
Setting this bit to ‘1’ prevents further updates of FLMAP[1:0]. This bit can only be cleared by a Reset.
Bits 5:4 – FLMAP[1:0] Flash Section Mapped into Data Space
Select what part (in blocks of 32 KB) of the Flash will be mapped as part of the CPU data space and will be
accessible through LD/ST instructions.
This bit field is not under Configuration Change Protection.
Description: Flash mapping
Value
0x0
0x1
0x2
0x3
Name
Mapped Flash Section (128 KB)
SECTION0
SECTION1
SECTION2
SECTION3
0-32
32-64
64-96
96-128
Bit 2 – APPDATAWP Application Data Section Write Protection
Writing this bit to ‘1’ prevents further updates to the Application Data section. This bit can only be cleared by a Reset.
Bit 1 – BOOTRP Boot Section Read Protection
Writing this bit to ‘1’ will protect the BOOT section from reading and instruction fetching. If a read is issued from the
other Flash sections, it will return ‘0’. An instruction fetch from the BOOT section will return a NOP instruction. This bit
can only be written from the BOOT section, and it can only be cleared by a Reset. The read protection will only take
effect when leaving the BOOT section after the bit is written.
Bit 0 – APPCODEWP Application Code Section Write Protection
Writing this bit to ‘1’ prevents further updates to the Application Code section. This bit can only be cleared by a
Reset.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 86
AVR128DB28/32/48/64
NVMCTRL - Nonvolatile Memory Controller
11.5.3
Status
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
STATUS
0x02
0x00
-
6
R/W
0
5
ERROR[2:0]
R/W
0
4
3
R/W
0
2
1
EEBUSY
R
0
0
FBUSY
R
0
Bits 6:4 – ERROR[2:0] Error Code
The Error Code bit field will show the last error occurring. This bit field can be cleared by writing it to ‘0’.
Value
Name
Description
0x0
NONE
No error
0x1
INVALIDCMD
The selected command is not supported
0x2
WRITEPROTECT Attempt to write a section that is protected
0x3
CMDCOLLISION A new write/erase command was selected while a write/erase command is already
ongoing
Other
—
Reserved
Bit 1 – EEBUSY EEPROM Busy
This bit will read ‘1’ when an EEPROM programming operation is ongoing.
Bit 0 – FBUSY Flash Busy
This bit will read ‘1’ when a Flash programming operation is ongoing.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 87
AVR128DB28/32/48/64
NVMCTRL - Nonvolatile Memory Controller
11.5.4
Interrupt Control
Name:
Offset:
Reset:
Property:
Bit
7
INTCTRL
0x03
0x00
-
6
5
4
3
Access
Reset
2
1
0
EEREADY
R/W
0
Bit 0 – EEREADY EEPROM Ready Interrupt
Writing a ‘1’ to this bit enables the interrupt which indicates that the EEPROM is ready for new write/erase
operations.
This is a level interrupt that will be triggered only when the EEREADY bit in the INTFLAGS register is set to ‘1’. The
interrupt must not be enabled before triggering an EEPROM write/erase operation, as the EEREADY bit will not be
cleared before this command is issued. The interrupt must be disabled in the interrupt handler.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 88
AVR128DB28/32/48/64
NVMCTRL - Nonvolatile Memory Controller
11.5.5
Interrupt Flags
Name:
Offset:
Reset:
Property:
Bit
7
INTFLAGS
0x04
0x00
-
6
5
4
3
2
1
0
EEREADY
R/W
0
Access
Reset
Bit 0 – EEREADY EEREADY Interrupt Flag
This flag is set continuously as long as the EEPROM is not busy. This flag is cleared by writing a ‘1’ to it.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 89
AVR128DB28/32/48/64
NVMCTRL - Nonvolatile Memory Controller
11.5.6
Data
Name:
Offset:
Reset:
Property:
DATA
0x06
0x00
-
The NVMCTRL.DATAL and NVMCTRL.DATAH register pair represents the 16-bit value, NVMCTRL.DATA.
The low byte [7:0] (suffix L) is accessible at the original offset.
The high byte [15:8] (suffix H) can be accessed at offset + 0x01.
Bit
15
14
13
12
11
10
9
8
R/W
0
R/W
0
R/W
0
R/W
0
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
DATA[15:8]
Access
Reset
Bit
R/W
0
R/W
0
R/W
0
R/W
0
7
6
5
4
DATA[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 15:0 – DATA[15:0] Data Register
The Data register will contain the last read value from Flash, EEPROM, or NVMCTRL. For EEPROM access, only
DATA[7:0] is used.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 90
AVR128DB28/32/48/64
NVMCTRL - Nonvolatile Memory Controller
11.5.7
Address
Name:
Offset:
Reset:
Property:
ADDR
0x08
0x00
-
NVMCTRL.ADDR0, NVMCTRL.ADDR1 and NVMCTRL.ADDR2 represent the 24-bit value NVMCTRL.ADDR.
The low byte [7:0] (suffix 0) is accessible at the original offset.
The high byte [15:8] (suffix 1) can be accessed at offset +0x01.
The extended byte [23:16] (suffix 2) can be accessed at offset +0x02.
Bit
Access
Reset
Bit
23
22
21
R/W
0
R/W
0
R/W
0
15
14
13
20
19
ADDR[23:16]
R/W
R/W
0
0
12
18
17
16
R/W
0
R/W
0
R/W
0
11
10
9
8
R/W
0
R/W
0
R/W
0
R/W
0
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
ADDR[15:8]
Access
Reset
Bit
R/W
0
R/W
0
R/W
0
R/W
0
7
6
5
4
ADDR[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 23:0 – ADDR[23:0] Address
The Address register contains the address of the last memory location that has been accessed. Only the number of
bits required to access the memory is used.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 91
AVR128DB28/32/48/64
CLKCTRL - Clock Controller
12.
CLKCTRL - Clock Controller
12.1
Features
•
•
•
•
•
12.2
All Clocks and Clock Sources are Automatically Enabled when Requested by Peripherals
Internal Oscillators:
– Up to 24 MHz Internal High-Frequency Oscillator (OSCHF)
– 32.768 kHz Ultra Low-Power Oscillator (OSC32K)
– Up to 48 MHz Phase-Locked Loop (PLL), with 2x or 3x clock multiplier
Auto-Tuning for Improved Internal Oscillator Accuracy
External Clock Options:
– 32.768 kHz Crystal Oscillator (XOSC32K)
– High-Frequency Crystal Oscillator (XOSCHF)
– External clock
Main Clock Features:
– Safe run-time switching
– Prescaler with a division factor ranging from 1 to 64
– Clock Failure Detection with automatic clock switching to internal source
Overview
The Clock Controller (CLKCTRL) controls, distributes and prescales the clock signals from the available oscillators.
The CLKCTRL supports internal and external clock sources.
The CLKCTRL is based on an automatic clock request system, implemented in all peripherals on the device. The
peripherals will automatically request the clocks needed. The request is routed to the correct clock source if multiple
clock sources are available.
The Main Clock (CLK_MAIN) is used by the CPU, RAM and all peripherals connected to the I/O bus. The main clock
source can be selected and prescaled. Some peripherals can share the same clock source as the main clock, or run
asynchronously to the main clock domain.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 92
AVR128DB28/32/48/64
CLKCTRL - Clock Controller
12.2.1
Block Diagram - CLKCTRL
Figure 12-1. CLKCTRL Block Diagram
RTC
CPU
NVM
RAM
CLKOUT
Peripherals
TCD
CLK_CPU
CLK_TCD
CLK_RTC
TCD CLKSEL
BOD
Interrupt
WDT
CFD
CLK_BOD
DIV8
RTC CLKSEL
CLK_PER
Main Clock
Prescaler
CFDSRC
PLL
CLK_MAIN
CLK_WDT
PLLSRC
Main Clock Switch
DIV32
XOSC32KSEL
32.768 kHz
OSC32K
XOSCSEL
32.768 kHz
XOSC32K
XOSCHF
XTAL32K1 XTAL32K2
XTALHF1
OSCHF
XTALHF2
The clock system consists of the main clock and clocks derived from the main clock, as well as several asynchronous
clocks:
• Main Clock (CLK_MAIN) is always running in Active and Idle sleep modes. If requested, it will also run in
Standby sleep mode.
• CLK_MAIN is prescaled and distributed by the clock controller:
– CLK_CPU is used by the CPU and the Nonvolatile Memory Controller (NVMCTRL)
– CLK_PER is used by SRAM and all peripherals that are not listed under asynchronous clocks and can also
be routed to the CLKOUT pin
– All the clock sources can be used as main clock
• Clocks running asynchronously to the main clock domain:
– CLK_RTC is used by the Real-Time Counter (RTC) and the Periodic Interrupt Timer (PIT). It will be
requested when the RTC/PIT is enabled. The clock source for CLK_RTC may be changed only if the
peripheral is disabled.
– CLK_WDT is used by the Watchdog Timer (WDT). It will be requested when the WDT is enabled.
– CLK_BOD is used by the Brown-out Detector (BOD). It will be requested when the BOD is enabled in the
Sampled mode. The alternative clock source is controlled by a fuse.
– CLK_TCD is used by the Timer Counter type D (TCD). It will be requested when the TCD is enabled. The
clock source may be changed only if the peripheral is disabled.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 93
AVR128DB28/32/48/64
CLKCTRL - Clock Controller
– Clock Failure Detector (CFD) is an asynchronous mechanism to detect a failure on an external crystal or
clock source
The clock source for the main clock domain is configured by writing to the Clock Select (CLKSEL) bit field in the Main
Clock Control A (CLKCTRL.MCLKCTRLA) register. This register has Configuration Change Protection (CCP), and
the appropriate key must be written to the CCP register before writing to the CLKSEL bit field. The asynchronous
clock sources are configured by the registers in the respective peripheral.
12.2.2
Signal Description
Signal
Type
Description
CLKOUT
Digital output
CLK_PER output
XTALHF1
Analog input
Input for external clock source (EXTCLK) or one pin of a high-frequency crystal
XTALHF2
Analog input
Input for one pin of a high-frequency crystal
XTAL32K1
Analog input
Input for external 32.768 kHz clock source or one pin of a 32.768 kHz crystal
XTAL32K2
Analog input
Input for one pin of a 32.768 kHz crystal
For more details, refer to the I/O Multiplexing section.
12.3
Functional Description
12.3.1
Initialization
To initialize a clock source as the main clock, these steps should be followed:
1.
Optional: Force the clock to always run by writing the Run Standby (RUNSTDBY) bit in the respective clock
source CTRLA register to ‘1’.
2.
Configure the clock source as needed in the corresponding clock source CTRLA register and, if applicable,
enable the clock source by writing a ‘1’ to the Enable bit.
3.
Optional: If RUNSTDBY is ‘1’, wait for the clock source to stabilize by polling the respective status bit in
CLKCTRL.MCLKSTATUS.
The following sub-steps need to be performed in an order such that the main clock frequency never exceeds
the allowed maximum clock frequency. Refer to the Electrical Characteristics section for further information.
4.1.
If required, divide the clock source frequency by writing to the Prescaler Division (PDIV) bit field and
enable the main clock prescaler by writing a ‘1’ to the Prescaler Enable (PEN) bit in
CLKCTRL.MCLKCTRLB.
4.2.
Select the configured clock source as the main clock in the Clock Select (CLKSEL) bit field in
CLKCTRL.MCLKCTRLA.
Wait for Main Clock to change by polling the Main Clock Oscillator Changing (SOSC) bit in the Main Clock
Status (CLKCTRL.MCLKSTATUS) register.
Optional: Clear the RUNSTDBY bit in the clock source CTRLA register.
4.
5.
6.
12.3.2
Main Clock Selection and Prescaler
All available oscillators and the external clock (EXTCLK) can be used as the main clock source for the Main Clock
(CLK_MAIN). The main clock source is selectable from software and can be safely changed during normal operation.
The Configuration Change Protection mechanism prevents unsafe clock switching. For more details, refer to the
Configuration Change Protection section.
The Clock Failure Detection mechanism ensures safe switching to internal clock source upon clock failure when
enabled.
Upon the selection of an external clock source, a switch to the chosen clock source will occur only if edges are
detected. Until a sufficient number of clock edges are detected, the switch will not occur, and it will not be possible to
change to another clock source again without executing a Reset.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 94
AVR128DB28/32/48/64
CLKCTRL - Clock Controller
An ongoing clock source switch is indicated by the Main Clock Oscillator Changing (SOSC) bit in the Main Clock
Status (CLKCTRL.MCLKSTATUS) register. The stability of the external clock sources is indicated by the respective
Status (EXTS and XOSC32KS) bits in CLKCTRL.MCLKSTATUS.
CAUTION
If an external clock source fails while used as the CLK_MAIN source, only the Watchdog Timer (WDT) can
provide a System Reset.
The CLK_MAIN is fed into the prescaler before being used by the peripherals (CLK_PER) in the device. The
prescaler divides CLK_MAIN by a factor from 1 to 64.
Figure 12-2. Main Clock and Prescaler
OSCHF
32.768 kHz Osc.
32.768 kHz Crystal Osc.
XOSCHF/External clock
12.3.3
CLK_MAIN
Main Clock Prescaler
CLK_PER
(Div 1, 2, 4, 8, 16, 32,
64, 6, 10, 24, 48)
Main Clock After Reset
After any Reset, the Main Clock (CLK_MAIN) is provided either by the OSCHF, running at the default frequency of 4
MHz, or the OSC32K, depending on the Clock Select (CLKSEL) bit field configuration of the Oscillator Configuration
(FUSE.OSCCFG) fuse. Refer to the description of the FUSE.OSCCFG fuse for details of the possible frequencies
after Reset.
12.3.4
Clock Sources
All the internal clock sources are automatically enabled when they are requested by a peripheral. The crystal
oscillators, based on an external crystal, must be enabled before it can serve as a clock source.
• The XOSC32K oscillator is enabled by writing a ‘1’ to the ENABLE bit in the 32.768 kHz Crystal Oscillator
Control A (CLKCTRL.XOSC32KCTRLA) register
• The XOSCHF oscillator is enabled by writing a ‘1’ to the ENABLE bit in the High Frequency Crystal Oscillator
Control A (CLKCTRL.XOSCHFCTRLA) register
After reset, the device starts up running from the internal high-frequency oscillator or the internal 32.768 kHz
oscillator.
The respective oscillator status bits in the Main Clock Status (CLKCTRL.MCLKSTATUS) register indicate if the clock
source is running and stable.
12.3.4.1 Internal Oscillators
The internal oscillators do not require any external components to run. Refer to the Electrical Characteristics section
for accuracy and electrical specifications.
12.3.4.1.1 Internal High-Frequency Oscillator (OSCHF)
The OSCHF supports output frequencies of 1, 2, 3, 4 MHz, and multiples of 4, up to 24 MHz, which can be used as
main clock, peripheral clock, or as input to the Phase-Locked Loop (PLL).
12.3.4.1.2 32.768 kHz Oscillator (OSC32K)
The 32.768 kHz oscillator is optimized for Ultra Low-Power (ULP) operation. Power consumption is decreased at the
cost of decreased accuracy compared to an external crystal oscillator.
This oscillator provides a 1.024 kHz or 32.768 kHz clock for the Real-Time Counter (RTC), the Watchdog Timer
(WDT) and the Brown-out Detector (BOD). Additionally, this oscillator can also provide a 32.768 kHz clock to the
Main Clock (CLK_MAIN).
For the start-up time of this oscillator, refer to the Electrical Characteristics section.
12.3.4.2 External Clock Sources
These external clock sources are available:
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 95
AVR128DB28/32/48/64
CLKCTRL - Clock Controller
•
•
•
•
The XTALHF1 and XTALHF2 pins are dedicated to driving a high-frequency crystal oscillator (XOSCHF)
Instead of a crystal oscillator, XTALHF1 can be configured to accept an external clock source
The XTAL32K1 and XTAL32K2 pins are dedicated to driving a 32.768 kHz crystal oscillator (XOSC32K)
Instead of a crystal oscillator, XTAL32K1 can be configured to accept an external clock source
12.3.4.2.1 High Frequency Crystal Oscillator (XOSCHF)
This oscillator supports two input options:
•
•
A crystal is connected to the XTALHF1 and XTALHF2 pins
An external clock running at up to 32 MHz connected to XTALHF1
The input option must be configured by writing to the Source Select (SELHF) bit in the XOSCHF Control A
(CLKCTRL.XOSCHFCTRLA) register.
The maximum frequency of the crystal must be configured by writing to the Frequency Range (FRQRANGE) bit field
in XOSCHFCTRLA. This is to ensure sufficient power is delivered to the oscillator to drive the crystal.
The XOSCHF is enabled by writing a ‘1’ to the ENABLE bit in XOSCHFCTRLA. When enabled, the configuration of
the general purpose input/output (GPIO) pins used by the XOSCHF is overridden as XTALHF1 and XTALHF2 pins.
The oscillator needs to be enabled to start running when requested.
The start-up time of a given crystal oscillator can be accommodated by writing to the Crystal Start-up Time (CSUTHF)
bit field in XOSCHFCTRLA.
When XOSCHF is configured to use an external clock on XTALHF1, the start-up time is fixed to two cycles.
12.3.4.2.2 32.768 kHz Crystal Oscillator (XOSC32K)
This oscillator supports two input options:
• A crystal is connected to the XTAL32K1 and XTAL32K2 pins
• An external clock running at 32.768 kHz, connected to XTAL32K1
The input option must be configured by writing the Source Select (SEL) bit in the XOSC32K Control A
(CLKCTRL.XOSC32KCTRLA) register.
The XOSC32K is enabled by writing a ‘1’ to the ENABLE bit in CLKCTRL.XOSC32KCTRLA. When enabled, the
configuration of the general purpose input/output (GPIO) pins used by the XOSC32K is overridden as XTAL32K1 and
XTAL32K2 pins. The oscillator needs to be enabled to start running when requested.
The start-up time of a given crystal oscillator can be accommodated by writing to the Crystal Start-up Time (CSUT) bit
field in XOSC32KCTRLA.
When XOSC32K is configured to use an external clock on XTAL32K1, the start-up time is fixed to two cycles.
12.3.5
Phase-Locked Loop (PLL)
The PLL can be used to increase the frequency of the clock source defined by the SOURCE bit in the PLL Control A
(CLKCTRL.PLLCTRLA) register. The minimum input frequency of the PLL is 16 MHz, and the maximum output
frequency is 48 MHz.
Initialization:
1. Enable the clock source to be used as input.
2. Configure SOURCE in CLKCTRL.PLLCTRLA to the desired clock source.
3. Enable the PLL by writing the desired multiplication factor to the Frequency Select (MULFAC) bit field in
PLLCTRLA.
4. Wait for the PLL Status (PLLS) bit in the Main Clock Status (CLKCTRL.MCLKSTATUS) register to become ‘1’,
indicating that the PLL has locked in on the desired frequency.
For available connections, refer to the Block Diagram - CLKCTRL figure in the Clock Controller (CLKCTRL) section.
12.3.6
Auto-Tune
The auto-tune feature can be used to improve the accuracy of the internal oscillator. The internal 1 MHz clock is
compared to a 1.024 kHz reference from the crystal. If an error is detected, the frequency calibration will be adjusted
according to the error.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 96
AVR128DB28/32/48/64
CLKCTRL - Clock Controller
The check against the reference clock runs for as long as an error is detected. When the oscillator is tuned in, and no
error is detected, the auto-tune feature will disable itself for 64 ms before it starts a new check.
Figure 12-3. OSCHF Auto-Tune Block Diagram
XOSC
32.768 kHz
12.3.7
AUTO-TUNE
Control Logic
"tune up/down"
OSCHF
Clock Failure Detection
The Clock Failure Detection (CFD) allows the device to continue operating if an external crystal oscillator or clock
source should fail. The CFD is enabled by writing a ‘1’ to the Clock Failure Detection Enable (CFDEN) bit in the Main
Clock Control C (CLKCTRL.MCLKCTRLC) register. See the Clock Failure Detection Block Diagram for monitorable
oscillators and clocks sources.
12.3.7.1 CFD Operation
The Clock Failure Detection (CFD) module detects a failed oscillator or clock source by checking for edges on the
selected oscillator/clock during a CFD period. A CFD period is ten cycles on the OSC32K internal oscillator. In the
first eight cycles of the CFD period, edge detectors will detect edges on the monitored oscillator/clock. The two
remaining cycles are used to check and issue a CFD condition if no edges are present, and to reset the edge
detectors at the end of the CFD period.
Figure 12-4. Clock Failure Detection Block Diagram
POSITIVE
EDGE
DETECTOR
CLK_MAIN
XOSCHF
XOSC32K
CFD
CONDITION
RESET
NEGATIVE
EDGE
DETECTOR
CHECK
RESET
OSC32K
CFD
COUNTER
When the CFD module is enabled, it will monitor the selected source from the Clock Failure Detection Source
(CFDSRC) bit field in the Main Clock Control C (CLKCTRL.MCLKCTRLC). In sleep, the CFD will only be enabled if
the selected source is active.
If a CFD condition occurs, the CFD interrupt flag in the Main Clock Interrupt Flags (CLKCTRL.MCLKINTFLAGS)
register is set, and if the interrupt is enabled, an interrupt request is issued. The Interrupt Type (INTTYP) bit in the
Main clock Interrupt control (CLKCTRL.MCLKINTCTRL) register determines if a normal interrupt or a non-maskable
interrupt (NMI) will be issued. If the NMI is selected, the interrupt will change from vector number 3 (CLKCTRL) to
vector number 1 (NMI). If more than one interrupt source is set to NMI, it is necessary to check the vector when an
interrupt triggers to see which source experienced an interrupt.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 97
AVR128DB28/32/48/64
CLKCTRL - Clock Controller
If the selected source is the main clock and it fails, everything running on it will stop. Therefore, the CFD condition will
overwrite the Clock Selection (CLKSEL) bit field in the Main Clock Control A(CLKCTRL.MCLKCTRLA) register to
select the start-up clock source, and the Main Clock Control B (CLKCTRL.MCLKCTRLB) register will be written to the
reset value. The start-up clock source is changed back to its reset frequency.
When the CLKSEL is overridden by a CFD event, the Main Clock Out (CLKOUT) bit in CLKCTRL.MCLKCTRLA will
be set to ‘0’, disabling the CLKOUT signal.
The start-up clock source is defined as the clock the system runs on after a power-on reset. This start-up clock
source is selectable by fuse(s).
Figure 12-5. Clock Failure Detection Timing Diagram
CLOCK
FAILURE
Detect edges
Check for
edge
CFD
Counter
Reset
Detect edges
Check for
edge
CFD
Counter
Reset
CFD Condition
8 Reference Clock Cycles
12.3.7.2 Condition Clearing
The Clock Failure Detection (CFD) condition is cleared after a Reset, the monitored source starts toggling again or
the CFD flag in the Main Clock Interrupt Flags (CLKCTRL.MCLKINTFLAGS) register is set. Note that as long as the
failure condition is met, the interrupt will trigger every ten OSC32K cycles. If these repeated interrupts are not
desired, write a ‘0’ to the Clock Failure Detection Interrupt Enable (CFD) bit in the Main Clock Interrupt Control
(CLKCTRL.MCLKCTRL) register. If it is the main clock that is being monitored, changing back to the default start-up
clock will make the main clock start toggling again, clearing the condition.
12.3.7.3 CFD Test
The Clock Failure Detection Test (CFDTST) bit in the Main Clock Control C (CLKCTL.CTRLC) register can be used
to trigger a clock failure in the clock failure detector. Depending on the use-case, there are two different modes of
testing the clock failure detector.
12.3.7.3.1 Testing the Clock Failure Without Influencing the Main Clock
This mode is intended to use run-time. To not influence the main clock when writing to the Clock Failure Detection
Test (CFDTST) in the Main Clock Control C (CLKCTRL.MCLKCTRLC) register, the Clock Failure Detection Source
(CFDSRC) in CLKCTRL.MCLKCTRLC must be configured to a clock source different than the main clock. CFDSRC
must be different from ‘0’. The CFD Interrupt Flag in the Main Clock Interrupt Flags (CLKCTRL.MCLKINTFLAGS)
register will be set, but the main clock will not change to the start-up clock source.
If the clock failure detector is monitoring the main clock and a run-time check of the clock failure detector is needed, it
is necessary to do the following steps:
1. Disable the clock failure detector by writing a ‘0’ to the Clock Failure Detection Enable (CFDEN) bit in
CLKCTRL.MCLKCTRLC, and change the source to the oscillator directly by writing a number other than a ‘0’
to the CFDSRC bit.
2. Write a ‘1’ to the CFD Interrupt Flag in CLKCTRL.MCLKINTFLAGS to clear the flag.
3.
Write a ‘1’ to the CFDTST bit and enable the clock failure detector again by writing a ‘1’ to the CFDEN bit.
4.
5.
Wait for the CFD bit in CLKCTRL.MCLKINTFLAGS to be set to check that the clock failure works.
Disable the clock failure detector by writing a ‘0’ to the CFDEN bit, and change the source to the main clock
again by writing a ‘0’ to the CFDSRC.
6.
Enable the clock failure detector again by writing a ‘1’ to the CFDEN bit, and write a ‘0’ to the CFDTST bit.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 98
AVR128DB28/32/48/64
CLKCTRL - Clock Controller
12.3.7.3.2 Testing the Clock Failure and Changing the Main Clock to the Start-up Clock Source
If the Clock Failure Detection Source (CFDSRC) bit field in the Main Clock Control C (CLKCTRL.MCLKCTRLC)
register has the value 0x0 and the main clock is monitored, writing a ‘1’ to the Clock Failure Detection Test
(CDFTST) bit in MCLKCTRLC will trigger a fault that will change the main clock to the start-up clock source.
12.3.8
Sleep Mode Operation
When a clock source is not used or requested, it will stop. It is possible to request a clock source directly by writing a
‘1’ to the Run Standby (RUNSTDBY) bit in the respective oscillator’s Control A (CLKCTRL.oscillatorCTRLA) register.
This will cause the oscillator to run constantly, except for Power-Down sleep mode. Additionally, when this bit is
written to a ‘1’, the oscillator start-up time is eliminated when the clock source is requested by a peripheral.
The main clock will always run in Active and Idle sleep modes. In Standby sleep mode, the main clock will run only if
any peripheral is requesting it, or RUNSTDBY in the respective oscillator’s CLKCTRL.oscillatorCTRLA register is
written to a ‘1’.
In Power-Down sleep mode, the main clock will stop after all nonvolatile memory (NVM) operations are completed.
Refer to the Sleep Controller section for more details on sleep mode operation.
In sleep, the Clock Failure Detection (CFD) will only be enabled if the selected source is active. After a Reset, the
CFD will not start looking for failure until a time equivalent to the monitored Oscillator Start-up Timer (SUT) has
expired.
12.3.9
Configuration Change Protection
This peripheral has registers that are under Configuration Change Protection (CCP). To write to these registers, a
certain key must first be written to the CPU.CCP register, followed by a write access to the protected bits within four
CPU instructions.
Attempting to write to a protected register without following the appropriate CCP unlock sequence leaves the
protected register unchanged.
The following registers are under CCP:
Table 12-1. CLKCTRL - Registers Under Configuration Change Protection
Register
Key
CLKCTRL.MCLKCTRLA
IOREG
CLKCTRL.MCLKCTRLB
IOREG
CLKCTRL.MCLKCTRLC
IOREG
CLKCTRL.MCLKINTCTRL
IOREG
CLKCTRL.OSCHFCTRLA
IOREG
CLKCTRL.PLLCTRLA
IOREG
CLKCTRL.OSC32KCTRLA
IOREG
CLKCTRL.XOSC32KCTRLA
IOREG
CLKCTRL.XOSCHFCTRLA
IOREG
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 99
AVR128DB28/32/48/64
CLKCTRL - Clock Controller
12.4
Register Summary
Offset
Name
Bit Pos.
7
0x00
0x01
0x02
0x03
0x04
0x05
0x06
...
0x07
0x08
0x09
0x0A
...
0x0F
0x10
0x11
...
0x17
0x18
0x19
...
0x1B
0x1C
0x1D
...
0x1F
0x20
MCLKCTRLA
MCLKCTRLB
MCLKCTRLC
MCLKINTCTRL
MCLKINTFLAGS
MCLKSTATUS
7:0
7:0
7:0
7:0
7:0
7:0
CLKOUT
12.5
6
5
4
3
2
1
0
CLKSEL[3:0]
PDIV[3:0]
CFDSRC[1:0]
CFDTST
INTTYPE
PLLS
EXTS
XOSC32KS
OSC32KS
OSCHFS
PEN
CFDEN
CFD
CFD
SOSC
Reserved
OSCHFCTRLA
OSCHFTUNE
7:0
7:0
RUNSTDBY
FRQSEL[3:0]
TUNE[7:0]
7:0
RUNSTDBY
7:0
RUNSTDBY
7:0
RUNSTDBY
CSUT[1:0]
7:0
RUNSTDBY
CSUTHF[1:0]
AUTOTUNE
Reserved
PLLCTRLA
SOURCE
MULFAC[1:0]
Reserved
OSC32KCTRLA
Reserved
XOSC32KCTRLA
SEL
LPMODE
ENABLE
SELHF
ENABLE
Reserved
XOSCHFCTRLA
FRQRANGE[1:0]
Register Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 100
AVR128DB28/32/48/64
CLKCTRL - Clock Controller
12.5.1
Main Clock Control A
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
CLKOUT
R/W
0
MCLKCTRLA
0x00
0x00
Configuration Change Protection
6
5
4
3
R/W
0
2
1
CLKSEL[3:0]
R/W
R/W
0
0
0
R/W
0
Bit 7 – CLKOUT Main Clock Out
This bit controls whether the main clock is available on the Main Clock Out (CLKOUT) pin or not, when the main
clock is running.
This bit is cleared when a ‘0’ is written to it or when a Clock Failure Detection (CFD) condition with the main clock as
source occurs.
This bit is set when a ‘1’ is written to it.
Value
Description
0
The main clock is not available on the CLKOUT pin
1
The main clock is available on the CLKOUT pin
Bits 3:0 – CLKSEL[3:0] Clock Select
This bit field controls the source for the Main Clock (CLK_MAIN).
Value
Name
Description
0x0
OSCHF
Internal high-frequency oscillator
0x1
OSC32K
32.768 kHz internal oscillator
0x2
XOSC32K 32.768 kHz external crystal oscillator
0x3
EXTCLK
External clock or external crystal, depending on the SELHF bit in XOSCHFCTRLA
Other
Reserved
Reserved
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 101
AVR128DB28/32/48/64
CLKCTRL - Clock Controller
12.5.2
Main Clock Control B
Name:
Offset:
Reset:
Property:
Bit
7
MCLKCTRLB
0x01
0x00
Configuration Change Protection
6
5
4
3
2
1
R/W
0
R/W
0
PDIV[3:0]
Access
Reset
R/W
0
R/W
0
0
PEN
R/W
0
Bits 4:1 – PDIV[3:0] Prescaler Division
This bit field controls the division ratio of the Main Clock (CLK_MAIN) prescaler, when the Prescaler (PEN) bit is ‘1’.
Value
Name
Description
0x0
2X
Divide by 2
0x1
4X
Divide by 4
0x2
8X
Divide by 8
0x3
16X
Divide by 16
0x4
32X
Divide by 32
0x5
64X
Divide by 64
0x8
6X
Divide by 6
0x9
10X
Divide by 10
0xA
12X
Divide by 12
0xB
24X
Divide by 24
0xC
48X
Divide by 48
Other
Reserved
Note: Configuration of the input frequency (CLK_MAIN) and prescaler settings must not exceed the allowed
maximum frequency of the peripheral clock (CLK_PER) or CPU clock (CLK_CPU). Refer to the Electrical
Characteristics section for further information.
Bit 0 – PEN Prescaler Enable
This bit controls whether the Main Clock (CLK_MAIN) prescaler is enabled or not.
Value
Description
0
The CLK_MAIN prescaler is disabled
1
The CLK_MAIN prescaler is enabled, and the division ratio is controlled by the Prescaler Division
(PDIV) bit field
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 102
AVR128DB28/32/48/64
CLKCTRL - Clock Controller
12.5.3
Main Clock Control C
Name:
Offset:
Reset:
Property:
Bit
7
MCLKCTRLC
0x02
0x00
Configuration Change Protection
6
Access
Reset
5
4
3
2
CFDSRC[1:0]
R/W
R/W
0
0
1
CFDTST
R/W
0
0
CFDEN
R/W
0
Bits 3:2 – CFDSRC[1:0] Clock Failure Detection Source
This bit field controls which clock source to monitor, when the Clock Failure Detection Enable (CFDEN) bit is ‘1’.
Value
Name
Description
0x0
CLKMAIN
Main Clock
0x1
XOSCHF
External High Frequency Oscillator
0x2
XOSC32K
External 32.768 kHz Oscillator
Other
Reserved
Reserved
Note: This bit field is read-only when the CFDEN bit is ‘1’, and both the Clock Failure Detection (CFD) interrupt
enable bit and Interrupt Type (INTTYPE) bit in the Main Clock Interrupt Control (CLKCTRL.MCLKINTCTRL) are ‘1’.
This bit will remain read-only until a System Reset occurs.
Bit 1 – CFDTST Clock Failure Detection Test
This bit controls testing of the Clock Failure Detection (CFD) functionality.
Writing a ‘0’ to this bit will clear the bit, and the ongoing CFD test fail condition.
Writing a ‘1’ to this bit will set the bit and force a CFD fail condition.
Value
Description
0
No ongoing test of the CFD functionality
1
A CFD fail condition has been forced
Bit 0 – CFDEN Clock Failure Detection Enable
This bit controls whether Clock Failure Detection (CFD) is enabled or not.
Value
Description
0
CFD is disabled
1
CFD is enabled
Note: This bit is read-only when this bit is ‘1’, and both the Clock Failure Detection (CFD) interrupt enable bit and
Interrupt Type (INTTYPE) bit in the Main Clock Interrupt Control (CLKCTRL.MCLKINTCTRL) are ‘1’. This bit will
remain read-only until a System Reset occurs.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 103
AVR128DB28/32/48/64
CLKCTRL - Clock Controller
12.5.4
Main Clock Interrupt Control
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
INTTYPE
R/W
0
MCLKINTCTRL
0x03
0x00
Configuration Change Protection
6
5
4
3
2
1
0
CFD
R/W
0
Bit 7 – INTTYPE Interrupt Type
This bit controls the type of the Clock Failure Detection (CFD) interrupt.
Value
Name
Description
0
INT
Regular Interrupt
1
NMI
Non-Maskable Interrupt
Note: This bit is read-only when the Clock Failure Detection Enable (CFDEN) bit in the Main Clock Control C
(CLKCTRL.MCLKCTRLC) register is ‘1’, and both the Clock Failure Detection (CFD) interrupt enable bit and this bit
are ‘1.’ This bit will remain read-only until a System Reset occurs.
Bit 0 – CFD Clock Failure Detection Interrupt Enable
This bit controls whether the Clock Failure Detection (CFD) interrupt is enabled or not.
Value
Description
0
The CFD interrupt is disabled
1
The CFD interrupt is enabled
Note: This bit is read-only when the Clock Failure Detection Enable (CFDEN) bit in the Main Clock Control C
(CLKCTRL.MCLKCTRLC) register is ‘1’, and both the Interrupt Type (INTTYPE) bit and this bit are ‘1.’ This bit will
remain read-only until a System Reset occurs.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 104
AVR128DB28/32/48/64
CLKCTRL - Clock Controller
12.5.5
Main Clock Interrupt Flags
Name:
Offset:
Reset:
Property:
Bit
7
MCLKINTFLAGS
0x04
0x0
-
6
5
4
3
Access
Reset
2
1
0
CFD
R/W
0
Bit 0 – CFD Clock Failure Detection Interrupt Flag
This flag is cleared by writing a ‘1’ to it.
This flag is set when a clock failure is detected.
Writing a ‘0’ to this bit has no effect.
Writing a ‘1’ to this bit will clear the Clock Failure Detection (CFD) interrupt flag.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 105
AVR128DB28/32/48/64
CLKCTRL - Clock Controller
12.5.6
Main Clock Status
Name:
Offset:
Reset:
Property:
Bit
7
MCLKSTATUS
0x05
0x00
-
6
Access
Reset
5
PLLS
R
0
4
EXTS
R
0
3
XOSC32KS
R
0
2
OSC32KS
R
0
1
OSCHFS
R
0
0
SOSC
R
0
Bit 5 – PLLS PLL Status
Value
Description
0
PLL is not running
1
PLL is running
Bit 4 – EXTS External Crystal/Clock Status
Value
Description
0
The external high-frequency crystal is not stable, when the Source Select (SELHF) bit in the External
High-Frequency Oscillator Control A (CLKCTRL.XOSCHFCTRLA) register is ‘0’.
The external high-frequency clock is not running, when the SELHF bit is ‘1’.
1
The external high-frequency crystal is stable, when the SELHF bit is ‘0’.
The external high-frequency clock is running, when the SELHF bit is ‘1’.
Bit 3 – XOSC32KS XOSC32K Status
Value
Description
0
The external 32.768 kHz crystal is not stable, when the Source Select (SEL) bit in the 32.768 Crystal
Oscillator Control A (CLKCTRL.XOSC32K) register is ‘0’.
The external 32.768 kHz clock is not running, when the SEL bit is ‘1’.
1
The external 32.768 kHz crystal is stable, when the SEL bit is ‘0’.
The external 32.768 kHz clock is running, when the SEL bit is ‘1’.
Bit 2 – OSC32KS OSC32K Status
Value
Description
0
OSC32K is not stable
1
OSC32K is stable
Bit 1 – OSCHFS Internal High-Frequency Oscillator Status
Value
Description
0
OSCHF is not stable
1
OSCHF is stable
Bit 0 – SOSC Main Clock Oscillator Changing
Value
Description
0
The clock source for CLK_MAIN is not undergoing a switch
1
The clock source for CLK_MAIN is undergoing a switch and will change as soon as the new source is
stable
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 106
AVR128DB28/32/48/64
CLKCTRL - Clock Controller
12.5.7
Internal High-Frequency Oscillator Control A
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
OSCHFCTRLA
0x08
0x0C
Configuration Change Protection
7
RUNSTDBY
R/W
0
6
5
R/W
0
4
3
FRQSEL[3:0]
R/W
R/W
0
1
2
1
R/W
1
0
AUTOTUNE
R/W
0
Bit 7 – RUNSTDBY Run Standby
This bit controls whether the Internal High-Frequency Oscillator (OSCHF) is always running or not.
Value
Description
0
The OSCHF oscillator will only run when requested by a peripheral or by the main clock (1)
1
The OSCHF oscillator will always run in Active, Idle and Standby sleep modes (2)
Notes:
1. The requesting peripheral, or the main clock, must take the oscillator start-up time into account.
2. The oscillator signal is only available if requested, and will be available after two OSCHF cycles.
Bits 5:2 – FRQSEL[3:0] Frequency Select
This bit field controls the output frequency of the Internal High-Frequency (OSCHF) oscillator.
Value
Name
Description
0x0
0x1
0x2
0x3
0x4
0x5
0x6
0x7
0x8
0x9
Other
1M
2M
3M
4M
8M
12M
16M
20M
24M
-
1 MHz output
2 MHz output
3 MHz output
4 MHz output (default)
Reserved
8 MHz output
12 MHz output
16 MHz output
20 MHz output
24 MHz output
Reserved
Bit 0 – AUTOTUNE Auto-Tune Enable
This bit controls whether the 32.768 kHz crystal auto-tune functionality of the Internal High-Frequency Oscillator
(OSCHF) is enabled or not.
Value
Description
0
The auto-tune functionality of the OSCHF oscillator is disabled
1
The auto-tune functionality of the OSCHF oscillator is enabled
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 107
AVR128DB28/32/48/64
CLKCTRL - Clock Controller
12.5.8
Internal High-Frequency Oscillator Frequency Tune
Name:
Offset:
Reset:
Property:
Bit
7
OSCHFTUNE
0x09
0x00
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
TUNE[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – TUNE[7:0] User Frequency Tuning
This bit field controls the manual tuning of the output frequency of the Internal High-Frequency Oscillator (OSCHF).
The frequency can be tuned 32 steps down or 31 steps up from the oscillator's target frequency. Thus, the register’s
acceptable input value range is -32 to +31.
Writing to bit 6 and 7 has no effect, as bit 5 will be mirrored to bit 6 and 7, due to the 6-bit value in this bit field being
represented in a signed (two’s complement) form.
Note: If the Auto-Tune Enable (AUTOTUNE) bit in the Internal High-Frequency Oscillator Control A
(CLKCTRL.OSCHFCTRLA) register is enabled, the TUNE value is locked. This bit field is updated with the latest tune
value from the auto-tune operation when AUTOTUNE is disabled.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 108
AVR128DB28/32/48/64
CLKCTRL - Clock Controller
12.5.9
PLL Control A
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
RUNSTDBY
R/W
0
PLLCTRLA
0x10
0x00
Configuration Change Protection
6
SOURCE
R/W
0
5
4
3
2
1
0
MULFAC[1:0]
R/W
R/W
0
0
Bit 7 – RUNSTDBY Run Standby
This bit controls whether the Phase-Locked Loop (PLL) is always running or not.
Value
Description
0
The PLL will only run if requested by a peripheral (1)
1
The PLL will always run in Active, Idle and Standby sleep modes (2)
Notes:
1. The requesting peripheral must take the PLL start-up time and PLL source start-up time into account.
2. The oscillator signal will only be available if requested and will be available after two PLL cycles.
Bit 6 – SOURCE Select Source for PLL
This bit controls the Phase-Locked Loop (PLL) clock source.
Value
Name
Description
0
1
OSCHF
XOSCHF
High-frequency internal oscillator as PLL source
High-frequency external clock or external high-frequency oscillator as
PLL source
Bits 1:0 – MULFAC[1:0] Multiplication Factor
This bit field controls the multiplication factor for the Phased-Locked Loop (PLL).
Value
Name
Description
0x0
0x1
0x2
0x3
DISABLE
2x
3x
-
PLL is disabled
2 x multiplication factor
3 x multiplication factor
Reserved
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 109
AVR128DB28/32/48/64
CLKCTRL - Clock Controller
12.5.10 32.768 kHz Oscillator Control A
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
RUNSTDBY
R/W
0
OSC32KCTRLA
0x18
0x00
Configuration Change Protection
6
5
4
3
2
1
0
Bit 7 – RUNSTDBY Run Standby
This bit controls whether the 32.768 kHz Oscillator (OSC32K) is always running or not.
Value
Description
0
The OSC32K oscillator will only run when requested by a peripheral or by the main clock (1)
1
The OSC32K oscillator will always run in Active, Idle, Standby and Power-Down sleep modes (2)
Notes:
1. The requesting peripheral, or the main clock, must take the oscillator start-up time into account.
2. The oscillator signal is only available if requested and will be available after four OSC32K cycles.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 110
AVR128DB28/32/48/64
CLKCTRL - Clock Controller
12.5.11 32.768 kHz Crystal Oscillator Control A
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
RUNSTDBY
R/W
0
XOSC32KCTRLA
0x1C
0x00
Configuration Change Protection
6
5
4
3
CSUT[1:0]
R/W
0
R/W
0
2
SEL
R/W
0
1
LPMODE
R/W
0
0
ENABLE
R/W
0
Bit 7 – RUNSTDBY Run Standby
This bit controls whether the 32.768 kHz Crystal Oscillator (XOSC32K) is always running or not, and in which modes,
when the ENABLE bit is ‘1’.
Value
Description
0
The XOSC32K oscillator will only run when requested by a peripheral or by the main clock, in Active
and Idle sleep mode(1)
1
The XOSC32K oscillator will always run in Active, Idle, Standby and Power-Down sleep modes(2)
Notes:
1. The requesting peripheral, or the main clock, must take the oscillator start-up time into account.
2. The oscillator signal is only available if requested, and will be available after a maximum of three XOSC32K
cycles, if the initial crystal start-up time has already ended.
Bits 5:4 – CSUT[1:0] Crystal Start-Up Time
This bit field controls the start-up time of the 32.768 kHz Crystal Oscillator (XOSC32K), when the Source Select
(SELHF) is ‘0’.
Value
Name
Description
0x0
1K
1k cycles
0x1
16K
16k cycles
0x2
32K
32k cycles
0x3
64K
64k cycles
Note: This bit field is read-only when the ENABLE bit or the XOSC32K Status (XOSCS) bit in the Main Clock Status
(CLKCTRL.MCLKSTATUS) register is ‘1’.
Bit 2 – SEL Source Select
This bit controls the source of the 32.768 kHz Crystal Oscillator (XOSC32K).
Value
Description
0
External crystal the XTAL32K1 and XTAL32K2 pins
1
External clock on the XTAL32K1 pin
Note: This bit field is read-only when the ENABLE bit or the XOSC32K Status (XOSCS) bit in the Main Clock Status
(CLKCTRL.MCLKSTATUS) register is ‘1’.
Bit 1 – LPMODE Low-Power Mode
This bit controls whether the 32.768 kHz Crystal Oscillator (XOSC32K) is in Low-Power mode or not.
Value
Description
0
The Low-Power mode is disabled
1
The Low-Power mode is enabled
Bit 0 – ENABLE Enable
This bit controls whether the 32.768 kHz Crystal Oscillator (XOSC32K) is enabled or not.
Value
Description
0
The XOSC32K oscillator is disabled
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 111
AVR128DB28/32/48/64
CLKCTRL - Clock Controller
Value
1
Description
The XOSC32K oscillator is enabled, and overrides normal port operation for the respective oscillator
pins
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 112
AVR128DB28/32/48/64
CLKCTRL - Clock Controller
12.5.12 External High-Frequency Oscillator Control A
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
RUNSTDBY
R/W
0
XOSCHFCTRLA
0x20
0x00
Configuration Change Protection
6
5
4
CSUTHF[1:0]
R/W
R/W
0
0
3
2
FRQRANGE[1:0]
R/W
R/W
0
0
1
SELHF
R/W
0
0
ENABLE
R/W
0
Bit 7 – RUNSTDBY Run Standby
This bit controls whether the External High-Frequency Oscillator (XOSCHF) is always running or not, when the
ENABLE bit is ‘1’.
Value
Description
0
The XOSCHF oscillator will only run when requested by a peripheral or by the main clock (1)
1
The XOSCHF oscillator will always run in Active, Idle and Standby sleep modes (2)
Notes:
1. The requesting peripheral, or the main clock, must take the oscillator start-up time into account.
2. The oscillator signal is only available if requested, and will be available after two XOSCHF cycles, if the initial
crystal start-up time has already ended.
Bits 5:4 – CSUTHF[1:0] Crystal Start-up Time
This bit field controls the start-up time for the External High-Frequency Oscillator (XOSCHF), when the Source Select
(SELHF) bit is ‘0’.
Value
Name
Description
0x0
256
256 XOSCHF cycles
0x1
1K
1K XOSCHF cycles
0x2
4K
4K XOSCHF cycles
0x3
Reserved
Note: This bit field is read-only when the ENABLE bit or the External Crystal/Clock Status (XOSCHFS) bit in the
Main Clock Status (MCLKSTATUS) register is ‘1’.
Bits 3:2 – FRQRANGE[1:0] Frequency Range
This bit field controls the maximum frequency supported for the external crystal. The larger the range selected, the
higher the current consumption by the oscillator.
Value
Name
Description
0x0
8M
Max. 8 MHz XTAL frequency
0x1
16M
Max. 16 MHz XTAL frequency
0x2
24M
Max. 24 MHz XTAL frequency
0x3
32M
Max. 32 MHz XTAL frequency
Note: If a crystal with a frequency larger than the maximum supported CLK_CPU frequency is used and used as the
main clock, it is necessary to divide it down by writing the appropriate configuration to the PDIV bit field in the Main
Clock Control B register.
Bit 1 – SELHF Source Select
This bit controls the source of the External High-Frequency Oscillator (XOSCHF).
Value
Name
Description
0
CRYSTAL
External Crystal on the XTALHF1 and XTALHF2 pins
1
EXTCLOCK
External Clock on the XTALHF1 pin
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 113
AVR128DB28/32/48/64
CLKCTRL - Clock Controller
Note: This bit field is read-only when the ENABLE bit or the External Crystal/Clock Status (XOSCHFS) bit in the
Main Clock Status (MCLKSTATUS) register is ‘1’.
Bit 0 – ENABLE Enable
This bit controls whether the External High-Frequency Oscillator (XOSCHF) is enabled or not.
Value
Description
0
The XOSCHF oscillator is disabled
1
The XOSCHF oscillator is enabled, and overrides normal port operation for the respective oscillator
pins
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 114
AVR128DB28/32/48/64
SLPCTRL - Sleep Controller
13.
SLPCTRL - Sleep Controller
13.1
Features
•
•
•
13.2
Power Management for Adjusting Power Consumption and Functions
Three Sleep Modes:
– Idle
– Standby
– Power-Down
Configurable Standby Mode where Peripherals Can Be Configured as ON or OFF
Overview
Sleep modes are used to shut down peripherals and clock domains in the device in order to save power. The Sleep
Controller (SLPCTRL) controls and handles the transitions between Active and sleep modes.
There are four modes available: One Active mode in which software is executed, and three sleep modes. The
available sleep modes are Idle, Standby and Power-Down.
All sleep modes are available and can be entered from the Active mode. In Active mode, the CPU is executing
application code. When the device enters sleep mode, the program execution is stopped. The application code
decides which sleep mode to enter and when.
Interrupts are used to wake the device from sleep. The available interrupt wake-up sources depend on the configured
sleep mode. When an interrupt occurs, the device will wake up and execute the Interrupt Service Routine before
continuing normal program execution from the first instruction after the SLEEP instruction. Any Reset will take the
device out of sleep mode.
The content of the register file, SRAM and registers, is kept during sleep. If a Reset occurs during sleep, the device
will reset, start and execute from the Reset vector.
13.2.1
Block Diagram
Figure 13-1. Sleep Controller in the System
SLEEP Instruction
SLPCTRL
Interrupt Request
CPU
Sleep State
Interrupt Request
Peripheral
13.3
13.3.1
Functional Description
Initialization
To put the device into a sleep mode, follow these steps:
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 115
AVR128DB28/32/48/64
SLPCTRL - Sleep Controller
1.
Configure and enable the interrupts that are able to wake the device from sleep.
Also, enable global interrupts.
WARNING
2.
If there are no interrupts enabled when going to sleep, the device cannot wake up again. Only a
Reset will allow the device to continue operation.
Select which sleep mode to enter and enable the Sleep Controller by writing to the Sleep Mode (SMODE) bit
field and the Enable (SEN) bit in the Control A (SLPCTRL.CTRLA) register.
The SLEEP instruction must be executed to make the device go to sleep.
13.3.2
Voltage Regulator Configuration
A voltage regulator is used to regulate the core voltage. The regulator can be configured to balance power
consumption, wake-up time from Sleep, and maximum clock speed.
The Voltage Regulator Control (SLPCTRL.VREGCTRL) register is used to configure the regulator start-up time and
power consumption. The Power Mode Select (PMODE) bit field in SLPCTRL.VREGCTRL can be set to make the
regulator switch to Normal mode when OSC32K is the only oscillator enabled and if the device is in sleep mode. In
Normal mode, the regulator consumes less power, but can supply only a limited amount of current, permitting only a
low clock frequency.
The user may select one of the following Voltage Regulator Power modes:
Table 13-1. Voltage Regulator Power Modes Description
Voltage Regulator Power Mode
13.3.3
Description
Normal (AUTO)
Maximum performance in Active mode and Idle mode
Performance (FULL)
Maximum performance in all modes (Active and Sleep) and fast start-up from all
sleep modes
Operation
13.3.3.1 Sleep Modes
There are three sleep modes that can be enabled to reduce power consumption.
Idle
Standby
PowerDown
The CPU stops executing code. All peripherals are running and all interrupt sources can wake the device.
The user can configure peripherals to be enabled or not, using the respective RUNSTDBY bit. This
means that the power consumption is highly dependent on what functionality is enabled, and thus may
vary between the Idle and Power-Down modes.
Operation in Standby mode is enabled for a peripheral by writing a ‘1’ to the RUNSTDBY bit within the
control register of each peripheral.
The Watchdog Timer (WDT) and the Periodic Interrupt Timer (PIT) are active.
The only wake-up sources are the pin change interrupt, TWI address match, and CCL (if filter and edgedetect are disabled).
For temperature ranges above 70°C, the HTLLEN bit in the SLPCTRL.VREGCTRL register can be used
to reduce power consumption (High-Temperature Low Leakage Enable).
Important: The TWI address match and CCL wake-up sources are not functional when HighTemperature Low Leakage Enable is activated and must be disabled by the user to avoid
unpredictable behavior.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 116
AVR128DB28/32/48/64
SLPCTRL - Sleep Controller
Table 13-2. Sleep Mode Activity Overview for Peripherals
Peripheral
Active in Sleep Mode
Idle
Standby
Power-Down
HTLLEN=0
HTLLEN=1
CPU
RTC
X
X(1,2)
X(2)
X(2)
WDT
X
X
X
X
BOD
X
X
X
X
EVSYS
X
X
X
X
CCL
X
X(1)
ACn
ADCn
DACn
OPAMP
ZCDn
TCAn
TCBn
All other peripherals
X
Notes:
1. For the peripheral to run in Standby sleep mode, the RUNSTDBY bit of the corresponding peripheral must be
set.
2. In Standby sleep mode, only the RTC functionality requires the RUNSTDBY bit to be set.
In Power-Down sleep mode, only the PIT functionality is available.
Table 13-3. Sleep Mode Activity Overview for Clock Sources
Clock Source
Active in Sleep Mode
Idle
Standby
Main clock source
X
X(1)
RTC clock source
X
WDT oscillator
Power-Down
HTLLEN=0
HTLLEN=1
X(1,2)
X(2)
X(2)
X
X
X
X
X
X
X
X
CCL clock source
X
X(1)
TCD clock source
X
BOD
oscillator(3)
Notes:
1. For the clock source to run in Standby sleep mode, the RUNSTDBY bit of the corresponding peripheral must
be set.
2. In Standby sleep mode, only the RTC functionality requires the RUNSTDBY bit to be set.
In Power-Down sleep mode, only the PIT functionality is available.
3. The Sampled mode only.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 117
AVR128DB28/32/48/64
SLPCTRL - Sleep Controller
Table 13-4. Sleep Mode Wake-up Sources
Wake-Up Sources
Active in Sleep Mode
Idle
Standby
Power-Down
HTLLEN=0
HTLLEN=1
PORT Pin interrupt
X
X
X(1)
X(1)
BOD VLM interrupt
X
X
X
X
MVIO interrupts
X
X
X
X
RTC interrupts
X
X(2,3)
X(3)
X(3)
TWI Address Match interrupt
X
X
X
CCL interrupts
X
X
X(4)
USART Start-Of-Frame interrupt
X
X
TCAn interrupts
TCBn interrupts
ADCn interrupts
ACn interrupts
ZCD interrupts
All other interrupts
X
Notes:
1. The I/O pin must be configured according to Asynchronous Sensing Pin Properties in the PORT section.
2. For the peripheral to run in Standby sleep mode, the RUNSTDBY bit of the corresponding peripheral must be
set.
3. In Standby sleep mode, only the RTC functionality requires the RUNSTDBY bit to be set.
In Power-Down sleep mode, only the PIT functionality is available.
4. CCL will only wake up the device if the path through LUTn is asynchronous (FILTSEL=0x0 and
EDGEDET=0x0 in the CCL.LUTnCTRLA register).
13.3.3.2 Wake-up Time
The normal wake-up time for the device is six main clock cycles (CLK_PER), plus the time it takes to start the main
clock source and the time it takes to start the regulator, if it has been switched off:
• In Idle mode, the main clock source is kept running to eliminate additional wake-up time
• In Standby mode, the main clock might be running depending on the peripheral configuration
• In Power-Down mode, only the OSC32K oscillator and the Real-Time Clock (RTC) may be running if it is used
by the Brown-out Detector (BOD), Watchdog Timer (WDT) or Periodic Interrupt Timer (PIT). All the other clock
sources will be OFF.
Table 13-5. Sleep Modes and Start-up Time
Sleep Mode
Start-up Time
Idle
Six clock cycles
Standby
Six clock cycles + one (OSC start-up + Regulator start-up)
Power-Down
Six clock cycles + one (OSC start-up + Regulator start-up)
The start-up time for the different clock sources is described in the CLKCTRL - Clock Controller section.
In addition to the normal wake-up time, it is possible to make the device wait until the BOD is ready before executing
code. This is done by writing 0x3 to the BOD operation mode in Active and Idle (ACTIVE) bit field in the BOD
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 118
AVR128DB28/32/48/64
SLPCTRL - Sleep Controller
Configuration (FUSE.BODCFG) fuse. If the BOD is ready before the normal wake-up time, the total wake-up time will
be the same. If the BOD takes longer than the normal wake-up time, the wake-up time will be extended until the BOD
is ready. This ensures correct supply voltage whenever code is executed.
13.3.4
Debug Operation
During run-time debugging, this peripheral will continue normal operation. The SLPCTRL is only affected by a break
in the debug operation: If the SLPCTRL is in a sleep mode when a break occurs, the device will wake up, and the
SLPCTRL will go to Active mode, even if there are no pending interrupt requests.
If the peripheral is configured to require periodic service by the CPU through interrupts or similar, improper operation
or data loss may result during halted debugging.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 119
AVR128DB28/32/48/64
SLPCTRL - Sleep Controller
13.4
Register Summary
Offset
Name
Bit Pos.
0x00
0x01
CTRLA
VREGCTRL
7:0
7:0
13.5
7
6
5
4
3
2
1
SMODE[2:0]
HTLLEN
0
SEN
PMODE[2:0]
Register Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 120
AVR128DB28/32/48/64
SLPCTRL - Sleep Controller
13.5.1
Control A
Name:
Offset:
Reset:
Property:
Bit
7
CTRLA
0x00
0x00
-
6
Access
Reset
5
4
3
R/W
0
2
SMODE[2:0]
R/W
0
1
R/W
0
0
SEN
R/W
0
Bits 3:1 – SMODE[2:0] Sleep Mode
Writing these bits selects the desired sleep mode when the Sleep Enable (SEN) bit is written to ‘1’ and the SLEEP
instruction is executed.
Value
Name
Description
0x0
IDLE
Idle mode enabled
0x1
STANDBY
Standby mode enabled
0x2
PDOWN
Power-Down mode enabled
Other
Reserved
Bit 0 – SEN Sleep Enable
This bit must be written to ‘1’ before the SLEEP instruction is executed to make the microcontroller enter the selected
sleep mode.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 121
AVR128DB28/32/48/64
SLPCTRL - Sleep Controller
13.5.2
Voltage Regulator Control Register
Name:
Offset:
Reset:
Property:
Bit
VREGCTRL
0x01
0x00
Configuration Change Protection
7
6
Access
Reset
5
4
HTLLEN
R/W
0
3
2
R/W
0
1
PMODE[2:0]
R/W
0
0
R/W
0
Bit 4 – HTLLEN High-Temperature Low Leakage Enable
This bit is write-protected.
When writing this bit to ‘1’, the leakage is reduced when operating at a temperature above 70°C. This bit has an
effect only in Power-Down mode when PMODE is set to AUTO. To have effect, it must be written before executing
the SLEEP instruction.
WARNING
Value
0
1
This feature allows an additional power-saving feature, but the TWI address match and CCL wake-up
sources must be disabled in software before setting this bit.
Name
OFF
ON
Description
High-temperature low leakage disabled
High-temperature low leakage enabled
Bits 2:0 – PMODE[2:0] Power Mode Select
This bit field is write-protected.
Value
Name Description
0x0
AUTO In Auto mode, the regulator will run at the full performance unless the 32.768 kHz oscillator is
selected, then it runs in a lower power mode.
0x1
FULL Full performance voltage regulator drive strength in all modes.
Other
Reserved
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 122
AVR128DB28/32/48/64
RSTCTRL - Reset Controller
14.
RSTCTRL - Reset Controller
14.1
Features
•
•
•
•
14.2
Returns the Device to an Initial State after a Reset
Identifies the Previous Reset Source
Power Supply Reset Sources:
– Power-on Reset (POR)
– Brown-out Detector (BOD) Reset
User Reset Sources:
– External Reset (RESET)
– Watchdog Timer (WDT) Reset
– Software Reset (SWRST)
– Unified Program and Debug Interface (UPDI) Reset
Overview
The Reset Controller (RSTCTRL) manages the Reset of the device. When receiving a Reset request, it sets the
device to an initial state and allows the Reset source to be identified by the software. The Reset controller can also
be used to issue a Software Reset (SWRST).
14.2.1
Block Diagram
Figure 14-1. Reset System Overview
RESET SOURCES
VDD
POR
Pull-up
resistor
RESET
RESET CONTROLLER
FILTER
BOD
UPDI
External Reset
WDT
UPDI
All other
peripherals
UPDI
CPU(SWRST)
14.2.2
Signal Description
Signal
RESET
Description
External Reset (active-low)
© 2020 Microchip Technology Inc.
Type
Digital input
Preliminary Datasheet
DS40002247A-page 123
AVR128DB28/32/48/64
RSTCTRL - Reset Controller
...........continued
Signal
Description
UPDI
Unified Program and Debug Interface
14.3
Functional Description
14.3.1
Initialization
Type
Digital input
The RSTCTRL is always enabled, but some of the Reset sources must be enabled individually (either by Fuses or by
software) before they can request a Reset.
After a Reset from any source, the registers in the device with automatic loading from the Fuses or from the
Signature Row are updated.
14.3.2
Operation
14.3.2.1 Reset Sources
After any Reset, the source that caused the Reset is found in the Reset Flag (RSTCTRL.RSTFR) register. The user
can identify the previous Reset source by reading this register in the software application.
There are two types of Resets based on the source:
• Power Supply Reset Sources:
– Power-on Reset (POR)
– Brown-out Detector (BOD) Reset
• User Reset Sources:
– External Reset (RESET)
– Watchdog Timer (WDT) Reset
– Software Reset (SWRST)
– Unified Program and Debug Interface (UPDI) Reset
14.3.2.1.1 Power-on Reset (POR)
The purpose of the Power-on Reset (POR) is to ensure a safe start-up of logic and memories. It is generated by an
on-chip detection circuit and is always enabled. The POR is activated when the VDD rises and reaches the POR
threshold voltage (VPOT), starting the reset sequence. Before VPOT has been reached, the external RESET pin keeps
the device in reset, either through the internal pull-up to VDD or by external control.
Figure 14-2. MCU Start-Up, RESET Tied to VDD
VDD
RESET
TIME-OUT
VPOT
VRST
tTOUT
INTERNAL
RESET
14.3.2.1.2 Brown-out Detector (BOD) Reset
The Brown-out Detector (BOD) needs to be enabled by the user. The BOD is preventing code execution when the
voltage drops below a set threshold. This will ensure the voltage level needed for the oscillator to run at the speed
required by the application and will avoid code corruption due to low-voltage level.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 124
AVR128DB28/32/48/64
RSTCTRL - Reset Controller
The BOD issues a System Reset and is not released until the voltage level increases above the set threshold. The
on-chip BOD circuit will monitor the VDD level during operation by comparing it to a fixed trigger level. The trigger
level for the BOD must be selected by the BOD Configuration (FUSE.BODCFG) fuse.
Figure 14-3. Brown-out Detector Reset
tBOD
VDD
VBOT-
VBOT+
tTOUT
TIME-OUT
INTERNAL
RESET
14.3.2.1.3 External Reset (RESET)
The RESET pin requires a noise filter that eliminates short, low-going pulses. Filtering the input assures that an
external Reset event is only issued when the RESET has been low for a minimum amount of time. See the Electrical
Characteristics section for the minimum pulse width of the RESET signal.
The external Reset is enabled by configuring the Reset Pin Configuration (RSTPINCFG) bits in the System
Configuration 0 (FUSE.SYSCFG0) fuse.
When enabled, the external Reset requests a Reset as long as the RESET pin is low. The device will stay in Reset
until the RESET pin is high again.
Figure 14-4. External Reset Characteristics
DD
tEXT
14.3.2.1.4 Watchdog Timer (WDT) Reset
The Watchdog Timer (WDT) is a system function which monitors the correct operation of the program. If the WDT is
not handled by software according to the programmed time-out period, a Watchdog Reset will be issued. More details
can be found in the WDT section.
14.3.2.1.5 Software Reset (SWRST)
The software Reset makes it possible to issue a System Reset from the software. The Reset is generated by writing
a ‘1’ to the Software Reset (SWRST) bit in the Software Reset (RSTCTRL.SWRR) register.
The Reset sequence will start immediately after the bit is written.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 125
AVR128DB28/32/48/64
RSTCTRL - Reset Controller
14.3.2.1.6 Unified Program and Debug Interface (UPDI) Reset
The Unified Program and Debug Interface (UPDI) contains a separate Reset source used to reset the device during
external programming and debugging. The Reset source is accessible only from external debuggers and
programmers. More details can be found in the UPDI section.
14.3.2.1.7 Domains Affected By Reset
The following logic domains are affected by the different Reset sources:
Table 14-1. Logic Domains Affected by Various Resets
Reset Source
POR
BOD
Software Reset
External Reset
Watchdog Reset
UPDI Reset
Fuses are
Reloaded
X
X
X
X
X
X
Reset of BOD
Configuration
X
Reset of UPDI Reset of Other Volatile
Logic
X
X
X
X
X
X
X
X
14.3.2.2 Reset Time
The Reset time can be split into two parts.
The first part is when any of the Reset sources are active. This part depends on the input to the Reset sources. The
external Reset is active as long as the RESET pin is low. The Power-on Reset (POR) and the Brown-out Detector
(BOD) are active as long as the supply voltage is below the Reset source threshold.
The second part is when all the Reset sources are released, and an internal Reset initialization of the device is done.
This time will be increased with the start-up time given by the Start-Up Time Setting (SUT) bit field in the System
Configuration 1 (FUSE.SYSCFG1) fuse. The internal Reset initialization time will also increase if the Cyclic
Redundancy Check Memory Scan (CRCSCAN) is configured to run at start-up. This configuration can be changed in
the CRC Source (CRCSRC) bit field in the System Configuration 0 (FUSE.SYSCFG0) fuse.
14.3.3
Sleep Mode Operation
The RSTCTRL operates in Active mode and in all sleep modes.
14.3.4
Configuration Change Protection
This peripheral has registers that are under Configuration Change Protection (CCP). To write to these registers, a
certain key must first be written to the CPU.CCP register, followed by a write access to the protected bits within four
CPU instructions.
Attempting to write to a protected register without following the appropriate CCP unlock sequence leaves the
protected register unchanged.
The following registers are under CCP:
Table 14-2. RSTCTRL - Registers Under Configuration Change Protection
Register
Key
RSTCTRL.SWRR
© 2020 Microchip Technology Inc.
IOREG
Preliminary Datasheet
DS40002247A-page 126
AVR128DB28/32/48/64
RSTCTRL - Reset Controller
14.4
Register Summary
Offset
Name
Bit Pos.
0x00
0x01
RSTFR
SWRR
7:0
7:0
14.5
7
6
5
4
3
2
1
0
UPDIRF
SWRF
WDRF
EXTRF
BORF
PORF
SWRST
Register Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 127
AVR128DB28/32/48/64
RSTCTRL - Reset Controller
14.5.1
Reset Flag Register
Name:
Offset:
Reset:
Property:
RSTFR
0x00
0xXX
-
The Reset flags can be cleared by writing a ‘1’ to the respective flag. All flags will be cleared by a Power-on Reset
(POR), with the exception of the Power-on Reset (PORF) flag. All flags will be cleared by a Brown-out Reset (BOR),
with the exception of the Power-on Reset (PORF) and Brown-out Reset (BORF) flags.
Bit
7
6
Access
Reset
5
UPDIRF
R/W
x
4
SWRF
R/W
x
3
WDRF
R/W
x
2
EXTRF
R/W
x
1
BORF
R/W
x
0
PORF
R/W
x
Bit 5 – UPDIRF UPDI Reset Flag
This bit is set to ‘1’ if a UPDI Reset has occurred.
Bit 4 – SWRF Software Reset Flag
This bit is set to ‘1’ if a Software Reset has occurred.
Bit 3 – WDRF Watchdog Reset Flag
This bit is set to ‘1’ if a Watchdog Reset has occurred.
Bit 2 – EXTRF External Reset Flag
This bit is set to ‘1’ if an External Reset has occurred.
Bit 1 – BORF Brown-out Reset Flag
This bit is set to ‘1’ if a Brown-out Reset has occurred.
Bit 0 – PORF Power-on Reset Flag
This bit is set to ‘1’ if a Power-on Reset has occurred.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 128
AVR128DB28/32/48/64
RSTCTRL - Reset Controller
14.5.2
Software Reset Register
Name:
Offset:
Reset:
Property:
Bit
7
SWRR
0x01
0x00
Configuration Change Protection
6
5
4
3
Access
Reset
2
1
0
SWRST
R/W
0
Bit 0 – SWRST Software Reset
When this bit is written to ‘1’, a Software Reset will occur.
This bit will always read as ‘0’.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 129
AVR128DB28/32/48/64
CPUINT - CPU Interrupt Controller
15.
CPUINT - CPU Interrupt Controller
15.1
Features
•
•
•
•
•
•
•
15.2
Short and Predictable Interrupt Response Time
Separate Interrupt Configuration and Vector Address for Each Interrupt
Interrupt Prioritizing by Level and Vector Address
Non-Maskable Interrupts (NMI) for Critical Functions
Two Interrupt Priority Levels: 0 (Normal) and 1 (High):
– One of the interrupt requests can optionally be assigned as a priority level 1 interrupt
– Optional round robin priority scheme for priority level 0 interrupts
Interrupt Vectors Optionally Placed in the Application Section or the Boot Loader Section
Selectable Compact Vector Table (CVT)
Overview
An interrupt request signals a change of state inside a peripheral and can be used to alter the program execution.
The peripherals can have one or more interrupts. All interrupts are individually enabled and configured. When an
interrupt is enabled and configured, it will generate an interrupt request when the interrupt condition occurs.
The CPU Interrupt Controller (CPUINT) handles and prioritizes the interrupt requests. When an interrupt is enabled
and the interrupt condition occurs, the CPUINT will receive the interrupt request. Based on the interrupt's priority level
and the priority level of any ongoing interrupt, the interrupt request is either acknowledged or kept pending until it has
priority. After returning from the interrupt handler, the program execution continues from where it was before the
interrupt occurred, and any pending interrupts are served after one instruction is executed.
The CPUINT offers NMI for critical functions, one selectable high-priority interrupt and an optional round robin
scheduling scheme for normal-priority interrupts. The round robin scheduling ensures that all interrupts are serviced
within a certain amount of time.
15.2.1
Block Diagram
Figure 15-1. CPUINT Block Diagram
Interrupt Controller
Priority
Decoder
Peripheral 1
INT REQ
CPU RETI
CPU INT ACK
CPU
CPU INT REQ
Peripheral n
INT REQ
STATUS
LVL0PRI
LVL1VEC
© 2020 Microchip Technology Inc.
Global
Interrupt
Enable
CPU.SREG
Preliminary Datasheet
Wake-up
SLPCTRL
DS40002247A-page 130
AVR128DB28/32/48/64
CPUINT - CPU Interrupt Controller
15.3
Functional Description
15.3.1
Initialization
An interrupt must be initialized in the following order:
1.
2.
3.
15.3.2
Configure the CPUINT if the default configuration is not adequate (optional):
– Vector handling is configured by writing to the respective bits (IVSEL and CVT) in the Control A
(CPUINT.CTRLA) register.
– Vector prioritizing by round robin is enabled by writing a ‘1’ to the Round Robin Priority Enable (LVL0RR)
bit in CPUINT.CTRLA.
– Select the Priority Level 1 vector by writing the interrupt vector number to the Interrupt Vector with Priority
Level 1 (CPUINT.LVL1VEC) register.
Configure the interrupt conditions within the peripheral and enable the peripheral’s interrupt.
Enable interrupts globally by writing a ‘1’ to the Global Interrupt Enable (I) bit in the CPU Status (CPU.SREG)
register.
Operation
15.3.2.1 Enabling, Disabling and Resetting
The global enabling of interrupts is done by writing a ‘1’ to the Global Interrupt Enable (I) bit in the CPU Status
(CPU.SREG) register. To disable interrupts globally, write a ‘0’ to the I bit in CPU.SREG.
The desired interrupt lines must also be enabled in the respective peripheral by writing to the peripheral’s Interrupt
Control (peripheral.INTCTRL) register.
The interrupt flags are not automatically cleared after the interrupt is executed. The respective INTFLAGS register
descriptions provide information on how to clear specific flags.
15.3.2.2 Interrupt Vector Locations
The interrupt vector placement is dependent on the value of the Interrupt Vector Select (IVSEL) bit in the Control A
(CPUINT.CTRLA) register. Refer to the IVSEL description in CPUINT.CTRLA for the possible locations.
If the program never enables an interrupt source, the interrupt vectors are not used, and the regular program code
can be placed at these locations.
15.3.2.3 Interrupt Response Time
The minimum interrupt response time is represented in the following table.
Table 15-1. Minimum Interrupt Response Time
Flash Size > 8 KB
Flash Size ≤ 8 KB
Finish ongoing instruction
One cycle
One cycle
Store PC to stack
Two cycles
Two cycles
Jump to interrupt handler
Three cycles (jmp)
Two cycles (rjmp)
After the Program Counter is pushed on the stack, the program vector for the interrupt is executed. See the following
figure.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 131
AVR128DB28/32/48/64
CPUINT - CPU Interrupt Controller
Figure 15-2. Interrupt Execution of Single-Cycle Instruction
(1)
If an interrupt occurs during the execution of a multi-cycle instruction, the instruction is completed before the interrupt
is served, as shown in the following figure.
Figure 15-3. Interrupt Execution of Multi-Cycle Instruction
(1)
If an interrupt occurs when the device is in a sleep mode, the interrupt execution response time is increased by five
clock cycles, as shown in the figure below. Also, the response time is increased by the start-up time from the selected
sleep mode.
Figure 15-4. Interrupt Execution From Sleep
(1)
A return from an interrupt handling routine takes four to five clock cycles, depending on the size of the Program
Counter. During these clock cycles, the Program Counter is popped from the stack, and the Stack Pointer is
incremented.
Note:
1. Devices with 8 KB of Flash or less use RJMP instead of JMP, which takes only two clock cycles.
15.3.2.4 Interrupt Priority
All interrupt vectors are assigned to one of three possible priority levels, as shown in the table below. An interrupt
request from a high-priority source will interrupt any ongoing interrupt handler from a normal-priority source. When
returning from the high-priority interrupt handler, the execution of the normal-priority interrupt handler will resume.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 132
AVR128DB28/32/48/64
CPUINT - CPU Interrupt Controller
Table 15-2. Interrupt Priority Levels
Priority
Level
Source
Highest
Non-Maskable Interrupt
Device-dependent and statically assigned
...
Level 1 (high priority)
One vector is optionally user selectable as level 1
Lowest
Level 0 (normal priority)
The remaining interrupt vectors
15.3.2.4.1 Non-Maskable Interrupts
A Non-Maskable Interrupt (NMI) will be executed regardless of the setting of the I bit in CPU.SREG. An NMI will
never change the I bit. No other interrupt can interrupt an NMI handler. If more than one NMI is requested at the
same time, the priority is static according to the interrupt vector address, where the lowest address has the highest
priority.
Which interrupts are non-maskable is device-dependent and not subject to configuration. Non-maskable interrupts
must be enabled before they can be used. Refer to the interrupt vector mapping of the device for available NMI lines.
15.3.2.4.2 High-Priority Interrupt
It is possible to assign one interrupt request to level 1 (high priority) by writing its interrupt vector number to the
CPUINT.LVL1VEC register. This interrupt request will have a higher priority than the other (normal priority) interrupt
requests. The priority level 1 interrupts will interrupt the level 0 interrupt handlers.
15.3.2.4.3 Normal-Priority Interrupts
All interrupt vectors other than NMI are assigned to priority level 0 (normal) by default. The user may override this by
assigning one of these vectors as a high-priority vector. The device will have many normal-priority vectors, and some
of these may be pending at the same time. Two different scheduling schemes are available to choose which of the
pending normal-priority interrupts to service first: Static or round robin.
IVEC is the interrupt vector mapping, as listed in the Peripherals and Architecture chapter. The following sections use
IVEC to explain the scheduling schemes. IVEC0 is the Reset vector, IVEC1 is the NMI vector, and so on. In a vector
table with n+1 elements, the vector with the highest vector number is denoted IVECn. Reset, non-maskable interrupts
and high-level interrupts are included in the IVEC map, but will always be prioritized over the normal-priority
interrupts.
Static Scheduling
If several level 0 interrupt requests are pending at the same time, the one with the highest priority is scheduled for
execution first. The following figure illustrates the default configuration, where the interrupt vector with the lowest
address has the highest priority.
Figure 15-5. Default Static Scheduling
Lowest Address
IVEC 0
Highest Priority
IVEC 1
:
:
:
Highest Address
IVEC n
Lowest Priority
Modified Static Scheduling
The default priority can be changed by writing a vector number to the CPUINT.LVL0PRI register. This vector number
will be assigned the lowest priority. The next interrupt vector in the IVEC will have the highest priority among the LVL0
interrupts, as shown in the following figure.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 133
AVR128DB28/32/48/64
CPUINT - CPU Interrupt Controller
Figure 15-6. Static Scheduling when CPUINT.LVL0PRI is Different From Zero
Lowest Address
IVEC 0
RESET
IVEC 1
NMI
:
:
:
IVEC Y
Lowest Priority
IVEC Y+1
Highest Priority
:
:
:
Highest Address
IVEC n
Here, value Y has been written to CPUINT.LVL0PRI, so that interrupt vector Y+1 has the highest priority. Note that, in
this case, the priorities will wrap so that the lowest address no longer has the highest priority. This does not include
RESET and NMI, which will always have the highest priority.
Refer to the interrupt vector mapping of the device for available interrupt requests and their interrupt vector number.
Round Robin Scheduling
The static scheduling may prevent some interrupt requests from being serviced. To avoid this, the CPUINT offers
round robin scheduling for normal-priority (LVL0) interrupts. In the round robin scheduling, the CPUINT.LVL0PRI
register stores the last acknowledged interrupt vector number. This register ensures that the last acknowledged
interrupt vector gets the lowest priority and is automatically updated by the hardware. The following figure illustrates
the priority order after acknowledging IVEC Y and after acknowledging IVEC Y+1.
Figure 15-7. Round Robin Scheduling
IVEC Y was the last acknowledged
interrupt
IVEC Y+1 was the last acknowledged
interrupt
IVEC 0
RESET
IVEC 0
RESET
IVEC 1
NMI
IVEC 1
NMI
:
:
:
:
:
:
IVEC Y
Lowest Priority
IVEC Y
IVEC Y+1
Highest Priority
IVEC Y+1
Lowest Priority
IVEC Y+2
Highest Priority
:
:
:
IVEC n
:
:
:
IVEC n
The round robin scheduling for LVL0 interrupt requests is enabled by writing a ‘1’ to the Round Robin Priority Enable
(LVL0RR) bit in the Control A (CPUINT.CTRLA) register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 134
AVR128DB28/32/48/64
CPUINT - CPU Interrupt Controller
15.3.2.5 Compact Vector Table
The Compact Vector Table (CVT) is a feature to allow writing of compact code by having all level 0 interrupts share
the same interrupt vector number. Thus, the interrupts share the same Interrupt Service Routine (ISR). This reduces
the number of interrupt handlers and thereby frees up memory that can be used for the application code.
When CVT is enabled by writing a ‘1’ to the CVT bit in the Control A (CPUINT.CTRLA) register, the vector table
contains these three interrupt vectors:
1. The non-maskable interrupts (NMI) at vector address 1.
2. The Priority Level 1 (LVL1) interrupt at vector address 2.
3. All priority level 0 (LVL0) interrupts at vector address 3.
This feature is most suitable for devices with limited memory and applications using a small number of interrupt
generators.
15.3.3
Debug Operation
When using a level 1 priority interrupt, it is important to make sure the Interrupt Service Routine is configured
correctly as it may cause the application to be stuck in an interrupt loop with level 1 priority.
By reading the CPUINT STATUS (CPUINT.STATUS) register, it is possible to see if the application has executed the
correct RETI (interrupt return) instruction. The CPUINT.STATUS register contains state information, which ensures
that the CPUINT returns to the correct interrupt level when the RETI instruction is executed at the end of an interrupt
handler. Returning from an interrupt will return the CPUINT to the state it had before entering the interrupt.
15.3.4
Configuration Change Protection
This peripheral has registers that are under Configuration Change Protection (CCP). To write to these registers, a
certain key must first be written to the CPU.CCP register, followed by a write access to the protected bits within four
CPU instructions.
Attempting to write to a protected register without following the appropriate CCP unlock sequence leaves the
protected register unchanged.
The following registers are under CCP:
Table 15-3. CPUINT - Registers under Configuration Change Protection
Register
Key
IVSEL in CPUINT.CTRLA
IOREG
CVT in CPUINT.CTRLA
IOREG
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 135
AVR128DB28/32/48/64
CPUINT - CPU Interrupt Controller
15.4
Register Summary
Offset
Name
Bit Pos.
0x00
0x01
0x02
0x03
CTRLA
STATUS
LVL0PRI
LVL1VEC
7:0
7:0
7:0
7:0
15.5
7
6
5
IVSEL
CVT
4
3
NMIEX
2
1
0
LVL1EX
LVL0RR
LVL0EX
LVL0PRI[7:0]
LVL1VEC[7:0]
Register Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 136
AVR128DB28/32/48/64
CPUINT - CPU Interrupt Controller
15.5.1
Control A
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
CTRLA
0x00
0x00
Configuration Change Protection
6
IVSEL
R/W
0
5
CVT
R/W
0
4
3
2
1
0
LVL0RR
R/W
0
Bit 6 – IVSEL Interrupt Vector Select
This bit is protected by the Configuration Change Protection mechanism.
Value
Description
0
Interrupt vectors are placed at the start of the application section of the Flash
1
Interrupt vectors are placed at the start of the boot section of the Flash
Bit 5 – CVT Compact Vector Table
This bit is protected by the Configuration Change Protection mechanism.
Value
Description
0
Compact Vector Table function is disabled
1
Compact Vector Table function is enabled
Bit 0 – LVL0RR Round Robin Priority Enable
This bit is not protected by the Configuration Change Protection mechanism.
Value
Description
0
Priority is fixed for priority level 0 interrupt requests: The lowest interrupt vector address has the
highest priority
1
The round robin priority scheme is enabled for priority level 0 interrupt requests
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 137
AVR128DB28/32/48/64
CPUINT - CPU Interrupt Controller
15.5.2
Status
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
NMIEX
R
0
STATUS
0x01
0x00
-
6
5
4
3
2
1
LVL1EX
R
0
0
LVL0EX
R
0
Bit 7 – NMIEX Non-Maskable Interrupt Executing
This flag is set if a non-maskable interrupt is executing. The flag is cleared when returning (RETI) from the interrupt
handler.
Bit 1 – LVL1EX Level 1 Interrupt Executing
This flag is set when a priority level 1 interrupt is executing, or when the interrupt handler has been interrupted by an
NMI. The flag is cleared when returning (RETI) from the interrupt handler.
Bit 0 – LVL0EX Level 0 Interrupt Executing
This flag is set when a priority level 0 interrupt is executing, or when the interrupt handler has been interrupted by a
priority level 1 interrupt or an NMI. The flag is cleared when returning (RETI) from the interrupt handler.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 138
AVR128DB28/32/48/64
CPUINT - CPU Interrupt Controller
15.5.3
Interrupt Priority Level 0
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
LVL0PRI
0x02
0x00
-
7
6
5
R/W
0
R/W
0
R/W
0
4
3
LVL0PRI[7:0]
R/W
R/W
0
0
2
1
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – LVL0PRI[7:0] Interrupt Priority Level 0
This register is used to modify the priority of the LVL0 interrupts. See the section Normal-Priority Interrupts for more
information.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 139
AVR128DB28/32/48/64
CPUINT - CPU Interrupt Controller
15.5.4
Interrupt Vector with Priority Level 1
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
LVL1VEC
0x03
0x00
-
7
6
5
R/W
0
R/W
0
R/W
0
4
3
LVL1VEC[7:0]
R/W
R/W
0
0
2
1
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – LVL1VEC[7:0] Interrupt Vector with Priority Level 1
This bit field contains the number of the single vector with increased priority level 1 (LVL1). If this bit field has the
value 0x00, no vector has LVL1. Consequently, the LVL1 interrupt is disabled.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 140
AVR128DB28/32/48/64
EVSYS - Event System
16.
EVSYS - Event System
16.1
Features
•
•
•
•
•
•
•
16.2
System for Direct Peripheral-to-Peripheral Signaling
Peripherals Can Directly Produce, Use, and React to Peripheral Events
Short and Predictable Response Time
Up to 10 Parallel Event Channels Available
Each Channel is Driven by One Event Generator and Can Have Multiple Event Users
Events Can be Sent and/or Received by Most Peripherals and by Software
The Event System Works in Active, Idle, and Standby Sleep Modes
Overview
The Event System (EVSYS) enables direct peripheral-to-peripheral signaling. It allows a change in one peripheral
(the event generator) to trigger actions in other peripherals (the event users) through event channels, without using
the CPU. It is designed to provide a short and predictable response time between peripherals, allowing for
autonomous peripheral control and interaction, and for synchronized timing of actions in several peripheral modules.
Thus, it is a powerful tool for reducing the complexity, size, and execution time of the software.
A change of the event generator’s state is referred to as an event and usually corresponds to one of the peripheral’s
interrupt conditions. Events can be forwarded directly to other peripherals using the dedicated event routing network.
The routing of each channel is configured in software, including event generation and use.
Only one event signal can be routed on each channel. Multiple peripherals can use events from the same channel.
The EVSYS can connect peripherals such as ADCs, analog comparators, I/O PORT pins, the real-time counter,
timer/counters, and the configurable custom logic peripheral. Events can also be generated from software.
16.2.1
Block Diagram
Figure 16-1. Block Diagram
Event Channel n
SWEVENTx[n]
From Event
Generators
.
.
.
0
D QD Q
CLK_PER
CHANNELn
1
Is
Async?
To Channel
MUX for Async
Event User
To Channel
MUX for Sync
Event User
EVOUTx pin
The block diagram shows the operation of an event channel. A multiplexer controlled by Channel n Generator
Selection (EVSYS.CHANNELn) register at the input selects which of the event sources to route onto the event
channel. Each event channel has two subchannels: one asynchronous and one synchronous. A synchronous user
will listen to the synchronous subchannel, and an asynchronous user will listen to the asynchronous subchannel.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 141
AVR128DB28/32/48/64
EVSYS - Event System
An event signal from an asynchronous source will be synchronized by the Event System before being routed to the
synchronous subchannel. An asynchronous event signal to be used by a synchronous consumer must last for at least
one peripheral clock cycle to ensure that it will propagate through the synchronizer. The synchronizer will delay such
an event between two and three clock cycles, depending on when the event occurs.
Figure 16-2. Example of Event Source, Generator, User, and Action
Event Generator
Event User
Timer/Counter
ADC
Compare Match
Channel Sweep
Event
Routing
Network
Over/Underflow
|
Single
Conversion
Error
Event Action Selection
Event Source
16.2.2
Event Action
Signal Description
Signal
EVOUTx
Type
Digital output
16.3
Functional Description
16.3.1
Initialization
Description
Event output, one output per I/O Port
To utilize events, the Event System, the generating peripheral, and the peripheral(s) using the event must be set up
accordingly:
1.
2.
3.
4.
16.3.2
Configure the generating peripheral appropriately. For example, if the generating peripheral is a timer, set the
prescaling, the Compare register, etc., so that the desired event is generated.
Configure the event user peripheral(s) appropriately. For example, if the ADC is the event user, set the ADC
prescaler, resolution, conversion time, etc., as desired, and configure the ADC conversion to start at the
reception of an event.
Configure the Event System to route the desired source. In this case, the Timer/Compare match to the desired
event channel. This may, for example, be Channel 0, which is accomplished by writing to the Channel 0
Generator Selection (EVSYS.CHANNEL0) register.
Configure the ADC to listen to this channel by writing to the corresponding User x Channel MUX
(EVSYS.USERx) register.
Operation
16.3.2.1 Event User Multiplexer Setup
Each event user has one dedicated event user multiplexer selecting which event channel to listen to. The application
configures these multiplexers by writing to the corresponding EVSYS.USERx register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 142
AVR128DB28/32/48/64
EVSYS - Event System
16.3.2.2 Event System Channel
An event channel can be connected to one of the event generators.
The source for each event channel is configured by writing to the respective Channel n Generator Selection
(EVSYS.CHANNELn) register.
16.3.2.3 Event Generators
Each event channel has several possible event generators, but only one can be selected at a time. The event
generator for a channel is selected by writing to the respective Channel n Generator Selection (EVSYS.CHANNELn)
register. By default, the channels are not connected to any event generator. For details on event generation, refer to
the documentation of the corresponding peripheral.
A generated event is either synchronous or asynchronous to the device peripheral clock (CLK_PER). Asynchronous
events can be generated outside the normal edges of the peripheral clock, making the system respond faster than
the selected clock frequency would suggest. Asynchronous events can also be generated while the device is in a
Sleep mode when the peripheral clock is not running.
Any generated event is classified as either a pulse event or a level event. In both cases, the event can be either
synchronous or asynchronous, with properties according to the table below.
Table 16-1. Properties of Generated Events
Event Type
Sync/Async
Pulse
Level
Description
Sync
An event generated from CLK_PER that lasts one clock cycle
Async
An event generated from a clock other than CLK_PER lasting
one clock cycle
Sync
An event generated from CLK_PER that lasts multiple clock
cycles
Async
An event generated without a clock (for example, a pin or a
comparator), or an event generated from a clock other than
CLK_PER that lasts multiple clock cycles
The properties of both the generated event and the intended event user must be considered in order to ensure
reliable and predictable operation.
The table below shows the available event generators for this device family.
Table 16-2. Event Generators
Generator Name
Description
Event
Type
Generating
Clock Domain
Length of event
Peripheral Event
UPDI
SYNCH
SYNCH character
Level
CLK_PDI
SYNCH character on PDI
RX input synchronized to
CLK_PDI
MVIO
VDDIO2OK
VDDIO2 is OK
Level
Asynchronous
High as long as VDDIO2
is OK
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 143
AVR128DB28/32/48/64
EVSYS - Event System
...........continued
Generator Name
Description
Event
Type
Generating
Clock Domain
Length of event
OVF
Overflow
Pulse
CLK_RTC
One CLK_RTC period
CMP
Compare Match
PIT_DIV8192
Prescaled RTC clock
divided by 8192
PIT_DIV4096
Prescaled RTC clock
divided by 4096
Given by prescaled RTC
clock divided by 4096
PIT_DIV2048
Prescaled RTC clock
divided by 2048
Given by prescaled RTC
clock divided by 2048
PIT_DIV1024
Prescaled RTC clock
divided by 1024
Given by prescaled RTC
clock divided by 1024
PIT_DIV512
Prescaled RTC clock
divided by 512
Given by prescaled RTC
clock divided by 512
PIT_DIV256
Prescaled RTC clock
divided by 256
Given by prescaled RTC
clock divided by 256
PIT_DIV128
Prescaled RTC clock
divided by 128
Given by prescaled RTC
clock divided by 128
PIT_DIV64
Prescaled RTC clock
divided by 64
Given by prescaled RTC
clock divided by 64
CCL
LUTn
LUT output level
Level
Asynchronous
Depends on CCL
configuration
ACn
OUT
Comparator output level
Level
Asynchronous
Given by AC output level
ADCn
RESRDY
Result ready
Pulse
CLK_PER
One CLK_PER period
ZCDn
OUT
ZCD output level
Level
Asynchronous
Given by ZCD output level
OPAMPn
READY
Op amp ready
Pulse
CLK_PER
One CLK_PER period
PORTx
PINn
Pin level
Level
Asynchronous
Given by pin level
USARTn
XCK
USART Baud clock
Level
CLK_PER
Minimum two CLK_PER
periods
SPIn
SCK
SPI Master clock
Level
CLK_PER
Minimum two CLK_PER
periods
OVF_LUNF
Overflow/Low byte timer
underflow
HUNF
High byte timer underflow
Pulse
CLK_PER
One CLK_PER period
Peripheral Event
RTC
TCAn
CMP0_LCMP0 Compare channel 0
match/Low byte timer
compare channel 0 match
Level
Given by prescaled RTC
clock divided by 8192
CMP1_LCMP1 Compare channel 1
match/Low byte timer
compare channel 1 match
CMP2_LCMP2 Compare channel 2
match/Low byte timer
compare channel 2 match
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 144
AVR128DB28/32/48/64
EVSYS - Event System
...........continued
Generator Name
Description
Peripheral Event
TCBn
TCDn
CAPT
CAPT flag set
OVF
Overflow
CMPBCLR
Counter matches
CMPBCLR
CMPASET
Counter matches
CMPASET
CMPBSET
Counter matches
CMPBSET
PROGEV
Programmable event
output
Event
Type
Generating
Clock Domain
Length of event
Pulse
CLK_PER
One CLK_PER period
Pulse
CLK_TCD
One CLK_TCD period
16.3.2.4 Event Users
The event channel to listen to is selected by configuring the event user. An event user may require the event signal to
be either synchronous or asynchronous to the peripheral clock. An asynchronous event user can respond to events
in Sleep modes when clocks are not running. Such events can be responded to outside the normal edges of the
peripheral clock, making the event user respond faster than the clock frequency would suggest. For details on the
requirements of each peripheral, refer to the documentation of the corresponding peripheral.
Most event users implement edge or level detection to trigger actions in the corresponding peripheral based on the
incoming event signal. In both cases, a user can either be synchronous, which requires that the incoming event is
generated from the peripheral clock (CLK_PER), or asynchronous, if not. Some asynchronous event users do not
apply event input detection but use the event signal directly. The different event user properties are described in
general in the table below.
Table 16-3. Properties of Event Users
Input Detection
Edge
Level
No detection
Async/Sync
Description
Sync
An event user is triggered by an event edge and requires that the
incoming event is generated from CLK_PER.
Async
An event user is triggered by an event edge and has
asynchronous detection or an internal synchronizer.
Sync
An event user is triggered by an event level and requires that the
incoming event is generated from CLK_PER.
Async
An event user is triggered by an event level and has
asynchronous detection or an internal synchronizer.
Async
An event user will use the event signal directly.
The table below shows the available event users for this device family.
Table 16-4. Event Users
USER Name
Description
Input Detection
Async/Sync
Peripheral Input
CCL
LUTnx
LUTn input x or clock signal
No detection
Async
ADCn
START
ADC start on event
Edge
Async
EVSYS
EVOUTx
Forward event signal to pin
No detection
Async
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 145
AVR128DB28/32/48/64
EVSYS - Event System
...........continued
USER Name
Description
Input Detection
Async/Sync
IrDA mode input
Level
Sync
Count on positive event edge
Edge
Count on any event edge
Edge
Count while event signal is high
Level
Event level controls count direction
Level
Event level controls count direction
Level
Restart counter on positive event edge
Edge
Restart counter on any event edge
Edge
Restart counter while event signal is high
Level
Time-out check
Edge
Input capture on event
Edge
Input capture frequency measurement
Edge
Input capture pulse-width measurement
Edge
Input capture frequency and pulse-width measurement
Edge
Single-shot
Edge
Both
COUNT
Count on event
Edge
Sync
INPUTA
Fault or capture
Level or edge
Async
Enable OPAMP
Edge
Async
DISABLE Disable OPAMP
Edge
Sync
DUMP
OPAMP VOUT Dump for Integrator Mode
Level
Sync
DRIVE
OPAMP Drive in Normal Mode
Level
Sync
Peripheral Input
USARTn
IRDA
CNTA
TCAn
CNTB
TCBn
TCDn
OPAMPn
CAPT
Sync
Sync
Sync
INPUTB
ENABLE
16.3.2.5 Synchronization
Events can be either synchronous or asynchronous to the peripheral clock. Each Event System channel has two
subchannels: one asynchronous and one synchronous.
The asynchronous subchannel is identical to the event output from the generator. If the event generator generates a
signal asynchronous to the peripheral clock, the signal on the asynchronous subchannel will be asynchronous. If the
event generator generates a signal synchronous to the peripheral clock, the signal on the asynchronous subchannel
will also be synchronous.
The synchronous subchannel is identical to the event output from the generator, if the event generator generates a
signal synchronous to the peripheral clock. If the event generator generates a signal asynchronous to the peripheral
clock, this signal is first synchronized before being routed onto the synchronous subchannel. Depending on when it
occurs, synchronization will delay the event by two to three clock cycles. The Event System automatically performs
this synchronization if an asynchronous generator is selected for an event channel.
16.3.2.6 Software Event
The application can generate a software event. Software events on Channel n are issued by writing a ‘1’ to the
Software Event Channel Select (CHANNEL[n]) bit in the Software Events (EVSYS.SWEVENTx) register. A software
event appears as a pulse on the Event System channel, inverting the current event signal for one clock cycle.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 146
AVR128DB28/32/48/64
EVSYS - Event System
Event users see software events as no different from those produced by event generating peripherals.
16.3.3
Sleep Mode Operation
When configured, the Event System will work in all Sleep modes. Software events represent one exception since
they require a peripheral clock.
Asynchronous event users are able to respond to an event without their clock running in Standby Sleep mode.
Synchronous event users require their clock to be running to be able to respond to events. Such users will only work
in Idle Sleep mode or in Standby Sleep mode, if configured to run in Standby mode by setting the RUNSTDBY bit in
the appropriate register.
Asynchronous event generators are able to generate an event without their clock running, that is, in Standby Sleep
mode. Synchronous event generators require their clock to be running to be able to generate events. Such
generators will only work in Idle Sleep mode or in Standby Sleep mode, if configured to run in Standby mode by
setting the RUNSTDBY bit in the appropriate register.
16.3.4
Debug Operation
This peripheral is unaffected by entering Debug mode.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 147
AVR128DB28/32/48/64
EVSYS - Event System
16.4
Register Summary
Offset
Name
Bit Pos.
0x00
0x01
0x02
...
0x0F
0x10
0x11
0x12
0x13
0x14
0x15
0x16
0x17
0x18
0x19
0x1A
...
0x1F
0x20
...
SWEVENTA
SWEVENTB
7:0
7:0
SWEVENTA[7:0]
SWEVENTB[7:0]
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
CHANNEL0[7:0]
CHANNEL1[7:0]
CHANNEL2[7:0]
CHANNEL3[7:0]
CHANNEL4[7:0]
CHANNEL5[7:0]
CHANNEL6[7:0]
CHANNEL7[7:0]
CHANNEL8[7:0]
CHANNEL9[7:0]
USERCCLLUT0A
7:0
USER[7:0]
USEROPAMP2DRI
VE
7:0
USER[7:0]
0x55
16.5
7
6
5
4
3
2
1
0
Reserved
CHANNEL0
CHANNEL1
CHANNEL2
CHANNEL3
CHANNEL4
CHANNEL5
CHANNEL6
CHANNEL7
CHANNEL8
CHANNEL9
Reserved
Register Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 148
AVR128DB28/32/48/64
EVSYS - Event System
16.5.1
Software Events
Name:
Offset:
Reset:
Property:
SWEVENTx
0x00 + x*0x01 [x=0..1]
0x00
-
Write bits in this register to create a software event on the corresponding event channels.
Bits 0-7 in the EVSYS.SWEVENTA register correspond to event channels 0-7. If the number of available event
channels is between eight and 15, these are available in the EVSYS.SWEVENTB register, where bit n corresponds to
event channel 8+n.
Refer to the Peripheral Overview section for the available number of Event System channels.
Bit
7
6
5
Access
Reset
W
0
W
0
W
0
4
3
SWEVENTx[7:0]
W
W
0
0
2
1
0
W
0
W
0
W
0
Bits 7:0 – SWEVENTx[7:0] Software Event Channel Select
Writing a bit in this bit group to ‘1’ will generate a single-pulse event on the corresponding event channel by inverting
the signal on the event channel for one peripheral clock cycle.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 149
AVR128DB28/32/48/64
EVSYS - Event System
16.5.2
Channel n Generator Selection
Name:
Offset:
Reset:
Property:
CHANNELn
0x10 + n*0x01 [n=0..9]
0x00
-
Each channel can be connected to one event generator. Not all generators can be connected to all channels. Refer to
the table below to see which generator sources can be routed onto each channel and the generator value to be
written to EVSYS.CHANNELn to achieve this routing. Writing the value 0x00 to EVSYS.CHANNELn turns the
channel off.
Bit
Access
Reset
7
6
5
R/W
0
R/W
0
R/W
0
4
3
CHANNELn[7:0]
R/W
R/W
0
0
2
1
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – CHANNELn[7:0] Channel Generator Selection
The specific generator name corresponding to each bit group configuration is given by combining Peripheral and
Output from the table below in the following way: PERIPHERAL_OUTPUT.
Generator
Value
Async/Sync
Description
Channel Availability
Name
Peripheral
Output
0x01
UPDI
SYNCH
Sync
Rising edge of SYNCH character detection
All channels
0x05
MVIO
VDDIO2OK
Async
VDDIO2 OK
All Channels
0x06
RTC
OVF
Async
Counter overflow
All channels
0x07
CMP
0x08
PIT_DIV8192
Prescaled RTC clock divided by 8192
0x09
PIT_DIV4096
Prescaled RTC clock divided by 4096
Compare match
0x0A
PIT_DIV2048
Prescaled RTC clock divided by 2048
0x0B
PIT_DIV1024
Prescaled RTC clock divided by 1024
0x08
PIT_DIV512
Prescaled RTC clock divided by 512
0x09
PIT_DIV256
Prescaled RTC clock divided by 256
0x0A
PIT_DIV128
Prescaled RTC clock divided by 128
0x0B
PIT_DIV64
0x10
CCL
LUT0
Odd numbered channels only
Prescaled RTC clock divided by 64
Async
LUT output level
All channels
OUT
Async
Comparator output level
All channels
RESRDY
Sync
Result ready
All channels
0x11
LUT1
0x12
LUT2
0x13
LUT3
0x14
LUT4(1)
LUT5(1)
0x15
Even numbered channels only
0x20
AC0
0x21
AC1
0x22
AC2
0x24
ADC0
0x30
ZCD0
0x31
ZCD1(1)
ZCD2(1)
OUT
Async
ZCD output level
All channels
0x32
READY
Sync
OPAMP Ready
All Channels
PIN0-PIN7
Async
Pin level(2)
CHANNEL0 and CHANNEL1 only
PIN0-PIN7
Async
PIN level(2)
CHANNEL2 and CHANNEL3 only
PIN0-PIN7
Async
Pin level (2)
CHANNEL4 and CHANNEL5 only
PIN0-PIN7
Async
Pin level (2)
CHANNEL6 and CHANNEL7 only
0x34
OPAMP0
0x35
OPAMP1
0x36
OPAMP2
0x40-0x47
PORTA
0x48-0x4F
PORTB(1)
0x40-0x47
PORTC
0x48-0x4F
PORTD
0x40-0x47
PORTE (1)
0x48-0x4F
PORTF
0x40-0x47
PORTG(1)
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 150
AVR128DB28/32/48/64
EVSYS - Event System
...........continued
Generator
Value
Description
Peripheral
USART0
0x61
USART1
0x62
0x64
USART2
USART3(1)
USART4(1)
0x65
USART5(1)
0x68
SPI0
0x69
SPI1
0x80
XCK
Sync
Clock signal in SPI Master mode and synchronous USART Master
mode
All channels
SCK
Sync
SPI master clock signal
All channels
TCA0
OVF_LUNF
Sync
Overflow/Low byte timer underflow
All channels
Output
0x81
HUNF
High byte timer underflow
0x84
CMP0_LCMP0
Compare channel 0 match/Low byte timer compare channel 0
match
0x85
CMP1_LCMP1
Compare channel 1 match/Low byte timer compare channel 1
match
0x86
CMP2_LCMP2
Compare channel 2 match/Low byte timer compare channel 2
match
0x88
TCA1(1)
OVF_LUNF
Sync
Overflow/Low byte timer underflow
0x89
HUNF
High byte timer underflow
0x8C
CMP0_LCMP0
Compare channel 0 match/Low byte timer compare channel 0
match
0x8D
CMP1_LCMP1
Compare channel 1 match/Low byte timer compare channel 1
match
0x8E
CMP2_LCMP2
Compare channel 2 match/Low byte timer compare channel 2
match
0xA0
TCB0
0xA1
0xA2
0xA4
0xA6
0xA7
0xA8
0xA9
CAPT
Sync
OVF
TCB1
0xA3
0xA5
CAPT
Sync
OVF
TCB2
TCB(1)
TCB4 (1)
CAPT
OVF
CAPT
OVF
CAPT
OVF
CAPT Interrupt flag set(3)
CAPT Interrupt flag set(3)
Sync
Sync
Sync
CAPT interrupt flag set(3)
Counter overflow
CAPT interrupt flag set(3)
Counter overflow
CAPT interrupt flag set (3)
Counter overflow
CMPBCLR
Counter matches CMPBCLR
CMPASET
Counter matches CMPASET
TCD0
CMPBSET
PROGEV
All channels
All channels
Counter overflow
0xB0
0xB3
All channels
Counter overflow
0xB1
0xB2
Channel Availability
Name
0x60
0x63
Async/Sync
Async
Counter matches CMPBSET
All channels
All channels
All channels
All channels
Programmable event output
Notes:
1. Not all peripheral instances are available for all pin-counts. Refer to the Peripherals and Architecture section
for details.
2. Event from PORT pin will be zero if the input driver is disabled.
3. The operational mode of the timer decides when the CAPT flag is raised. See the 16-bit Timer/Counter Type B
(TCB) section for details.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 151
AVR128DB28/32/48/64
EVSYS - Event System
16.5.3
User Channel MUX
Name:
Offset:
Reset:
Property:
USER
0x20 + n*0x01 [n=0..53]
0x00
-
Each event user can be connected to one channel and several users can be connected to the same channel. The
following table lists all Event System users with their corresponding user ID number and name. The user name is
given by combining USER with Peripheral and Input from the table below in the following way:
USERPERIPHERALINPUT.
USER #
User Name
Peripheral Input
Async/
Sync
Description
0x00
LUT0A
CCL LUT0 event input A
0x01
LUT0B
CCL LUT0 event input B
0x02
LUT1A
CCL LUT1 event input A
0x03
LUT1B
CCL LUT1 event input B
0x04
LUT2A
CCL LUT2 event input A
0x05
0x06
CCL
LUT2B
LUT3A
Async
CCL LUT2 event input B
CCL LUT3 event input A
0x07
LUT3B
CCL LUT3 event input B
0x08
LUT4A(1)
CCL LUT4 event input A
0x09
LUT4B(1)
CCL LUT4 event input B
0x0A
LUT5A(1)
CCL LUT5 event input A
0x0B
LUT5B (1)
CCL LUT5 event input B
0x0C
ADC0
START
Async
ADC start on event
0x0D
EVOUTA
Event output A
0x0E
EVOUTB(1)
Event output B
0x0F
EVOUTC
Event output C
0x10
EVSYS
EVOUTD
Async
Event output D
0x11
EVOUTE (1)
Event output E
0x12
EVOUTF (1)
Event output F
0x13
EVOUTG (1)
Event output G
0x14
USART0
IRDA
USART0 IrDA event input
0x15
USART1
IRDA
USART1 IrDA event input
0x16
USART2
IRDA
0x17
USART3(1)
IRDA
0x18
USART4(1) IRDA
USART4 IrDA event input
0x19
USART5
USART5 IrDA event input
IRDA
© 2020 Microchip Technology Inc.
Sync
USART2 IrDA event input
USART3 IrDA event input
Preliminary Datasheet
DS40002247A-page 152
AVR128DB28/32/48/64
EVSYS - Event System
...........continued
USER #
0x1A
0x1B
0x1C
0x1D
0x1E
0x1F
0x20
0x21
0x22
0x23
0x24
0x25
0x26
0x27
0x28
0x29
User Name
Peripheral Input
TCA0
TCA1(1)
TCB0
TCB1
TCB2
TCB3(1)
TCB4(1)
TCD0
CNTA
CNTB
CNTA
CNTB
Async/
Sync
Sync
Sync
Description
Count on event or control count direction
Restart on event or control count direction
Count on event or control count direction
Restart on event or control count direction
CAPT
Both(2)
Start, stop, capture, restart or clear counter
COUNT
Sync
Count on event
CAPT
Both(2)
Start, stop, capture, restart or clear counter
COUNT
Sync
Count on event
CAPT
Both(2)
Start, stop, capture, restart or clear counter
COUNT
Sync
Count on event
CAPT
Both(2)
Start, stop, capture, restart or clear counter
COUNT
Sync
Count on event
CAPT
Both(2)
Start, stop, capture, restart or clear counter
COUNT
Sync
Count on event
INPUTA
INPUTB
Async
Fault or capture
Fault or capture
0x2A
OPAMP0
ENABLE
Async
Enable OPAMP
0x2B
OPAMP0
DISABLE
Sync
Disable OPAMP
0x2C
OPAMP0
DUMP
Async
OPAMP VOUT dump for Integrator mode
0x2D
OPAMP0
DRIVE
Async
OPAMP drive in Normal mode
0x2E
OPAMP1
ENABLE
Async
Enable OPAMP
0x2F
OPAMP1
DISABLE
Sync
Disable OPAMP
0x30
OPAMP1
DUMP
Async
OPAMP VOUT dump for Integrator mode
0x31
OPAMP1
DRIVE
Async
OPAMP drive in Normal mode
0x32
OPAMP2
ENABLE
Async
Enable OPAMP
0x33
OPAMP2
DISABLE
Sync
Disable OPAMP
0x34
OPAMP2
DUMP
Async
OPAMP VOUT dump for Integrator mode
0x35
OPAMP2
DRIVE
Async
OPAMP drive in Normal mode
Notes:
1. Not all peripheral instances are available for all pin-counts. Refer to the Peripherals and Architecture section
for details.
2. Depends on timer operational mode.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 153
AVR128DB28/32/48/64
EVSYS - Event System
Bit
7
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
USER[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – USER[7:0] User Channel Selection
Configures which Event System channel the user is connected to.
Value
Description
0
OFF, no channel is connected to this Event System user
n
The event user is connected to CHANNEL(n-1)
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 154
AVR128DB28/32/48/64
PORTMUX - Port Multiplexer
17.
PORTMUX - Port Multiplexer
17.1
Overview
The Port Multiplexer (PORTMUX) can either enable or disable the functionality of the pins, or change between default
and alternative pin positions. Available options are described in detail in the PORTMUX register map and depend on
the actual pin and its properties.
For available pins and functionality, refer to the I/O Multiplexing and Considerations section.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 155
AVR128DB28/32/48/64
PORTMUX - Port Multiplexer
17.2
Register Summary
Offset
Name
Bit Pos.
0x00
0x01
0x02
0x03
0x04
0x05
0x06
0x07
0x08
0x09
0x0A
EVSYSROUTEA
CCLROUTEA
USARTROUTEA
USARTROUTEB
SPIROUTEA
TWIROUTEA
TCAROUTEA
TCBROUTEA
TCDROUTEA
ACROUTEA
ZCDROUTEA
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
17.3
7
6
EVOUTG
USART3[1:0]
5
4
EVOUTF
EVOUTE
LUT5
LUT4
USART2[1:0]
TCA1[2:0]
TCB4
3
2
EVOUTD
EVOUTC
LUT3
LUT2
USART1[1:0]
USART5[1:0]
SPI1[1:0]
TWI1[1:0]
TCB3
TCB2
AC2
ZCD2
1
0
EVOUTB
EVOUTA
LUT1
LUT0
USART0[1:0]
USART4[1:0]
SPI0[1:0]
TWI0[1:0]
TCA0[2:0]
TCB1
TCB0
TCD0[2:0]
AC1
AC0
ZCD1
ZCD0
Register Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 156
AVR128DB28/32/48/64
PORTMUX - Port Multiplexer
17.3.1
EVSYS Pin Position
Name:
Offset:
Reset:
Property:
Bit
7
Access
Reset
EVSYSROUTEA
0x00
0x00
-
6
EVOUTG
R/W
0
5
EVOUTF
R/W
0
4
EVOUTE
R/W
0
3
EVOUTD
R/W
0
2
EVOUTC
R/W
0
1
EVOUTB
R/W
0
0
EVOUTA
R/W
0
Bit 6 – EVOUTG Event Output G
This bit controls pin position for event output G.
Value
Name
Description
0
1
DEFAULT
ALT1
EVOUT on PG2
EVOUT on PG7
Bit 5 – EVOUTF Event Output F
This bit controls pin position for event output F.
Value
Name
Description
0
1
DEFAULT
-
EVOUT on PF2
Reserved
Bit 4 – EVOUTE Event Output E
This bit controls pin position for event output E.
Value
Name
Description
0
1
DEFAULT
ALT1
EVOUT on PE2
EVOUT on PE7
Bit 3 – EVOUTD Event Output D
This bit controls pin position for event output D.
Value
Name
Description
0
1
DEFAULT
ALT1
EVOUT on PD2
EVOUT on PD7
Bit 2 – EVOUTC Event Output C
This bit controls pin position for event output C.
Value
Name
Description
0
1
DEFAULT
ALT1
EVOUT on PC2
EVOUT on PC7
Bit 1 – EVOUTB Event Output B
This bit controls pin position for event output B.
Value
Name
Description
0
1
DEFAULT
ALT1
EVOUT on PB2
EVOUT on PB7
Bit 0 – EVOUTA Event Output A
This bit controls pin position for event output A.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 157
AVR128DB28/32/48/64
PORTMUX - Port Multiplexer
Value
Name
Description
0
1
DEFAULT
ALT1
EVOUT on PA2
EVOUT on PA7
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 158
AVR128DB28/32/48/64
PORTMUX - Port Multiplexer
17.3.2
CCL LUTn Pin Position
Name:
Offset:
Reset:
Property:
Bit
7
CCLROUTEA
0x01
0x00
-
6
Access
Reset
5
LUT5
R/W
0
4
LUT4
R/W
0
3
LUT3
R/W
0
2
LUT2
R/W
0
1
LUT1
R/W
0
0
LUT0
R/W
0
Bit 5 – LUT5 CCL LUT 5 Signals
This bit field controls the pin positions for CCL LUT 5 signals
Value
Name
0
1
DEFAULT
ALT1
Description
OUT
IN0
IN1
IN2
PG3
PG6
PG0
PG0
PG1
PG1
PG2
PG2
Bit 4 – LUT4 CCL LUT 4 Signals
This bit field controls the pin positions for CCL LUT 4 signals.
Value
0
1
Name
DEFAULT
ALT1
Description
OUT
IN0
IN1
IN2
PB3
PB6
PB0
PB0
PB1
PB1
PB2
PB2
Bit 3 – LUT3 CCL LUT 3 Signals
This bit field controls the pin positions for CCL LUT 3 signals.
Value
0
1
Name
DEFAULT
-
Description
OUT
IN0
IN1
IN2
PF3
Reserved
PF0
Reserved
PF1
Reserved
PF2
Reserved
Bit 2 – LUT2 CCL LUT 2 Signals
This bit field controls the pin positions for CCL LUT 2 signals.
Value
0
1
Name
DEFAULT
ALT1
Description
OUT
IN0
IN1
IN2
PD3
PD6
PD0
PD0
PD1
PD1
PD2
PD2
Bit 1 – LUT1 CCL LUT 1 Signals
This bit field controls the pin positions for CCL LUT 1 signals.
Value
Name
0
1
DEFAULT
ALT1
© 2020 Microchip Technology Inc.
Description
OUT
IN0
IN1
IN2
PC3
PC6
PC0
PC0
PC1
PC1
PC2
PC2
Preliminary Datasheet
DS40002247A-page 159
AVR128DB28/32/48/64
PORTMUX - Port Multiplexer
Bit 0 – LUT0 CCL LUT 0 Signals
This bit field controls the pin positions for CCL LUT 0 signals.
Value
Name
0
1
DEFAULT
ALT1
© 2020 Microchip Technology Inc.
Description
OUT
IN0
IN1
IN2
PA3
PA6
PA0
PA0
PA1
PA1
PA2
PA2
Preliminary Datasheet
DS40002247A-page 160
AVR128DB28/32/48/64
PORTMUX - Port Multiplexer
17.3.3
USARTn Pin Position
Name:
Offset:
Reset:
Property:
Bit
USARTROUTEA
0x02
0x00
-
7
6
USART3[1:0]
R/W
R/W
0
0
Access
Reset
5
4
USART2[1:0]
R/W
R/W
0
0
3
2
USART1[1:0]
R/W
R/W
0
0
1
0
USART0[1:0]
R/W
R/W
0
0
Bits 7:6 – USART3[1:0] USART 3 Signals
This bit field controls the pin positions for USART 3 signals.
Value
Name
0x0
0x1
0x2
0x3
DEFAULT
ALT1
NONE
Description
TxD
RxD
XCK
XDIR
PB0
PB1
PB4
PB5
Reserved
Not connected to any pins
PB2
PB6
PB3
PB7
XCK
XDIR
PF2
Not available
PF3
Not available
Bits 5:4 – USART2[1:0] USART 2 Signals
This bit field controls the pin positions for USART 2 signals.
Value
Name
0x0
0x1
0x2
0x3
DEFAULT
ALT1
NONE
Description
TxD
RxD
PF0
PF1
PF4
PF5
Reserved
Not connected to any pins
Bits 3:2 – USART1[1:0] USART 1 Signals
This bit field controls the pin positions for USART 1 signals.
Value
Name
0x0
0x1
0x2
0x3
DEFAULT
ALT1
NONE
Description
RxD
XCK
XDIR
PC0
PC1
PC4
PC5
Reserved
Not connected to any pins
TxD
PC2
PC6
PC3
PC7
RxD
XCK
XDIR
PA0
PA1
PA4
PA5
Reserved
Not connected to any pins
PA2
PA6
PA3
PA7
Bits 1:0 – USART0[1:0] USART 0 Signals
This bit field controls the pin positions for USART 0 signals.
Value
Name
0x0
0x1
0x2
0x3
DEFAULT
ALT1
NONE
Description
TxD
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 161
AVR128DB28/32/48/64
PORTMUX - Port Multiplexer
17.3.4
USARTn Pin Position
Name:
Offset:
Reset:
Property:
Bit
7
USARTROUTEB
0x03
0x00
-
6
5
4
Access
Reset
3
2
USART5[1:0]
R/W
R/W
0
0
1
0
USART4[1:0]
R/W
R/W
0
0
Bits 3:2 – USART5[1:0] USART 5 Signals
This bit field controls the pin positions for USART 5 signals.
Value
Name
0x0
0x1
0x2
0x3
DEFAULT
ALT1
NONE
Description
RxD
XCK
XDIR
PG0
PG1
PG4
PG5
Reserved
Not connected to any pins
TxD
PG2
PG6
PG3
PG7
RxD
XCK
XDIR
PE0
PE1
PE4
PE5
Reserved
Not connected to any pins
PE2
PE6
PE3
PE7
Bits 1:0 – USART4[1:0] USART 4 Signals
This bit field controls the pin positions for USART 4 signals.
Value
Name
0x0
0x1
0x2
0x3
DEFAULT
ALT1
NONE
Description
TxD
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 162
AVR128DB28/32/48/64
PORTMUX - Port Multiplexer
17.3.5
SPIn Pin Position
Name:
Offset:
Reset:
Property:
Bit
7
SPIROUTEA
0x04
0x00
-
6
5
4
3
2
1
SPI1[1:0]
Access
Reset
R/W
0
0
SPI0[1:0]
R/W
0
R/W
0
R/W
0
Bits 3:2 – SPI1[1:0] SPI 1 Signals
This bit field controls the pin positions for SPI 1 signals.
Value
Name
0x0
0x1
0x2
0x3
DEFAULT
ALT1
ALT2
NONE
Description
MOSI
MISO
PC0
PC1
PC4
PC5
PB4
PB5
Not connected to any pins
SCK
SS
PC2
PC6
PB6
PC3
PC7
PB7
Bits 1:0 – SPI0[1:0] SPI 0 Signals
This bit field controls the pin positions for SPI 0 signals.
Value
Name
0x0
0x1
0x2
0x3
DEFAULT
ALT1
ALT2
NONE
Description
MOSI
© 2020 Microchip Technology Inc.
MISO
SCK
SS
PA4
PA5
PE0
PE1
PG4
PG5
Not connected to any pins
PA6
PE2
PG6
PA7
PE3
PG7
Preliminary Datasheet
DS40002247A-page 163
AVR128DB28/32/48/64
PORTMUX - Port Multiplexer
17.3.6
TWIn Pin Position
Name:
Offset:
Reset:
Property:
Bit
7
TWIROUTEA
0x05
0x00
-
6
5
4
3
2
1
TWI1[1:0]
Access
Reset
R/W
0
0
TWI0[1:0]
R/W
0
R/W
0
R/W
0
Bits 3:2 – TWI1[1:0] TWI 1 Signals
This bit field controls the pin positions for TWI 1 signals.
Value
Name
Description
Master/Slave
0x0
0x1
0x2
0x3
DEFAULT
ALT1
ALT2
-
Dual mode (Slave)
SDA
SCL
SDA
SCL
PF2
PF2
PB2
Reserved
PF3
PF3
PB3
PB2
PB6(1)
PB6(1)
PB3
PB7(1)
PB7(1)
Note:
1. Normal and Fast Mode only. Standard General Purpose I/O driver specification apply.
Bits 1:0 – TWI0[1:0] TWI 0 Signals
This bit field controls the pin positions for TWI 0 signals.
Value
Name
Description
Master/Slave
0x0
0x1
0x2
0x3
DEFAULT
ALT1
ALT2
-
Dual mode (Slave)
SDA
SCL
SDA
SCL
PA2
PA2
PC2
Reserved
PA3
PA3
PC3
PC2
PC6(1)
PC6(1)
PC3
PC7(1)
PC7(1)
Note:
1. Normal and Fast Mode only. Standard General Purpose I/O driver specification apply.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 164
AVR128DB28/32/48/64
PORTMUX - Port Multiplexer
17.3.7
TCAn Pin Position
Name:
Offset:
Reset:
Property:
Bit
7
TCAROUTEA
0x06
0x00
-
6
Access
Reset
5
4
TCA1[2:0]
R/W
0
R/W
0
3
2
R/W
0
R/W
0
1
TCA0[2:0]
R/W
0
0
R/W
0
Bits 5:3 – TCA1[2:0] TCA1 Signals
This bit field controls the pin positions for TCA1 signals.
Value
Name
0x0
0x1
0x2
0x3
Other
PORTB
PORTC
PORTE
PORTG
-
Description
WO0
PB0
PC4
PE4
PG0
Reserved
WO1
WO2
WO3
WO4
WO5
PB1
PC5
PE5
PG1
PB2
PC6
PE6
PG2
PB3
PG3
PB4
PG4
PB5
PG5
Bits 2:0 – TCA0[2:0] TCA0 Signals
This bit field controls the pin positions for TCA0 signals.
Value
Name
0x0
0x1
0x2
0x3
0x4
0x5
0x6
0x7
PORTA
PORTB
PORTC
PORTD
PORTE
PORTF
PORTG
-
Description
WO0
© 2020 Microchip Technology Inc.
PA0
PB0
PC0
PD0
PE0
PF0
PG0
Reserved
WO1
WO2
WO3
WO4
WO5
PA1
PB1
PC1
PD1
PE1
PF1
PG1
PA2
PB2
PC2
PD2
PE2
PF2
PG2
PA3
PB3
PC3
PD3
PE3
PF3
PG3
PA4
PB4
PC4
PD4
PE4
PF4
PG4
PA5
PB5
PC5
PD5
PE5
PF5
PG5
Preliminary Datasheet
DS40002247A-page 165
AVR128DB28/32/48/64
PORTMUX - Port Multiplexer
17.3.8
TCBn Pin Position
Name:
Offset:
Reset:
Property:
Bit
7
TCBROUTEA
0x07
0x00
-
6
5
Access
Reset
4
TCB4
R/W
0
3
TCB3
R/W
0
2
TCB2
R/W
0
1
TCB1
R/W
0
0
TCB0
R/W
0
Bit 4 – TCB4 TCB4 Output
This bit controls pin position for TCB4 output.
Value
Name
Description
0
1
DEFAULT
ALT1
WO on PG3
WO on PC6
Bit 3 – TCB3 TCB3 Output
This bit controls pin position for TCB3 output.
Value
Name
Description
0
1
DEFAULT
ALT1
WO on PB5
WO on PC1
Bit 2 – TCB2 TCB2 Output
This bit controls pin position for TCB2 output.
Value
Name
Description
0
1
DEFAULT
ALT1
WO on PC0
WO on PB4
Bit 1 – TCB1 TCB1 Output
This bit controls pin position for TCB1 output.
Value
Name
Description
0
1
DEFAULT
ALT1
WO on PA3
WO on PF5
Bit 0 – TCB0 TCB0 Output
This bit controls pin position for TCB0 output.
Value
Name
Description
0
1
DEFAULT
ALT1
WO on PA2
WO on PF4
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 166
AVR128DB28/32/48/64
PORTMUX - Port Multiplexer
17.3.9
TCDn Pin Position
Name:
Offset:
Reset:
Property:
Bit
7
TCDROUTEA
0x08
0x00
-
6
5
4
3
2
Access
Reset
1
TCD0[2:0]
R/W
0
R/W
0
0
R/W
0
Bits 2:0 – TCD0[2:0] TCD0 Signals
This bit field controls the pin positions for TCD0 signals.
Value
Name
0x0
0x1
0x2
0x3
Other
DEFAULT
ALT1
ALT2
ALT3
-
© 2020 Microchip Technology Inc.
Description
WOA
WOB
WOC
WOD
PA4
PB4
PF0
PG4
Reserved
PA5
PB5
PF1
PG5
PA6
PB6
PF2
PG6
PA7
PB7
PF3
PG7
Preliminary Datasheet
DS40002247A-page 167
AVR128DB28/32/48/64
PORTMUX - Port Multiplexer
17.3.10 ACn Pin Position
Name:
Offset:
Reset:
Property:
Bit
7
ACROUTEA
0x09
0x00
-
6
5
4
3
2
AC2
R/W
0
Access
Reset
1
AC1
R/W
0
0
AC0
R/W
0
Bit 2 – AC2 Analog Comparator 2 Output
This bit controls pin position for AC2 output.
Value
Name
Description
0
1
DEFAULT
ALT1
OUT on PA7
OUT on PC6
Bit 1 – AC1 Analog comparator 1 Output
This bit controls pin position for AC1 output.
Value
Name
Description
0
1
DEFAULT
ALT1
OUT on PA7
OUT on PC6
Bit 0 – AC0 Analog Comparator 0 Output
This bit controls pin position for AC0 output.
Value
Name
Description
0
1
DEFAULT
ALT1
OUT on PA7
OUT on PC6
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 168
AVR128DB28/32/48/64
PORTMUX - Port Multiplexer
17.3.11 ZCDn Pin Position
Name:
Offset:
Reset:
Property:
Bit
7
ZCDROUTEA
0x0A
0x00
-
6
5
4
3
2
ZCD2
R/W
0
Access
Reset
1
ZCD1
R/W
0
0
ZCD0
R/W
0
Bit 2 – ZCD2 Zero Cross Detector 2 Output
This bit controls pin position for ZCD2 output.
Value
Name
Description
0
1
DEFAULT
ALT1
OUT on PA7
OUT on PC7
Bit 1 – ZCD1 Zero Cross Detector 1 Output
This bit controls pin position for ZCD1 output.
Value
Name
Description
0
1
DEFAULT
ALT1
OUT on PA7
OUT on PC7
Bit 0 – ZCD0 Zero Cross Detector 0 Output
This bit controls pin position for ZCD0 output.
Value
Name
Description
0
1
DEFAULT
ALT1
OUT on PA7
OUT on PC7
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 169
AVR128DB28/32/48/64
PORT - I/O Pin Configuration
18.
PORT - I/O Pin Configuration
18.1
Features
•
•
•
•
•
18.2
General Purpose Input and Output Pins with Individual Configuration:
– Pull-up
– Inverted I/O
– Input voltage threshold
Interrupts and Events:
– Sense both edges
– Sense rising edges
– Sense falling edges
– Sense low level
Optional Slew Rate Control per I/O Port
Asynchronous Pin Change Sensing that Can Wake the Device From all Sleep Modes
Efficient and Safe Access to Port Pins
– Hardware Read-Modify-Write (RMW) through dedicated toggle/clear/set registers
– Mapping of often-used PORT registers into bit-accessible I/O memory space (virtual ports)
Overview
The I/O pins of the device are controlled by instances of the PORT peripheral registers. Each PORT instance has up
to eight I/O pins. The PORTs are named PORTA, PORTB, PORTC, etc. Refer to the I/O Multiplexing and
Considerations section to see which pins are controlled by what instance of PORT. The base addresses of the PORT
instances and the corresponding Virtual PORT instances are listed in the Peripherals and Architecture section.
Each PORT pin has a corresponding bit in the Data Direction (PORTx.DIR) and Data Output Value (PORTx.OUT)
registers to enable that pin as an output and to define the output state. For example, pin PA3 is controlled by DIR[3]
and OUT[3] of the PORTA instance.
The input value of a PORT pin is synchronized to the Peripheral Clock (CLK_PER) and then made accessible as the
data input value (PORTx.IN). The value of the pin can be read whether the pin is configured as input or output.
The PORT also supports asynchronous input sensing with interrupts and events for selectable pin change conditions.
Asynchronous pin change sensing means that a pin change can trigger an interrupt and wake the device from sleep,
including sleep modes where CLK_PER is stopped.
All pin functions are individually configurable per pin. The pins have hardware Read-Modify-Write functionality for a
safe and correct change of the drive values and/or input and sense configuration.
The PORT pin configuration controls input and output selection of other device functions.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 170
AVR128DB28/32/48/64
PORT - I/O Pin Configuration
18.2.1
Block Diagram
Figure 18-1. PORT Block Diagram
Pull-up Enable
DIRn
Q
D
Peripheral Override
R
OUTn
Q
D
Pxn
Peripheral Override
R
Invert Enable
Synchronizer
INn
Synchronous
Input
Q
D Q
R
D
R
Sense Configuration
Interrupt
Generator
Interrupt
Asynchronous
Input/Event
Input Level Select
Input Disable
Peripheral Override
Analog
Input/Output
18.2.2
Signal Description
Signal
Type
Description
Pxn
I/O pin
I/O pin n on PORTx
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 171
AVR128DB28/32/48/64
PORT - I/O Pin Configuration
18.3
Functional Description
18.3.1
Initialization
After Reset, all outputs are tri-stated, and digital input buffers enabled even if there is no clock running.
The following steps are all optional when initializing PORT operation:
•
•
•
•
Enable or disable the output driver for pin Pxn by respectively writing ‘1’ to bit n in the PORTx.DIRSET or
PORTx.DIRCLR register
Set the output driver for pin Pxn to high or low level respectively by writing ‘1’ to bit n in the PORTx.OUTSET or
PORTx.OUTCLR register
Read the input of pin Pxn by reading bit n in the PORTx.IN register
Configure the individual pin configurations and interrupt control for pin Pxn in PORTx.PINnCTRL
Important: For lowest power consumption, disable the digital input buffer of unused pins and pins that
are used as analog inputs or outputs.
Specific pins, such as those used to connect a debugger, may be configured differently, as required by their special
function.
18.3.2
Operation
18.3.2.1 Basic Functions
Each pin group x has its own set of PORT registers. I/O pin Pxn can be controlled by the registers in PORTx.
To use pin number n as an output, write bit n of the PORTx.DIR register to ‘1’. This can be done by writing bit n in the
PORTx.DIRSET register to ‘1’, which will avoid disturbing the configuration of other pins in that group. The nth bit in
the PORTx.OUT register must be written to the desired output value.
Similarly, writing a PORTx.OUTSET bit to ‘1’ will set the corresponding bit in the PORTx.OUT register to ‘1’. Writing a
bit in PORTx.OUTCLR to ‘1’ will clear that bit in PORTx.OUT to ‘0’. Writing a bit in PORTx.OUTTGL or PORTx.IN to
‘1’ will toggle that bit in PORTx.OUT.
To use pin n as an input, bit n in the PORTx.DIR register must be written to ‘0’ to disable the output driver. This can
be done by writing bit n in the PORTx.DIRCLR register to ‘1’, which will avoid disturbing the configuration of other
pins in that group. The input value can be read from bit n in the PORTx.IN register as long as the ISC bit is not set to
INPUT_DISABLE.
Writing a bit to ‘1’ in PORTx.DIRTGL will toggle that bit in PORTx.DIR and toggle the direction of the corresponding
pin.
18.3.2.2 Port Configuration
The Port Control (PORTx.PORTCTRL) register is used to configure the slew rate limitation for all the PORTx pins.
The slew rate limitation is enabled by writing a ‘1’ to the Slew Rate Limit Enable (SLR) bit in PORTx.PORTCTRL.
Refer to the Electrical Characteristics section for further details.
18.3.2.3 Pin Configuration
The Pin n Control (PORTx.PINnCTRL) register is used to configure inverted I/O, pull-up, and input sensing of a pin.
The control register for pin n is at the byte address PORTx + 0x10 + �.
All input and output on the respective pin n can be inverted by writing a ‘1’ to the Inverted I/O Enable (INVEN) bit in
PORTx.PINnCTRL. When INVEN is ‘1’, the PORTx.IN/OUT/OUTSET/OUTTGL registers will have inverted operation
for this pin.
Toggling the INVEN bit causes an edge on the pin, which can be detected by all peripherals using this pin, and is
seen by interrupts or events if enabled.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 172
AVR128DB28/32/48/64
PORT - I/O Pin Configuration
The Input Level Select (INLVL) bit controls the input voltage threshold for pin n in PORTx.PINnCTRL. A selection of
Schmitt trigger thresholds derived from the supply voltage or TTL levels are available.
The input threshold is important in determining the value of bit n in the PORTx.IN register and also the level at which
an interrupt condition occurs if that feature is enabled.
The input pull-up of pin n is enabled by writing a ‘1’ to the Pull-up Enable (PULLUPEN) bit in PORTx.PINnCTRL. The
pull-up is disconnected when the pin is configured as an output, even if PULLUPEN is ‘1’.
Pin interrupts can be enabled for pin n by writing to the Input/Sense Configuration (ISC) bit field in
PORTx.PINnCTRL. Refer to 18.3.3 Interrupts for further details.
The digital input buffer for pin n can be disabled by writing the INPUT_DISABLE setting to ISC. This can reduce
power consumption and may reduce noise if the pin is used as analog input. While configured to INPUT_DISABLE,
bit n in PORTx.IN will not change since the input synchronizer is disabled.
18.3.2.4 Multi-Pin Configuration
The multi-pin configuration function is used to configure multiple port pins in one operation. The wanted pin
configuration is first written to the PORTx.PINCONFIG register, followed by a register write with the selected pins to
modify. This allows changing the configuration (PORTx.PINnCTRL) for up to eight pins in one write.
Tip: The PORTx.PINCONFIG register is mirrored on all ports, which allows the use of a single setting
across multiple ports. The PORTx.PINCTRLUPD/SET/CLR registers are not mirrored and need to be
applied to each port.
For the multi-pin configuration, port pins can be configured and modified by writing to the following registers.
Table 18-1. Multi-Pin Configuration Registers
Register
Description
PORTx.PINCONFIG
PINnCTRL (ISC, PULLUPEN, INLVL and INVEN) setting to load simultaneously to
multiple PINnCTRL registers
PORTx.PINCTRLUPD
Writing a ‘1’ to bit n in the PINCTRLUPD register will copy the PINCONFIG register
content to the PINnCTRL register
PORTx.PINCTRLSET(1)
Writing a ‘1’ to bit n in the PINCTRLSET register will set the individual bits in the
PINnCTRL register, according to the bits set to ‘1’ in the PINCONFIG register
PORTx.PINCTRLCLR(2)
Writing a ‘1’ to bit n in the PINCTRLCLRn register will clear the individual bits in the
PINnCTRL register, according to the bits set to ‘1’ in the PINCONFIG register
Notes:
1. Using PINCTRLSET to configure ISC bit fields that are non-zero will result in a bitwise OR with the
PINCONFIG and PINnCTRL registers. This may give an unexpected setting.
2. Using PINCTRLCLR to configure ISC bit fields that are non-zero will result in a bitwise inverse AND with the
PINCONFIG and PINnCTRL registers. This may give an unexpected setting.
18.3.2.5 Virtual Ports
The Virtual PORT registers map the most frequently used regular PORT registers into the I/O Register space with
single-cycle bit access. Access to the Virtual PORT registers has the same outcome as access to the regular
registers but allows for memory specific instructions, such as bit manipulation instructions, which cannot be used in
the extended I/O Register space where the regular PORT registers reside. The following table shows the mapping
between the PORT and VPORT registers.
Table 18-2. Virtual Port Mapping
Regular PORT Register
Mapped to Virtual PORT Register
PORTx.DIR
VPORTx.DIR
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 173
AVR128DB28/32/48/64
PORT - I/O Pin Configuration
...........continued
Regular PORT Register
Mapped to Virtual PORT Register
PORTx.OUT
VPORTx.OUT
PORTx.IN
VPORTx.IN
PORTx.INTFLAGS
VPORTx.INTFLAGS
18.3.2.6 Peripheral Override
Peripherals such as USARTs, ADCs and timers may be connected to I/O pins. Such peripherals will usually have a
primary and, optionally, one or more alternate I/O pin connections, selectable by PORTMUX or a multiplexer inside
the peripheral. By configuring and enabling such peripherals, the general purpose I/O pin behavior normally
controlled by PORT will be overridden in a peripheral dependent way. Some peripherals may not override all the
PORT registers, leaving the PORT module to control some aspects of the I/O pin operation.
Refer to the description of each peripheral for information on the peripheral override. Any pin in a PORT that is not
overridden by a peripheral will continue to operate as a general purpose I/O pin.
18.3.2.7 Multi-Voltage I/O
One or more PORT pin groups are connected to the VDDIO2 power domain, allowing a different I/O supply voltage
on these pins. Refer to the Multi-Voltage I/O (MVIO) section for further information.
18.3.3
Interrupts
Table 18-3. Available Interrupt Vectors and Sources
Name
Vector Description Conditions
PORTx PORT interrupt
INTn in PORTx.INTFLAGS is raised as configured by the Input/Sense Configuration
(ISC) bit in PORTx.PINnCTRL
Each PORT pin n can be configured as an interrupt source. Each interrupt can be individually enabled or disabled by
writing to ISC in PORTx.PINnCTRL.
When an interrupt condition occurs, the corresponding interrupt flag is set in the Interrupt Flags register of the
peripheral (peripheral.INTFLAGS).
An interrupt request is generated when the corresponding interrupt source is enabled and the interrupt flag is set. The
interrupt request remains active until the interrupt flag is cleared. See the peripheral's INTFLAGS register for details
on how to clear interrupt flags.
When setting or changing interrupt settings, take these points into account:
• If an Inverted I/O Enable (INVEN) bit is toggled in the same cycle as ISC is changed, the edge caused by the
inversion toggling may not cause an interrupt request
• Changing INLVL for a pin must be performed while relevant interrupts and peripheral modules are disabled.
Changing the threshold while a module is active may generate a temporary state transition on the input,
regardless of the actual voltage level on that pin.
• If an input is disabled by writing to ISC while synchronizing an interrupt, that interrupt may be requested on reenabling the input, even if it is re-enabled with a different interrupt setting
• If the interrupt setting is changed by writing to ISC while synchronizing an interrupt, that interrupt may not be
requested
18.3.3.1 Asynchronous Sensing Pin Properties
All PORT pins support asynchronous input sensing with interrupts for selectable pin change conditions. Fully
asynchronous pin change sensing can trigger an interrupt and wake the device from all sleep modes, including
modes where the Peripheral Clock (CLK_PER) is stopped, while partially asynchronous pin change sensing is limited
as per the table below. See the I/O Multiplexing and Considerations section for further details on which pins support
fully asynchronous pin change sensing.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 174
AVR128DB28/32/48/64
PORT - I/O Pin Configuration
Table 18-4. Behavior Comparison of Sense Pins
Property
Partially Asynchronous Pins
Waking the device from sleep
modes with CLK_PER running
From all interrupt sense configurations
Waking the device from sleep
modes with CLK_PER stopped
Only from BOTHEDGES or LEVEL
interrupt sense configurations
Minimum pulse-width to trigger an
interrupt with CLK_PER running
Minimum one CLK_PER cycle
Minimum pulse-width to trigger an
interrupt with CLK_PER stopped
The pin value must be kept until
CLK_PER has restarted(1)
Interrupt “dead-time”
No new interrupt for three CLK_PER
cycles after the previous
Fully Asynchronous Pins
From all interrupt sense
configurations
Less than one CLK_PER cycle
Note:
1. If a partially asynchronous input pin is used for wake-up from sleep with CLK_PER stopped, the required level
must be held long enough for the MCU to complete the wake-up to trigger the interrupt. If the level disappears,
the MCU can wake up without any interrupt generated.
18.3.4
Events
PORT can generate the following events:
Table 18-5. Event Generators in PORTx
Generator Name
Peripheral
Event
PORTx
PINn
Description
Event Type
Generating Clock Domain
Length of Event
Pin level
Level
Asynchronous
Given by pin level
All PORT pins are asynchronous event system generators. PORT has as many event generators as there are PORT
pins in the device. Each event system output from PORT is the value present on the corresponding pin if the digital
input buffer is enabled. If a pin input buffer is disabled, the corresponding event system output is zero.
PORT has no event inputs. Refer to the Event System (EVSYS) section for more details regarding event types and
Event System configuration.
18.3.5
Sleep Mode Operation
Except for interrupts and input synchronization, all pin configurations are independent of sleep modes. All pins can
wake the device from sleep, see the PORT Interrupt section for further details.
Peripherals connected to the PORTs can be affected by sleep modes, described in the respective peripherals’ data
sheet section.
Important: The PORTs will always use the Peripheral Clock (CLK_PER). Input synchronization will halt
when this clock stops.
18.3.6
Debug Operation
When the CPU is halted in Debug mode, the PORT continues normal operation. If the PORT is configured in a way
that requires it to be periodically serviced by the CPU through interrupts or similar, improper operation or data loss
may result during debugging.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 175
AVR128DB28/32/48/64
PORT - I/O Pin Configuration
18.4
Register Summary - PORTx
Offset
Name
Bit Pos.
0x00
0x01
0x02
0x03
0x04
0x05
0x06
0x07
0x08
0x09
0x0A
0x0B
0x0C
0x0D
0x0E
0x0F
0x10
0x11
0x12
0x13
0x14
0x15
0x16
0x17
DIR
DIRSET
DIRCLR
DIRTGL
OUT
OUTSET
OUTCLR
OUTTGL
IN
INTFLAGS
PORTCTRL
PINCONFIG
PINCTRLUPD
PINCTRLSET
PINCTRLCLR
Reserved
PIN0CTRL
PIN1CTRL
PIN2CTRL
PIN3CTRL
PIN4CTRL
PIN5CTRL
PIN6CTRL
PIN7CTRL
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
18.5
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7
6
5
4
3
2
1
0
DIR[7:0]
DIRSET[7:0]
DIRCLR[7:0]
DIRTGL[7:0]
OUT[7:0]
OUTSET[7:0]
OUTCLR[7:0]
OUTTGL[7:0]
IN[7:0]
INT[7:0]
SRL
INVEN
INLVL
PULLUPEN
PINCTRLUPD[7:0]
PINCTRLSET[7:0]
PINCTRLCLR[7:0]
ISC[2:0]
INVEN
INVEN
INVEN
INVEN
INVEN
INVEN
INVEN
INVEN
INLVL
INLVL
INLVL
INLVL
INLVL
INLVL
INLVL
INLVL
PULLUPEN
PULLUPEN
PULLUPEN
PULLUPEN
PULLUPEN
PULLUPEN
PULLUPEN
PULLUPEN
ISC[2:0]
ISC[2:0]
ISC[2:0]
ISC[2:0]
ISC[2:0]
ISC[2:0]
ISC[2:0]
ISC[2:0]
Register Description - PORTx
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 176
AVR128DB28/32/48/64
PORT - I/O Pin Configuration
18.5.1
Data Direction
Name:
Offset:
Reset:
Property:
Bit
7
DIR
0x00
0x00
-
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
DIR[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – DIR[7:0] Data Direction
This bit field controls the output driver for each PORTx pin.
This bit field does not control the digital input buffer. The digital input buffer for pin n (Pxn) can be configured in the
Input/Sense Configuration (ISC) bit field in the Pin n Control (PORTx.PINnCTRL) register.
The available configuration for each bit n in this bit field is shown in the table below.
Value
Description
0
Pxn is configured as an input-only pin, and the output driver is disabled
1
Pxn is configured as an output pin, and the output driver is enabled
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 177
AVR128DB28/32/48/64
PORT - I/O Pin Configuration
18.5.2
Data Direction Set
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
DIRSET
0x01
0x00
-
7
6
5
R/W
0
R/W
0
R/W
0
4
3
DIRSET[7:0]
R/W
R/W
0
0
2
1
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – DIRSET[7:0] Data Direction Set
This bit field controls the output driver for each PORTx pin, without using a read-modify-write operation.
Writing a ‘0’ to bit n in this bit field has no effect.
Writing a ‘1’ to bit n in this bit field will set the corresponding bit in PORTx.DIR, which will configure pin n (Pxn) as an
output pin and enable the output driver.
Reading this bit field will return the value of PORTx.DIR.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 178
AVR128DB28/32/48/64
PORT - I/O Pin Configuration
18.5.3
Data Direction Clear
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
DIRCLR
0x02
0x00
-
7
6
5
R/W
0
R/W
0
R/W
0
4
3
DIRCLR[7:0]
R/W
R/W
0
0
2
1
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – DIRCLR[7:0] Data Direction Clear
This bit field controls the output driver for each PORTx pin, without using a read-modify-write operation.
Writing a ‘0’ to bit n in this bit field has no effect.
Writing a ‘1’ to bit n in this bit field will clear the corresponding bit in PORTx.DIR, which will configure pin n (Pxn) as
an input-only pin and disable the output driver.
Reading this bit field will return the value of PORTx.DIR.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 179
AVR128DB28/32/48/64
PORT - I/O Pin Configuration
18.5.4
Data Direction Toggle
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
DIRTGL
0x03
0x00
-
7
6
5
R/W
0
R/W
0
R/W
0
4
3
DIRTGL[7:0]
R/W
R/W
0
0
2
1
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – DIRTGL[7:0] Data Direction Toggle
This bit field controls the output driver for each PORTx pin, without using a read-modify-write operation.
Writing a ‘0’ to bit n in this bit field has no effect.
Writing a ‘1’ to bit n in this bit field will toggle the corresponding bit in PORTx.DIR.
Reading this bit field will return the value of PORTx.DIR.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 180
AVR128DB28/32/48/64
PORT - I/O Pin Configuration
18.5.5
Output Value
Name:
Offset:
Reset:
Property:
Bit
7
OUT
0x04
0x00
-
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
OUT[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – OUT[7:0] Output Value
This bit field controls the output driver level for each PORTx pin.
This configuration only has an effect when the output driver (PORTx.DIR) is enabled for the corresponding pin.
The available configuration for each bit n in this bit field is shown in the table below.
Value
Description
0
The pin n (Pxn) output is driven low
1
The Pxn output is driven high
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 181
AVR128DB28/32/48/64
PORT - I/O Pin Configuration
18.5.6
Output Value Set
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
OUTSET
0x05
0x00
-
7
6
5
R/W
0
R/W
0
R/W
0
4
3
OUTSET[7:0]
R/W
R/W
0
0
2
1
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – OUTSET[7:0] Output Value Set
This bit field controls the output driver level for each PORTx pin, without using a read-modify-write operation.
Writing a ‘0’ to bit n in this bit field has no effect.
Writing a ‘1’ to bit n in this bit field will set the corresponding bit in PORTx.OUT, which will configure the output for pin
n (Pxn) to be driven high.
Reading this bit field will return the value of PORTx.OUT.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 182
AVR128DB28/32/48/64
PORT - I/O Pin Configuration
18.5.7
Output Value Clear
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
OUTCLR
0x06
0x00
-
7
6
5
R/W
0
R/W
0
R/W
0
4
3
OUTCLR[7:0]
R/W
R/W
0
0
2
1
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – OUTCLR[7:0] Output Value Clear
This bit field controls the output driver level for each PORTx pin, without using a read-modify-write operation.
Writing a ‘0’ to bit n in this bit field has no effect.
Writing a ‘1’ to bit n in this bit field will clear the corresponding bit in PORTx.OUT, which will configure the output for
pin n (Pxn) to be driven low.
Reading this bit field will return the value of PORTx.OUT.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 183
AVR128DB28/32/48/64
PORT - I/O Pin Configuration
18.5.8
Output Value Toggle
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
OUTTGL
0x07
0x00
-
7
6
5
R/W
0
R/W
0
R/W
0
4
3
OUTTGL[7:0]
R/W
R/W
0
0
2
1
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – OUTTGL[7:0] Output Value Toggle
This bit field controls the output driver level for each PORTx pin, without using a read-modify-write operation.
Writing a ‘0’ to bit n in this bit field has no effect.
Writing a ‘1’ to bit n in this bit field will toggle the corresponding bit in PORTx.OUT.
Reading this bit field will return the value of PORTx.OUT.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 184
AVR128DB28/32/48/64
PORT - I/O Pin Configuration
18.5.9
Input Value
Name:
Offset:
Reset:
Property:
Bit
7
IN
0x08
0x00
-
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
IN[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – IN[7:0] Input Value
This bit field shows the state of the PORTx pins when the digital input buffer is enabled.
Writing a ‘0’ to bit n in this bit field has no effect.
Writing a ‘1’ to bit n in this bit field will toggle the corresponding bit in PORTx.OUT.
If the digital input buffer is disabled, the input is not sampled, and the bit value will not change. The digital input buffer
for pin n (Pxn) can be configured in the Input/Sense Configuration (ISC) bit field in the Pin n Control
(PORTx.PINnCTRL) register.
The available states of each bit n in this bit field is shown in the table below.
Value
Description
0
The voltage level on Pxn is low
1
The voltage level on Pxn is high
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 185
AVR128DB28/32/48/64
PORT - I/O Pin Configuration
18.5.10 Interrupt Flags
Name:
Offset:
Reset:
Property:
Bit
7
INTFLAGS
0x09
0x00
-
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
INT[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – INT[7:0] Pin Interrupt Flag
Pin interrupt flag n is cleared by writing a ‘1’ to it.
Pin interrupt flag n is set when the change or state of pin n (Pxn) matches the pin's Input/Sense Configuration (ISC)
in PORTx.PINnCTRL.
Writing a ‘0’ to bit n in this bit field has no effect.
Writing a ‘1’ to bit n in this bit field will clear Pin interrupt flag n.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 186
AVR128DB28/32/48/64
PORT - I/O Pin Configuration
18.5.11 Port Control
Name:
Offset:
Reset:
Property:
PORTCTRL
0x0A
0x00
-
This register contains the slew rate limit enable bit for this port.
Bit
7
6
5
4
3
Access
Reset
2
1
0
SRL
R/W
0
Bit 0 – SRL Slew Rate Limit Enable
This bit controls the slew rate limitation for all pins in PORTx.
Value
Description
0
Slew rate limitation is disabled for all pins in PORTx
1
Slew rate limitation is enabled for all pins in PORTx
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 187
AVR128DB28/32/48/64
PORT - I/O Pin Configuration
18.5.12 Multi-Pin Configuration
Name:
Offset:
Reset:
Property:
PINCONFIG
0x0B
0x00
-
The multi-pin configuration write enables the configuration of several pins of a port in a single cycle, for faster
configuration of the port module. Especially with large pin count devices, this function can significantly speed up port
pin configuration operations.
Writing to this register may be followed by a write to either of the Multi-Pin Control (PORTx.PINCTRLUPD/SET/CLR)
registers to update the Pin n Control (PORTx.PINnCTRL) registers for PORTx.
This register is mirrored across all PORTx modules.
Bit
Access
Reset
7
INVEN
R/W
0
6
INLVL
R/W
0
5
4
3
PULLUPEN
R/W
0
2
R/W
0
1
ISC[2:0]
R/W
0
0
R/W
0
Bit 7 – INVEN Inverted I/O Enable
This bit controls whether the input and output for pin n are inverted or not.
Value
Description
0
Input and output values are not inverted
1
Input and output values are inverted
Bit 6 – INLVL Input Level Select
This bit controls the input voltage threshold for pin n, used for port input reads and interrupt conditions.
Value
Name
Description
0
ST
Schmitt Trigger derived from supply level
1
TTL
TTL Levels
Bit 3 – PULLUPEN Pull-up Enable
This bit controls whether the internal pull-up of pin n is enabled or not when the pin is configured as input-only.
Value
Description
0
Pull-up disabled
1
Pull-up enabled
Bits 2:0 – ISC[2:0] Input/Sense Configuration
This bit field controls the input and sense configuration of pin n. The sense configuration determines how a port
interrupt can be triggered.
Value
Name
Description
0x0
INTDISABLE
Interrupt disabled but digital input buffer enabled
0x1
BOTHEDGES
Interrupt enabled with sense on both edges
0x2
RISING
Interrupt enabled with sense on rising edge
0x3
FALLING
Interrupt enabled with sense on falling edge
0x4
INPUT_DISABLE
Interrupt and digital input buffer disabled(1)
0x5
LEVEL
Interrupt enabled with sense on low level
other
—
Reserved
Note:
1. If the digital input buffer for pin n is disabled, bit n in the Input Value (PORTx.IN) register will not be updated.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 188
AVR128DB28/32/48/64
PORT - I/O Pin Configuration
18.5.13 Multi-Pin Control Update Mask
Name:
Offset:
Reset:
Property:
PINCTRLUPD
0x0C
0x00
-
The multi-pin configuration write enables the configuration of several pins of a port in a single cycle, for faster
configuration of the port module. Especially with large pin count devices, this function can significantly speed up port
pin configuration operations.
Bit
Access
Reset
7
6
5
R/W
0
R/W
0
R/W
0
4
3
PINCTRLUPD[7:0]
R/W
R/W
0
0
2
1
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – PINCTRLUPD[7:0] Multi-Pin Control Update Mask
This bit field controls the copy of the Multi-Pin Configuration (PORTx.PINCONFIG) register content to the individual
Pin n Control (PORTx.PINnCTRL) registers, without using an individual write operation for each register.
Writing a ‘0’ to bit n in this bit field has no effect.
Writing a ‘1’ to bit n in this bit field will copy the PORTx.PINCONFIG register content to the corresponding
PORTx.PINnCTRL register.
Reading this bit field will always return zero.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 189
AVR128DB28/32/48/64
PORT - I/O Pin Configuration
18.5.14 Multi-Pin Control Set Mask
Name:
Offset:
Reset:
Property:
PINCTRLSET
0x0D
0x00
-
The multi-pin configuration write enables the configuration of several pins of a port in a single cycle, for faster
configuration of the port module. Especially with large pin count devices, this function can significantly speed up port
pin configuration operations.
Bit
Access
Reset
7
6
5
R/W
0
R/W
0
R/W
0
4
3
PINCTRLSET[7:0]
R/W
R/W
0
0
2
1
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – PINCTRLSET[7:0] Multi-Pin Control Set Mask
This bit field controls the setting of bits in the individual Pin n Control (PORTx.PINnCTRL) registers, without using an
individual read-modify-write operation for each register.
Writing a ‘0’ to bit n in this bit field has no effect.
Writing a ‘1’ to bit n in this bit field will set the individual bits in the PORTx.PINnCTRL register, according to the bits
set to ‘1’ in the Multi-Pin Configuration (PORTx.PINCONFIG) register.
Reading this bit field will always return zero.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 190
AVR128DB28/32/48/64
PORT - I/O Pin Configuration
18.5.15 Multi-Pin Control Clear Mask
Name:
Offset:
Reset:
Property:
PINCTRLCLR
0x0E
0x00
-
The multi-pin configuration write enables the configuration of several pins of a port in a single cycle, for faster
configuration of the port module. Especially with large pin count devices, this function can significantly speed up port
pin configuration operations.
Bit
Access
Reset
7
6
5
R/W
0
R/W
0
R/W
0
4
3
PINCTRLCLR[7:0]
R/W
R/W
0
0
2
1
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – PINCTRLCLR[7:0] Multi-Pin Control Clear Mask
This bit field controls the clearing of bits in the individual Pin n Control (PORTx.PINnCTRL) registers, without using
an individual read-modify-write operation for each register.
Writing a ‘0’ to bit n in this bit field has no effect.
Writing a ‘1’ to bit n in this bit field will clear the individual bits in the PORTx.PINnCTRL register, according to the bits
set to ‘1’ in the Multi-Pin Configuration (PORTx.PINCONFIG) register.
Reading this bit field will always return zero.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 191
AVR128DB28/32/48/64
PORT - I/O Pin Configuration
18.5.16 Pin n Control
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
INVEN
R/W
0
PINnCTRL
0x10 + n*0x01 [n=0..7]
0x00
-
6
INLVL
R/W
0
5
4
3
PULLUPEN
R/W
0
2
R/W
0
1
ISC[2:0]
R/W
0
0
R/W
0
Bit 7 – INVEN Inverted I/O Enable
This bit controls whether the input and output for pin n are inverted or not.
Value
Description
0
Input and output values are not inverted
1
Input and output values are inverted
Bit 6 – INLVL Input Level Select
This bit controls the input voltage threshold for pin n, used for port input reads and interrupt conditions.
Value
Name
Description
0
ST
Schmitt Trigger derived from supply level
1
TTL
TTL Levels
Bit 3 – PULLUPEN Pull-up Enable
This bit controls whether the internal pull-up of pin n is enabled or not when the pin is configured as input-only.
Value
Description
0
Pull-up disabled
1
Pull-up enabled
Bits 2:0 – ISC[2:0] Input/Sense Configuration
This bit field controls the input and sense configuration of pin n. The sense configuration determines how a port
interrupt can be triggered.
Value
Name
Description
0x0
INTDISABLE
Interrupt disabled but digital input buffer enabled
0x1
BOTHEDGES
Interrupt enabled with sense on both edges
0x2
RISING
Interrupt enabled with sense on rising edge
0x3
FALLING
Interrupt enabled with sense on falling edge
0x4
INPUT_DISABLE
Interrupt and digital input buffer disabled(1)
0x5
LEVEL
Interrupt enabled with sense on low level
other
—
Reserved
Note:
1. If the digital input buffer for pin n is disabled, bit n in the Input Value (PORTx.IN) register will not be updated.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 192
AVR128DB28/32/48/64
PORT - I/O Pin Configuration
18.6
Register Summary - VPORTx
Offset
Name
Bit Pos.
0x00
0x01
0x02
0x03
DIR
OUT
IN
INTFLAGS
7:0
7:0
7:0
7:0
18.7
7
6
5
4
3
2
1
0
DIR[7:0]
OUT[7:0]
IN[7:0]
INT[7:0]
Register Description - VPORTx
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 193
AVR128DB28/32/48/64
PORT - I/O Pin Configuration
18.7.1
Data Direction
Name:
Offset:
Reset:
Property:
DIR
0x00
0x00
-
Access to the Virtual PORT registers has the same outcome as access to the regular registers but allows for memory
specific instructions, such as bit manipulation instructions, which cannot be used in the extended I/O Register space
where the regular PORT registers reside.
Bit
7
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
DIR[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – DIR[7:0] Data Direction
This bit field controls the output driver for each PORTx pin.
This bit field does not control the digital input buffer. The digital input buffer for pin n (Pxn) can be configured in the
Input/Sense Configuration (ISC) bit field in the Pin n Control (PORTx.PINnCTRL) register.
The available configuration for each bit n in this bit field is shown in the table below.
Value
Description
0
Pxn is configured as an input-only pin, and the output driver is disabled
1
Pxn is configured as an output pin, and the output driver is enabled
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 194
AVR128DB28/32/48/64
PORT - I/O Pin Configuration
18.7.2
Output Value
Name:
Offset:
Reset:
Property:
OUT
0x01
0x00
-
Access to the Virtual PORT registers has the same outcome as access to the regular registers but allows for memory
specific instructions, such as bit manipulation instructions, which cannot be used in the extended I/O Register space
where the regular PORT registers reside.
Bit
7
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
OUT[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – OUT[7:0] Output Value
This bit field controls the output driver level for each PORTx pin.
This configuration only has an effect when the output driver (PORTx.DIR) is enabled for the corresponding pin.
The available configuration for each bit n in this bit field is shown in the table below.
Value
Description
0
The pin n (Pxn) output is driven low
1
The Pxn output is driven high
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 195
AVR128DB28/32/48/64
PORT - I/O Pin Configuration
18.7.3
Input Value
Name:
Offset:
Reset:
Property:
IN
0x02
0x00
-
Access to the Virtual PORT registers has the same outcome as access to the regular registers but allows for memory
specific instructions, such as bit manipulation instructions, which cannot be used in the extended I/O Register space
where the regular PORT registers reside.
Bit
7
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
IN[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – IN[7:0] Input Value
This bit field shows the state of the PORTx pins when the digital input buffer is enabled.
Writing a ‘0’ to bit n in this bit field has no effect.
Writing a ‘1’ to bit n in this bit field will toggle the corresponding bit in PORTx.OUT.
If the digital input buffer is disabled, the input is not sampled, and the bit value will not change. The digital input buffer
for pin n (Pxn) can be configured in the Input/Sense Configuration (ISC) bit field in the Pin n Control
(PORTx.PINnCTRL) register.
The available states of each bit n in this bit field is shown in the table below.
Value
Description
0
The voltage level on Pxn is low
1
The voltage level on Pxn is high
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 196
AVR128DB28/32/48/64
PORT - I/O Pin Configuration
18.7.4
Interrupt Flags
Name:
Offset:
Reset:
Property:
INTFLAGS
0x03
0x00
-
Access to the Virtual PORT registers has the same outcome as access to the regular registers but allows for memory
specific instructions, such as bit manipulation instructions, which cannot be used in the extended I/O Register space
where the regular PORT registers reside.
Bit
7
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
INT[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – INT[7:0] Pin Interrupt Flag
Pin interrupt flag n is cleared by writing a ‘1’ to it.
Pin interrupt flag n is set when the change or state of pin n (Pxn) matches the pin's Input/Sense Configuration (ISC)
in PORTx.PINnCTRL.
Writing a ‘0’ to bit n in this bit field has no effect.
Writing a ‘1’ to bit n in this bit field will clear Pin interrupt flag n.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 197
AVR128DB28/32/48/64
MVIO - Multi-Voltage I/O
19.
MVIO - Multi-Voltage I/O
19.1
Features
•
•
•
•
•
•
•
19.2
A Subset of the Device I/O Pins Can Be Powered by VDDIO2
The VDDIO2 Supply Can Ramp Up and Down Independently of the VDD Supply
Single- or Dual-Supply Configuration Determined by Fuse
PORT Access and Peripheral Override Independent of the Supply Configuration
VDDIO2 Supply Status Bit
Interrupt and Event for VDDIO2 Supply Status Change
ADC Channel for Measuring VDDIO2 Supply Voltage
Overview
The MVIO feature allows a subset of the I/O pins to be powered by a different I/O voltage domain than the rest of the
I/O pins. This eliminates the need of having external level shifters for communication or control of external
components running on a different voltage level. The MVIO-capable I/O pads are supplied by a voltage applied to the
VDDIO2 power pin(s), while the regular I/O pins are supplied by the voltage applied to the VDD pin(s).
The MVIO can be configured in one of two supply modes:
•
•
Single-Supply mode, where the MVIO-capable I/O pins are powered at the same voltage level as the non-MVIO
capable pins, i.e., VDD. The user must connect the VDDIO2 pin(s) to the VDD pin(s).
Dual-Supply mode, where the MVIO-capable I/O pins are supplied by the VDDIO2 voltage, which may be different
from the voltage supplied to the VDD pin(s).
A configuration fuse determines the MVIO supply mode. The loss or gain of power on VDDIO2 is signaled by a status
register bit. This status bit has corresponding interrupt and event functionality.
The MVIO pins are capable of the same digital behavior as regular I/O pins, e.g., GPIO, serial communication
(USART, SPI, I2C), or connected to PWM peripherals. The input Schmitt trigger levels are scaled according to the
VDDIO2 voltage, as described in the Electrical Characteristics section of the data sheet.
A divided-down VDDIO2 voltage is available as input to the ADC.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 198
AVR128DB28/32/48/64
MVIO - Multi-Voltage I/O
19.2.1
Block Diagram
Figure 19-1. MVIO Block Diagram
VDDIO2
Voltage
Monitor
VDDIO2
VDDIO2S
Voltage Divider
ADC Input
Edge Detector
VDDIO2F
(Int Req.)
VDDCORE
Enable
PORTy[7:0]
VDDIO
Voltage
Monitor
VDDCORE
VDD
PORT
PORTx[7:0]
19.2.2
Signal Description
Signal
Description
Type
VDD
Power pin for VDDIO and other power domains
Supply
VDDIO2
Power pin for VDDIO2
Supply
PORTx[7:0]
PORT pins powered by VDDIO
Input/Output
PORTy[7:0]
PORT pins powered by VDDIO2
Input/Output
19.3
Functional Description
19.3.1
Initialization
Initialize the MVIO in Dual-Supply configuration by following these steps:
1. Program the Multi-Voltage System Configuration (MVSYSCFG) fuse to the Dual-Supply configuration.
2. Optional: Write the VDDIO2 Interrupt Enable (VDDIO2IE) bit to ‘1’ in the Interrupt Control (MVIO.INTCTRL)
register.
3. Read the VDDIO2 Status (VDDIO2S) bit in the Status (MVIO.STATUS) register to check if the VDDIO2 voltage
is within the acceptable range for operation.
4. Configure and use the PORT pins powered by VDDIO2.
If the MVSYSCFG fuse is programmed to the Single-Supply configuration, the VDDIO2 Status bit is read as ‘1’, and
the VDDIO2 Interrupt Flag is read as ‘0’.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 199
AVR128DB28/32/48/64
MVIO - Multi-Voltage I/O
19.3.2
Operation
19.3.2.1 Power Sequencing
The system supports the following power ramp scenarios for MVIO when configured in Dual-Supply mode:
• Supply ramp of VDDIO before VDDIO2
• Supply ramp of VDDIO2 before VDDIO
• VDDIO2 loses and regains power
• VDDIO loses and regains power
When either voltage domain loses power, the MVIO I/O pins are tri-stated. If VDDIO2 regains power, the pins will
reload the current configuration of the PORT registers. If VDDIO loses power, the device will reset, and the PORTs will
have to be reinitialized. Refer to the Electrical Characteristics section for VDD and VDDIO2 power supply thresholds.
19.3.2.2 Voltage Measurement
19.3.3
VDDIO2 is available as an internal input channel to the ADC. The voltage is divided by 10 to allow the use of any
internal ADC reference. To measure VDDIO2, follow these steps:
• Configure the voltage reference for the ADC
• Select VDDIO2 as the positive input to the ADC
• Run a single-ended ADC conversion
• Calculate the voltage using the following equation:
ADC Result × �REF × 10
�DDIO2 =
ADC Resolution
Events
The MVIO can generate the following events:
Table 19-1. Event Generators in MVIO
Generator Name
Peripheral Event
MVIO
VDDIO2OK
Description
Event
Type
Generating Clock
Domain
Length of Event
VDDIO2 level is
above threshold
Level
Asynchronous
Given by the VDDIO2 Status
(VDDIO2S) bit in the Status
(MVIO.STATUS) register
The MVIO has no event users. Refer to the Event System (EVSYS) section for more details regarding event types
and Event System configuration.
19.3.4
Interrupts
Table 19-2. Available Interrupt Vectors and Sources
Name
Vector Description
Interrupt Flag
MVIO
VDDIO2 interrupt
VDDIO2IF
Conditions
VDDIO2S toggles
A change in the VDDIO2 Status (VDDIO2S) bit in the Status (MVIO.STATUS) register can be configured to trigger an
interrupt. This can be enabled or disabled by writing to the VDDIO2 Interrupt Enable (VDDIO2IE) bit in the Interrupt
Control (MVIO.INTCTRL) register.
When an interrupt condition occurs, the corresponding interrupt flag is set in the peripheral’s Interrupt Flags
(peripheral.INTFLAGS) register.
An interrupt source is enabled or disabled by writing to the corresponding enable bit in the peripheral’s Interrupt
Control (peripheral.INTCTRL) register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 200
AVR128DB28/32/48/64
MVIO - Multi-Voltage I/O
An interrupt request is generated when the corresponding interrupt source is enabled, and the interrupt flag is set.
The interrupt request remains active until the interrupt flag is cleared. See the peripheral’s INTFLAGS register for
details on how to clear interrupt flags.
19.3.5
Sleep Mode Operation
When enabled by the Multi-Voltage System Configuration (MVSYSCFG) fuse, the module will operate in all sleep
modes.
19.3.6
Debug Operation
When the CPU is halted in Debug mode, the MVIO continues normal operation. If the MVIO is configured in a way
that requires it to be periodically serviced by the CPU through interrupts or similar, improper operation or data loss
may result during debugging.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 201
AVR128DB28/32/48/64
MVIO - Multi-Voltage I/O
19.4
Register Summary
Offset
Name
Bit Pos.
0x00
0x01
0x02
INTCTRL
INTFLAGS
STATUS
7:0
7:0
7:0
19.5
7
6
5
4
3
2
1
0
VDDIO2IE
VDDIO2IF
VDDIO2S
Register Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 202
AVR128DB28/32/48/64
MVIO - Multi-Voltage I/O
19.5.1
Interrupt Control
Name:
Offset:
Reset:
Property:
Bit
7
INTCTRL
0x00
0x00
-
6
5
4
3
2
Access
Reset
1
0
VDDIO2IE
R/W
0
Bit 0 – VDDIO2IE VDDIO2 Interrupt Enable
This bit controls whether the interrupt for a VDDIO2 Status change is enabled or not.
Value
Description
0
The VDDIO2 interrupt is disabled
1
The VDDIO2 interrupt is enabled
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 203
AVR128DB28/32/48/64
MVIO - Multi-Voltage I/O
19.5.2
Interrupt Flags
Name:
Offset:
Reset:
Property:
Bit
7
INTFLAGS
0x01
0x00
-
6
5
4
3
2
Access
Reset
1
0
VDDIO2IF
R/W
0
Bit 0 – VDDIO2IF VDDIO2 Interrupt Flag
This flag is cleared by writing a ‘1’ to it.
This flag is set when the VDDIO2 Status (VDDIO2S) bit in MVIO.STATUS changes value.
Writing a ‘0’ to this bit has no effect.
Writing a ‘1’ to this bit will clear the VDDIO2 Interrupt Flag.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 204
AVR128DB28/32/48/64
MVIO - Multi-Voltage I/O
19.5.3
Status
Name:
Offset:
Reset:
Property:
Bit
7
STATUS
0x02
0x00
-
6
5
4
3
Access
Reset
2
1
0
VDDIO2S
R
0
Bit 0 – VDDIO2S VDDIO2 Status
This bit shows the state of the VDDIO2 voltage level.
Writing to this bit has no effect.
Value
Description
0
The VDDIO2 supply voltage is below the acceptable range for operation. The MVIO pins are tri-stated.
1
The VDDIO2 supply voltage is within the acceptable range for operation. The MVIO pin configurations
are loaded from the corresponding PORT registers.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 205
AVR128DB28/32/48/64
BOD - Brown-out Detector
20.
BOD - Brown-out Detector
20.1
Features
•
•
•
•
•
20.2
Brown-out Detector Monitors the Power Supply to Avoid Operation Below a Programmable Level
Three Available Modes:
– Enabled mode (continuously active)
– Sampled mode
– Disabled
Separate Selection of Mode for Active and Sleep Modes
Voltage Level Monitor (VLM) with Interrupt
Programmable VLM Level Relative to the BOD Level
Overview
The Brown-out Detector (BOD) monitors the power supply and compares the supply voltage with the programmable
brown-out threshold level. The brown-out threshold level defines when to generate a System Reset. The Voltage
Level Monitor (VLM) monitors the power supply and compares it to a threshold higher than the BOD threshold. The
VLM can then generate an interrupt as an “early warning” when the supply voltage is approaching the BOD threshold.
The VLM threshold level is expressed as a percentage above the BOD threshold level.
The BOD is controlled mainly by fuses and has to be enabled by the user. The mode used in Standby sleep mode
and Power-Down sleep mode can be altered in normal program execution. The VLM is controlled by I/O registers as
well.
When activated, the BOD can operate in Enabled mode, where the BOD is continuously active, or in Sampled mode,
where the BOD is activated briefly at a given period to check the supply voltage level.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 206
AVR128DB28/32/48/64
BOD - Brown-out Detector
20.2.1
Block Diagram
Figure 20-1. BOD Block Diagram
VDD
BOD
+
BOD
Reset
BOD Threshold
-
VLM
+
VLM Threshold
20.3
Functional Description
20.3.1
Initialization
VLM
Interrupt
-
The BOD settings are loaded from fuses during Reset. The BOD level and operating mode in Active mode and Idle
sleep mode are set by fuses and cannot be changed by software. The operating mode in Standby and Power-Down
sleep mode is loaded from fuses and can be changed by software.
The Voltage Level Monitor function can be enabled by writing a ‘1’ to the VLM Interrupt Enable (VLMIE) bit in the
Interrupt Control (BOD.INTCTRL) register. The VLM interrupt is configured by writing the VLM Configuration
(VLMCFG) bits in BOD.INTCTRL. An interrupt is requested when the supply voltage crosses the VLM threshold from
either above or below.
The VLM functionality will follow the BOD mode. If the BOD is disabled, the VLM will not be enabled, even if the
VLMIE is ‘1’. If the BOD is using Sampled mode, the VLM will also be sampled. When enabling the VLM interrupt, the
interrupt flag will always be set if VLMCFG equals 0x2, and may be set if VLMCFG is configured to 0x0 or 0x1.
The VLM threshold is defined by writing the VLM Level (VLMLVL) bits in the VLM Control (BOD.VLMCTRL) register.
20.3.2
Interrupts
Table 20-1. Available Interrupt Vectors and Sources
Name
VLM
Vector Description
Conditions
Voltage Level Monitor Supply voltage crossing the VLM threshold as configured by the VLM Configuration
(VLMCFG) bit field in the Interrupt Control (BOD.INTCTRL) register
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 207
AVR128DB28/32/48/64
BOD - Brown-out Detector
The VLM interrupt will not be executed if the CPU is halted in Debug mode.
When an interrupt condition occurs, the corresponding interrupt flag is set in the peripheral’s Interrupt Flags
(peripheral.INTFLAGS) register.
An interrupt source is enabled or disabled by writing to the corresponding enable bit in the peripheral’s Interrupt
Control (peripheral.INTCTRL) register.
An interrupt request is generated when the corresponding interrupt source is enabled, and the interrupt flag is set.
The interrupt request remains active until the interrupt flag is cleared. See the peripheral’s INTFLAGS register for
details on how to clear interrupt flags.
20.3.3
Sleep Mode Operation
The BOD configuration in the different sleep modes is defined by fuses. The mode used in Active mode and Idle
sleep mode is defined by the ACTIVE fuses in FUSE.BODCFG, which is loaded into the ACTIVE bit field in the
Control A (BOD.CTRLA) register. The mode used in Standby sleep mode and Power-Down sleep mode is defined by
SLEEP in FUSE.BODCFG, which is loaded into the SLEEP bit field in the Control A (BOD.CTRLA) register.
The operating mode in Active mode and Idle sleep mode (i.e., ACTIVE in BOD.CTRLA) cannot be altered by
software. The operating mode in Standby sleep mode and Power-Down sleep mode can be altered by writing to the
SLEEP bit field in the Control A (BOD.CTRLA) register.
When the device is going into Standby or Power-Down sleep mode, the BOD will change the operation mode as
defined by SLEEP in BOD.CTRLA. When the device is waking up from Standby or Power-Down sleep mode, the
BOD will operate in the mode defined by the ACTIVE bit field in the Control A (BOD.CTRLA) register.
20.3.4
Configuration Change Protection
This peripheral has registers that are under Configuration Change Protection (CCP). To write to these registers, a
certain key must first be written to the CPU.CCP register, followed by a write access to the protected bits within four
CPU instructions.
Attempting to write to a protected register without following the appropriate CCP unlock sequence leaves the
protected register unchanged.
The following registers are under CCP:
Table 20-2. Registers Under Configuration Change Protection
Register
The SLEEP and SAMPFREQ bits in the BOD.CTRLA register
© 2020 Microchip Technology Inc.
Preliminary Datasheet
Key
IOREG
DS40002247A-page 208
AVR128DB28/32/48/64
BOD - Brown-out Detector
20.4
Register Summary
Offset
Name
Bit Pos.
0x00
0x01
0x02
...
0x07
0x08
0x09
0x0A
0x0B
CTRLA
CTRLB
7:0
7:0
20.5
7
6
5
4
SAMPFREQ
3
2
ACTIVE[1:0]
1
0
SLEEP[1:0]
LVL[2:0]
Reserved
VLMCTRL
INTCTRL
INTFLAGS
STATUS
7:0
7:0
7:0
7:0
VLMCFG[1:0]
VLMLVL[1:0]
VLMIE
VLMIF
VLMS
Register Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 209
AVR128DB28/32/48/64
BOD - Brown-out Detector
20.5.1
Control A
Name:
Offset:
Reset:
Property:
Bit
7
CTRLA
0x00
0x00
Configuration Change Protection
6
Access
Reset
5
4
SAMPFREQ
R
0
3
2
1
ACTIVE[1:0]
R
0
0
SLEEP[1:0]
R
0
R/W
0
R/W
0
Bit 4 – SAMPFREQ Sample Frequency
This bit controls the BOD sample frequency.
The Reset value is loaded from the SAMPFREQ bit in FUSE.BODCFG.
This bit is not under Configuration Change Protection (CCP).
Value
Description
0x0
Sample frequency is 128 Hz
0x1
Sample frequency is 32 Hz
Bits 3:2 – ACTIVE[1:0] Active
These bits select the BOD operation mode when the device is in Active mode or Idle sleep mode.
The Reset value is loaded from the ACTIVE bits in FUSE.BODCFG.
These bits are not under Configuration Change Protection (CCP).
Value
Name
Description
0x0
DIS
Disabled
0x1
ENABLED Enabled in Continuous mode
0x2
SAMPLE
Enabled in Sampled mode
0x3
ENWAKE
Enabled in Continuous mode. Execution is halted at wake-up until BOD is running.
Bits 1:0 – SLEEP[1:0] Sleep
These bits select the BOD operation mode when the device is in Standby or Power-Down sleep mode. The Reset
value is loaded from the SLEEP bits in FUSE.BODCFG.
Value
Name
Description
0x0
DIS
Disabled
0x1
ENABLED
Enabled in Continuous mode
0x2
SAMPLED
Enabled in Sampled mode
0x3
Reserved
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 210
AVR128DB28/32/48/64
BOD - Brown-out Detector
20.5.2
Control B
Name:
Offset:
Reset:
Property:
CTRLB
0x01
Loaded from fuse
-
Bit
7
6
5
4
3
2
Access
Reset
R
0
R
0
R
0
R
0
R
0
R
x
1
LVL[2:0]
R
x
0
R
x
Bits 2:0 – LVL[2:0] BOD Level
This bit field controls the BOD threshold level.
The Reset value is loaded from the BOD Level (LVL) bits in the BOD Configuration Fuse (FUSE.BODCFG).
Value
0x0
0x1
0x2
0x3
Other
Name
BODLEVEL0
BODLEVEL1
BODLEVEL2
BODLEVEL3
—
Typical Values
1.9V
2.45V
2.7V
2.85V
Reserved
Note: Refer to the Reset, WDT, Oscillator, Start-up Timer, Power-up Timer, Brown-out Detector Specifications
section for BOD level characterization.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 211
AVR128DB28/32/48/64
BOD - Brown-out Detector
20.5.3
VLM Control
Name:
Offset:
Reset:
Property:
Bit
7
VLMCTRL
0x08
0x00
-
6
5
4
3
2
Access
Reset
1
0
VLMLVL[1:0]
R/W
R/W
0
0
Bits 1:0 – VLMLVL[1:0] VLM Level
These bits select the VLM threshold relative to the BOD threshold (LVL in BOD.CTRLB).
Value
Name
Description
0x00
OFF
VLM disabled
0x01
5ABOVE
VLM threshold 5% above the BOD threshold
0x02
15ABOVE
VLM threshold 15% above the BOD threshold
0x03
25ABOVE
VLM threshold 25% above the BOD threshold
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 212
AVR128DB28/32/48/64
BOD - Brown-out Detector
20.5.4
Interrupt Control
Name:
Offset:
Reset:
Property:
Bit
7
INTCTRL
0x09
0x00
-
6
5
4
3
Access
Reset
2
1
VLMCFG[1:0]
R/W
R/W
0
0
0
VLMIE
R/W
0
Bits 2:1 – VLMCFG[1:0] VLM Configuration
These bits select which incidents will trigger a VLM interrupt.
Value
Name
Description
0x0
FALLING
VDD falls below VLM threshold
0x1
RISING
VDD rises above VLM threshold
0x2
BOTH
VDD crosses VLM threshold
Other
Reserved
Bit 0 – VLMIE VLM Interrupt Enable
Writing a ‘1’ to this bit enables the VLM interrupt.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 213
AVR128DB28/32/48/64
BOD - Brown-out Detector
20.5.5
VLM Interrupt Flags
Name:
Offset:
Reset:
Property:
Bit
7
INTFLAGS
0x0A
0x00
-
6
5
4
3
Access
Reset
2
1
0
VLMIF
R/W
0
Bit 0 – VLMIF VLM Interrupt Flag
This flag is set when a trigger from the VLM is given, as configured by the VLMCFG bit in the BOD.INTCTRL register.
The flag is only updated when the BOD is enabled.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 214
AVR128DB28/32/48/64
BOD - Brown-out Detector
20.5.6
VLM Status
Name:
Offset:
Reset:
Property:
Bit
7
STATUS
0x0B
0x00
-
6
5
4
3
Access
Reset
2
1
0
VLMS
R/W
0
Bit 0 – VLMS VLM Status
This bit is only valid when the BOD is enabled.
Value
Name
Description
0
ABOVE
The voltage is above the VLM threshold level
1
BELOW
The voltage is below the VLM threshold level
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 215
AVR128DB28/32/48/64
VREF - Voltage Reference
21.
VREF - Voltage Reference
21.1
Features
•
•
21.2
Programmable Voltage Reference Sources:
– One reference for Analog to Digital Converter 0 (ADC0)
– One reference for Digital to Analog Converter 0 (DAC0)
– One reference shared between all Analog Comparators (ACs)
Each Reference Source Supports the Following Voltages:
– 1.024V
– 2.048V
– 4.096V
– 2.500V
– VDD
– VREFA
Overview
The Voltage Reference (VREF) peripheral provides control registers for the voltage reference sources used by
several peripherals. The user can select the reference voltages for the ADC0, DAC0 and ACs by writing to the
appropriate registers in the VREF peripheral.
A voltage reference source is enabled automatically when requested by a peripheral. The user can enable the
reference voltage sources, and thus, override the automatic disabling of unused sources by writing to the respective
ALWAYSON bit in VREF.ADC0REF, VREF.DAC0REF and VREF.ACREF. This will decrease the start-up time at the
cost of increased power consumption.
21.2.1
Block Diagram
Figure 21-1. VREF Block Diagram
Reference reque st
ALWAYSON
REFSEL[2:0]
Bandgap
Reference
Gen erator
Ban dgap
ena ble
21.3
Functional Description
21.3.1
Initialization
1.024V
2.048V
4.096V
2.500V
VDD
VREFA
BUF
Inte rnal
Reference
The default configuration will enable the respective source when the ADC0, DAC0, or any of the ACs are requesting
a reference voltage. The default reference voltage is 1.024V but can be configured by writing to the respective
Reference Select (REFSEL) bit field in the ADC0 Reference (ADC0REF), DAC0 Reference (DAC0REF) or Analog
Comparators (ACREF) registers.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 216
AVR128DB28/32/48/64
VREF - Voltage Reference
21.4
Register Summary
Offset
Name
Bit Pos.
7
0x00
0x01
0x02
0x03
0x04
ADC0REF
Reserved
DAC0REF
Reserved
ACREF
7:0
ALWAYSON
REFSEL[2:0]
7:0
ALWAYSON
REFSEL[2:0]
7:0
ALWAYSON
REFSEL[2:0]
21.5
6
5
4
3
2
1
0
Register Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 217
AVR128DB28/32/48/64
VREF - Voltage Reference
21.5.1
ADC0 Reference
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
ALWAYSON
R/W
0
ADC0REF
0x00
0x00
-
6
5
4
3
2
R/W
0
1
REFSEL[2:0]
R/W
0
0
R/W
0
Bit 7 – ALWAYSON Reference Always On
This bit controls whether the ADC0 reference is always on or not.
Value
Description
0
The reference is automatically enabled when needed
1
The reference is always on
Bits 2:0 – REFSEL[2:0] Reference Select
This bit field controls the reference voltage level for ADC0.
Value
Name
Description
0x0
1V024
Internal 1.024V reference(1)
0x1
2V048
Internal 2.048V reference(1)
0x2
4V096
Internal 4.096V reference(1)
0x3
2V500
Internal 2.500V reference(1)
0x4
Reserved
0x5
VDD
VDD as reference
0x6
VREFA
External reference from the VREFA pin
0x7
Reserved
Note:
1. The values given for internal references are only typical. Refer to the Electrical Characteristics section for
further details.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 218
AVR128DB28/32/48/64
VREF - Voltage Reference
21.5.2
DAC0 Reference
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
ALWAYSON
R/W
0
DAC0REF
0x02
0x00
-
6
5
4
3
2
R/W
0
1
REFSEL[2:0]
R/W
0
0
R/W
0
Bit 7 – ALWAYSON Reference Always On
This bit controls whether the DAC0 reference is always on or not.
Value
Description
0
The reference is automatically enabled when needed
1
The reference is always on
Bits 2:0 – REFSEL[2:0] Reference Select
This bit field controls the reference voltage level for DAC0.
Value
Name
Description
0x0
1V024
Internal 1.024V reference(1)
0x1
2V048
Internal 2.048V reference(1)
0x2
4V096
Internal 4.096V reference(1)
0x3
2V500
Internal 2.500V reference(1)
0x4
Reserved
0x5
VDD
VDD as reference
0x6
VREFA
External reference from the VREFA pin
0x7
Reserved
Note:
1. The values given for internal references are only typical. Refer to the Electrical Characteristics section for
further details.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 219
AVR128DB28/32/48/64
VREF - Voltage Reference
21.5.3
Analog Comparator Reference
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
ALWAYSON
R/W
0
ACREF
0x04
0x00
-
6
5
4
3
2
R/W
0
1
REFSEL[2:0]
R/W
0
0
R/W
0
Bit 7 – ALWAYSON Reference Always On
This bit controls whether the ACs reference is always on or not.
Value
Description
0
The reference is automatically enabled when needed
1
The reference is always on
Bits 2:0 – REFSEL[2:0] Reference Select
This bit field controls the reference voltage level for ACs.
Value
Name
Description
0x0
1V024
Internal 1.024V reference(1)
0x1
2V048
Internal 2.048V reference(1)
0x2
4V096
Internal 4.096V reference(1)
0x3
2V500
Internal 2.500V reference(1)
0x4
Reserved
0x5
VDD
VDD as reference
0x6
VREFA
External reference from the VREFA pin
0x7
Reserved
Note:
1. The values given for internal references are only typical. Refer to the Electrical Characteristics section for
further details.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 220
AVR128DB28/32/48/64
WDT - Watchdog Timer
22.
WDT - Watchdog Timer
22.1
Features
•
•
•
•
•
•
22.2
Issues a System Reset if the Watchdog Timer is not Cleared Before its Time-out Period
Operates Asynchronously from the Peripheral Clock Using an Independent Oscillator
Uses the 1.024 kHz Output of the 32.768 kHz Ultra Low-Power Oscillator (OSC32K)
11 Selectable Time-out Periods, from 8 ms to 8s
Two Operation Modes:
– Normal mode
– Window mode
Configuration Lock to Prevent Unwanted Changes
Overview
The Watchdog Timer (WDT) is a system function for monitoring the correct program operation. When enabled, the
WDT is a constantly running timer with a configurable time-out period. If the WDT is not reset within the time-out
period, it will issue a system Reset. This allows the system to recover from situations such as runaway or deadlocked
code. The WDT is reset by executing the WDR (Watchdog Timer Reset) instruction from software.
In addition to the Normal mode as described above, the WDT has a Window mode. The Window mode defines a time
slot or “window” inside the time-out period during which the WDT must be reset. If the WDT is reset outside this
window, either too early or too late, a system Reset will be issued. Compared to the Normal mode, the Window mode
can catch situations where a code error causes constant WDR execution.
When enabled, the WDT will run in Active mode and all sleep modes. Since it is asynchronous (that is running from a
CPU independent clock source), it will continue to operate and be able to issue a system Reset, even if the main
clock fails.
The WDT has a Configuration Change Protection (CCP) mechanism and a lock functionality, ensuring the WDT
settings cannot be changed by accident.
22.2.1
Block Diagram
Figure 22-1. WDT Block Diagram
WINDOW
CTRLA
PERIOD
CLK_WDT
WDR
(Instruction)
COUNT
22.3
Functional Description
22.3.1
Initialization
1.
+
>
Closed
window
=
System
Reset
Time-out
The WDT is enabled when a non-zero value is written to the Period (PERIOD) bit field in the Control A
(WDT.CTRLA) register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 221
AVR128DB28/32/48/64
WDT - Watchdog Timer
2.
Optional: Write a non-zero value to the Window (WINDOW) bit field in WDT.CTRLA to enable the Window
mode operation.
All bits in the Control A register and the Lock (LOCK) bit in the Status (WDT.STATUS) register are write-protected by
the Configuration Change Protection (CCP) mechanism.
A fuse (FUSE.WDTCFG) defines the Reset value of the WDT.CTRLA register. If the value of the PERIOD bit field in
the FUSE.WDTCFG fuse is different than zero, the WDT is enabled and the LOCK bit in the WDT.STATUS register is
set at boot time.
22.3.2
Clocks
A 1.024 kHz clock (CLK_WDT) is sourced from the internal Ultra Low-Power Oscillator, OSC32K. Due to the ultra
low-power design, the oscillator is less accurate than other oscillators featured in the device, and hence the exact
time-out period may vary from device to device. This variation must be taken into consideration when designing
software that uses the WDT, to ensure that the time-out periods used are valid for all devices. Refer to the Electrical
Characteristics section for more specific information.
The WDT clock (CLK_WDT) is asynchronous to the peripheral clock. Due to this asynchronicity, writing to the WDT
Control A (WDT.CTRLA) register will require synchronization between the clock domains. Refer to 22.3.6
Synchronization for further details.
22.3.3
Operation
22.3.3.1 Normal Mode
In the Normal mode operation, a single time-out period is set for the WDT. If the WDT is not reset from software using
the WDR instruction during the defined time-out period, the WDT will issue a system Reset.
A new WDT time-out period starts each time the WDT is reset by software using the WDR instruction.
There are 11 possible WDT time-out periods (TOWDT), selectable from 8 ms to 8s by writing to the Period (PERIOD)
bit field in the Control A (WDT.CTRLA) register.
The figure below shows a typical timing scheme for the WDT operating in Normal mode.
Figure 22-2. Normal Mode Operation
WDT Count
Timely WDT Reset (WDR)
WDT Time-out
System Reset
Here:
TO WDT = 15.625 ms
5
10
15
20
25
30
TOWDT
35
t [ms]
The Normal mode is enabled as long as the Window (WINDOW) bit field in the WDT.CTRLA register is ‘0x0’.
22.3.3.2 Window Mode
In the Window mode operation, the WDT uses two different time-out periods:
• The closed window time-out period (TOWDTW) defines a duration, from 8 ms to 8s, where the WDT cannot be
reset. If the WDT is reset during this period, the WDT will issue a system Reset.
• The open window time-out period (TOWDT), which is also 8 ms to 8s, defines the duration of the open period
during which the WDT can (and needs to) be reset. The open period will always follow the closed period, so the
total duration of the time-out period is the sum of the closed window and the open window time-out periods.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 222
AVR128DB28/32/48/64
WDT - Watchdog Timer
When enabling the Window mode or when going out of the Debug mode, the window is activated after the first WDR
instruction.
The figure below shows a typical timing scheme for the WDT operating in Window mode.
Figure 22-3. Window Mode Operation
WDT Count
Open
Timely WDT Reset (WDR)
Closed
WDR too early:
System Reset
Here:
TOWDTW =TOWDT = 8 ms
5
10
15
20
TOWDTW
25
30
TOWDT
35
t [ms]
The Window mode is enabled by writing a non-zero value to the Window (WINDOW) bit field in the Control A
(WDT.CTRLA) register. The Window mode is disabled by writing the WINDOW bit field to ‘0x0’.
22.3.3.3 Preventing Unintentional Changes
The WDT provides two security mechanisms to avoid unintentional changes to the WDT settings:
•
•
The CCP mechanism, employing a timed write procedure for changing the WDT control registers. Refer to
22.3.7 Configuration Change Protection for further details.
Locking the configuration by writing a ‘1’ to the Lock (LOCK) bit in the Status (WDT.STATUS) register. When this
bit is ‘1’, the Control A (WDT.CTRLA) register cannot be changed. The LOCK bit can only be written to ‘1’ in
software, while the device needs to be in Debug mode to be able to write it to ‘0’. Consequently, the WDT
cannot be disabled from software.
Note: The WDT configuration is loaded from fuses after Reset. If the PERIOD bit field is set to a non-zero value, the
LOCK bit is automatically set in WDT.STATUS.
22.3.4
Sleep Mode Operation
The WDT will continue to operate in any sleep mode where the source clock is active.
22.3.5
Debug Operation
When run-time debugging, this peripheral will continue normal operation. Halting the CPU in Debugging mode will
halt the normal operation of the peripheral.
When halting the CPU in Debug mode, the WDT counter is reset.
When starting the CPU and when the WDT is operating in Window mode, the first closed window time-out period will
be disabled, and a Normal mode time-out period is executed.
22.3.6
Synchronization
The Control A (WDT.CTRLA) register is synchronized when written, due to the asynchronicity between the WDT
clock domain and the peripheral clock domain. The Synchronization Busy (SYNCBUSY) flag in the STATUS
(WDT.STATUS) register indicates if there is an ongoing synchronization.
Writing to WDT.CTRLA while SYNCBUSY = 1 is not allowed.
The following bit fields are synchronized when written:
• The Period (PERIOD) bit field in Control A (WDT.CTRLA) register
• The Window (WINDOW) bit field in Control A (WDT.CTRLA) register
The WDR instruction will need two to three cycles of the WDT clock to be synchronized.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 223
AVR128DB28/32/48/64
WDT - Watchdog Timer
22.3.7
Configuration Change Protection
This peripheral has registers that are under Configuration Change Protection (CCP). To write to these registers, a
certain key must first be written to the CPU.CCP register, followed by a write access to the protected bits within four
CPU instructions.
Attempting to write to a protected register without following the appropriate CCP unlock sequence leaves the
protected register unchanged.
The following registers are under CCP:
Table 22-1. WDT - Registers Under Configuration Change Protection
Register
Key
WDT.CTRLA
IOREG
LOCK bit in WDT.STATUS
IOREG
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 224
AVR128DB28/32/48/64
WDT - Watchdog Timer
22.4
Register Summary
Offset
Name
Bit Pos.
7
0x00
0x01
CTRLA
STATUS
7:0
7:0
LOCK
22.5
6
5
4
WINDOW[3:0]
3
2
1
0
PERIOD[3:0]
SYNCBUSY
Register Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 225
AVR128DB28/32/48/64
WDT - Watchdog Timer
22.5.1
Control A
Name:
Offset:
Reset:
Property:
Bit
7
Access
Reset
R/W
x
CTRLA
0x00
From FUSE.WDTCFG
Configuration Change Protection
6
5
WINDOW[3:0]
R/W
R/W
x
x
4
3
R/W
x
R/W
x
2
1
PERIOD[3:0]
R/W
R/W
x
x
0
R/W
x
Bits 7:4 – WINDOW[3:0] Window
Writing a non-zero value to these bits enables the Window mode and selects the duration of the closed period
accordingly.
The bits are optionally lock-protected:
• If the LOCK bit in WDT.STATUS is ‘1’, all bits are change-protected (Access = R)
•
If the LOCK bit in WDT.STATUS is ‘0’, all bits can be changed (Access = R/W)
Value
0x0
0x1
0x2
0x3
0x4
0x5
0x6
0x7
0x8
0x9
0xA
0xB
Other
Name
OFF
8CLK
16CLK
32CLK
64CLK
128CLK
256CLK
512CLK
1KCLK
2KCLK
4KCLK
8KCLK
-
Description
7.8125 ms
15.625 ms
31.25 ms
62.5 ms
0.125s
0.250s
0.500s
1.0s
2.0s
4.0s
8.0s
Reserved
Note: Refer to the Electrical Characteristics section for specific information regarding the accuracy of the 32.768
kHz Ultra Low-Power Oscillator (OSC32K).
Bits 3:0 – PERIOD[3:0] Period
Writing a non-zero value to this bit enables the WDT and selects the time-out period in the Normal mode accordingly.
In the Window mode, these bits select the duration of the open window.
The bits are optionally lock-protected:
• If the LOCK bit in WDT.STATUS is ‘1’, all bits are change-protected (Access = R)
•
If the LOCK bit in WDT.STATUS is ‘0’, all bits can be changed (Access = R/W)
Value
0x0
0x1
0x2
0x3
0x4
0x5
0x6
0x7
0x8
0x9
0xA
Name
OFF
8CLK
16CLK
32CLK
64CLK
128CLK
256CLK
512CLK
1KCLK
2KCLK
4KCLK
© 2020 Microchip Technology Inc.
Description
7.8125 ms
15.625 ms
31.25 ms
62.5 ms
0.125s
0.250s
0.500s
1.0s
2.0s
4.0s
Preliminary Datasheet
DS40002247A-page 226
AVR128DB28/32/48/64
WDT - Watchdog Timer
Value
0xB
Other
Name
8KCLK
-
Description
8.0s
Reserved
Note: Refer to the Electrical Characteristics section for specific information regarding the accuracy of the 32.768
kHz Ultra Low-Power Oscillator (OSC32K).
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 227
AVR128DB28/32/48/64
WDT - Watchdog Timer
22.5.2
Status
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
LOCK
R/W
0
STATUS
0x01
0x00
Configuration Change Protection
6
5
4
3
2
1
0
SYNCBUSY
R
0
Bit 7 – LOCK Lock
Writing this bit to ‘1’ write-protects the WDT.CTRLA register.
It is only possible to write this bit to ‘1’. This bit can be cleared in Debug mode only.
If the PERIOD bits in WDT.CTRLA are different from zero after boot code, the lock will automatically be set.
This bit is under CCP.
Bit 0 – SYNCBUSY Synchronization Busy
This bit is set after writing to the WDT.CTRLA register, while the data is being synchronized from the peripheral clock
domain to the WDT clock domain.
This bit is cleared after the synchronization is finished.
This bit is not under CCP.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 228
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.
TCA - 16-bit Timer/Counter Type A
23.1
Features
•
•
•
•
•
•
•
•
•
23.2
16-Bit Timer/Counter
Three Compare Channels
Double-Buffered Timer Period Setting
Double-Buffered Compare Channels
Waveform Generation:
– Frequency generation
– Single-slope PWM (Pulse-Width Modulation)
– Dual-slope PWM
Count on Event
Timer Overflow Interrupts/Events
One Compare Match per Compare Channel
Two 8-Bit Timer/Counters in Split Mode
Overview
The flexible 16-bit PWM Timer/Counter type A (TCA) provides accurate program execution timing, frequency and
waveform generation, and command execution.
A TCA consists of a base counter and a set of compare channels. The base counter can be used to count clock
cycles or events, or let events control how it counts clock cycles. It has direction control and period setting that can
be used for timing. The compare channels can be used together with the base counter to do compare match control,
frequency generation, and pulse-width waveform modulation.
Depending on the mode of operation, the counter is cleared, reloaded, incremented, or decremented at each timer/
counter clock or event input.
A timer/counter can be clocked and timed from the peripheral clock, with optional prescaling, or from the Event
System. The Event System can also be used for direction control or to synchronize operations.
By default, the TCA is a 16-bit timer/counter. The timer/counter has a Split mode feature that splits it into two 8-bit
timer/counters with three compare channels each.
A block diagram of the 16-bit timer/counter with closely related peripheral modules (in grey) is shown in the figure
below.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 229
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
Figure 23-1. 16-bit Timer/Counter and Closely Related Peripherals
Timer/Counter
Base Counter
Counter
Control Logic
Compare Channel 0
Compare Channel 1
Compare Channel 2
Comparator
Buffer
23.2.1
Waveform
Generation
CLK_PER
Event
System
PORTS
Timer Period
Prescaler
Block Diagram
The figure below shows a detailed block diagram of the timer/counter.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 230
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
Figure 23-2. Timer/Counter Block Diagram
Base Counter
Clock Select
CTRLA
PERBUF
Mode
CTRLB
PER
EVCTRL
Event
Action
‘‘count’’
‘‘clear’’
‘‘load’’
‘‘direction’’
Counter
CNT
=
=0
OVF
(INT Req. and Event)
Control Logic
Event
TOP
UPDATE
BV
BOTTOM
Compare Unit n
BV
CMPnBUF
Control Logic
CMPn
=
Waveform
Generation
‘‘match’’
WOn Out
CMPn
(INT Req. and Event)
The Counter (TCAn.CNT) register, Period and Compare (TCAn.PER and TCAn.CMPn) registers, and their
corresponding buffer registers (TCAn.PERBUF and TCAn.CMPBUFn) are 16-bit registers. All buffer registers have a
Buffer Valid (BV) flag that indicates when the buffer contains a new value.
During normal operation, the counter value is continuously compared to zero and the period (PER) value to
determine whether the counter has reached TOP or BOTTOM. The counter value can also be compared to the
TCAn.CMPn registers.
The timer/counter can generate interrupt requests, events, or change the waveform output after being triggered by
the Counter (TCAn.CNT) register reaching TOP, BOTTOM, or CMPn. The interrupt requests, events, or waveform
output changes will occur on the next CLK_TCA cycle after the triggering.
CLK_TCA is either the prescaled peripheral clock or events from the Event System, as shown in the figure below.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 231
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
Figure 23-3. Timer/Counter Clock Logic
CLK_PER
Prescaler
Event System
Event
CLKSEL
EVACT
(Encoding)
CLK_TCA
CNT
CNTxEI
23.2.2
Signal Description
Signal
Description
Type
WOn
Digital output
Waveform output
23.3
Functional Description
23.3.1
Definitions
The following definitions are used throughout the documentation:
Table 23-1. Timer/Counter Definitions
Name
Description
BOTTOM The counter reaches BOTTOM when it becomes 0x0000.
MAX
The counter reaches MAXimum when it becomes all ones.
TOP
The counter reaches TOP when it becomes equal to the highest value in the count sequence.
The update condition is met when the timer/counter reaches BOTTOM or TOP, depending on the
UPDATE Waveform Generator mode. Buffered registers with valid buffer values will be updated unless the Lock
Update (LUPD) bit in the TCAn.CTRLE register has been set.
CNT
Counter register value.
CMP
Compare register value.
PER
Period register value.
In general, the term timer is used when the timer/counter is counting periodic clock ticks. The term counter is used
when the input signal has sporadic or irregular ticks. The latter can be the case when counting events.
23.3.2
Initialization
To start using the timer/counter in a basic mode, follow these steps:
1. Write a TOP value to the Period (TCAn.PER) register.
2. Enable the peripheral by writing a ‘1’ to the Enable (ENABLE) bit in the Control A (TCAn.CTRLA) register.
The counter will start counting clock ticks according to the prescaler setting in the Clock Select (CLKSEL) bit
field in TCAn.CTRLA.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 232
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
3.
4.
23.3.3
Optional: By writing a ‘1’ to the Enable Counter Event Input A (CNTAEI) bit in the Event Control
(TCAn.EVCTRL) register, events are counted instead of clock ticks.
The counter value can be read from the Counter (CNT) bit field in the Counter (TCAn.CNT) register.
Operation
23.3.3.1 Normal Operation
In normal operation, the counter is counting clock ticks in the direction selected by the Direction (DIR) bit in the
Control E (TCAn.CTRLE) register, until it reaches TOP or BOTTOM. The clock ticks are given by the peripheral clock
(CLK_PER), prescaled according to the Clock Select (CLKSEL) bit field in the Control A (TCAn.CTRLA) register.
When TOP is reached while the counter is counting up, the counter will wrap to ‘0’ at the next clock tick. When
counting down, the counter is reloaded with the Period (TCAn.PER) register value when BOTTOM is reached.
Figure 23-4. Normal Operation
CNT written
MAX
‘‘update’’
CNT
TOP
BOTTOM
DIR
It is possible to change the counter value in the Counter (TCAn.CNT) register when the counter is running. The write
access to TCAn.CNT has higher priority than count, clear or reload, and will be immediate. The direction of the
counter can also be changed during normal operation by writing to DIR in TCAn.CTRLE.
23.3.3.2 Double Buffering
The Period (TCAn.PER) register value and the Compare n (TCAn.CMPn) register values are all double-buffered
(TCAn.PERBUF and TCAn.CMPnBUF).
Each buffer register has a Buffer Valid (BV) flag (PERBV, CMPnBV) in the Control F (TCAn.CTRLF) register, which
indicates that the buffer register contains a valid (new) value that can be copied into the corresponding Period or
Compare register. When the Period register and Compare n registers are used for a compare operation, the BV flag
is set when data are written to the buffer register and cleared on an UPDATE condition. This is shown for a Compare
(CMPn) register in the figure below.
Figure 23-5. Period and Compare Double Buffering
‘‘write enable’’
BV
UPDATE
EN
EN
‘‘data write’’
CMPnBUF
CMPn
CNT
=
‘‘match’’
Both the TCAn.CMPn and TCAn.CMPnBUF registers are available as I/O registers. This allows initialization and
bypassing of the buffer register and the double-buffering function.
23.3.3.3 Changing the Period
The Counter period is changed by writing a new TOP value to the Period (TCAn.PER) register.
No Buffering: If double-buffering is not used, any period update is immediate.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 233
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
Figure 23-6. Changing the Period Without Buffering
Counter wraparound
MAX
‘‘update’’
‘‘write’’
CNT
BOTTOM
New TOP written to
New TOP written to
PER that is higher
PER that is lower
than current CNT.
than current CNT.
A counter wraparound can occur in any mode of operation when counting up without buffering, as the TCAn.CNT and
TCAn.PER registers are continuously compared. If a new TOP value is written to TCAn.PER that is lower than the
current TCAn.CNT, the counter will wrap first, before a compare match occurs.
Figure 23-7. Unbuffered Dual-Slope Operation
Counter wraparound
MAX
‘‘update’’
‘‘write’’
CNT
BOTTOM
New TOP written to
PER that is higher
than current CNT.
New TOP written to
PER that is lower
than current CNT.
With Buffering: When double-buffering is used, the buffer can be written at any time and still maintain correct
operation. The TCAn.PER is always updated on the UPDATE condition, as shown for dual-slope operation in the
figure below. This prevents wraparound and the generation of odd waveforms.
Figure 23-8. Changing the Period Using Buffering
MAX
‘‘update’’
‘‘write’’
CNT
BOTTOM
New Period written to
PERB that is higher
than current CNT.
New Period written to
PERB that is lower
than current CNT.
New PER is updated
with PERB value.
Note: Buffering is used in figures illustrating TCA operation if not otherwise specified.
23.3.3.4 Compare Channel
Each Compare Channel n continuously compares the counter value (TCAn.CNT) with the Compare n (TCAn.CMPn)
register. If TCAn.CNT equals TCAn.CMPn, the Comparator n signals a match. The match will set the Compare
Channel’s interrupt flag at the next timer clock cycle, and the optional interrupt is generated.
The Compare n Buffer (TCAn.CMPnBUF) register provides double-buffer capability equivalent to that for the period
buffer. The double-buffering synchronizes the update of the TCAn.CMPn register with the buffer value to either the
TOP or BOTTOM of the counting sequence, according to the UPDATE condition. The synchronization prevents the
occurrence of odd-length, non-symmetrical pulses for glitch-free output.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 234
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
The value in CMPnBUF is moved to CMPn at the UPDATE condition and is compared to the counter value
(TCAn.CNT) from the next count.
23.3.3.4.1 Waveform Generation
The compare channels can be used for waveform generation on the corresponding port pins. The following
requirements must be met to make the waveform visible on the connected port pin:
1.
2.
3.
4.
A Waveform Generation mode must be selected by writing the Waveform Generation Mode (WGMODE) bit
field in the TCAn.CTRLB register.
The compare channels used must be enabled (CMPnEN = 1 in TCAn.CTRLB). This will override the output
value for the corresponding pin. An alternative pin can be selected by configuring the Port Multiplexer
(PORTMUX). Refer to the PORTMUX section for details.
The direction for the associated port pin n must be configured in the Port peripheral as an output.
Optional: Enable the inverted waveform output for the associated port pin n. Refer to the PORT section for
details.
23.3.3.4.2 Frequency (FRQ) Waveform Generation
For frequency generation, the period time (T) is controlled by the TCAn.CMP0 register instead of the Period
(TCAn.PER) register. The corresponding waveform generator output is toggled on each compare match between the
TCAn.CNT and TCAn.CMPn registers.
Figure 23-9. Frequency Waveform Generation
Period (T)
Direction change
CNT written
MAX
‘‘update’’
TOP
CNT
BOTTOM
WG Output
The waveform frequency (fFRQ) is defined by the following equation:
�FRQ =
f CLK_PER
2� CMPn+1
where N represents the prescaler divider used (see CLKSEL bit field in the TCAn.CTRLA register), and fCLK_PER is
the peripheral clock frequency.
The maximum frequency of the waveform generated is half of the peripheral clock frequency (fCLK_PER/2) when
TCAn.CMP0 is written to 0x0000 and no prescaling is used (N = 1, CLKSEL = 0x0 in TCAn.CTRLA).
23.3.3.4.3 Single-Slope PWM Generation
For single-slope Pulse-Width Modulation (PWM) generation the period (T) is controlled by the TCAn.PER register,
while the values of the TCAn.CMPn registers control the duty cycles of the generated waveforms. The figure below
shows how the counter counts from BOTTOM to TOP and then restarts from BOTTOM. The waveform generator
output is set at BOTTOM and cleared on the compare match between the TCAn.CNT and TCAn.CMPn registers.
CMPn = BOTTOM will produce a static low signal on WOn while CMPn > TOP will produce a static high signal on
WOn.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 235
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
Figure 23-10. Single-Slope Pulse-Width Modulation
Period (T)
CMPn=BOTTOM
CMPn>TOP
MAX
TOP
‘‘update’’
‘‘match’’
CNT
CMPn
BOTTOM
Output WOn
Note: The representation in the figure above is valid for when CMPn is updated using CMPnBUF.
The TCAn.PER register defines the PWM resolution. The minimum resolution is 2 bits (TCAn.PER = 0x0002), and
the maximum resolution is 16 bits (TCAn.PER = MAX-1).
The following equation calculates the exact resolution in bits for single-slope PWM (RPWM_SS):
�PWM_SS =
log PER+2
log 2
�PWM_SS =
�CLK_PER
� PER+1
The single-slope PWM frequency (fPWM_SS) depends on the period setting (TCAn.PER), the peripheral clock
frequency fCLK_PER and the TCA prescaler (the CLKSEL bit field in the TCAn.CTRLA register). It is calculated by the
following equation where N represents the prescaler divider used:
23.3.3.4.4 Dual-Slope PWM
For dual-slope PWM generation, the period (T) is controlled by TCAn.PER, while the values of TCAn.CMPn control
the duty cycle of the WG output.
The figure below shows how, for dual-slope PWM, the counter counts repeatedly from BOTTOM to TOP and then
from TOP to BOTTOM. The waveform generator output is set at BOTTOM, cleared on compare match when upcounting and set on compare match when down-counting.
CMPn = BOTTOM will produce a static low signal on WOn, while CMPn = TOP will produce a static high signal on
WOn.
Figure 23-11. Dual-Slope Pulse-Width Modulation
Period (T)
CMPn=BOTTOM
‘‘update’’
‘‘match’’
CMPn=TOP
MAX
CMPn
CNT
TOP
BOTTOM
Waveform Output WOn
Note: The representation in the figure above is valid for when CMPn is updated using CMPnBUF.
Using dual-slope PWM results in half the maximum operation frequency compared to single-slope PWM operation,
due to twice the number of timer increments per period.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 236
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
The Period (TCAn.PER) register defines the PWM resolution. The minimum resolution is 2 bits (TCAn.PER =
0x0003), and the maximum resolution is 16 bits (TCAn.PER = MAX).
The following equation calculates the exact resolution in bits for dual-slope PWM (RPWM_DS):
�PWM_DS =
log PER+1
log 2
�PWM_DS =
�CLK_PER
2� ⋅ PER
The PWM frequency depends on the period setting in the TCAn.PER register, the peripheral clock frequency
(fCLK_PER) and the prescaler divider selected in the CLKSEL bit field in the TCAn.CTRLA register. It is calculated by
the following equation:
N represents the prescaler divider used.
23.3.3.4.5 Port Override for Waveform Generation
To make the waveform generation available on the port pins, the corresponding port pin direction must be set as
output (PORTx.DIR[n] = 1). The TCA will override the port pin values when the compare channel is enabled
(CMPnEN = 1 in TCAn.CTRLB) and a Waveform Generation mode is selected.
The figure below shows the port override for TCA. The timer/counter compare channel will override the port pin
output value (OUT) on the corresponding port pin. Enabling inverted I/O on the port pin (INVEN = 1 in PORT.PINn)
inverts the corresponding WG output.
Figure 23-12. Port Override for Timer/Counter Type A
OUT
WOn
Waveform
CMPnEN
INVEN
23.3.3.5 Timer/Counter Commands
A set of commands can be issued by software to immediately change the state of the peripheral. These commands
give direct control of the UPDATE, RESTART and RESET signals. A command is issued by writing the respective
value to the Command (CMD) bit field in the Control E (TCAn.CTRLESET) register.
An UPDATE command has the same effect as when an UPDATE condition occurs, except that the UPDATE
command is not affected by the state of the Lock Update (LUPD) bit in the Control E (TCAn.CTRLE) register.
The software can force a restart of the current waveform period by issuing a RESTART command. In this case, the
counter, direction, and all compare outputs are set to ‘0’.
A RESET command will set all timer/counter registers to their initial values. A RESET command can be issued only
when the timer/counter is not running (ENABLE = 0 in the TCAn.CTRLA register).
23.3.3.6 Split Mode - Two 8-Bit Timer/Counters
Split Mode Overview
To double the number of timers and PWM channels in the TCA, a Split mode is provided. In this Split mode, the 16-bit
timer/counter acts as two separate 8-bit timers, which each have three compare channels for PWM generation. The
Split mode will only work with single-slope down-count. Event controlled operation is not supported in Split mode.
Activating Split mode results in changes to the functionality of some registers and register bits. The modifications are
described in a separate register map (see 23.6 Register Summary - Split Mode).
Split Mode Differences Compared to Normal Mode
• Count:
– Down-count only
– Low Byte Timer Counter (TCAn.LCNT) register and High Byte Timer Counter (TCAn.HCNT) register are
independent
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 237
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
•
•
•
•
•
Waveform Generation:
– Single-slope PWM only (WGMODE = SINGLESLOPE in TCAn.CTRLB)
Interrupt:
– No change for Low Byte Timer Counter (TCAn.LCNT) register
– Underflow interrupt for High Byte Timer Counter (TCAn.HCNT) register
– No compare interrupt or flag for High Byte Compare n (TCAn.HCMPn) register
Event Actions: Not Compatible
Buffer Registers and Buffer Valid Flags: Unused
Register Access: Byte Access to All Registers
Block Diagram
Figure 23-13. Timer/Counter Block Diagram Split Mode
Base Counter
HPER
LPER
Clock Select
CTRLA
‘‘count high’’
‘‘load high’’
Counter
HCNT
‘‘count low’’
‘‘load low’’
LCNT
HUNF
Control Logic
(INT Req. and Event)
LUNF
(INT Req. and Event)
=0
BOTTOML
BOTTOMH
=0
Compare Unit n
LCMPn
=
Waveform
Generation
WOn Out
LCMPn
‘‘match’’
(INT Req. and Event)
Compare Unit n
HCMPn
=
Waveform
Generation
WO[n+3] Out
‘‘match’’
Split Mode Initialization
When shifting between Normal mode and Split mode, the functionality of some registers and bits changes, but their
values do not. For this reason, disabling the peripheral (ENABLE = 0 in TCAn.CTRLA) and doing a hard Reset (CMD
= RESET in TCAn.CTRLESET) is recommended when changing the mode to avoid unexpected behavior.
To start using the timer/counter in basic Split mode after a hard Reset, follow these steps:
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 238
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
1.
2.
3.
4.
23.3.4
Enable Split mode by writing a ‘1’ to the Split mode enable (SPLITM) bit in the Control D (TCAn.CTRLD)
register.
Write a TOP value to the Period (TCAn.PER) registers.
Enable the peripheral by writing a ‘1’ to the Enable (ENABLE) bit in the Control A (TCAn.CTRLA) register.
The counter will start counting clock ticks according to the prescaler setting in the Clock Select (CLKSEL) bit
field in the TCAn.CTRLA register.
The counter values can be read from the Counter bit field in the Counter (TCAn.CNT) registers.
Events
The TCA can generate the events described in the table below. All event generators except TCAn_HUNF are shared
between Normal mode and Split mode operation, and the generator name indicates what specific signal the
generator represents in each mode in the following way: OVF_LUNF corresponds to overflow in Normal mode and
Low byte timer underflow in Split mode. The same applies to CMPn_LCMPn.
Table 23-2. Event Generators in TCA
Generator Name
Peripheral
Event
Description
Event
Type
Generating
Clock Domain
Pulse
CLK_PER
One CLK_PER
period
Pulse
CLK_PER
One CLK_PER
period
Pulse
CLK_PER
One CLK_PER
period
Pulse
CLK_PER
One CLK_PER
period
Pulse
CLK_PER
One CLK_PER
period
Normal mode: Overflow
OVF_LUNF
Split mode: Low byte timer
underflow
Normal mode: Not available
HUNF
CMP0_LCMP0
TCAn
CMP1_LCMP1
CMP2_LCMP2
Split mode: High byte timer
underflow
Normal mode: Compare Channel 0
match
Split mode: Low byte timer
Compare Channel 0 match
Normal mode: Compare Channel 1
match
Split mode: Low byte timer
Compare Channel 1 match
Normal mode: Compare Channel 2
match
Split mode: Low byte timer
Compare Channel 2 match
Length of Event
The conditions for generating an event are identical to those that will raise the corresponding interrupt flag in the
TCAn.INTFLAGS register for both Normal mode and Split mode.
The TCA has two event users for detecting and acting upon input events. The table below describes the event users
and their associated functionality.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 239
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
Table 23-3. Event Users in TCA
User Name
Description
Peripheral Input
Input Detection Async/Sync
Count on a positive event edge
Edge
Sync
Count on any event edge
Edge
Sync
Level
Sync
The event level controls the count direction, up when low and
down when high
Level
Sync
The event level controls count direction, up when low and
down when high
Level
Sync
Edge
Sync
Restart counter on any event edge
Edge
Sync
Restart counter while the event signal is high
Level
Sync
CNTA Count while the event signal is high
TCAn
CNTB Restart counter on a positive event edge
The specific actions described in the table above are selected by writing to the Event Action (EVACTA, EVACTB) bits
in the Event Control (TCAn.EVCTRL) register. Input events are enabled by writing a ‘1’ to the Enable Counter Event
Input (CNTAEI and CNTBEI) bits in the TCAn.EVCTRL register.
If both EVACTA and EVACTB are configured to control the count direction, the event signals will be OR’ed to
determine the count direction. Both event inputs must then be low for the counter to count upwards.
Event inputs are not used in Split mode.
Refer to the Event System (EVSYS) section for more details regarding event types and Event System configuration.
23.3.5
Interrupts
Table 23-4. Available Interrupt Vectors and Sources in Normal Mode
Name
OVF
Vector Description
Overflow or underflow interrupt
Conditions
The counter has reached TOP or BOTTOM
CMP0 Compare Channel 0 interrupt
Match between the counter value and the Compare 0 register
CMP1 Compare Channel 1 interrupt
Match between the counter value and the Compare 1 register
CMP2 Compare Channel 2 interrupt
Match between the counter value and the Compare 2 register
Table 23-5. Available Interrupt Vectors and Sources in Split Mode
Name
Vector Description
Conditions
LUNF
Low-byte Underflow interrupt Low byte timer reaches BOTTOM
HUNF
High-byte Underflow interrupt High byte timer reaches BOTTOM
LCMP0 Compare Channel 0 interrupt
Match between the counter value and the low byte of the Compare 0
register
LCMP1 Compare Channel 1 interrupt
Match between the counter value and the low byte of the Compare 1
register
LCMP2 Compare Channel 2 interrupt
Match between the counter value and the low byte of the Compare 2
register
When an interrupt condition occurs, the corresponding interrupt flag is set in the peripheral’s Interrupt Flags
(peripheral.INTFLAGS) register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 240
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
An interrupt source is enabled or disabled by writing to the corresponding enable bit in the peripheral’s Interrupt
Control (peripheral.INTCTRL) register.
An interrupt request is generated when the corresponding interrupt source is enabled, and the interrupt flag is set.
The interrupt request remains active until the interrupt flag is cleared. See the peripheral’s INTFLAGS register for
details on how to clear interrupt flags.
23.3.6
Sleep Mode Operation
TCA is by default disabled in Standby Sleep mode. It will be halted as soon as the Sleep mode is entered.
The module can stay fully operational in Standby Sleep mode if the Run Standby (RUNSTDBY) bit in the
TCAn.CTRLA register is written to ‘1’.
All operation is halted in Power-Down Sleep mode.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 241
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.4
Register Summary - Normal Mode
Offset
Name
Bit Pos.
7
0x00
0x01
0x02
0x03
0x04
0x05
0x06
0x07
0x08
0x09
0x0A
0x0B
0x0C
...
0x0D
0x0E
0x0F
0x10
...
0x1F
CTRLA
CTRLB
CTRLC
CTRLD
CTRLECLR
CTRLESET
CTRLFCLR
CTRLFSET
Reserved
EVCTRL
INTCTRL
INTFLAGS
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
RUNSTDBY
0x20
CNT
0x22
...
0x25
Reserved
0x26
PER
0x28
CMP0
0x2A
CMP1
0x2C
CMP2
0x2E
...
0x35
Reserved
0x36
PERBUF
0x38
CMP0BUF
0x3A
CMP1BUF
0x3C
CMP2BUF
23.5
7:0
7:0
7:0
6
5
4
3
CMP2EN
CMP1EN
CMP0EN
ALUPD
2
1
CLKSEL[2:0]
CMP2OV
CMD[1:0]
CMD[1:0]
CMP2BV
CMP1BV
CMP2BV
CMP1BV
EVACTB[2:0]
CMP2
CMP2
CMP1
CMP1
CNTBEI
CMP0
CMP0
EVACTA[2:0]
0
ENABLE
WGMODE[2:0]
CMP1OV
LUPD
LUPD
CMP0BV
CMP0BV
CMP0OV
SPLITM
DIR
DIR
PERBV
PERBV
CNTAEI
OVF
OVF
Reserved
DBGCTRL
TEMP
7:0
7:0
TEMP[7:0]
DBGRUN
7:0
15:8
CNT[7:0]
CNT[15:8]
7:0
15:8
7:0
15:8
7:0
15:8
7:0
15:8
PER[7:0]
PER[15:8]
CMP[7:0]
CMP[15:8]
CMP[7:0]
CMP[15:8]
CMP[7:0]
CMP[15:8]
Reserved
7:0
PERBUF[7:0]
15:8
7:0
15:8
7:0
15:8
7:0
15:8
PERBUF[15:8]
CMPBUF[7:0]
CMPBUF[15:8]
CMPBUF[7:0]
CMPBUF[15:8]
CMPBUF[7:0]
CMPBUF[15:8]
Register Description - Normal Mode
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 242
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.5.1
Control A
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
RUNSTDBY
R/W
0
CTRLA
0x00
0x00
-
6
5
4
3
R/W
0
2
CLKSEL[2:0]
R/W
0
1
R/W
0
0
ENABLE
R/W
0
Bit 7 – RUNSTDBY Run Standby
Writing a ‘1’ to this bit will enable the peripheral to run in Standby Sleep mode.
Bits 3:1 – CLKSEL[2:0] Clock Select
These bits select the clock frequency for the timer/counter.
Value
Name
Description
0x0
DIV1
fTCA = fCLK_PER
0x1
DIV2
fTCA = fCLK_PER/2
0x2
DIV4
fTCA = fCLK_PER/4
0x3
DIV8
fTCA = fCLK_PER/8
0x4
DIV16
fTCA = fCLK_PER/16
0x5
DIV64
fTCA = fCLK_PER/64
0x6
DIV256
fTCA = fCLK_PER/256
0x7
DIV1024
fTCA = fCLK_PER/1024
Bit 0 – ENABLE Enable
Value
Description
0
The peripheral is disabled
1
The peripheral is enabled
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 243
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.5.2
Control B - Normal Mode
Name:
Offset:
Reset:
Property:
Bit
7
Access
Reset
CTRLB
0x01
0x00
-
6
CMP2EN
R/W
0
5
CMP1EN
R/W
0
4
CMP0EN
R/W
0
3
ALUPD
R/W
0
2
R/W
0
1
WGMODE[2:0]
R/W
0
0
R/W
0
Bits 4, 5, 6 – CMPEN Compare n Enable
In the FRQ and PWM Waveform Generation modes the Compare n Enable (CMPnEN) bits will make the waveform
output available on the pin corresponding to WOn, overriding the value in the corresponding PORT output register.
The corresponding pin direction must be configured as an output in the PORT peripheral.
Value
Description
0
Waveform output WOn will not be available on the corresponding pin
1
Waveform output WOn will override the output value of the corresponding pin
Bit 3 – ALUPD Auto-Lock Update
The Auto-Lock Update bit controls the Lock Update (LUPD) bit in the TCAn.CTRLE register. When ALUPD is written
to ‘1’, LUPD will be set to ‘1’ until the Buffer Valid (CMPnBV) bits of all enabled compare channels are ‘1’. This
condition will clear LUPD.
It will remain cleared until the next UPDATE condition, where the buffer values will be transferred to the CMPn
registers and LUPD will be set to ‘1’ again. This makes sure that the CMPnBUF register values are not transferred to
the CMPn registers until all enabled compare buffers are written.
Value
Description
0
LUPD in TCA.CTRLE is not altered by the system
1
LUPD in TCA.CTRLE is set and cleared automatically
Bits 2:0 – WGMODE[2:0] Waveform Generation Mode
These bits select the Waveform Generation mode and control the counting sequence of the counter, TOP value,
UPDATE condition, Interrupt condition, and the type of waveform generated.
No waveform generation is performed in the Normal mode of operation. For all other modes, the waveform generator
output will only be directed to the port pins if the corresponding CMPnEN bit has been set. The port pin direction must
be set as output.
Table 23-6. Timer Waveform Generation Mode
Value
0x0
0x1
0x2
0x3
0x4
0x5
0x6
0x7
Group Configuration
NORMAL
FRQ
SINGLESLOPE
DSTOP
DSBOTH
DSBOTTOM
Mode of Operation
Normal
Frequency
Reserved
Single-slope PWM
Reserved
Dual-slope PWM
Dual-slope PWM
Dual-slope PWM
TOP
UPDATE
OVF
PER
CMP0
PER
PER
PER
PER
TOP(1)
TOP(1)
TOP(1)
BOTTOM
TOP
TOP and BOTTOM
BOTTOM
TOP(1)
BOTTOM
BOTTOM
BOTTOM
BOTTOM
Note:
1. When counting up.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 244
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.5.3
Control C - Normal Mode
Name:
Offset:
Reset:
Property:
Bit
7
CTRLC
0x02
0x00
-
6
5
4
3
Access
Reset
2
CMP2OV
R/W
0
1
CMP1OV
R/W
0
0
CMP0OV
R/W
0
Bit 2 – CMP2OV Compare Output Value 2
See CMP0OV.
Bit 1 – CMP1OV Compare Output Value 1
See CMP0OV.
Bit 0 – CMP0OV Compare Output Value 0
The CMPnOV bits allow direct access to the waveform generator’s output compare value when the timer/counter is
not enabled. This is used to set or clear the WG output value when the timer/counter is not running.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 245
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.5.4
Control D
Name:
Offset:
Reset:
Property:
Bit
7
CTRLD
0x03
0x00
-
6
5
4
3
Access
Reset
2
1
0
SPLITM
R/W
0
Bit 0 – SPLITM Enable Split Mode
This bit sets the timer/counter in Split mode operation. It will then work as two 8-bit timer/counters. The register map
will change compared to normal 16-bit mode.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 246
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.5.5
Control Register E Clear - Normal Mode
Name:
Offset:
Reset:
Property:
CTRLECLR
0x04
0x00
-
This register can be used instead of a Read-Modify-Write (RMW) to clear individual bits by writing a ‘1’ to its bit
location.
Bit
7
6
5
4
3
2
CMD[1:0]
Access
Reset
R/W
0
R/W
0
1
LUPD
R/W
0
0
DIR
R/W
0
Bits 3:2 – CMD[1:0] Command
These bits are used for software control of update, restart and Reset of the timer/counter. The command bits are
always read as ‘0’.
Value
Name
Description
0x0
NONE
No command
0x1
UPDATE
Force update
0x2
RESTART
Force restart
0x3
RESET
Force hard Reset (ignored if the timer/counter is enabled)
Bit 1 – LUPD Lock Update
Lock update can be used to ensure that all buffers are valid before an update is performed.
Value
Description
0
The buffered registers are updated as soon as an UPDATE condition has occurred
1
No update of the buffered registers is performed, even though an UPDATE condition has occurred
Bit 0 – DIR Counter Direction
Normally this bit is controlled in hardware by the Waveform Generation mode or by event actions, but it can also be
changed from software.
Value
Description
0
The counter is counting up (incrementing)
1
The counter is counting down (decrementing)
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 247
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.5.6
Control Register E Set - Normal Mode
Name:
Offset:
Reset:
Property:
CTRLESET
0x05
0x00
-
This register can be used instead of a Read-Modify-Write (RMW) to set individual bits by writing a ‘1’ to its bit
location.
Bit
7
6
5
4
3
2
CMD[1:0]
Access
Reset
R/W
0
R/W
0
1
LUPD
R/W
0
0
DIR
R/W
0
Bits 3:2 – CMD[1:0] Command
These bits are used for software control of update, restart and Reset the timer/counter. The command bits are always
read as ‘0’.
Value
Name
Description
0x0
NONE
No command
0x1
UPDATE
Force update
0x2
RESTART
Force restart
0x3
RESET
Force hard Reset (ignored if the timer/counter is enabled)
Bit 1 – LUPD Lock Update
Locking the update ensures that all buffers are valid before an update is performed.
Value
Description
0
The buffered registers are updated as soon as an UPDATE condition has occurred
1
No update of the buffered registers is performed, even though an UPDATE condition has occurred
Bit 0 – DIR Counter Direction
Normally this bit is controlled in hardware by the Waveform Generation mode or by event actions, but it can also be
changed from software.
Value
Description
0
The counter is counting up (incrementing)
1
The counter is counting down (decrementing)
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 248
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.5.7
Control Register F Clear
Name:
Offset:
Reset:
Property:
CTRLFCLR
0x06
0x00
-
This register can be used instead of a Read-Modify-Write (RMW) to clear individual bits by writing a ‘1’ to its bit
location.
Bit
7
6
5
Access
Reset
4
3
CMP2BV
R/W
0
2
CMP1BV
R/W
0
1
CMP0BV
R/W
0
0
PERBV
R/W
0
Bit 3 – CMP2BV Compare 2 Buffer Valid
See CMP0BV.
Bit 2 – CMP1BV Compare 1 Buffer Valid
See CMP0BV.
Bit 1 – CMP0BV Compare 0 Buffer Valid
The CMPnBV bits are set when a new value is written to the corresponding TCAn.CMPnBUF register. These bits are
automatically cleared on an UPDATE condition.
Bit 0 – PERBV Period Buffer Valid
This bit is set when a new value is written to the TCAn.PERBUF register. This bit is automatically cleared on an
UPDATE condition.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 249
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.5.8
Control Register F Set
Name:
Offset:
Reset:
Property:
CTRLFSET
0x07
0x00
-
This register can be used instead of a Read-Modify-Write (RMW) to set individual bits by writing a ‘1’ to its bit
location.
Bit
7
6
5
Access
Reset
4
3
CMP2BV
R/W
0
2
CMP1BV
R/W
0
1
CMP0BV
R/W
0
0
PERBV
R/W
0
Bit 3 – CMP2BV Compare 2 Buffer Valid
See CMP0BV.
Bit 2 – CMP1BV Compare 1 Buffer Valid
See CMP0BV.
Bit 1 – CMP0BV Compare 0 Buffer Valid
The CMPnBV bits are set when a new value is written to the corresponding TCAn.CMPnBUF register. These bits are
automatically cleared on an UPDATE condition.
Bit 0 – PERBV Period Buffer Valid
This bit is set when a new value is written to the TCAn.PERBUF register. This bit is automatically cleared on an
UPDATE condition.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 250
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.5.9
Event Control
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
R/W
0
EVCTRL
0x09
0x00
-
6
EVACTB[2:0]
R/W
0
5
R/W
0
4
CNTBEI
R/W
0
3
R/W
0
2
EVACTA[2:0]
R/W
0
1
R/W
0
0
CNTAEI
R/W
0
Bits 7:5 – EVACTB[2:0] Event Action B
These bits define what action the counter will take upon certain event conditions.
Value
Name
Description
0x0
NONE
No action
0x1
Reserved
0x2
Reserved
0x3
UPDOWN
Count prescaled clock cycles or count events according to setting for event
input A. The event signal controls the count direction, up when low and down
when high.
0x4
RESTART_POSEDGE Restart counter on positive event edge
0x5
RESTART_ANYEDGE Restart counter on any event edge
0x6
RESTART_HIGHLVL Restart counter while the event signal is high
Other
Reserved
Bit 4 – CNTBEI Enable Counter Event Input B
Value
Description
0
Counter Event input B is disabled
1
Counter Event input B is enabled according to EVACTB bit field
Bits 3:1 – EVACTA[2:0] Event Action A
These bits define what action the counter will take upon certain event conditions.
Value
Name
Description
0x0
CNT_POSEDGE Count on positive event edge
0x1
CNT_ANYEDGE Count on any event edge
0x2
CNT_HIGHLVL Count prescaled clock cycles while the event signal is high
0x3
UPDOWN
Count prescaled clock cycles. The event signal controls the count direction, up when
low and down when high.
Other
Reserved
Bit 0 – CNTAEI Enable Counter Event Input A
Value
Description
0
Counter Event input A is disabled
1
Counter Event input A is enabled according to EVACTA bit field
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 251
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.5.10 Interrupt Control Register - Normal Mode
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
INTCTRL
0x0A
0x00
-
6
CMP2
R/W
0
5
CMP1
R/W
0
4
CMP0
R/W
0
3
2
1
0
OVF
R/W
0
Bit 6 – CMP2 Compare Channel 2 Interrupt Enable
See CMP0.
Bit 5 – CMP1 Compare Channel 1 Interrupt Enable
See CMP0.
Bit 4 – CMP0 Compare Channel 0 Interrupt Enable
Writing the CMPn bit to ‘1’ enables the interrupt from Compare Channel n.
Bit 0 – OVF Timer Overflow/Underflow Interrupt Enable
Writing the OVF bit to ‘1’ enables the overflow/underflow interrupt.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 252
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.5.11 Interrupt Flag Register - Normal Mode
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
INTFLAGS
0x0B
0x00
-
6
CMP2
R/W
0
5
CMP1
R/W
0
4
CMP0
R/W
0
3
2
1
0
OVF
R/W
0
Bit 6 – CMP2 Compare Channel 2 Interrupt Flag
See the CMP0 flag description.
Bit 5 – CMP1 Compare Channel 1 Interrupt Flag
See the CMP0 flag description.
Bit 4 – CMP0 Compare Channel 0 Interrupt Flag
The Compare Interrupt (CMPn) flag is set on a compare match on the corresponding compare channel.
For all modes of operation, the CMPn flag will be set when a compare match occurs between the Count (CNT)
register and the corresponding Compare n (CMPn) register. The CMPn flag is not cleared automatically. It will be
cleared only by writing a ‘1’ to its bit location.
Bit 0 – OVF Overflow/Underflow Interrupt Flag
This flag is set either on a TOP (overflow) or BOTTOM (underflow) condition, depending on the WGMODE setting.
The OVF flag is not cleared automatically. It will be cleared only by writing a ‘1’ to its bit location.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 253
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.5.12 Debug Control Register
Name:
Offset:
Reset:
Property:
Bit
7
DBGCTRL
0x0E
0x00
-
6
5
4
3
2
Access
Reset
1
0
DBGRUN
R/W
0
Bit 0 – DBGRUN Run in Debug
Value
Description
0
The peripheral is halted in Break Debug mode and ignores events.
1
The peripheral will continue to run in Break Debug mode when the CPU is halted.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 254
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.5.13 Temporary Bits for 16-Bit Access
Name:
Offset:
Reset:
Property:
TEMP
0x0F
0x00
-
The Temporary register is used by the CPU for 16-bit single-cycle access to the 16-bit registers of this peripheral. The
register is common for all the 16-bit registers of this peripheral and can be read and written by software. For more
details on reading and writing 16-bit registers, refer to Accessing 16-Bit Registers in the Memories section. There is
one common Temporary register for all the 16-bit registers of this peripheral.
Bit
7
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
TEMP[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – TEMP[7:0] Temporary Bits for 16-bit Access
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 255
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.5.14 Counter Register - Normal Mode
Name:
Offset:
Reset:
Property:
CNT
0x20
0x00
-
The TCAn.CNTL and TCAn.CNTH register pair represents the 16-bit value, TCAn.CNT. The low byte [7:0] (suffix L) is
accessible at the original offset. The high byte [15:8] (suffix H) can be accessed at offset + 0x01.
CPU and UPDI write access has priority over internal updates of the register.
Bit
15
14
13
12
11
10
9
8
R/W
0
R/W
0
R/W
0
R/W
0
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
CNT[15:8]
Access
Reset
Bit
R/W
0
R/W
0
R/W
0
R/W
0
7
6
5
4
CNT[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 15:8 – CNT[15:8] Counter High Byte
These bits hold the MSB of the 16-bit Counter register.
Bits 7:0 – CNT[7:0] Counter Low Byte
These bits hold the LSB of the 16-bit Counter register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 256
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.5.15 Period Register - Normal Mode
Name:
Offset:
Reset:
Property:
PER
0x26
0xFFFF
-
TCAn.PER contains the 16-bit TOP value in the timer/counter in all modes of operation, except Frequency Waveform
Generation (FRQ).
The TCAn.PERL and TCAn.PERH register pair represents the 16-bit value, TCAn.PER. The low byte [7:0] (suffix L)
is accessible at the original offset. The high byte [15:8] (suffix H) can be accessed at offset + 0x01.
Bit
15
14
13
12
11
10
9
8
R/W
1
R/W
1
R/W
1
R/W
1
3
2
1
0
R/W
1
R/W
1
R/W
1
R/W
1
PER[15:8]
Access
Reset
Bit
R/W
1
R/W
1
R/W
1
R/W
1
7
6
5
4
PER[7:0]
Access
Reset
R/W
1
R/W
1
R/W
1
R/W
1
Bits 15:8 – PER[15:8] Periodic High Byte
These bits hold the MSB of the 16-bit Period register.
Bits 7:0 – PER[7:0] Periodic Low Byte
These bits hold the LSB of the 16-bit Period register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 257
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.5.16 Compare n Register - Normal Mode
Name:
Offset:
Reset:
Property:
CMPn
0x28 + n*0x02 [n=0..2]
0x00
-
This register is continuously compared to the counter value. Normally, the outputs from the comparators are used to
generate waveforms.
TCAn.CMPn registers are updated with the buffer value from their corresponding TCAn.CMPnBUF register when an
UPDATE condition occurs.
The TCAn.CMPnL and TCAn.CMPnH register pair represents the 16-bit value, TCAn.CMPn. The low byte [7:0]
(suffix L) is accessible at the original offset. The high byte [15:8] (suffix H) can be accessed at offset + 0x01.
Bit
15
14
13
12
11
10
9
8
R/W
0
R/W
0
R/W
0
R/W
0
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
CMP[15:8]
Access
Reset
Bit
R/W
0
R/W
0
R/W
0
R/W
0
7
6
5
4
CMP[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 15:8 – CMP[15:8] Compare High Byte
These bits hold the MSB of the 16-bit Compare register.
Bits 7:0 – CMP[7:0] Compare Low Byte
These bits hold the LSB of the 16-bit Compare register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 258
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.5.17 Period Buffer Register
Name:
Offset:
Reset:
Property:
PERBUF
0x36
0xFFFF
-
This register serves as the buffer for the Period (TCAn.PER) register. Writing to this register from the CPU or UPDI
will set the Period Buffer Valid (PERBV) bit in the TCAn.CTRLF register.
The TCAn.PERBUFL and TCAn.PERBUFH register pair represents the 16-bit value, TCAn.PERBUF. The low byte
[7:0] (suffix L) is accessible at the original offset. The high byte [15:8] (suffix H) can be accessed at offset + 0x01.
Bit
Access
Reset
Bit
Access
Reset
15
14
13
R/W
1
R/W
1
R/W
1
7
6
5
R/W
1
R/W
1
R/W
1
12
11
PERBUF[15:8]
R/W
R/W
1
1
4
3
PERBUF[7:0]
R/W
R/W
1
1
10
9
8
R/W
1
R/W
1
R/W
1
2
1
0
R/W
1
R/W
1
R/W
1
Bits 15:8 – PERBUF[15:8] Period Buffer High Byte
These bits hold the MSB of the 16-bit Period Buffer register.
Bits 7:0 – PERBUF[7:0] Period Buffer Low Byte
These bits hold the LSB of the 16-bit Period Buffer register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 259
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.5.18 Compare n Buffer Register
Name:
Offset:
Reset:
Property:
CMPnBUF
0x38 + n*0x02 [n=0..2]
0x00
-
This register serves as the buffer for the associated Compare n (TCAn.CMPn) register. Writing to this register from
the CPU or UPDI will set the Compare Buffer valid (CMPnBV) bit in the TCAn.CTRLF register.
The TCAn.CMPnBUFL and TCAn.CMPnBUFH register pair represents the 16-bit value, TCAn.CMPnBUF. The low
byte [7:0] (suffix L) is accessible at the original offset. The high byte [15:8] (suffix H) can be accessed at offset
+ 0x01.
Bit
Access
Reset
Bit
Access
Reset
15
14
13
R/W
0
R/W
0
R/W
0
7
6
5
R/W
0
R/W
0
R/W
0
12
11
CMPBUF[15:8]
R/W
R/W
0
0
4
3
CMPBUF[7:0]
R/W
R/W
0
0
10
9
8
R/W
0
R/W
0
R/W
0
2
1
0
R/W
0
R/W
0
R/W
0
Bits 15:8 – CMPBUF[15:8] Compare High Byte
These bits hold the MSB of the 16-bit Compare Buffer register.
Bits 7:0 – CMPBUF[7:0] Compare Low Byte
These bits hold the LSB of the 16-bit Compare Buffer register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 260
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.6
Register Summary - Split Mode
Offset
Name
Bit Pos.
7
0x00
0x01
0x02
0x03
0x04
0x05
0x06
...
0x09
0x0A
0x0B
0x0C
...
0x0D
0x0E
0x0F
...
0x1F
0x20
0x21
0x22
...
0x25
0x26
0x27
0x28
0x29
0x2A
0x2B
0x2C
0x2D
CTRLA
CTRLB
CTRLC
CTRLD
CTRLECLR
CTRLESET
7:0
7:0
7:0
7:0
7:0
7:0
RUNSTDBY
23.7
6
5
4
3
HCMP2EN
HCMP2OV
HCMP1EN
HCMP1OV
HCMP0EN
HCMP0OV
2
1
CLKSEL[2:0]
LCMP2EN
LCMP2OV
LCMP1EN
LCMP1OV
CMD[1:0]
CMD[1:0]
0
ENABLE
LCMP0EN
LCMP0OV
SPLITM
CMDEN[1:0]
CMDEN[1:0]
Reserved
INTCTRL
INTFLAGS
7:0
7:0
LCMP2
LCMP2
LCMP1
LCMP1
LCMP0
LCMP0
HUNF
HUNF
LUNF
LUNF
Reserved
DBGCTRL
7:0
DBGRUN
Reserved
LCNT
HCNT
7:0
7:0
LCNT[7:0]
HCNT[7:0]
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
LPER[7:0]
HPER[7:0]
LCMP[7:0]
HCMP[7:0]
LCMP[7:0]
HCMP[7:0]
LCMP[7:0]
HCMP[7:0]
Reserved
LPER
HPER
LCMP0
HCMP0
LCMP1
HCMP1
LCMP2
HCMP2
Register Description - Split Mode
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 261
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.7.1
Control A
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
RUNSTDBY
R/W
0
CTRLA
0x00
0x00
-
6
5
4
3
R/W
0
2
CLKSEL[2:0]
R/W
0
1
R/W
0
0
ENABLE
R/W
0
Bit 7 – RUNSTDBY Run Standby
Writing a ‘1’ to this bit will enable the peripheral to run in Standby Sleep mode.
Bits 3:1 – CLKSEL[2:0] Clock Select
These bits select the clock frequency for the timer/counter.
Value
Name
Description
0x0
DIV1
fTCA = fCLK_PER
0x1
DIV2
fTCA = fCLK_PER/2
0x2
DIV4
fTCA = fCLK_PER/4
0x3
DIV8
fTCA = fCLK_PER/8
0x4
DIV16
fTCA = fCLK_PER/16
0x5
DIV64
fTCA = fCLK_PER/64
0x6
DIV256
fTCA = fCLK_PER/256
0x7
DIV1024
fTCA = fCLK_PER/1024
Bit 0 – ENABLE Enable
Value
Description
0
The peripheral is disabled
1
The peripheral is enabled
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 262
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.7.2
Control B - Split Mode
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
CTRLB
0x01
0x00
-
6
HCMP2EN
R/W
0
5
HCMP1EN
R/W
0
4
HCMP0EN
R/W
0
3
2
LCMP2EN
R/W
0
1
LCMP1EN
R/W
0
0
LCMP0EN
R/W
0
Bit 6 – HCMP2EN High byte Compare 2 Enable
See HCMP0EN.
Bit 5 – HCMP1EN High byte Compare 1 Enable
See HCMP0EN.
Bit 4 – HCMP0EN High byte Compare 0 Enable
Setting the HCMPnEN bit in the FRQ or PWM Waveform Generation mode of operation will override the port output
register for the corresponding WO[n+3] pin.
Bit 2 – LCMP2EN Low byte Compare 2 Enable
See LCMP0EN.
Bit 1 – LCMP1EN Low byte Compare 1 Enable
See LCMP0EN.
Bit 0 – LCMP0EN Low byte Compare 0 Enable
Setting the LCMPnEN bit in the FRQ or PWM Waveform Generation mode of operation will override the port output
register for the corresponding WOn pin.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 263
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.7.3
Control C - Split Mode
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
CTRLC
0x02
0x00
-
6
HCMP2OV
R/W
0
5
HCMP1OV
R/W
0
4
HCMP0OV
R/W
0
3
2
LCMP2OV
R/W
0
1
LCMP1OV
R/W
0
0
LCMP0OV
R/W
0
Bit 6 – HCMP2OV High byte Compare 2 Output Value
See HCMP0OV.
Bit 5 – HCMP1OV High byte Compare 1 Output Value
See HCMP0OV.
Bit 4 – HCMP0OV High byte Compare 0 Output Value
The HCMPnOV bit allows direct access to the output compare value of the waveform generator when the timer/
counter is not enabled. This is used to set or clear the WO[n+3] output value when the timer/counter is not running.
Bit 2 – LCMP2OV Low byte Compare 2 Output Value
See LCMP0OV.
Bit 1 – LCMP1OV Low byte Compare 1 Output Value
See LCMP0OV.
Bit 0 – LCMP0OV Low byte Compare 0 Output Value
The LCMPnOV bit allows direct access to the output compare value of the waveform generator when the timer/
counter is not enabled. This is used to set or clear the WOn output value when the timer/counter is not running.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 264
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.7.4
Control D
Name:
Offset:
Reset:
Property:
Bit
7
CTRLD
0x03
0x00
-
6
5
4
3
Access
Reset
2
1
0
SPLITM
R/W
0
Bit 0 – SPLITM Enable Split Mode
This bit sets the timer/counter in Split mode operation. It will then work as two 8-bit timer/counters. The register map
will change compared to normal 16-bit mode.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 265
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.7.5
Control Register E Clear - Split Mode
Name:
Offset:
Reset:
Property:
CTRLECLR
0x04
0x00
-
This register can be used instead of a Read-Modify-Write (RMW) to clear individual bits by writing a ‘1’ to its bit
location.
Bit
7
6
5
4
3
2
CMD[1:0]
Access
Reset
R/W
0
R/W
0
1
0
CMDEN[1:0]
R/W
R/W
0
0
Bits 3:2 – CMD[1:0] Command
These bits are used for software control of restart and reset of the timer/counter. The command bits are always read
as ‘0’.
Value
Name
Description
0x0
NONE
No command
0x1
Reserved
0x2
RESTART
Force restart
0x3
RESET
Force hard Reset (ignored if the timer/counter is enabled)
Bits 1:0 – CMDEN[1:0] Command Enable
These bits configure what timer/counters the command given by the CMD-bits will be applied to.
Value
Name
Description
0x0
NONE None
0x1
Reserved
0x2
Reserved
0x3
BOTH
Command (CMD) will be applied to both low byte and high byte timer/counter
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 266
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.7.6
Control Register E Set - Split Mode
Name:
Offset:
Reset:
Property:
CTRLESET
0x05
0x00
-
This register can be used instead of a Read-Modify-Write (RMW) to set individual bits by writing a ‘1’ to its bit
location.
Bit
7
6
5
4
3
2
CMD[1:0]
Access
Reset
R/W
0
R/W
0
1
0
CMDEN[1:0]
R/W
R/W
0
0
Bits 3:2 – CMD[1:0] Command
This bit field used for software control of restart and reset of the timer/counter. The command bits are always read as
‘0’. The CMD bit field must be used together with the Command Enable (CMDEN) bits. Using the RESET command
requires that both low byte and high byte timer/counter are selected with CMDEN.
Value
Name
Description
0x0
NONE
No command
0x1
Reserved
0x2
RESTART
Force restart
0x3
RESET
Force hard Reset (ignored if the timer/counter is enabled)
Bits 1:0 – CMDEN[1:0] Command Enable
These bits configure what timer/counters the command given by the CMD-bits will be applied to.
Value
Name
Description
0x0
NONE None
0x1
Reserved
0x2
Reserved
0x3
BOTH
Command (CMD) will be applied to both low byte and high byte timer/counter
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 267
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.7.7
Interrupt Control Register - Split Mode
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
INTCTRL
0x0A
0x00
-
6
LCMP2
R/W
0
5
LCMP1
R/W
0
4
LCMP0
R/W
0
3
2
1
HUNF
R/W
0
0
LUNF
R/W
0
Bit 6 – LCMP2 Low byte Compare Channel 2 Interrupt Enable
See LCMP0.
Bit 5 – LCMP1 Low byte Compare Channel 1 Interrupt Enable
See LCMP0.
Bit 4 – LCMP0 Low byte Compare Channel 0 Interrupt Enable
Writing the LCMPn bit to ‘1’ enables the low byte Compare Channel n interrupt.
Bit 1 – HUNF High byte Underflow Interrupt Enable
Writing the HUNF bit to ‘1’ enables the high byte underflow interrupt.
Bit 0 – LUNF Low byte Underflow Interrupt Enable
Writing the LUNF bit to ‘1’ enables the low byte underflow interrupt.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 268
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.7.8
Interrupt Flag Register - Split Mode
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
INTFLAGS
0x0B
0x00
-
6
LCMP2
R/W
0
5
LCMP1
R/W
0
4
LCMP0
R/W
0
3
2
1
HUNF
R/W
0
0
LUNF
R/W
0
Bit 6 – LCMP2 Low byte Compare Channel 2 Interrupt Flag
See LCMP0 flag description.
Bit 5 – LCMP1 Low byte Compare Channel 1 Interrupt Flag
See LCMP0 flag description.
Bit 4 – LCMP0 Low byte Compare Channel 0 Interrupt Flag
The Low byte Compare Interrupt (LCMPn) flag is set on a compare match on the corresponding compare channel in
the low byte timer.
For all modes of operation, the LCMPn flag will be set when a compare match occurs between the Low Byte Timer
Counter (TCAn.LCNT) register and the corresponding Compare n (TCAn.LCMPn) register. The LCMPn flag will not
be cleared automatically and has to be cleared by software. This is done by writing a ‘1’ to its bit location.
Bit 1 – HUNF High byte Underflow Interrupt Flag
This flag is set on a high byte timer BOTTOM (underflow) condition. HUNF is not automatically cleared and needs to
be cleared by software. This is done by writing a ‘1’ to its bit location.
Bit 0 – LUNF Low byte Underflow Interrupt Flag
This flag is set on a low byte timer BOTTOM (underflow) condition. LUNF is not automatically cleared and needs to
be cleared by software. This is done by writing a ‘1’ to its bit location.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 269
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.7.9
Debug Control Register
Name:
Offset:
Reset:
Property:
Bit
7
DBGCTRL
0x0E
0x00
-
6
5
4
3
2
Access
Reset
1
0
DBGRUN
R/W
0
Bit 0 – DBGRUN Run in Debug
Value
Description
0
The peripheral is halted in Break Debug mode and ignores events.
1
The peripheral will continue to run in Break Debug mode when the CPU is halted.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 270
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.7.10 Low Byte Timer Counter Register - Split Mode
Name:
Offset:
Reset:
Property:
LCNT
0x20
0x00
-
TCAn.LCNT contains the counter value for the low byte timer. CPU and UPDI write access has priority over count,
clear or reload of the counter.
Bit
7
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
LCNT[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – LCNT[7:0] Counter Value for Low Byte Timer
These bits define the counter value of the low byte timer.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 271
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.7.11 High Byte Timer Counter Register - Split Mode
Name:
Offset:
Reset:
Property:
HCNT
0x21
0x00
-
TCAn.HCNT contains the counter value for the high byte timer. CPU and UPDI write access has priority over count,
clear or reload of the counter.
Bit
7
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
HCNT[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – HCNT[7:0] Counter Value for High Byte Timer
These bits define the counter value in high byte timer.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 272
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.7.12 Low Byte Timer Period Register - Split Mode
Name:
Offset:
Reset:
Property:
LPER
0x26
0xFF
-
The TCAn.LPER register contains the TOP value for the low byte timer.
Bit
7
6
5
4
3
2
1
0
R/W
1
R/W
1
R/W
1
R/W
1
LPER[7:0]
Access
Reset
R/W
1
R/W
1
R/W
1
R/W
1
Bits 7:0 – LPER[7:0] Period Value Low Byte Timer
These bits hold the TOP value for the low byte timer.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 273
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.7.13 High Byte Period Register - Split Mode
Name:
Offset:
Reset:
Property:
HPER
0x27
0xFF
-
The TCAn.HPER register contains the TOP value for the high byte timer.
Bit
7
6
5
4
3
2
1
0
R/W
1
R/W
1
R/W
1
R/W
1
HPER[7:0]
Access
Reset
R/W
1
R/W
1
R/W
1
R/W
1
Bits 7:0 – HPER[7:0] Period Value High Byte Timer
These bits hold the TOP value for the high byte timer.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 274
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.7.14 Compare Register n For Low Byte Timer - Split Mode
Name:
Offset:
Reset:
Property:
LCMPn
0x28 + n*0x02 [n=0..2]
0x00
-
The TCAn.LCMPn register represents the compare value of Compare Channel n for the low byte timer. This register
is continuously compared to the counter value of the low byte timer, TCAn.LCNT. Normally, the outputs from the
comparators are then used to generate waveforms.
Bit
7
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
LCMP[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – LCMP[7:0] Compare Value of Channel n
These bits hold the compare value of channel n that is compared to TCAn.LCNT.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 275
AVR128DB28/32/48/64
TCA - 16-bit Timer/Counter Type A
23.7.15 High Byte Compare Register n - Split Mode
Name:
Offset:
Reset:
Property:
HCMPn
0x29 + n*0x02 [n=0..2]
0x00
-
The TCAn.HCMPn register represents the compare value of Compare Channel n for the high byte timer. This register
is continuously compared to the counter value of the high byte timer, TCAn.HCNT. Normally, the outputs from the
comparators are then used to generate waveforms.
Bit
7
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
HCMP[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – HCMP[7:0] Compare Value of Channel n
These bits hold the compare value of channel n that is compared to TCAn.HCNT.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 276
AVR128DB28/32/48/64
TCB - 16-bit Timer/Counter Type B
24.
TCB - 16-bit Timer/Counter Type B
24.1
Features
•
•
•
24.2
16-bit Counter Operation Modes:
– Periodic interrupt
– Time-out check
– Input capture
• On event
• Frequency measurement
• Pulse-width measurement
• Frequency and pulse-width measurement
• 32-bit capture
– Single-shot
– 8-bit Pulse-Width Modulation (PWM)
Noise Canceler on Event Input
Synchronize Operation with TCAn
Overview
The capabilities of the 16-bit Timer/Counter type B (TCB) include frequency and waveform generation, and input
capture on event with time and frequency measurement of digital signals. The TCB consists of a base counter and
control logic that can be set in one of eight different modes, each mode providing unique functionality. The base
counter is clocked by the peripheral clock with optional prescaling.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 277
AVR128DB28/32/48/64
TCB - 16-bit Timer/Counter Type B
24.2.1
Block Diagram
Figure 24-1. Timer/Counter Type B Block Diagram
Clock Select
CTRLA
Mode
CTRLB
EVCTRL
Event Action
Count
Counter
Clear
CNT
Restart
Events
Control
Logic
CAPT
(Interrupt Request
and Events)
OVF
= MAX
(Interrupt Request
and Events)
BOTTOM
=0
CCMP
=
Waveform
Generation
Match
WO
The timer/counter can be clocked from the Peripheral Clock (CLK_PER), from a 16-bit Timer/Counter type A
(CLK_TCAn) or the Event System (EVSYS).
Figure 24-2. Timer/Counter Clock Logic
CTRLA
CLK_PER
DIV2
CLK_TCB
CLK_TCAn
CNT
Events
Control
Logic
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 278
AVR128DB28/32/48/64
TCB - 16-bit Timer/Counter Type B
The Clock Select (CLKSEL) bit field in the Control A (TCBn.CTRLA) register selects one of the prescaler outputs
directly, or an event channel as the clock (CLK_TCB) input.
Setting the timer/counter to use the clock from a TCAn allows the timer/counter to run in sync with that TCAn.
By using the EVSYS, any event source, such as an external clock signal on any I/O pin, may be used as the counter
clock input or as a control logic input. When an event action controlled operation is used, the clock selection must be
set to use an event channel as the counter input.
24.2.2
Signal Description
Signal
Description
WO
Type
Digital Asynchronous Output
24.3
Functional Description
24.3.1
Definitions
Waveform Output
The following definitions are used throughout the documentation:
Table 24-1. Timer/Counter Definitions
Name
Description
BOTTOM
The counter reaches BOTTOM when it becomes 0x0000
MAX
The counter reaches the maximum when it becomes 0xFFFF
TOP
The counter reaches TOP when it becomes equal to the highest value in the count sequence
CNT
Count (TCBn.CNT) register value
CCMP
Capture/Compare (TCBn.CCMP) register value
Note: In general, the term ‘timer’ is used when the timer/counter is counting periodic clock ticks. The term ‘counter’
is used when the input signal has sporadic or irregular ticks.
24.3.2
Initialization
By default, the TCB is in Periodic Interrupt mode. Follow these steps to start using it:
1. Write a TOP value to the Compare/Capture (TCBn.CCMP) register.
2. Optional: Write the Compare/Capture Output Enable (CCMPEN) bit in the Control B (TCBn.CTRLB) register to
‘1’. This will make the waveform output available on the corresponding pin, overriding the value in the
corresponding PORT output register. The corresponding pin direction must be configured as an output in the
PORT peripheral.
3. Enable the counter by writing a ‘1’ to the ENABLE bit in the Control A (TCBn.CTRLA) register.
The counter will start counting clock ticks according to the prescaler setting in the Clock Select (CLKSEL) bit
field in the Control A (TCBn.CTRLA) register.
4. The counter value can be read from the Count (TCBn.CNT) register. The peripheral will generate a CAPT
interrupt and event when the CNT value reaches TOP.
4.1.
If the Compare/Capture register is modified to a value lower than the current CNT, the peripheral will
count to MAX and wrap around.
4.2.
At MAX, an OVF interrupt and event will be generated.
24.3.3
Operation
24.3.3.1 Modes
The timer can be configured to run in one of the eight different modes described in the sections below. The event
pulse needs to be longer than one peripheral clock cycle to ensure edge detection.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 279
AVR128DB28/32/48/64
TCB - 16-bit Timer/Counter Type B
24.3.3.1.1 Periodic Interrupt Mode
In the Periodic Interrupt mode, the counter counts to the capture value and restarts from BOTTOM. A CAPT interrupt
and event is generated when the CNT is equal to TOP. If TOP is updated to a value lower than CNT, upon reaching
MAX, an OVF interrupt and event is generated, and the counter restarts from BOTTOM.
Figure 24-3. Periodic Interrupt Mode
CAPT
(Interrupt Request
MAX
and Event)
OVF
(Interrupt Request
and Event)
TOP
CNT
BOTTOM
TOP changed to
a value lower than CNT
OVF set, and CNT set to
BOTTOM
24.3.3.1.2 Time-Out Check Mode
In the Time-Out Check mode, the peripheral starts counting on the first signal edge and stops on the next signal edge
detected on the event input channel. CNT remains stationary after the Stop edge (Freeze state). In Freeze state, the
counter will restart on a new Start edge.
This mode requires TCB to be configured as an event user, and is explained in Events section.
Start or Stop edge is determined by the Event Edge (EDGE) bit in the Event Control (TCBn.EVCTRL) register. If CNT
reaches TOP before the second edge, a CAPT interrupt and event will be generated. If TOP is updated to a value
lower than the CNT upon reaching MAX, an OVF interrupt and the simultaneous event is generated, and the counter
restarts from BOTTOM. In Freeze state, reading the Count (TCBn.CNT) register or Compare/Capture (TCBn.CCMP)
register, or writing the Run (RUN) bit in the Status (TCBn.STATUS) register has no effect.
Figure 24-4. Time-Out Check Mode
CAPT
Event Input
(Interrupt Request
and Event)
OVF
Event Detector
(Interrupt Request
and Event)
MAX
TOP
CNT
BOTTOM
TOP changed to a value lower
than CNT
© 2020 Microchip Technology Inc.
Preliminary Datasheet
OVF set, and CNT
set to BOTTOM
DS40002247A-page 280
AVR128DB28/32/48/64
TCB - 16-bit Timer/Counter Type B
24.3.3.1.3 Input Capture on Event Mode
In the Input Capture on Event mode, the counter will count from BOTTOM to MAX continuously. When an event is
detected, the CNT is transferred to the Capture/Compare (TCBn.CCMP) register, and a CAPT interrupt and event is
generated. The Event edge detector can be configured to trigger a capture on either rising or falling edges.
This mode requires TCB to be configured as an event user, and is explained in Events section.
The figure below shows the input capture unit configured to capture on the falling edge of the event input signal. The
CAPT Interrupt flag is automatically cleared after the low byte of the Compare/Capture (TCBn.CCMP) register has
been read. An OVF interrupt and event is generated when the CNT is MAX.
Figure 24-5. Input Capture on Event
CAPT
(Interrupt Request
Event Input
and Event)
OVF
(Interrupt Request
and Event)
Event Detector
MAX
CNT
BOTTOM
Copy CNT to
CCMP and CAPT
OVF set, and CNT
set to BOTTOM
Copy CNT to
CCMP and CAPT
Important: It is recommended to write 0x0000 to the Count (TCBn.CNT) register when entering this
mode from any other mode.
24.3.3.1.4 Input Capture Frequency Measurement Mode
In the Input Capture Frequency Measurement mode, the TCB captures the counter value and restarts on either a
positive or negative edge of the event input signal.
This mode requires TCB to be configured as an event user, and is explained in Events section.
The CAPT Interrupt flag is automatically cleared after the low byte of the Compare/Capture (TCBn.CCMP) register
has been read. An OVF interrupt and event is generated when the CNT value is MAX.
The figure below illustrates this mode when configured to act on a rising edge.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 281
AVR128DB28/32/48/64
TCB - 16-bit Timer/Counter Type B
Figure 24-6. Input Capture Frequency Measurement
CAPT
(Interrupt Request
and Event)
Event Input
OVF
(Interrupt Request
and Event)
Event Detector
MAX
CNT
BOTTOM
OVF set, and CNT
set to BOTTOM
Copy CNT to CCMP,
CAPT and restart
Copy CNT to CCMP,
CAPT and restart
24.3.3.1.5 Input Capture Pulse-Width Measurement Mode
In the Input Capture Pulse-Width Measurement mode, the input capture pulse-width measurement will restart the
counter on a positive edge, and capture on the next falling edge before an interrupt request is generated. The CAPT
Interrupt flag is automatically cleared after the low byte of the Compare/Capture (TCBn.CCMP) register has been
read. An OVF interrupt and event is generated when the CNT is MAX. The timer will automatically switch between
rising and falling edge detection, but a minimum edge separation of two clock cycles is required for correct behavior.
This mode requires TCB to be configured as an event user, and is explained in Events section.
Figure 24-7. Input Capture Pulse-Width Measurement
CAPT
(Interrupt Request
Event Input
and Event)
OVF
(Interrupt Request
and Event)
Edge Detector
MAX
CNT
BOTTOM
Start counter
Copy CNT to
CCMP and CAPT
Restart
counter
Copy CNT to
CCMP and CAPT
OVF set, and CNT
set to BOTTOM
24.3.3.1.6 Input Capture Frequency and Pulse-Width Measurement Mode
In the Input Capture Frequency and Pulse-Width Measurement mode, the timer will start counting when a positive
edge is detected on the event input signal. The count value is captured on the following falling edge. The counter
stops when the second rising edge of the event input signal is detected. This will set the CAPT interrupt flag.
This mode requires TCB to be configured as an event user, and is explained in Events section.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 282
AVR128DB28/32/48/64
TCB - 16-bit Timer/Counter Type B
The CAPT Interrupt flag is automatically cleared after the low byte of the Compare/Capture (TCBn.CCMP) register
has been read, and the timer/counter is ready for a new capture sequence. Therefore, the Count (TCBn.CNT)
register must be read before the Compare/Capture (TCBn.CCMP) register, since it is reset to BOTTOM at the next
positive edge of the event input signal. An OVF interrupt and event is generated when the CNT value is MAX.
Figure 24-8. Input Capture Frequency and Pulse-Width Measurement
Ignored until CPU
reads CCMP register
Trigger next capture
sequence
CAPT
(Interrupt Request
Event Input
and Event)
Event Detector
MAX
CNT
BOTTOM
Start
counter
Copy CNT to
CCMP
Stop counter and
CAPT
CPU reads the
CCMP register
24.3.3.1.7 Single-Shot Mode
The Single-Shot mode can be used to generate a pulse with a duration defined by the Compare (TCBn.CCMP)
register, every time a rising or falling edge is observed on a connected event channel.
This mode requires TCB to be configured as an event user, and is explained in Events section.
When the counter is stopped, the output pin is set low. If an event is detected on the connected event channel, the
timer will reset and start counting from BOTTOM to TOP while driving its output high. The RUN bit in the Status
(TCBn.STATUS) register can be read to see if the counter is counting or not. When CNT reaches the CCMP register
value, the counter will stop, and the output pin will go low for at least one counter clock cycle (TCB_CLK), and a new
event arriving during this time will be ignored. After this, there is a delay of two peripheral clock cycles (PER_CLK)
from when a new event is received until the output is set high.
The counter will start counting as soon as the peripheral is enabled, even without triggering by an event, or if the
Event Edge (EDGE) bit in the Event Control (TCBn.EVCTRL) register is modified while the peripheral is enabled.
This is prevented by writing TOP to the Counter register. Similar behavior is seen if the Event Edge (EDGE) bit in the
Event Control (TCBn.EVCTRL) register is ‘1’ while the module is enabled. Writing TOP to the Counter register
prevents this as well.
If the Event Asynchronous (ASYNC) bit in the Control B (TCBn.CTRLB) register is written to ‘1’, the timer will react
asynchronously to an incoming event. An edge on the event will immediately cause the output signal to be set. The
counter will still start counting two clock cycles after the event is received.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 283
AVR128DB28/32/48/64
TCB - 16-bit Timer/Counter Type B
Figure 24-9. Single-Shot Mode
Ignored
Ignored
CAPT
(Interrupt Request
and Event)
Edge Detector
TOP
CNT
BOTTOM
Output
Event starts
counter
Counter reaches
TOP value
Event starts
counter
Counter reaches
TOP value
24.3.3.1.8 8-Bit PWM Mode
The TCB can be configured to run in 8-bit PWM mode, where each of the register pairs in the 16-bit Compare/
Capture (TCBn.CCMPH and TCBn.CCMPL) register are used as individual Compare registers. The period (T) is
controlled by CCMPL, while CCMPH controls the duty cycle of the waveform. The counter will continuously count
from BOTTOM to CCMPL, and the output will be set at BOTTOM and cleared when the counter reaches CCMPH.
CCMPH is the number of cycles for which the output will be driven high. CCMPL+1 is the period of the output pulse.
Figure 24-10. 8-Bit PWM Mode
Period (T)
CCMPH=BOTTOM
CCMPH=TOP
CCMPH>TOP
CAPT
(Interrupt Request
and Event)
MAX
OVF
(Interrupt Request
and Event)
TOP
CNT
CCMPL
CCMPH
BOTTOM
Output
24.3.3.2 Output
Timer synchronization and output logic level are dependent on the selected Timer Mode (CNTMODE) bit field in
Control B (TCBn.CTRLB) register. In Single-Shot mode, the timer/counter can be configured so that the signal
generation happens asynchronously to an incoming event (ASYNC = 1 in TCBn.CTRLB). The output signal is then
set immediately at the incoming event instead of being synchronized to the TCB clock. Even though the output is set
immediately, it will take two to three CLK_TCB cycles before the counter starts counting.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 284
AVR128DB28/32/48/64
TCB - 16-bit Timer/Counter Type B
Writing the Compare/Capture Output Enable (CCMPEN) bit in TCBn.CTRLB to ‘1’ enables the waveform output. This
will make the waveform output available on the corresponding pin, overriding the value in the corresponding PORT
output register. The corresponding pin direction must be configured as an output in the PORT peripheral.
The different configurations and their impact on the output are listed in the table below.
Table 24-2. Output Configuration
CCMPEN
CNTMODE
ASYNC
0
The output is high when the counter starts, and the output is low
when the counter stops
1
The output is high when the event arrives, and the output is low
when the counter stops
Single-Shot mode
1
0
Output
8-bit PWM mode
Not applicable 8-bit PWM mode
Other modes
Not applicable
Not applicable
Not applicable No output
The Compare/Capture Pin Initial Value bit (CCMPINIT) in the Control
B (TCBn.CTRLB) register selects the initial output level
It is not recommended to change modes while the peripheral is enabled, as this can produce an unpredictable output.
There is a possibility that an interrupt flag is set during the timer configuration. It is recommended to clear the Timer/
Counter Interrupt Flags (TCBn.INTFLAGS) register after configuring the peripheral.
24.3.3.3 32-Bit Input Capture
Two 16-bit Timer/Counter Type B (TCBn) can be combined to work as a true 32-bit input capture:
One TCB is counting the two LSBs. Once this counter reaches MAX, an overflow (OVF) event is generated, and the
counter wraps around. The second TCB is configured to count these OVF events and thus provides the two MSBs.
The 32-bit counter value is concatenated from the two counter values.
To function as a 32-bit counter, the two TCBs and the system have to be set up as described in the following
paragraphs.
System Configuration
• Configure a source (TCA, events, CLK_PER) for the count input for the LSB TCB, according to the application
requirements
• Configure the event system to route the OVF events from the LSB TCB (event generator) to the TCB intended
for counting the MSB (event user)
• Configure the event system to route the same capture event (CAPT) generator to both TCBs
Configuration of the LSB Counter
• Select the configured count input by writing the Clock Select (CLKSEL) bit field in the Control A (CTRLA)
register
• Write the Timer Mode (CNTMODE) bit field in the Control B (CTRLB) register to select the Input Capture on
Event mode
• Ensure that the Cascade Two Timer/Counters (CASCADE) bit in CTRLA is ‘0’
Configuration of the MSB Counter
• Enable the 32-bit mode by writing the Cascade Two Timer/Counters bit (CASCADE) in CTRLA to ‘1’
•
•
Select events as clock input by writing to the Clock Select (CLKSEL) bit field in the Control A (CTRLA) register
Write the Timer Mode (CNTMODE) bit field in the Control B (CTRLB) register to select the Input Capture on
Event mode
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 285
AVR128DB28/32/48/64
TCB - 16-bit Timer/Counter Type B
Capturing a 32-Bit Counter Value
To acquire a 32-bit counter value, send a CAPT event to both TCBs. Both TCBs are running in Input Capture on
Event mode, so each will capture the current counter value (CNT) in the respective Capture/Compare (CCMP)
register. The 32-bit capture value is formed by concatenating the two CCMP registers.
Example 24-1. Using TCB0 as LSB Counter and TCB1 as MSB Counter
TCB0 is counting the count input, and TCB1 is counting the OVF signals from TCB0.
A CAPT event is generated and causes both TCB0 and TCB1 to copy their current CNT values to
their respective CCMP registers. The two different CASCADE bit values allow correct timing of the
CAPT event.
The captured 32-bit value is concatenated from TCB1.CCMP (MSB) and TCB0.CCMP (LSB).
Capture request
TCB0 - LSB Counter
CTRLA.CASCADE=0
CTRLB.CNTMODE=CAPT
CNT
Count
input
TCB1 - MSB Counter
CAPT
CTRLA.CLKSEL=EVENT
Event System
CTRLA.CASCADE=1
CTRLB.CNTMODE=CAPT
OVF
= MAX
CNT
TCB1.CCMP TCB0.CCMP
32-bit capture value: Byte 3 Byte 2 Byte 1 Byte 0
(MSB)
(LSB)
24.3.3.4 Noise Canceler
The Noise Canceler improves the noise immunity by using a simple digital filter scheme. When the Noise Filter
(FILTER) bit in the Event Control (TCBn.EVCTRL) register is enabled, the peripheral monitors the event channel and
keeps a record of the last four observed samples. If four consecutive samples are equal, the input is considered to be
stable, and the signal is fed to the edge detector.
When enabled, the Noise Canceler introduces an additional delay of four peripheral clock cycles between a change
applied to the input and the update of the Input Compare register.
The Noise Canceler uses the peripheral clock and is, therefore, not affected by the prescaler.
24.3.3.5 Synchronized with Timer/Counter Type A
The TCB can be configured to use the clock (CLK_TCA) of a Timer/Counter type A (TCAn) by writing to the Clock
Select bit field (CLKSEL) in the Control A register (TCBn.CTRLA). In this setting, the TCB will count on the same
clock source as selected in TCAn.
When the Synchronize Update (SYNCUPD) bit in the Control A (TCBn.CTRLA) register is written to ‘1’, the TCB
counter will restart when the TCAn counter restarts.
24.3.4
Events
The TCB can generate the events described in the following table:
Table 24-3. Event Generators in TCB
Generator Name
Peripheral
TCBn
Event
Description
CAPT CAPT flag set
OVF
OVF flag set
© 2020 Microchip Technology Inc.
Event Type
Generating Clock Domain
Pulse
CLK_PER
Preliminary Datasheet
Length of Event
One CLK_PER period
DS40002247A-page 286
AVR128DB28/32/48/64
TCB - 16-bit Timer/Counter Type B
The conditions for generating the CAPT and OVF events are identical to those that will raise the corresponding
interrupt flags in the Timer/Counter Interrupt Flags (TCBn.INTFLAGS) register. Refer to the Event System section for
more details regarding event users and Event System configuration.
The TCB can receive the events described in the following table:
Table 24-4. Event Users and Available Event Actions in TCB
User Name
Peripheral
Description
Input
Input Detection Async/Sync
Time-Out Check Count mode
Input Capture on Event Count mode
Input Capture Frequency Measurement Count mode
TCBn
CAPT
Input Capture Pulse-Width Measurement Count mode
Sync
Edge
Input Capture Frequency and Pulse-Width Measurement
Count mode
Single-Shot Count mode
Both
COUNT Event as clock source in combination with a count mode
Sync
CAPT and COUNT are TCB event users that detect and act upon input events.
The COUNT event user is enabled on the peripheral by modifying the Clock Select (CLKSEL) bit field in the Control A
(TCBn.CTRLA) register to EVENT and setting up the Event System accordingly.
If the Capture Event Input Enable (CAPTEI) bit in the Event Control (TCBn.EVCTRL) register is written to ‘1’,
incoming events will result in an event action as defined by the Event Edge (EDGE) bit in Event Control
(TCBn.EVCTRL) register and the Timer Mode (CNTMODE) bit field in Control B (TCBn.CTRLB) register. The event
needs to last for at least one CLK_PER cycle to be recognized.
If the Asynchronous mode is enabled for Single-Shot mode, the event is edge-triggered and will capture changes on
the event input shorter than one peripheral clock cycle.
24.3.5
Interrupts
Table 24-5. Available Interrupt Vectors and Sources
Name Vector Description
CAPT
TCB interrupt
OVF
Conditions
Depending on the operating mode. See the description of the CAPT bit in the
TCBn.INTFLAG register.
The timer/counter overflows from MAX to BOTTOM.
When an interrupt condition occurs, the corresponding interrupt flag is set in the peripheral’s Interrupt Flags
(peripheral.INTFLAGS) register.
An interrupt source is enabled or disabled by writing to the corresponding enable bit in the peripheral’s Interrupt
Control (peripheral.INTCTRL) register.
An interrupt request is generated when the corresponding interrupt source is enabled, and the interrupt flag is set.
The interrupt request remains active until the interrupt flag is cleared. See the peripheral’s INTFLAGS register for
details on how to clear interrupt flags.
24.3.6
Sleep Mode Operation
TCBn is, by default, disabled in Standby sleep mode. It will be halted as soon as the sleep mode is entered.
The module can stay fully operational in the Standby sleep mode if the Run Standby (RUNSTDBY) bit in the
TCBn.CTRLA register is written to ‘1’.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 287
AVR128DB28/32/48/64
TCB - 16-bit Timer/Counter Type B
All operations are halted in Power-Down sleep mode.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 288
AVR128DB28/32/48/64
TCB - 16-bit Timer/Counter Type B
24.4
Register Summary
Offset
Name
Bit Pos.
0x00
0x01
0x02
...
0x03
0x04
0x05
0x06
0x07
0x08
0x09
CTRLA
CTRLB
EVCTRL
INTCTRL
INTFLAGS
STATUS
DBGCTRL
TEMP
0x0A
CNT
0x0C
CCMP
24.5
7
6
5
4
3
7:0
7:0
RUNSTDBY
ASYNC
CASCADE
CCMPINIT
SYNCUPD
CCMPEN
7:0
7:0
7:0
7:0
7:0
7:0
7:0
15:8
7:0
15:8
FILTER
2
1
CLKSEL[2:0]
0
ENABLE
CNTMODE[2:0]
Reserved
EDGE
OVF
OVF
CAPTEI
CAPT
CAPT
RUN
DBGRUN
TEMP[7:0]
CNT[7:0]
CNT[15:8]
CCMP[7:0]
CCMP[15:8]
Register Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 289
AVR128DB28/32/48/64
TCB - 16-bit Timer/Counter Type B
24.5.1
Control A
Name:
Offset:
Reset:
Property:
Bit
7
Access
Reset
CTRLA
0x00
0x00
-
6
RUNSTDBY
R/W
0
5
CASCADE
R/W
0
4
SYNCUPD
R/W
0
3
R/W
0
2
CLKSEL[2:0]
R/W
0
1
R/W
0
0
ENABLE
R/W
0
Bit 6 – RUNSTDBY Run Standby
Writing a ‘1’ to this bit will enable the peripheral to run in Standby sleep mode.
Bit 5 – CASCADE Cascade Two Timer/Counters
Writing this bit to ‘1’ enables cascading of two 16-bit Timer/Counters type B (TCBn) for 32-bit operation using the
Event System. This bit must be ‘1’ for the timer/counter used for the two Most Significant Bytes (MSB). When this bit
is ‘1’, the selected event source for capture (CAPT) is delayed by one peripheral clock cycle. This compensates the
carry propagation delay when cascading two counters via the Event System.
Bit 4 – SYNCUPD Synchronize Update
When this bit is written to ‘1’, the TCB will restart whenever TCAn is restarted or overflows. This can be used to
synchronize capture with the PWM period. If TCAn is selected as the clock source, the TCB will restart when that
TCAn is restarted. For other clock selections, it will restart together with TCA0.
Bits 3:1 – CLKSEL[2:0] Clock Select
Writing these bits selects the clock source for this peripheral.
Value
Name
Description
0x0
0x1
0x2
0x3
0x4-0x6
0x07
DIV1
DIV2
TCA0
TCA1
EVENT
CLK_PER
CLK_PER / 2
CLK_TCA from TCA0
CLK_TCA from TCA1
Reserved
Positive edge on event input
Bit 0 – ENABLE Enable
Writing this bit to ‘1’ enables the Timer/Counter type B peripheral.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 290
AVR128DB28/32/48/64
TCB - 16-bit Timer/Counter Type B
24.5.2
Control B
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
CTRLB
0x01
0x00
-
6
ASYNC
R/W
0
5
CCMPINIT
R/W
0
4
CCMPEN
R/W
0
3
2
R/W
0
1
CNTMODE[2:0]
R/W
0
0
R/W
0
Bit 6 – ASYNC Asynchronous Enable
Writing this bit to ‘1’ will allow asynchronous updates of the TCB output signal in Single-Shot mode.
Value
Description
0
The output will go HIGH when the counter starts after synchronization
1
The output will go HIGH when an event arrives
Bit 5 – CCMPINIT Compare/Capture Pin Initial Value
This bit is used to set the initial output value of the pin when a pin output is used. This bit has no effect in 8-bit PWM
mode and Single-Shot mode.
Value
Description
0
Initial pin state is LOW
1
Initial pin state is HIGH
Bit 4 – CCMPEN Compare/Capture Output Enable
Writing this bit to ‘1’ enables the waveform output. This will make the waveform output available on the corresponding
pin, overriding the value in the corresponding PORT output register. The corresponding pin direction must be
configured as an output in the PORT peripheral.
Value
Description
0
Waveform output is not enabled on the corresponding pin.
1
Waveform output will override the output value of the corresponding pin.
Bits 2:0 – CNTMODE[2:0] Timer Mode
Writing to this bit field selects the Timer mode.
Value
Name
Description
0x0
INT
Periodic Interrupt mode
0x1
TIMEOUT
Time-out Check mode
0x2
CAPT
Input Capture on Event mode
0x3
FRQ
Input Capture Frequency Measurement mode
0x4
PW
Input Capture Pulse-Width Measurement mode
0x5
FRQPW
Input Capture Frequency and Pulse-Width Measurement mode
0x6
SINGLE
Single-Shot mode
0x7
PWM8
8-Bit PWM mode
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 291
AVR128DB28/32/48/64
TCB - 16-bit Timer/Counter Type B
24.5.3
Event Control
Name:
Offset:
Reset:
Property:
Bit
EVCTRL
0x04
0x00
-
7
Access
Reset
6
FILTER
R/W
0
5
4
EDGE
R/W
0
3
2
1
0
CAPTEI
R/W
0
Bit 6 – FILTER Input Capture Noise Cancellation Filter
Writing this bit to ‘1’ enables the Input Capture Noise Cancellation unit.
Bit 4 – EDGE Event Edge
This bit is used to select the event edge. The effect of this bit is dependent on the selected Count Mode (CNTMODE)
bit field in TCBn.CTRLB. “—” means that an event or edge has no effect in this mode.
Count Mode
Periodic Interrupt mode
Timeout Check mode
Input Capture on Event mode
Input Capture Frequency
Measurement mode
Input Capture Pulse-Width
Measurement mode
EDGE Positive Edge
Negative Edge
0
1
0
1
0
1
—
—
Stop counter
Start counter
—
Input Capture, interrupt
0
1
0
1
0
Input Capture Frequency and Pulse
Width Measurement mode
1
Single-Shot mode
8-Bit PWM mode
0
1
0
1
—
—
Start counter
Stop counter
Input Capture, interrupt
—
Input Capture, clear and restart
counter, interrupt
—
Input Capture, clear and restart
counter, interrupt
Clear and restart counter
Input Capture, interrupt
Input Capture, interrupt
Clear and restart counter
• On the 1st Positive: Clear and restart counter
• On the following Negative: Input Capture
• On the 2nd Positive: Stop counter, interrupt
• On the 1st Negative: Clear and restart counter
• On the following Positive: Input Capture
• On the 2nd Negative: Stop counter, interrupt
Start counter
—
—
Start counter
—
—
—
—
—
Bit 0 – CAPTEI Capture Event Input Enable
Writing this bit to ‘1’ enables the input capture event.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 292
AVR128DB28/32/48/64
TCB - 16-bit Timer/Counter Type B
24.5.4
Interrupt Control
Name:
Offset:
Reset:
Property:
Bit
7
INTCTRL
0x05
0x00
-
6
5
4
3
Access
Reset
2
1
OVF
R/W
0
0
CAPT
R/W
0
Bit 1 – OVF Overflow Interrupt Enable
Writing this bit to ‘1’ enables interrupt on overflow.
Bit 0 – CAPT Capture Interrupt Enable
Writing this bit to ‘1’ enables interrupt on capture.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 293
AVR128DB28/32/48/64
TCB - 16-bit Timer/Counter Type B
24.5.5
Interrupt Flags
Name:
Offset:
Reset:
Property:
Bit
INTFLAGS
0x06
0x00
-
7
6
5
4
3
2
Access
Reset
1
OVF
R/W
0
0
CAPT
R/W
0
Bit 1 – OVF Overflow Interrupt Flag
This bit is set when an overflow interrupt occurs. The flag is set whenever the timer/counter wraps from MAX to
BOTTOM.
The bit is cleared by writing a ‘1’ to the bit position.
Bit 0 – CAPT Capture Interrupt Flag
This bit is set when a capture interrupt occurs. The interrupt conditions are dependent on the Counter Mode
(CNTMODE) bit field in the Control B (TCBn.CTRLB) register.
This bit is cleared by writing a ‘1’ to it or when the Capture register is read in Capture mode.
Table 24-6. Interrupt Sources Set Conditions by Counter Mode
Counter Mode
Interrupt Set Condition
TOP
Value
CAPT
Periodic Interrupt mode
Timeout Check mode
Single-Shot mode
Set when the counter reaches TOP
Set when the counter reaches TOP
Set when the counter reaches TOP
CCMP
CNT == TOP
Input Capture Frequency
Measurement mode
Set on edge when the Capture register is
loaded and the counter restarts; the flag clears
when the capture is read
Set when an event occurs and the Capture
register is loaded; the flag clears when the
capture is read
-Set on edge when the Capture register is
Input Capture Pulse-Width
loaded; the previous edge initialized the count;
Measurement mode
the flag clears when the capture is read
Input Capture Frequency Set on the second edge (positive or negative)
and Pulse-Width
when the counter is stopped; the flag clears
Measurement mode
when the capture is read
8-Bit PWM mode
Set when the counter reaches CCMH
CCML
On Event, copy CNT to
CCMP, and restart
counting (CNT ==
BOTTOM)
Input Capture on Event
mode
© 2020 Microchip Technology Inc.
Preliminary Datasheet
On Event, copy CNT to
CCMP, and continue
counting
CNT == CCMH
DS40002247A-page 294
AVR128DB28/32/48/64
TCB - 16-bit Timer/Counter Type B
24.5.6
Status
Name:
Offset:
Reset:
Property:
Bit
7
STATUS
0x07
0x00
-
6
5
4
3
2
1
Access
Reset
0
RUN
R
0
Bit 0 – RUN Run
When the counter is running, this bit is set to ‘1’. When the counter is stopped, this bit is cleared to ‘0’.
The bit is read-only and cannot be set by UPDI.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 295
AVR128DB28/32/48/64
TCB - 16-bit Timer/Counter Type B
24.5.7
Debug Control
Name:
Offset:
Reset:
Property:
Bit
7
DBGCTRL
0x08
0x00
-
6
5
4
3
2
Access
Reset
1
0
DBGRUN
R/W
0
Bit 0 – DBGRUN Debug Run
Value
Description
0
The peripheral is halted in Break Debug mode and ignores events.
1
The peripheral will continue to run in Break Debug mode when the CPU is halted.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 296
AVR128DB28/32/48/64
TCB - 16-bit Timer/Counter Type B
24.5.8
Temporary Value
Name:
Offset:
Reset:
Property:
TEMP
0x09
0x00
-
The Temporary register is used by the CPU for 16-bit single-cycle access to the 16-bit registers of this peripheral. The
register is common for all the 16-bit registers of this peripheral and can be read and written by software. For more
details on reading and writing 16-bit registers, refer to Accessing 16-Bit Registers in the Memories section.
Bit
7
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
TEMP[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – TEMP[7:0] Temporary Value
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 297
AVR128DB28/32/48/64
TCB - 16-bit Timer/Counter Type B
24.5.9
Count
Name:
Offset:
Reset:
Property:
CNT
0x0A
0x00
-
The TCBn.CNTL and TCBn.CNTH register pair represents the 16-bit value TCBn.CNT. The low byte [7:0] (suffix L) is
accessible at the original offset. The high byte [15:8] (suffix H) can be accessed at offset + 0x01.
CPU and UPDI write access has priority over internal updates of the register.
Bit
15
14
13
12
11
10
9
8
R/W
0
R/W
0
R/W
0
R/W
0
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
CNT[15:8]
Access
Reset
Bit
R/W
0
R/W
0
R/W
0
R/W
0
7
6
5
4
CNT[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 15:8 – CNT[15:8] Count Value High
These bits hold the MSB of the 16-bit Counter register.
Bits 7:0 – CNT[7:0] Count Value Low
These bits hold the LSB of the 16-bit Counter register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 298
AVR128DB28/32/48/64
TCB - 16-bit Timer/Counter Type B
24.5.10 Capture/Compare
Name:
Offset:
Reset:
Property:
CCMP
0x0C
0x00
-
The TCBn.CCMPL and TCBn.CCMPH register pair represents the 16-bit value TCBn.CCMP. The low byte [7:0]
(suffix L) is accessible at the original offset. The high byte [15:8] (suffix H) can be accessed at offset + 0x01.
This register has different functions depending on the mode of operation:
• For Capture operation, these registers contain the captured value of the counter at the time the capture occurs
• In Periodic Interrupt/Time-Out and Single-Shot mode, this register acts as the TOP value
• In 8-bit PWM mode, TCBn.CCMPL and TCBn.CCMPH act as two independent registers: The period of the
waveform is controlled by CCMPL, while CCMPH controls the duty cycle.
Bit
Access
Reset
Bit
15
14
13
R/W
0
R/W
0
R/W
0
7
6
5
12
11
CCMP[15:8]
R/W
R/W
0
0
4
10
9
8
R/W
0
R/W
0
R/W
0
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
CCMP[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 15:8 – CCMP[15:8] Capture/Compare Value High Byte
These bits hold the MSB of the 16-bit compare, capture, and top value.
Bits 7:0 – CCMP[7:0] Capture/Compare Value Low Byte
These bits hold the LSB of the 16-bit compare, capture, and top value.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 299
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
25.
TCD - 12-Bit Timer/Counter Type D
25.1
Features
•
•
•
•
•
•
•
•
•
25.2
12-bit Timer/Counter
Programmable Prescaler
Double-Buffered Compare Registers
Waveform Generation:
– One Ramp mode
– Two Ramp mode
– Four Ramp mode
– Dual Slope mode
Two Separate Input Channels
Software and Input Based Capture
Programmable Filter for Input Events
Conditional Waveform Generation on External Events:
– Fault handling
– Input blanking
– Overload protection
– Fast emergency stop by hardware
Half-Bridge and Full-Bridge Output Support
Overview
The Timer/Counter type D (TCD) is a high-performance waveform generator that consists of an asynchronous
counter, a prescaler, and compare, capture and control logic.
The TCD contains a counter that can run on a clock which is asynchronous to the peripheral clock. It contains
compare logic that generates two independent outputs with optional dead time. It is connected to the Event System
for capture and deterministic Fault control. The timer/counter can generate interrupts and events on compare match
and overflow.
This device provides one instance of the TCD peripheral, TCD0.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 300
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
25.2.1
Block Diagram
Figure 25-1. Timer/Counter Block Diagram
Peripheral
clock
domain
TCD clock
domain
Counter and
Fractional
Accumulator
CMPASET
Compare/Capture
Unit A
CMPASET_
BUF
=
CMPACLR
SET A
CMPACLR_
BUF
Waveform
generator A
CLR A
=
Event Input A
CAPTUREA_
BUF
CMPBSET
CMPBSET_
BUF
PROGEV (Event)
TRIGA (INT Req.)
WOC
Compare/Capture
Unit B
=
SET B
CMPBCLR_
BUF
Waveform
generator B
CLR B
=
Event Input
Logic B
Event Input B
CAPTUREB
WOA
Event Input
Logic A
CAPTUREA
CMPBCLR
CMPASET/PROGEV
(Event)
CAPTUREB_
BUF
WOD
CMPBSET/PROGEV
(Event)
WOB
CMPBCLR/PROGEV
(Event)
TRIG OVF (INT Req.)
TRIGB (INT Req.)
The TCD core is asynchronous to the peripheral clock. The timer/counter consists of two compare/capture units,
each with a separate waveform output. There are also two extra waveform outputs which can be equal to the output
from one of the units. For each compare/capture unit, there is a pair of compare registers which are stored in the
respective peripheral registers (TCDn.CMPASET, TCDn.CMPACLR, TCDn.CMPBSET, TCDn.CMPBCLR).
During normal operation, the counter value is continuously compared to the compare registers. This is used to
generate both interrupts and events.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 301
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
The TCD can use the input events in ten different input modes, selected separately for the two input events. The
input mode defines how the input events will affect the outputs, and where in the TCD cycle the counter must go
when an event occurs.
The TCD can select between four different clock sources that can be prescaled. There are three different prescalers
with separate controls, as shown below.
Figure 25-2. Clock Selection and Prescalers Overview
CLKSEL
Counter
prescaler
OSCHF
PLL
EXTCLK
CLK_PER
CLK_TCD
Counter clock
(CLK_TCD_CNT)
Synchronization
prescaler
Synchronizer clock
(CLK_TCD_SYNC)
Delay
prescaler (1)
Delay clock
(CLK_TCD_DLY)
1. Used by input blanking/delay event out.
The TCD synchronizer clock is separate from the other module clocks, enabling faster synchronization between the
TCD domain and the I/O domain.
The total prescaling for the counter is:
SYNCPRESC_division_factor × CNTPRESC_division_factor
The delay prescaler is used to prescale the clock used for the input blanking/delayed event output functionality. The
prescaler can be configured independently allowing separate range and accuracy settings from the counter
functionality. The synchronization prescaler and counter prescaler can be configured from the Control A
(TCDn.CTRLA) register, while the delay prescaler can be configured from the Delay Control (TCDn.DLYCTRL)
register.
25.2.2
Signal Description
Signal
Description
Type
WOA
TCD waveform output A
Digital output
WOB
TCD waveform output B
Digital output
WOC
TCD waveform output C
Digital output
WOD
TCD waveform output D
Digital output
25.3
Functional Description
25.3.1
Definitions
The following definitions are used throughout the documentation:
Table 25-1. Timer/Counter Definitions
Name
Description
TCD cycle
The sequence of four states that the counter needs to go through before it has returned to the
same position.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 302
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
...........continued
25.3.2
Name
Description
Input blanking
The functionality to ignore an event input for a programmable time in a selectable part of the TCD
cycle.
Asynchronous
output control
Allows the event to override the output instantly when an event occurs. It is used for handling
non-recoverable Faults.
One ramp
The counter is reset to zero once during a TCD cycle.
Two ramp
The counter is reset to zero two times during a TCD cycle.
Four ramp
The counter is reset to zero four times during a TCD cycle.
Dual ramp
The counter counts both up and down between zero and a selected top value during a TCD
cycle.
Input mode
A predefined setting that changes the output characteristics, based on the given input events.
Initialization
To initialize the TCD:
1. Select the clock source and the prescaler from the Control A (TCDn.CTRLA) register.
2. Select the Waveform Generation Mode from the Control B (TCDn.CTRLB) register.
3. Optional: Configure the other static registers to the desired functionality.
4. Write the initial values in the Compare (TCDn.CMPxSET/CLR) registers.
5. Optional: Write the desired values to the other double-buffered registers.
6. Ensure that the Enable Ready (ENRDY) bit in the Status (TCDn.STATUS) register is set to ‘1’.
7.
25.3.3
Enable the TCD by writing a ‘1’ to the ENABLE bit in the Control A (TCDn.CTRLA) register.
Operation
25.3.3.1 Register Synchronization Categories
Most of the I/O registers need to be synchronized to the TCD core clock domain. This is done differently for different
register categories.
Table 25-2. Categorization of Registers
Enable and
Command
Registers
Double-Buffered
Registers
Static Registers
TCDn.CTRLA
(ENABLE bit)
TCDn.DLYCTRL
TCDn.CTRLA(1) (All bits TCDn.STATUS
except ENABLE bit)
TCDn.INTCTRL
TCDn.CTRLE
TCDn.DLYVAL
TCDn.CTRLB
TCDn.CAPTUREA
TCDn.INTFLAGS
TCDn.DITCTRL
TCDn.CTRLC
TCDn.CAPTUREB
TCDn.DITVAL
TCDn.CTRLD
TCDn.DBGCTRL
TCDn.EVCTRLA
TCDn.CMPASET
TCDn.EVCTRLB
TCDn.CMPACLR
TCDn.INPUTCTRLA
TCDn.CMPBSET
TCDn.INPUTCTRLB
TCDn.CMPBCLR
TCDn.FAULTCTRL(2)
© 2020 Microchip Technology Inc.
Read-Only Registers
Preliminary Datasheet
Normal I/O
Registers
DS40002247A-page 303
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
Notes:
1. The bits in the Control A (TCDn.CTRLA) register are enable-protected, except the ENABLE bit. They can only
be written when ENABLE is written to ‘0’ first.
2.
This register is protected by the Configuration Change Protection Mechanism, requiring a timed write
procedure for changing its value settings.
Enable and Command Registers
Because of the synchronization between the clock domains, it is only possible to change the ENABLE bit in the
Control A (TCDn.CTRLA) register, while the Enable Ready (ENRDY) bit in the Status (TCDn.STATUS) register is ‘1’.
The Control E (TCDn.CTRLE) register is automatically synchronized to the TCD core domain when the TCD is
enabled and as long as no synchronization is ongoing already. Check if the Command Ready (CCMDRDY) bit in
TCDn.STATUS is ‘1’ to ensure that it is possible to issue a new command. TCDn.CTRLE is a strobe register that will
clear itself when the command is sent.
Double-Buffered Registers
The double-buffered registers can be updated in normal I/O writes, while TCD is enabled and no synchronization
between the two clock domains is ongoing. Check that the CMDRDY bit in TCDn.STATUS is ‘1’ to ensure that it is
possible to update the double-buffered registers. The values will be synchronized to the TCD core domain when a
synchronization command is sent or when TCD is enabled.
Table 25-3. Issuing Synchronization Command
Synchronization Issuing Bit
Double Register Update
CTRLC.AUPDATE
Every time the CMPBCLRH register is written, the
synchronization occurs at the end of the TCD cycle.
CTRLE.SYNC (1)
Occurs once, as soon as the SYNC bit is synchronized
with the TDC domain.
CTRLE.SYNCEOC (1)
Occurs once at the end of the next TCD cycle.
Note:
1. If synchronization is already ongoing, the action has no effect.
Static Registers
Static registers cannot be updated while TCD is enabled. Therefore, these registers must be configured before
enabling TCD. To see if TCD is enabled, check if ENABLE in TCDn.CTRLA is read as ‘1’.
Normal I/O and Read-Only Registers
Normal I/O and read-only registers are not constrained by any synchronization between the domains. The read-only
registers inform about synchronization status and values synchronized from the core domain.
25.3.3.2 Waveform Generation Modes
The TCD provides four different Waveform Generation modes controlled by the Waveform Generation Mode
(WGMODE) bit field in the Control B (TCDn.CTRLB) register. The Waveform Generation modes are:
• One Ramp mode
• Two Ramp mode
• Four Ramp mode
• Dual Slope mode
The Waveform Generation modes determine how the counter is counting during a TCD cycle and how the compare
values influence the waveform. A TCD cycle is split into these states:
•
•
•
Dead time WOA (DTA)
On time WOA (OTA)
Dead time WOB (DTB)
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 304
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
•
On time WOB (OTB)
The Compare A Set (CMPASET), Compare A Clear (CMPACLR), Compare B Set (CMPBSET) and Compare B Clear
(CMPBCLR) compare values define when each state ends and the next begins.
25.3.3.2.1 One Ramp Mode
In One Ramp mode, the TCD counter counts up until it reaches the CMPBCLR value. Then, the TCD cycle is
completed, and the counter restarts from 0x000, beginning a new TCD cycle. The TCD cycle period is:
�TCD_cycle =
CMPBCLR + 1
�CLK_TCD_CNT
Figure 25-3. One Ramp Mode
TCD cycle
Dead time A
Compare
values
On time A
Dead time B
On time B
Counter
value
CMPBCLR
CMPBSET
CMPACLR
CMPASET
WOA
WOB
In the figure above, CMPASET < CMPACLR < CMPBSET < CMPBCLR. In One Ramp mode, this is required to avoid
overlapping outputs during the on time. The figure below is an example where CMPBSET < CMPASET < CMPACLR
< CMPBCLR, which has overlapping outputs during the on time.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 305
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
Figure 25-4. One Ramp Mode with CMPBSET < CMPASET
TCD cycle
Dead time A
Compare
values
On time A
On time B
Counter
value
CMPBCLR
CMPACLR
CMPASET
CMPBSET
WOA
WOB
A match with CMPBCLR will always result in all outputs being cleared. If any of the other compare values are bigger
than CMPBCLR, their associated effect will never occur. If the CMPACLR is smaller than the CMPASET value, the
clear value will not have any effect.
25.3.3.2.2 Two Ramp Mode
In Two Ramp mode, the TCD counter counts up until it reaches the CMPACLR value, then it resets and counts up
until it reaches the CMPBCLR value. Then, the TCD cycle is completed, and the counter restarts from 0x000,
beginning a new TCD cycle. The TCD cycle period is given by:
�TCD_cycle =
CMPACLR + 1 + CMPBCLR + 1
�CLK_TCD_CNT
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 306
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
Figure 25-5. Two Ramp Mode
TCD cycle
Dead time A
On time A
Dead time B
On time B
Counter
value
CMPBCLR
CMPACLR
CMPBSET
CMPASET
WOA
WOB
In the figure above, CMPASET < CMPACLR and CMPBSET < CMPBCLR. This causes the outputs to go high. There
are no restrictions on the CMPASET and CMPACLR compared to the CMPBSET and CMPBCLR values.
In Two Ramp mode, it is not possible to get overlapping outputs without using the override feature. Even if
CMPASET/CMPBSET > CMPACLR/CMPBCLR, the counter resets at CMPACLR/CMPBCLR and will never reach
CMPASET/CMPBSET.
25.3.3.2.3 Four Ramp Mode
In Four Ramp mode, the TCD cycle follows this pattern:
1. A TCD cycle begins with the TCD counter counting up from zero until it reaches the CMPASET value, and
resets to zero.
2. The counter counts up until it reaches the CMPACLR value, and resets to zero.
3. The counter counts up until it reaches the CMPBSET value, and resets to zero.
4. The counter counts up until it reaches the CMPBCLR value, and ends the TCD cycle by resetting to zero.
The TCD cycle period is given by:
�TCD_cycle =
CMPASET + 1 + CMPACLR + 1 + CMPBSET + 1 + CMPBCLR + 1
�CLK_TCD_CNT
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 307
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
Figure 25-6. Four Ramp Mode
TCD cycle
Dead time A
On time A
Dead time B
On time B
Counter
value
CMPBCLR
CMPACLR
CMPBSET
CMPASET
WOA
WOB
There are no restrictions regarding the compare values, because there are no dependencies between them.
In Four Ramp mode, it is not possible to get overlapping outputs without using the override feature.
25.3.3.2.4 Dual Slope Mode
In Dual Slope mode, a TCD cycle consists of the TCD counter counting down from CMPBCLR value to zero, and up
again to the CMPBCLR value. This gives a TCD cycle period:
2 × CMPBCLR + 1
�CLK_TCD_CNT
The WOA output is set when the TCD counter counts down and matches the CMPASET value. WOA is cleared when
the TCD counter counts up and matches the CMPASET value.
�TCD_cycle =
The WOB output is set when the TCD counter counts up and matches the CMPBSET value. WOB is cleared when
the TCD counter counts down and matches the CMPBSET value.
The outputs will overlap if CMPASET > CMPBSET.
CMPACLR is not used in Dual Slope mode. Writing a value to CMPACLR has no effect.
Figure 25-7. Dual Slope Mode
TCD cycle
On time B
CMPBCLR
Dead
time A
On time A
Dead
time B
On time B
Dead
time A
On time A
Counter
value
CMPBSET
CMPASET
WOA
WOB
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 308
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
When starting the TCD in Dual Slope mode, the TCD counter starts at the CMPBCLR value and counts down. In the
first cycle, the WOB will not be set until the TCD counter matches the CMPBSET value when counting up.
When the Disable at End of Cycle Strobe (DISEOC) bit in the Control E (TCDn.CTRLE) register is set, the TCD will
automatically be disabled at the end of the TCD cycle.
Figure 25-8. Dual Slope Mode Starting and Stopping
TCD cycle
CMPBCLR
Counter
value
CMPBSET
CMPASET
WOA
WOB
Stop
Start
25.3.3.3 Disabling TCD
Disabling the TCD can be done in two different ways:
1. By writing a ‘0’ to the ENABLE bit in the Control A (TCDn.CTRLA) register. This disables the TCD instantly
when synchronized to the TCD core domain.
2. By writing a ‘1’ to the Disable at End of Cycle Strobe (DISEOC) bit in the Control E (TCDn.CTRLE) register.
This disables the TCD at the end of the TCD cycle.
25.3.3.4 TCD Inputs
The TCD has two inputs connected to the Event System: input A and input B. Each input has a functionality
connected to the corresponding output (WOA and WOB). This functionality is controlled by the Event Control
(TCDn.EVCTRLA and TCDn.EVCTRLB) registers and the Input Control (TCDn.INPUTCTRLA and
TCDn.INPUTCTRLB) registers.
To enable the input events, write a ‘1’ to the Trigger Event Input Enable (TRIGEI) bit in the corresponding Event
Control (TCDn.EVCTRLA or TCDn.EVCTRLB) register. The inputs will be used as a Fault detect by default, but they
can also be used as a capture trigger. To enable a capture trigger, write a ‘1’ to the ACTION bit in the corresponding
Event Control (TCDn.EVCTRLA or TCDn.EVCTRLB) register. To disable Fault detect, the INPUTMODE bit field in
the corresponding Input Control (TCDn.INPUTCTRLA or TCDn.INPUTCTRLB) register must be written to ‘0’.
There are ten different input modes for the Fault detection. The two inputs have the same functionality, except for
input blanking which is only supported by input A. Input blanking is configured by the Delay Control
(TCDn.DLYCTRL) register and the Delay Value (TCDn.DLYVAL) register.
The inputs are connected to the Event System. The connections between the event source and the TCD input must
be configured in the Event System.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 309
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
Figure 25-9. TCD Input Overview
EVCTRLA.EDGE
Asynchonous overrride
EVCTRLA.ASYNC
Input Event A
INPUT
BLANKING
Input processing logic
(Input mode logic A)
Digital
Filter
EVCTRLA.FILTER
DLYPRESC
Change flow
INPUT
MODE
DLYTRIG
Synchronized
override
TC Core
(Timer/Counter,
compare values,
waveform generator)
DLYSEL
Output state
Output
control
INPUT
MODE
EVCTRLB.FILTER
Digital
Filter
Input Event B
EVCTRLB.EDGE
EVCTRLB.ASYNC
Change flow
Synchronized
override
Input processing logic
(Input mode logic B)
Asynchonous overrride
There is a delay of two/three clock cycles on the TCD synchronizer clock between receiving the input event,
processing it, and overriding the outputs. If using the asynchronous event detection, the outputs will override instantly
outside the input processing.
25.3.3.4.1 Input Blanking
Input blanking functionality masks out the input events for a programmable time in a selectable part of the TCD cycle.
Input blanking can be used to mask out ‘false’ input events triggered right after changes on the outputs occur.
Input blanking can be enabled by configuring the Delay Select (DLYSEL) bit field in the Delay Control
(TCDn.DLYCTRL) register. The trigger source is selected by the Delay Trigger (DLYTRIG) bit field in
TCDn.DLYCTRL.
Input blanking uses the delay clock. After a trigger, a counter counts up until the Delay Value (DLYVAL) bit field in the
Delay Value (TCDn.DLYVAL) register is reached. Afterward, input blanking is turned off. The TCD delay clock is a
prescaled version of the synchronizer clock (CLK_TCD_SYNC). The division factor is set by the Delay Prescaler
(DLYPRESC) bit field in the Delay Control (TCDn.DLYCTRL) register. The duration of the input blanking is given by:
�BLANK =
DLYPRESC_division_factor × DLYVAL
�CLK_TCD_SYNC
Input blanking uses the same logic as the programmable output event. For this reason, it is not possible to use both
at the same time.
25.3.3.4.2 Digital Filter
The digital filter for event input x is enabled by writing a ‘1’ to the FILTER bit in the corresponding Event Control
(TCDn.EVCTRLA or TCDn.EVCTRLB) register. When the digital filter is enabled, any pulse lasting less than four
counter clock cycles will be filtered out. Any change on the incoming event will, therefore, take four counter clock
cycles before it affects the input processing logic.
25.3.3.4.3 Asynchronous Event Detection
To enable asynchronous event detection on an input event, the Event Configuration (CFG) bit field in the
corresponding Event Control (TCDn.EVCTRLA or TCDn.EVCTRLB) register must be configured accordingly.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 310
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
The asynchronous event detection makes it possible to asynchronously override the output when the input event
occurs. What the input event will do depends on the input mode. The outputs have direct override while the counter
flow will be changed when the event is synchronized to the synchronizer clock (CLK_TCD_SYNC).
It is not possible to use asynchronous event detection and digital filter at the same time.
25.3.3.4.4 Software Commands
The following table displays the commands for the TCD module.
Table 25-4. Software Commands
Trigger
Software Command
The SYNCEOC bit in the TCDn.CTRLE register
Update the double-buffered registers at the end of the
TCD cycle
The SYNC bit in the TCDn.CTRLE register
Update the double-buffered registers
The RESTART bit in the TCDn.CTRLE register
Restart the TCD counter
The SCAPTUREA bit in the TCDn.CTRLE register
Capture to Capture A (TCDn.CAPTUREAL/H) register
The SCAPTUREB bit in the TCDn.CTRLE register
Capture to Capture B (TCDn.CAPTUREBL/H) register
25.3.3.4.5 Input Modes
The user can select between ten input modes. The selection is done by writing to the Input Mode (INPUTMODE) bit
field in the Input Control (TCDn.INPUTCTRLA and TCDn.INPUTCTRLB) registers.
Input Modes Validity
Not all input modes work in all Waveform Generation modes. The table below shows the Waveform Generation
modes in which the different input modes are valid.
Table 25-5. Input Modes Validity
INPUTMODE
One Ramp Mode
Two Ramp Mode
Four Ramp Mode
Dual Slope Mode
0
Valid
Valid
Valid
Valid
1
Valid
Valid
Valid
Do not use
2
Do not use
Valid
Valid
Do not use
3
Do not use
Valid
Valid
Do not use
4
Valid
Valid
Valid
Valid
5
Do not use
Valid
Valid
Do not use
6
Do not use
Valid
Valid
Do not use
7
Valid
Valid
Valid
Valid
8
Valid
Valid
Valid
Do not use
9
Valid
Valid
Valid
Do not use
10
Valid
Valid
Valid
Do not use
Input Mode 0: Input Has No Action
In Input mode 0, the inputs do not affect the outputs, but they can still trigger captures and interrupts if enabled.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 311
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
Figure 25-10. Input Mode 0
DTA
OTA
DTB
OTB
DTA
OTA
DTB
OTB
DTA
OTA
DTB
WOA
WOB
INPUT A
INPUT B
Input Mode 1: Stop Output, Jump to Opposite Compare Cycle, and Wait
An input event in Input mode 1 will stop the output signal, jump to the opposite dead time, and wait until the input
event goes low before the TCD counter continues.
If Input mode 1 is used on input A, an event will only have an effect if the TCD is in dead time A or on time A, and it
will only affect the WOA output. When the event is done, the TCD counter starts at dead time B.
Figure 25-11. Input Mode 1 on Input A
DTA
OTA
DTB
OTB
DTA
OTA
Wait
DTB
OTB
DTA
OTA
WOA
WOB
INPUT A
INPUT B
If Input mode 1 is used on input B, an event will only have an effect if the TCD is in dead time B or on time B, and it
will only affect the WOB output. When the event is done, the TCD counter starts at dead time A.
Figure 25-12. Input Mode 1 on Input B
DTA
OTA
DTB
OTB
Wait
DTA
OTA
DTB
OTB
DTA
OTA
WOA
WOB
INPUT A
INPUT B
Input Mode 2: Stop Output, Execute Opposite Compare Cycle, and Wait
An input event in Input mode 2 will stop the output signal, execute to the opposite dead time and on time, and then
wait until the input event goes low before the TCD counter continues. If the input is done before the opposite dead
time and on time have finished, there will be no waiting, but the opposite dead time and on time will continue.
If Input mode 2 is used on input A, an event will only have an effect if the TCD is in dead time A or on time A, and will
only affect the WOA output.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 312
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
Figure 25-13. Input Mode 2 on Input A
DTA
OTA
DTB
OTB
DTA
OTA
DTB
OTB
Wait
DTA
OTA
WOA
WOB
INPUT A
INPUT B
If Input mode 2 is used on input B, an event will only have an effect if the TCD is in dead time B or on time B, and it
will only affect the WOB output.
Figure 25-14. Input Mode 2 on Input B
DTA
OTA
DTB
OTB
DTA
OTA
Wait
DTB
OTB
DTA
OTA
WOA
WOB
INPUT A
INPUT B
Input Mode 3: Stop Output, Execute Opposite Compare Cycle while Fault Active
An input event in Input mode 3 will stop the output signal and start executing the opposite dead time and on time
repetitively, as long as the Fault/input is active. When the input is released, the ongoing dead time and/or on time will
finish, and then the normal flow will start.
If Input mode 3 is used on input A, an event will only have an effect if the TCD is in dead time A or on time A.
Figure 25-15. Input Mode 3 on Input A
DTA
OTA
DTB
OTB
DTA
OTA
DTB
OTB
DTB
OTB
DTA
OTA
WOA
WOB
INPUT A
INPUT B
If Input mode 3 is used on input B, an event will only have an effect if the TCD is in dead time B or on time B.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 313
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
Figure 25-16. Input Mode 3 on Input B
DTA
OTA
DTB
OTB
DTA
OTA
DTA
OTA
DTB
OTB
DTA
OTA
WOA
WOB
INPUT A
INPUT B
Input Mode 4: Stop all Outputs, Maintain Frequency
When Input mode 4 is used, both input A and input B will give the same functionality. An input event will deactivate
the outputs as long as the event is active. The TCD counter will not be affected by events in this input mode.
Figure 25-17. Input Mode 4
DTA
OTA
DTB
OTB
DTA
OTA
DTB
OTB
DTA
OTA
DTB
OTB
WOA
WOB
INPUT A/B
Input Mode 5: Stop all Outputs, Execute Dead Time while Fault Active
When Input mode 5 is used, both input A and input B give the same functionality. The input event stops the outputs
and starts on the opposite dead time if it occurs during an on time. If the event occurs during dead time, the dead
time will continue until the next on time is scheduled to start. Though, if the input is still active, the cycle will continue
with the other dead time. As long as the input event is active, alternating dead times will occur. When the input event
stops, the ongoing dead time will finish, and the next on time will continue in the normal flow.
Figure 25-18. Input Mode 5
DTA
OTA
DTB
OTB
DTA
OTA
DTB
DTA
DTB
DTA
DTB
OTB
WOA
WOB
INPUT A/B
Input Mode 6: Stop All Outputs, Jump to Next Compare Cycle, and Wait
When Input mode 6 is used, both input A and input B will give the same functionality. The input event stops the
outputs and jumps to the opposite dead time if it occurs during an on time. If the event occurs during dead time, the
dead time will continue until the next on time is scheduled to start. As long as the input event is active, the TCD
counter will wait. When the input event stops, the next dead time will start, and normal flow will continue.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 314
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
Figure 25-19. Input Mode 6
DTA
OTA
DTB
Wait
DTA
OTA
Wait
DTB
OTB
DTA
OTA
WOA
WOB
INPUT A/B
Input Mode 7: Stop all Outputs, Wait for Software Action
When Input mode 7 is used, both input A and input B will give the same functionality. The input events stop the
outputs and the TCD counter. It will be stopped until a Restart command is given. If the input event is still high when
the Restart command (RESTART bit in TCDn.CTRLE register) is given, it will stop again. When the TCD counter
restarts, it will always start on dead time A.
Figure 25-20. Input Mode 7
DTA
OTA
DTB
OTB
DTA
OTA
Wait
DTA
OTA
WOA
WOB
INPUT A/B
Software Restart
command
Input Mode 8: Stop Output on Edge, Jump to Next Compare Cycle
In Input mode 8, a positive edge on the input event while the corresponding output is ON will cause the output to stop
and the TCD counter to jump to the opposite dead time.
If Input mode 8 is used on input A and a positive edge on the input event occurs while in on time A, the TCD counter
jumps to dead time B.
Figure 25-21. Input Mode 8 on Input A
DTA
OTA
DTB
OTB
DTA
OTA
DTB
OTB
DTA
OTA
DTB
OTB
WOA
WOB
INPUT A
OR
INPUT A
If Input mode 8 is used on input B and a positive edge on the input event occurs while in on time B, the TCD counter
jumps to dead time A.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 315
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
Figure 25-22. Input Mode 8 on Input B
DTA
OTA
DTB
OTB DTA
OTA
DTB
OTB
DTA
OTA
DTB
OTB
WOA
WOB
INPUT B
OR
INPUT B
Input Mode 9: Stop Output on Edge, Maintain Frequency
In Input mode 9, a positive edge on the input event while the corresponding output is ON will cause the output to stop
during the rest of the on time. The TCD counter will not be affected by the event, only the output.
If Input mode 9 is used on input A and a positive edge on the input event occurs while in on time A, the output will be
OFF for the rest of the on time.
Figure 25-23. Input Mode 9 on Input A
DTA
OTA
DTB
OTB
DTA
OTA
DTB
OTB
DTA
OTA
WOA
WOB
INPUT A
INPUT B
If Input mode 9 is used on input B and a positive edge on the input event occurs while in on time B, the output will be
OFF for the rest of the on time.
Figure 25-24. Input Mode 9 on Input B
DTA
OTA
DTB
OTB
DTA
OTA
DTB
OTB
DTA
OTA
WOA
WOB
INPUT A
INPUT B
Input Mode 10: Stop Output at Level, Maintain Frequency
In Input mode 10, the input event will cause the corresponding output to stop, as long as the input is active. If the
input goes low while there must have been an on time on the corresponding output, the output will be deactivated for
the rest of the on time. The TCD counter is not affected by the event, only the output.
If Input mode 10 is used on input A and an input event occurs, the WOA will be OFF as long as the event lasts. If
released during an on time, it will be OFF for the rest of the on time.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 316
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
Figure 25-25. Input Mode 10 on Input A
DTA
OTA
DTB
OTB
DTA
OTA
DTB
OTB
DTA
OTA
WOA
WOB
INPUT A
INPUT B
If Input mode 10 is used on input B and an input event occurs, the WOB will be OFF as long as the event lasts. If
released during an on time, it will be OFF for the rest of the on time.
Figure 25-26. Input Mode 10 on Input B
DTA
OTA
DTB
OTB
DTA
OTA
DTB
OTB
DTA
OTA
WOA
WOB
INPUT A
INPUT B
Input Mode Summary
Table 25-6 summarizes the conditions, as illustrated in the timing diagrams of the preceding sections.
Table 25-6. Input Mode Summary
INPUTMODE
Trigger → Output Affected
Fault On/Active
Fault Release/Inactive
0
-
No action.
No action.
1
Input A→WOA
End the current on time and wait.
Start with dead time for
the other compare.
End the current on time, execute the
other compare cycle and wait.
Start with dead time for
the current compare.
Input B→WOB
Execute the current on time, then
execute the other compare cycle
repetitively.
Re-enable the current
compare cycle.
Input A→{WOA, WOB}
Deactivate the outputs.
Input B→WOB
2
Input A→WOA
Input B→WOB
3
4
Input A→WOA
Input B→{WOA, WOB}
5
Input A→{WOA, WOB}
Execute dead time only.
Input B→{WOA, WOB}
6
Input A→{WOA, WOB}
End on time and wait.
Input B→{WOA, WOB}
© 2020 Microchip Technology Inc.
Preliminary Datasheet
Start with dead time for
the other compare.
DS40002247A-page 317
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
...........continued
INPUTMODE
Trigger → Output Affected
Fault On/Active
Fault Release/Inactive
7
Input A→{WOA, WOB}
End on time and wait for software
action.
Start with dead time for
the current compare.
Input B→{WOA, WOB}
8
Input A→WOA
Input B→WOB
9
Input A→WOA
Input B→WOB
10
Input A→WOA
other
End the current on time and continue
with the other off time.
Block the current on time and
continue the sequence.
Input B→WOB
Deactivate on time until the end of
the sequence while the trigger is
active.
-
-
-
Note: When using different modes on each event input, take into consideration possible conflicts, keeping in mind
that TCD has a single counter, to avoid unexpected results.
25.3.3.5 Dithering
If it is not possible to achieve the desired frequency because of the prescaler/period selection limitations, dithering
can be used to approximate the desired frequency and reduce the waveform drift.
The dither accumulates the fractional error of the counter clock for each cycle. When the fractional error overflows, an
additional clock cycle is added to the selected part of the TCD cycle.
Example 25-1. Generate 75 kHz from a 10 MHz Clock
If the timer clock frequency is 10 MHz, it will give the timer a resolution of 100 ns. The desired
output frequency is 75 kHz, which means a period of 13333 ns. This period cannot be achieved
with a 100 ns resolution as it would require 133.33 cycles. The output period can be set to either
133 cycles (75.188 kHz) or 134 cycles (74.626 kHz).
It is possible to change the period between the two frequencies manually in the firmware to get an
average output frequency of 75 kHz (change every third period to 134 cycles). The dither can do
this automatically by accumulating the error (0.33 cycles). The accumulator calculates when the
accumulated error is larger than one clock cycle. When that happens, an additional cycle is added
to the timer period.
Figure 25-27. Dither Logic
Overflow
Dither value
ACCUMULATOR REGISTER
The user can select where in the TCD cycle the dither will be added by writing to the Dither Selection (DITHERSEL)
bits in the Dither Control (TCDn.DITCTRL) register:
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 318
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
•
•
•
•
On time B
On time A and B
Dead time B
Dead time A and B
How much the dithering will affect the TCD cycle time depends on what Waveform Generation mode is used (see
Table 25-7). Dithering is not supported in Dual Slope mode.
Table 25-7. Mode-Dependent Dithering Additions to TCD Cycle
WAVEGEN
DITHERSEL in TCDn.DITCTRL
Additional TCD Clock Cycles to TCD Cycle
One Ramp mode
On time B
1
On time A and B
1
Dead time B
0
Dead time A and B
0
On time B
1
On time A and B
2
Dead time B
0
Dead time A and B
0
On time B
1
On time A and B
2
Dead time B
1
Dead time A and B
2
On time B
Not supported
On time A and B
Not supported
Dead time B
Not supported
Dead time A and B
Not supported
Two Ramp mode
Four Ramp mode
Dual Slope mode
The differences in the number of TCD clock cycles added to the TCD cycle are caused by the different number of
compare values used by the TCD cycle. For example, in One Ramp mode, only CMPBCLR affects the TCD cycle
time.
For DITHERSEL configurations where no extra cycles are added to the TCD cycles, compensation is reached by
shortening the following output state.
Example 25-2. DITHERSEL in One Ramp Mode
In One Ramp mode with DITHERSEL selecting dead time B, the dead time B will be increased by
one cycle when dither overflow occurs, reducing on time B by one cycle.
25.3.3.6 TCD Counter Capture
The TCD counter is asynchronous to the peripheral clock, so it is not possible to read out the counter value directly. It
is possible to capture the TCD counter value, synchronized to the I/O clock domain, in two ways:
• Capture value on input events
• Software capture
The capture logic contains two separate capture blocks, CAPTUREA and CAPTUREB, that can capture and
synchronize the TCD counter value to the I/O clock domain. CAPTUREA/B can be triggered by input event A/B or by
software.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 319
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
The capture values can be obtained by reading first TCDn.CAPTUREAL/TCDn.CAPTUREBL and then
TCDn.CAPTUREAH/TCDn.CAPTUREBH registers.
Captures Triggered by Input Events
To enable the capture on an input event, write a ‘1’ to the ACTION bit in the respective Event Control
(TCDn.EVCTRLA or TCDn.EVCTRLB) register when configuring an event input.
When a capture has occurred, the TRIGA/B flag is raised in the Interrupt Flags (TCDn.INTFLAGS) register. The
corresponding TRIGA/B interrupt can be enabled by writing a ‘1’ to the respective Trigger Interrupt Enable (TRIGA or
TRIGB) bit in the Interrupt Control (TCDn.INTCTRL) register. By polling TRIGA or TRIGB in TCDn.INTFLAGS, the
user knows that a CAPTURE value is available, and can read out the value by reading first the TCDn.CAPTUREAL
or TCDn.CAPTUREBL register and then the TCDn.CAPTUREAH or TCDn.CAPTUREBH register.
Example 25-3. PWM Capture
To perform a PWM capture, connect both event A and event B to the same asynchronous event
channel that contains the PWM signal. To get information on the PWM signal, configure one event
input to capture the rising edge of the signal. Configure the other event input to capture the falling
edge of the signal.
TCD cycle
Dead time A
On time A
Dead time B
On time B
Counter
value
Compare
values
EVENT
CMPBCLR
EVENT
CMPBSET
EVENT
* OVF
CMPACLR
EVENT
CMPASET
WOA
WOB
INPUT A
TRIGA*
INPUT B
TRIGA*
* TRIGB
TRIGA*
* TRIGB
* TRIGB
Note:
▲ Event trigger
* Interrupt trigger
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 320
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
Capture Triggered by Software
The software can capture the TCD value by writing a ‘1’ to the respective Software Capture A/B Strobe
(SCAPTUREx) bit in the Control E (TCDn.CTRLE) register. When this command is executed and the Command
Ready (CMDRDY) bit in the Status (TCDn.STATUS) register reads ‘1’ again, the CAPTUREA/B value is available. It
can now be read by reading first the TCDn.CAPTUREAL or TCDn.CAPTUREBL register and then the
TCDn.CAPTUREAH or TCDn.CAPTUREBH register.
Using Capture Together with Input Modes
The capture functionality can be used together with input modes. The same event will then both capture the counter
value and trigger a change in the counter flow, depending on the input mode selected.
Example 25-4. Reset One Ramp Mode by Input Event Capture
In One Ramp mode, the counter can be reset by an input event capture. To achieve this, use input
event B and write 0x08 to the INPUTMODE bit field in the Input Control B (TCDn.INPUTCTRLB)
register.
DTA
OTA
DTB
DTA
OTA
Counter
value
CMPBCLR
CMPBSET
CMPACLR
CMPASET
INPUT B
25.3.3.7 Output Control
The outputs are configured by writing to the Fault Control (TCDn.FAULTCTRL) register.
The Compare x Enable (CMPxEN) bits in TCDn.FAULTCTRL enable the different outputs. The CMPx bits in
TCDn.FAULTCTRL set the output values when a Fault is triggered.
The TCD itself generates two different outputs, WOA and WOB. The two additional outputs, WOC and WOD, can be
configured by software to be connected to either WOA or WOB by writing the Compare C/D Output Select
(CMPCSEL and CMPDSEL) bits in the Control C (TCDn.CTRLC) register.
The user can override the outputs based on the TCD counter state by writing a ‘1’ to the Compare Output Value
Override (CMPOVR) bit in the Control C (TCDn.CTRLC) register. The user can then select the output values in the
different dead and on times by writing to the Compare Value (CMPAVAL and CMPBVAL) bit fields in the Control D
(TCDn.CTRLD) register.
When used in One Ramp mode, WOA will only use the setup for dead time A (DTA) and on time A (OTA) to set the
output. WOB will only use dead time B (DTB) and on time B (OTB) values to set the output.
When using the override feature together with Faults detection (input modes), the CMPA (and CMPC/D if WOC/D
equals WOA) bit in TCDn.FAULTCTRL must be equal to CMPAVAL[0] and [2] in CTRL. If not, the first cycle after a
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 321
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
Fault is detected can have the wrong polarity on the outputs. The same applies to CMPB in the TCDn.FAULTCTRL
(and CMPC/D if WOC/D equals WOB) bit, which must be equal to CMPBVAL[0] and [2] in TCDn.CTRLD.
Due to the asynchronous nature of the TCD and that input events can immediately affect the output signal, there is a
risk of nanosecond spikes occurring on the output without any load on the pin. The case occurs in any input mode
different from ‘0’ and when an input event is triggering. The spike value will always be in the direction of the CMPx
values given by the TCDn.FAULTCTRL register.
25.3.4
Events
The TCD can generate the events described in the following table:
Table 25-8. Event Generators in TCD
Generator Name
Peripheral
TCDn
Description
Event
CMPBCLR
The counter matches
CMPBCLR
CMPASET
The counter matches
CMPASET
Event
Type
Generating
Clock Domain
Pulse
CLK_TCD
Length of Event
One CLK_TCD_CNT period
CMPBSET The counter matches
CMPBSET
PROGEV
Programmable event output(1)
One CLK_TCD_SYNC period
Note:
1. The user can select the trigger and all the compare matches (including CMPACLR). Also, it is possible to delay
the output event from 0 to 255 TCD delay cycles.
The three events based on the counter match directly generate event strobes that last for one clock cycle on the TCD
counter clock. The programmable output event generates an event strobe that lasts for one clock cycle on the TCD
synchronizer clock.
The TCD can receive the events described in the following table:
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 322
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
Table 25-9. Event Users and Available Event Actions in TCD
User Name
Peripheral
Description
Input
Input Detection Async/Sync
Stop the output, jump to the opposite compare cycle
and wait.
Stop the output, execute the opposite compare cycle
and wait.
Stop the output, execute the opposite compare cycle
while the Fault is active.
Stop all outputs, maintain the frequency.
TCDn
Level
Stop all outputs, execute dead time while the Fault is
Input A/ Input B active.
Both
Stop all outputs, jump to the next compare cycle and
wait.
Stop all outputs, wait for software action.
Stop the output on the edge, jump to the next
compare cycle.
Edge
Stop the output on the edge, maintain the frequency.
Stop the output at level, maintain the frequency.
Level
Input A and Input B are TCD event users that detect and act upon the input events. Additional information about input
events and how to configure them can be found in the 25.3.3.4 TCD Inputs section. Refer to the Event System
(EVSYS) section for more details regarding event types and Event System configuration.
25.3.4.1 Programmable Output Events
The Programmable Output Event (PROGEV) uses the same logic as the input blanking for trigger selection and
delay. Therefore, it is not possible to configure the functionalities independently. If the input blanking functionality is
used, the output event cannot be delayed, and the trigger used for input blanking will also be used for the output
event.
PROGEV is configured in the TCDn.DLYCTRL and TCDn.DLYVAL registers. It is possible to delay the output event
by 0 to 255 TCD delay clock cycles. The delayed output event functionality uses the TCD delay clock and counts until
the DLYVAL value is reached before the trigger is sent out as an event. The TCD delay clock is a prescaled version
of the TCD synchronizer clock (CLK_TCD_SYNC), and the division factor is set by the DLYPRESC bits in the
TCDn.DLYCTRL register. The output event will be delayed by the TCD clock period x DLYPRESC division factor x
DLYVAL.
25.3.5
Interrupts
Table 25-10. Available Interrupt Vectors and Sources
Name
Vector Description
Conditions
OVF
Overflow interrupt
The TCD finishes one TCD cycle.
TRIG
Trigger interrupt
•
•
TRIGA: On event input A
TRIGB: On event input B
When an interrupt condition occurs, the corresponding interrupt flag is set in the Interrupt Flags (TCDn.INTFLAGS)
register.
An interrupt source is enabled or disabled by writing to the corresponding enable bit in the Interrupt Control
(TCDn.INTCTRL) register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 323
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
An interrupt request is generated when the corresponding interrupt source is enabled, and the interrupt flag is set.
The interrupt request remains active until the interrupt flag is cleared. See the peripheral’s INTFLAGS register for
details on how to clear interrupt flags.
When several interrupt request conditions are supported by an interrupt vector, the interrupt requests are ORed
together into one combined interrupt request to the interrupt controller. The user must read the peripheral’s
INTFLAGS register to determine which of the interrupt conditions are present.
25.3.6
Sleep Mode Operation
The TCD operates in Idle sleep mode and is stopped when entering Standby and Power-Down sleep modes.
25.3.7
Debug Operation
Halting the CPU in Debugging mode will halt the normal operation of the peripheral. This peripheral can be forced to
operate with the CPU halted by writing a ‘1’ to the Debug Run (DBGRUN) bit in the Debug Control
(TCDn.DBGCTRL) register.
When the Fault Detection (FAULTDET) bit in TCDn.DBGCTRL is written to ‘1’, and the CPU is halted in Debug mode,
an event/Fault is created on both input event channels. These events/Faults last as long as the break and can serve
as a safeguard in Debug mode, for example, by forcing external components off.
If the peripheral is configured to require periodic service by the CPU through interrupts or similar, improper operation
or data loss may result during halted debugging.
25.3.8
Configuration Change Protection
This peripheral has registers that are under Configuration Change Protection (CCP). To write to these registers, a
certain key must first be written to the CPU.CCP register, followed by a write access to the protected bits within four
CPU instructions.
Attempting to write to a protected register without following the appropriate CCP unlock sequence leaves the
protected register unchanged.
The following registers are under CCP:
Table 25-11. Registers under Configuration Change Protection in TCD
Register
Key
TCDn.FAULTCTRL
IOREG
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 324
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
25.4
Register Summary
Offset
Name
Bit Pos.
0x00
0x01
0x02
0x03
0x04
0x05
...
0x07
0x08
0x09
0x0A
...
0x0B
0x0C
0x0D
0x0E
0x0F
0x10
0x11
0x12
0x13
0x14
0x15
0x16
...
0x17
0x18
0x19
0x1A
...
0x1D
0x1E
0x1F
...
0x21
CTRLA
CTRLB
CTRLC
CTRLD
CTRLE
7:0
7:0
7:0
7:0
7:0
0x22
EVCTRLA
EVCTRLB
INTCTRL
INTFLAGS
STATUS
Reserved
INPUTCTRLA
INPUTCTRLB
FAULTCTRL
Reserved
DLYCTRL
DLYVAL
DITCTRL
DITVAL
CMPDSEL
3
CNTPRES[1:0]
CMPCSEL
CMPBVAL[3:0]
DISEOC
FIFTY
SCAPTUREB SCAPTUREA
2
1
0
SYNCPRES[1:0]
ENABLE
WGMODE[1:0]
AUPDATE
CMPOVR
CMPAVAL[3:0]
RESTART
SYNC
SYNCEOC
7:0
7:0
CFG[1:0]
CFG[1:0]
EDGE
EDGE
ACTION
ACTION
TRIGEI
TRIGEI
7:0
7:0
7:0
7:0
7:0
7:0
TRIGB
TRIGB
TRIGA
TRIGA
OVF
OVF
ENRDY
PWMACTB
CMPDxEN
7:0
7:0
PWMACTA
CMPCxEN
CMDRDY
CMPBxEN
CMPAxEN
CMPDx
DLYPRESC[1:0]
INPUTMODE[3:0]
INPUTMODE[3:0]
CMPCx
CMPBx
DLYTRIG[1:0]
CMPAx
DLYSEL[1:0]
DLYVAL[7:0]
7:0
7:0
DITHERSEL[1:0]
DITHER[3:0]
Reserved
DBGCTRL
7:0
FAULTDET
DBGRUN
Reserved
CAPTUREA
Reserved
0x28
CMPASET
25.5
CLKSEL[1:0]
4
Reserved
0x26
...
0x27
0x2E
5
Reserved
CAPTUREB
0x2C
6
Reserved
0x24
0x2A
7
CMPACLR
CMPBSET
CMPBCLR
7:0
15:8
7:0
15:8
CAPTUREA[7:0]
7:0
15:8
7:0
15:8
7:0
15:8
7:0
15:8
CMPASET[7:0]
CAPTUREA[11:8]
CAPTUREB[7:0]
CAPTUREB[11:8]
CMPASET[11:8]
CMPACLR[7:0]
CMPACLR[11:8]
CMPBSET[7:0]
CMPBSET[11:8]
CMPBCLR[7:0]
CMPBCLR[11:8]
Register Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 325
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
25.5.1
Control A
Name:
Offset:
Reset:
Property:
Bit
7
Access
Reset
CTRLA
0x00
0x00
Enable-protected
6
5
CLKSEL[1:0]
R/W
R/W
0
0
4
3
CNTPRES[1:0]
R/W
R/W
0
0
2
1
SYNCPRES[1:0]
R/W
R/W
0
0
0
ENABLE
R/W
0
Bits 6:5 – CLKSEL[1:0] Clock Select
The Clock Select bits select the clock source of the TCD clock.
Value
Name
Description
0x0
0x1
0x2
0x3
OSCHF
PLL
EXTCLK
CLKPER
Internal High-Frequency Oscillator
PLL
External Clock or external crystal oscillator
Main clock after prescaler (CLK_PER)
Bits 4:3 – CNTPRES[1:0] Counter Prescaler
The Counter Prescaler bits select the division factor of the TCD counter clock.
Value
Name
Description
0x0
DIV1
Division factor 1
0x1
DIV4
Division factor 4
0x2
DIV32
Division factor 32
0x3
Reserved
Bits 2:1 – SYNCPRES[1:0] Synchronization Prescaler
The Synchronization Prescaler bits select the division factor of the TCD clock.
Value
Name
Description
0x0
DIV1
Division factor 1
0x1
DIV2
Division factor 2
0x2
DIV4
Division factor 4
0x3
DIV8
Division factor 8
Bit 0 – ENABLE Enable
When writing to this bit, it will automatically be synchronized to the TCD clock domain.
This bit can be changed as long as the synchronization of this bit is not ongoing. See the Enable Ready (ENRDY) bit
in the Status (TCDn.STATUS) register.
This bit is not enable-protected.
Value
Name
Description
0
NO
The TCD is disabled.
1
YES
The TCD is enabled and running.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 326
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
25.5.2
Control B
Name:
Offset:
Reset:
Property:
Bit
7
CTRLB
0x01
0x00
-
6
5
4
3
Access
Reset
Bits 1:0 – WGMODE[1:0] Waveform Generation Mode
These bits select the waveform generation.
Value
Name
0x0
ONERAMP
0x1
TWORAMP
0x2
FOURRAMP
0x3
DS
© 2020 Microchip Technology Inc.
2
1
0
WGMODE[1:0]
R/W
R/W
0
0
Description
One Ramp mode
Two Ramp mode
Four Ramp mode
Dual Slope mode
Preliminary Datasheet
DS40002247A-page 327
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
25.5.3
Control C
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
CMPDSEL
R/W
0
CTRLC
0x02
0x00
-
6
CMPCSEL
R/W
0
5
4
3
FIFTY
R/W
0
2
1
AUPDATE
R/W
0
0
CMPOVR
R/W
0
Bit 7 – CMPDSEL Compare D Output Select
This bit selects which waveform will be connected to output D.
Value
Name
Description
0
PWMA
Waveform A
1
PWMB
Waveform B
Bit 6 – CMPCSEL Compare C Output Select
This bit selects which waveform will be connected to output C.
Value
Name
Description
0
PWMA
Waveform A
1
PWMB
Waveform B
Bit 3 – FIFTY Fifty Percent Waveform
If the two waveforms have identical characteristics, this bit can be written to ‘1’. This will cause any values written to
the TCDn.CMPBSET/TCDn.CLR register to also be written to the TCDn.CMPASET/TCDn.CLR register.
Bit 1 – AUPDATE Automatically Update
If this bit is written to ‘1’, synchronization at the end of the TCD cycle is automatically requested after the Compare B
Clear High (TCDn.CMPBCLRH) register is written.
If the fifty percent waveform is enabled by setting the FIFTY bit in this register, writing the Compare A Clear High
register will also request a synchronization at the end of the TCD cycle if the AUPDATE bit is set.
Bit 0 – CMPOVR Compare Output Value Override
When this bit is written to ‘1’, default values of the Waveform Outputs A and B are overridden by the values written in
the Compare x Value in Active state bit fields in the Control D register. See the 25.5.4 CTRLD register description for
more details.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 328
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
25.5.4
Control D
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
R/W
0
CTRLD
0x03
0x00
-
6
5
CMPBVAL[3:0]
R/W
R/W
0
0
4
3
R/W
0
R/W
0
2
1
CMPAVAL[3:0]
R/W
R/W
0
0
0
R/W
0
Bits 0:3, 4:7 – CMPVAL Compare x Value (in Active state)
These bits set the logical value of the PWMx signal for the corresponding states in the TCD cycle.
These settings are valid only if the Compare Output Value Override (CMPOVR) bit in the Control C (TCDn.CTRLC)
register is written to ‘1’.
Table 25-12. Two and Four Ramp Mode
CMPxVAL
DTA
OTA
DTB
OTB
PWMA
PWMB
CMPAVAL[0]
CMPBVAL[0]
CMPAVAL[1]
CMPBVAL[1]
CMPAVAL[2]
CMPBVAL[2]
CMPAVAL[3]
CMPBVAL[3]
When used in One Ramp mode, WOA will only use the setup for dead time A (DTA) and on time A (OTA) to set the
output. WOB will only use dead time B (DTB) and on time B (OTB) values to set the output.
Table 25-13. One Ramp Mode
CMPxVAL
DTA
OTA
DTB
OTB
PWMA
PWMB
CMPAVAL[1]
-
CMPAVAL[0]
-
CMPBVAL[3]
CMPBVAL[2]
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 329
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
25.5.5
Control E
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
DISEOC
R/W
0
CTRLE
0x04
0x00
-
6
5
4
SCAPTUREB
R/W
0
3
SCAPTUREA
R/W
0
2
RESTART
R/W
0
1
SYNC
R/W
0
0
SYNCEOC
R/W
0
Bit 7 – DISEOC Disable at End of TCD Cycle Strobe
When this bit is written to ‘1’, the TCD will automatically disable at the end of the TCD cycle.
Note that ENRDY in TCDn.STATUS will stay low until the TCD is disabled.
Writing to this bit has effect only if there is no ongoing synchronization of the ENABLE value in TCDn.CTRLA with the
TCD domain. See also the ENRDY bit in TCDn.STATUS.
Bit 4 – SCAPTUREB Software Capture B Strobe
When this bit is written to ‘1’, a software capture to the Capture B (TCDn.CAPTUREBL/H) register is triggered as
soon as synchronization to the TCD clock domain occurs.
Writing to this bit has effect only if there is no ongoing synchronization of a command. See also the CMDRDY bit in
TCDn.STATUS.
Bit 3 – SCAPTUREA Software Capture A Strobe
When this bit is written to ‘1’, a software capture to the Capture A (TCDn.CAPTUREAL/H) register is triggered as
soon as synchronization to the TCD clock domain occurs.
Writing to this bit has effect only if there is no ongoing synchronization of a command. See also the CMDRDY bit in
TCDn.STATUS.
Bit 2 – RESTART Restart Strobe
When this bit is written to ‘1’, a restart of the TCD counter is executed as soon as this bit is synchronized to the TCD
domain.
Writing to this bit has effect only if there is no ongoing synchronization of a command. See also the CMDRDY bit in
TCDn.STATUS.
Bit 1 – SYNC Synchronize Strobe
When this bit is written to ‘1’, the double-buffered registers will be loaded to the TCD domain as soon as this bit is
synchronized to the TCD domain.
Writing to this bit has effect only if there is no ongoing synchronization of a command. See also the CMDRDY bit in
TCDn.STATUS.
Bit 0 – SYNCEOC Synchronize End of TCD Cycle Strobe
When this bit is written to ‘1’, the double-buffered registers will be loaded to the TCD domain at the end of the next
TCD cycle.
Writing to this bit has effect only if there is no ongoing synchronization of a command. See also the CMDRDY bit in
TCDn.STATUS.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 330
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
25.5.6
Event Control A
Name:
Offset:
Reset:
Property:
Bit
EVCTRLA
0x08
0x00
-
7
6
CFG[1:0]
Access
Reset
R/W
0
R/W
0
5
4
EDGE
R/W
0
3
2
ACTION
R/W
0
1
0
TRIGEI
R/W
0
Bits 7:6 – CFG[1:0] Event Configuration
When the input capture noise canceler is activated (FILTERON), the event input is filtered. The filter function requires
four successive equal valued samples of the trigger pin to change its output. The input capture is, therefore, delayed
by four clock cycles when the noise canceler is enabled (FILTERON).
When the Asynchronous Event is enabled (ASYNCON), the event input will affect the output directly.
Value
Name
Description
0x0
NEITHER
Neither filter nor asynchronous event is enabled.
0x1
FILTERON
Input capture noise cancellation filter enabled.
0x2
ASYNCON
Asynchronous event output qualification enabled.
other
Reserved.
Bit 4 – EDGE Edge Selection
This bit is used to select the active edge or level for the event input.
Value
Name
Description
0
FALL_LOW The falling edge or low level of the event input triggers a Capture or Fault action.
1
RISE_HIGH The rising edge or high level of the event input triggers a Capture or Fault action.
Bit 2 – ACTION Event Action
This bit enables capturing on the event input. By default, the input will trigger a Fault, depending on the Input Control
register’s Input mode. It is also possible to trigger a capture on the event input.
Value
Name
Description
0
FAULT
Event triggers a Fault.
1
CAPTURE
Event triggers a Fault and capture.
Bit 0 – TRIGEI Trigger Event Input Enable
Writing this bit to ‘1’ enables event as the trigger for input A.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 331
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
25.5.7
Event Control B
Name:
Offset:
Reset:
Property:
Bit
EVCTRLB
0x09
0x00
-
7
6
CFG[1:0]
Access
Reset
R/W
0
R/W
0
5
4
EDGE
R/W
0
3
2
ACTION
R/W
0
1
0
TRIGEI
R/W
0
Bits 7:6 – CFG[1:0] Event Configuration
When the input capture noise canceler is activated (FILTERON), the event input is filtered. The filter function requires
four successive equal valued samples of the trigger pin to change its output. The input capture is, therefore, delayed
by four clock cycles when the noise canceler is enabled (FILTERON).
When the Asynchronous Event is enabled (ASYNCON), the event input will affect the output directly.
Value
Name
Description
0x0
NEITHER
Neither filter nor asynchronous event is enabled.
0x1
FILTERON
Input capture noise cancellation filter enabled.
0x2
ASYNCON
Asynchronous event output qualification enabled.
other
Reserved.
Bit 4 – EDGE Edge Selection
This bit is used to select the active edge or level for the event input.
Value
Name
Description
0
FALL_LOW The falling edge or low level of the event input triggers a Capture or Fault action.
1
RISE_HIGH The rising edge or high level of the event input triggers a Capture or Fault action.
Bit 2 – ACTION Event Action
This bit enables capturing on the event input. By default, the input will trigger a Fault, depending on the Input Control
register’s Input mode. It is also possible to trigger a capture on the event input.
Value
Name
Description
0
FAULT
Event triggers a Fault.
1
CAPTURE
Event triggers a Fault and capture.
Bit 0 – TRIGEI Trigger Event Input Enable
Writing this bit to ‘1’ enables event as a trigger for input B.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 332
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
25.5.8
Interrupt Control
Name:
Offset:
Reset:
Property:
Bit
7
INTCTRL
0x0C
0x00
-
6
Access
Reset
5
4
3
TRIGB
R/W
0
2
TRIGA
R/W
0
1
0
OVF
R/W
0
Bit 3 – TRIGB Trigger B Interrupt Enable
Writing this bit to ‘1’ enables the interrupt when trigger input B is received.
Bit 2 – TRIGA Trigger A Interrupt Enable
Writing this bit to ‘1’ enables the interrupt when trigger input A is received.
Bit 0 – OVF Counter Overflow
Writing this bit to ‘1’ enables the restart-of-sequence interrupt or overflow interrupt.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 333
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
25.5.9
Interrupt Flags
Name:
Offset:
Reset:
Property:
Bit
7
INTFLAGS
0x0D
0x00
-
6
Access
Reset
5
4
3
TRIGB
R/W
0
2
TRIGA
R/W
0
1
0
OVF
R/W
0
Bit 3 – TRIGB Trigger B Interrupt Flag
The Trigger B Interrupt (TRIGB) flag is set on a Trigger B or Capture B condition. The flag is cleared by writing a ‘1’ to
its bit location.
Bit 2 – TRIGA Trigger A Interrupt Flag
The Trigger A Interrupt (TRIGA) flag is set on a Trigger A or Capture A condition. The flag is cleared by writing a ‘1’ to
its bit location.
Bit 0 – OVF Overflow Interrupt Flag
The Overflow Flag (OVF) is set at the end of a TCD cycle. The flag is cleared by writing a ‘1’ to its bit location.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 334
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
25.5.10 Status
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
PWMACTB
R/W
0
STATUS
0x0E
0x00
-
6
PWMACTA
R/W
0
5
4
3
2
1
CMDRDY
R
0
0
ENRDY
R
0
Bit 7 – PWMACTB PWM Activity on B
This bit is set by hardware each time the WOB output toggles from ‘0’ to ‘1’ or from ‘1’ to ‘0’.
This status bit must be cleared by software by writing a ‘1’ to it before new PWM activity can be detected.
Bit 6 – PWMACTA PWM Activity on A
This bit is set by hardware each time the WOA output toggles from ‘0’ to ‘1’ or from ‘1’ to ‘0’.
This status bit must be cleared by software by writing a ‘1’ to it before new PWM activity can be detected.
Bit 1 – CMDRDY Command Ready
This status bit tells when a command is synced to the TCD domain and the system is ready to receive new
commands.
The following actions clear the CMDRDY bit:
1. TCDn.CTRLE SYNCEOC strobe.
2. TCDn.CTRLE SYNC strobe.
3. TCDn.CTRLE RESTART strobe.
4. TCDn.CTRLE SCAPTUREA Capture A strobe.
5. TCDn.CTRLE SCAPTUREB Capture B strobe.
6. TCDn.CTRLC AUPDATE written to ‘1’ and writing to the TCDn.CMPBCLRH register.
Bit 0 – ENRDY Enable Ready
This status bit tells when the ENABLE value in TCDn.CTRLA is synced to the TCD domain and is ready to be written
to again.
The following actions clear the ENRDY bit:
1. Writing to the ENABLE bit in TCDn.CTRLA.
2. TCDn.CTRLE DISEOC strobe.
3. Going into BREAK in an On-Chip Debugging (OCD) session while the Debug Run (DBGCTRL) bit in
TCDn.DBGCTRL is ‘0’.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 335
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
25.5.11 Input Control A
Name:
Offset:
Reset:
Property:
Bit
7
INPUTCTRLA
0x10
0x00
-
6
Access
Reset
5
4
3
R/W
0
2
1
INPUTMODE[3:0]
R/W
R/W
0
0
0
R/W
0
Bits 3:0 – INPUTMODE[3:0] Input Mode
Value
Name
Description
0x0
NONE
The input has no action.
0x1
JMPWAIT
Stop the output, jump to the opposite compare cycle, and wait.
0x2
EXECWAIT
Stop the output, execute the opposite compare cycle, and wait.
0x3
EXECFAULT
Stop the output, execute the opposite compare cycle while the Fault is active.
0x4
FREQ
Stop all outputs, maintain the frequency.
0x5
EXECDT
Stop all outputs, execute dead time while the Fault is active.
0x6
WAIT
Stop all outputs, jump to the next compare cycle, and wait.
0x7
WAITSW
Stop all outputs, wait for software action.
0x8
EDGETRIG
Stop the output on the edge, jump to the next compare cycle.
0x9
EDGETRIGFREQ Stop the output on the edge, maintain the frequency.
0xA
LVLTRIGFREQ
Stop the output at level, maintain the frequency.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 336
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
25.5.12 Input Control B
Name:
Offset:
Reset:
Property:
Bit
7
INPUTCTRLB
0x11
0x00
-
6
Access
Reset
5
4
3
R/W
0
2
1
INPUTMODE[3:0]
R/W
R/W
0
0
0
R/W
0
Bits 3:0 – INPUTMODE[3:0] Input Mode
Value
Name
Description
0x0
NONE
The input has no action.
0x1
JMPWAIT
Stop the output, jump to the opposite compare cycle, and wait.
0x2
EXECWAIT
Stop the output, execute the opposite compare cycle, and wait.
0x3
EXECFAULT
Stop the output, execute the opposite compare cycle while the Fault is active.
0x4
FREQ
Stop all outputs, maintain the frequency.
0x5
EXECDT
Stop all outputs, execute dead time while the Fault is active.
0x6
WAIT
Stop all outputs, jump to the next compare cycle, and wait.
0x7
WAITSW
Stop all outputs, wait for software action.
0x8
EDGETRIG
Stop the output on the edge, jump to the next compare cycle.
0x9
EDGETRIGFREQ Stop the output on the edge, maintain the frequency.
0xA
LVLTRIGFREQ
Stop the output at level, maintain the frequency.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 337
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
25.5.13 Fault Control
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
CMPDxEN
R/W
0
FAULTCTRL
0x12
0x00
Configuration Change Protection
6
CMPCxEN
R/W
0
5
CMPBxEN
R/W
0
4
CMPAxEN
R/W
0
3
CMPDx
R/W
0
2
CMPCx
R/W
0
1
CMPBx
R/W
0
0
CMPAx
R/W
0
Bits 4, 5, 6, 7 – CMPxEN Compare x Enable
These bits enable the waveform from compare as output on the pin.
Bits 0, 1, 2, 3 – CMPx Compare x Value
These bits set the default state of the compare waveform output.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 338
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
25.5.14 Delay Control
Name:
Offset:
Reset:
Property:
Bit
7
DLYCTRL
0x14
0x00
-
6
Access
Reset
5
4
DLYPRESC[1:0]
R/W
R/W
0
0
3
2
DLYTRIG[1:0]
R/W
R/W
0
0
1
0
DLYSEL[1:0]
R/W
R/W
0
0
Bits 5:4 – DLYPRESC[1:0] Delay Prescaler
These bits control the prescaler settings for the blanking or output event delay.
Value
Name
Description
0x0
DIV1
Prescaler division factor 1
0x1
DIV2
Prescaler division factor 2
0x2
DIV4
Prescaler division factor 4
0x3
DIV8
Prescaler division factor 8
Bits 3:2 – DLYTRIG[1:0] Delay Trigger
These bits control the trigger of the blanking or output event delay.
Value
Name
Description
0x0
CMPASET
CMPASET triggers delay
0x1
CMPACLR
CMPACLR triggers delay
0x2
CMPBSET
CMPBSET triggers delay
0x3
CMPBCLR
CMPASET triggers delay (end of cycle)
Bits 1:0 – DLYSEL[1:0] Delay Select
These bits control what function must be used by the delay trigger, the blanking or output event delay.
Value
Name
Description
0x0
OFF
Delay functionality not used
0x1
INBLANK
Input blanking enabled
0x2
EVENT
Event delay enabled
0x3
Reserved
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 339
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
25.5.15 Delay Value
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
DLYVAL
0x15
0x00
-
7
6
5
R/W
0
R/W
0
R/W
0
4
3
DLYVAL[7:0]
R/W
R/W
0
0
2
1
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – DLYVAL[7:0] Delay Value
These bits configure the blanking/output event delay time or event output synchronization delay in a number of
prescaled TCD cycles.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 340
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
25.5.16 Dither Control
Name:
Offset:
Reset:
Property:
Bit
7
DITCTRL
0x18
0x00
-
6
5
4
3
Access
Reset
2
1
0
DITHERSEL[1:0]
R/W
R/W
0
0
Bits 1:0 – DITHERSEL[1:0] Dither Select
This bit field selects which state of the TCD cycle will benefit from the dither function. See the 25.3.3.5 Dithering
section.
Value
Name
Description
0x0
ONTIMEB
On time ramp B
0x1
ONTIMEAB
On time ramp A and B
0x2
DEADTIMEB
Dead time ramp B
0x3
DEADTIMEAB
Dead time ramp A and B
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 341
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
25.5.17 Dither Value
Name:
Offset:
Reset:
Property:
Bit
7
DITVAL
0x19
0x00
-
6
Access
Reset
5
4
3
R/W
0
2
1
DITHER[3:0]
R/W
R/W
0
0
0
R/W
0
Bits 3:0 – DITHER[3:0] Dither Value
These bits configure the fractional adjustment of the on time or off time, according to the Dither Selection
(DITHERSEL) bits in the Dither Control (TCDn.DITCTRL) register. The DITHER value is added to a 4-bit accumulator
at the end of each TCD cycle. When the accumulator overflows, the frequency adjustment will occur.
The DITHER bits are double-buffered, so the new value is copied when an update condition occurs.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 342
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
25.5.18 Debug Control
Name:
Offset:
Reset:
Property:
Bit
7
DBGCTRL
0x1E
0x00
-
6
5
4
3
Access
Reset
2
FAULTDET
R/W
0
1
0
DBGRUN
R/W
0
Bit 2 – FAULTDET Fault Detection
This bit defines how the peripheral behaves when stopped in Debug mode.
Value
Name Description
0
NONE No Fault is generated if TCD is stopped in Debug mode.
1
FAULT A Fault is generated, and both trigger flags are set, if TCD is halted in Debug mode.
Bit 0 – DBGRUN Debug Run
When written to ‘1’, the peripheral will continue operating in Debug mode when the CPU is halted.
Value
Description
0
The peripheral is halted in Break Debug mode and ignores events.
1
The peripheral will continue to run in Break Debug mode when the CPU is halted.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 343
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
25.5.19 Capture A
Name:
Offset:
Reset:
Property:
CAPTUREA
0x22
0x00
-
The TCDn.CAPTUREAL and TCDn.CAPTUREAH register pair represents the 12-bit TCDn.CAPTUREA value.
For capture operation, these registers constitute the second buffer level and access point for the CPU. The
TCDn.CAPTUREA registers are updated with the buffer value when an update condition occurs. The CAPTURE A
register contains the TCD counter value when a trigger A or software capture A occurs.
The TCD counter value is synchronized to CAPTUREA by either software or an event.
The capture register is blocked for an update of new capture data until the higher byte of this register is read.
Bit
15
14
13
Access
Reset
12
11
R
0
Bit
7
6
5
Access
Reset
R
0
R
0
R
0
4
3
CAPTUREA[7:0]
R
R
0
0
10
9
CAPTUREA[11:8]
R
R
0
0
8
R
0
2
1
0
R
0
R
0
R
0
Bits 11:0 – CAPTUREA[11:0] Capture A Value
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 344
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
25.5.20 Capture B
Name:
Offset:
Reset:
Property:
CAPTUREB
0x24
0x00
-
The TCDn.CAPTUREBL and TCDn.CAPTUREBH register pair represents the 12-bit TCDn.CAPTUREB value.
For capture operation, these registers constitute the second buffer level and access point for the CPU. The
TCDn.CAPTUREB registers are updated with the buffer value when an update condition occurs. The CAPTURE B
register contains the TCD counter value when a trigger B or software capture B occurs.
The TCD counter value is synchronized to CAPTUREB by either software or an event.
The capture register is blocked for an update of new capture data until the higher byte of this register is read.
Bit
15
14
13
Access
Reset
12
11
R
0
Bit
7
6
5
Access
Reset
R
0
R
0
R
0
4
3
CAPTUREB[7:0]
R
R
0
0
10
9
CAPTUREB[11:8]
R
R
0
0
8
R
0
2
1
0
R
0
R
0
R
0
Bits 11:0 – CAPTUREB[11:0] Capture B Value
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 345
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
25.5.21 Compare Set A
Name:
Offset:
Reset:
Property:
CMPASET
0x28
0x00
-
The TCDn.CMPASETL and TCDn.CMPASETH register pair represents the 12-bit TCDn.CMPASET value. This
register is continuously compared to the counter value. Then, the outputs from the comparators are used for
generating waveforms.
Bit
15
14
13
12
Access
Reset
Bit
Access
Reset
11
R/W
0
7
6
5
R/W
0
R/W
0
R/W
0
4
3
CMPASET[7:0]
R/W
R/W
0
0
10
9
CMPASET[11:8]
R/W
R/W
0
0
8
R/W
0
2
1
0
R/W
0
R/W
0
R/W
0
Bits 11:0 – CMPASET[11:0] Compare A Set
These bits hold the value of the compare register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 346
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
25.5.22 Compare Set B
Name:
Offset:
Reset:
Property:
CMPBSET
0x2C
0x00
-
The TCDn.CMPBSETL and TCDn.CMPBSETH register pair represents the 12-bit TCDn.CMPBSET value. This
register is continuously compared to the counter value. Then, the outputs from the comparators are used for
generating waveforms.
Bit
15
14
13
12
Access
Reset
Bit
Access
Reset
11
R/W
0
7
6
5
R/W
0
R/W
0
R/W
0
4
3
CMPBSET[7:0]
R/W
R/W
0
0
10
9
CMPBSET[11:8]
R/W
R/W
0
0
8
R/W
0
2
1
0
R/W
0
R/W
0
R/W
0
Bits 11:0 – CMPBSET[11:0] Compare B Set
These bits hold the value of the compare register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 347
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
25.5.23 Compare Clear A
Name:
Offset:
Reset:
Property:
CMPACLR
0x2A
0x00
-
The TCDn.CMPACLRL and TCDn.CMPACLRH register pair represents the 12-bit TCDn.CMPACLR value. This
register is continuously compared to the counter value. Then, the outputs from the comparators are used for
generating waveforms.
Bit
15
14
13
12
Access
Reset
Bit
Access
Reset
11
R/W
0
7
6
5
R/W
0
R/W
0
R/W
0
4
3
CMPACLR[7:0]
R/W
R/W
0
0
10
9
CMPACLR[11:8]
R/W
R/W
0
0
8
R/W
0
2
1
0
R/W
0
R/W
0
R/W
0
Bits 11:0 – CMPACLR[11:0] Compare A Clear
These bits hold the value of the compare register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 348
AVR128DB28/32/48/64
TCD - 12-Bit Timer/Counter Type D
25.5.24 Compare Clear B
Name:
Offset:
Reset:
Property:
CMPBCLR
0x2E
0x00
-
The TCDn.CMPBCLRL and TCDn.CMPBCLRH register pair represents the 12-bit TCDn.CMPBCLR value. This
register is continuously compared to the counter value. Then, the outputs from the comparators are used for
generating waveforms.
Bit
15
14
13
12
Access
Reset
Bit
Access
Reset
11
R/W
0
7
6
5
R/W
0
R/W
0
R/W
0
4
3
CMPBCLR[7:0]
R/W
R/W
0
0
10
9
CMPBCLR[11:8]
R/W
R/W
0
0
8
R/W
0
2
1
0
R/W
0
R/W
0
R/W
0
Bits 11:0 – CMPBCLR[11:0] Compare B Clear
These bits hold the value of the compare register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 349
AVR128DB28/32/48/64
RTC - Real-Time Counter
26.
RTC - Real-Time Counter
26.1
Features
•
•
•
•
•
•
•
•
•
26.2
16-bit Resolution
Selectable Clock Sources
Programmable 15-bit Clock Prescaling
One Compare Register
One Period Register
Clear Timer on Period Overflow
Optional Interrupt/Event on Overflow and Compare Match
Periodic Interrupt and Event
Crystal Error Correction
Overview
The RTC peripheral offers two timing functions: the Real-Time Counter (RTC) and a Periodic Interrupt Timer (PIT).
The PIT functionality can be enabled independently of the RTC functionality.
RTC - Real-Time Counter
The RTC counts (prescaled) clock cycles in a Counter register and compares the content of the Counter register to a
Period register and a Compare register.
The RTC can generate both interrupts and events on compare match or overflow. It will generate a compare interrupt
and/or event at the first count after the counter equals the Compare register value, and an overflow interrupt and/or
event at the first count after the counter value equals the Period register value. The overflow will reset the counter
value to zero.
The RTC peripheral typically runs continuously, including in Low-Power Sleep modes, to keep track of time. It can
wake up the device from sleep modes and/or interrupt the device at regular intervals.
The reference clock is typically the 32.768 kHz output from an external crystal. The RTC can also be clocked from an
external clock signal, the 32.768 kHz Internal Oscillator (OSC32K), or the OSC32K divided by 32.
The RTC peripheral includes a 15-bit programmable prescaler that can scale down the reference clock before it
reaches the counter. A wide range of resolutions and time-out periods can be configured for the RTC. With a 32.768
kHz clock source, the maximum resolution is 30.5 μs, and time-out periods can be up to two seconds. With a
resolution of 1s, the maximum time-out period is more than 18 hours (65536 seconds).
The RTC also supports crystal error correction when operated using external crystal selection. An externally
calibrated value will be used for correction. The RTC can be adjusted by software with an accuracy of ±1 PPM, and
the maximum adjustment is ±127 PPM. The RTC correction operation will either speed up (by skipping count) or slow
down (by adding extra count) the prescaler to account for the crystal error.
PIT - Periodic Interrupt Timer
The PIT uses the same clock source (CLK_RTC) as the RTC function and can generate an interrupt request or a
level event on every nth clock period. The n can be selected from {4, 8, 16,... 32768} for interrupts and from {64, 128,
256,... 8192} for events.
26.2.1
Block Diagram
RTC Block Diagram
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 350
AVR128DB28/32/48/64
RTC - Real-Time Counter
EXTCLK
XTAL32K1
XTAL32K2
External Clock
32.768 kHz Crystal Osc.
32.768 kHz Int. Osc.
DIV32
PER
CLKSEL
CLK_RTC
Correction
counter
RTC
15-bit
prescaler
PIT
=
Overflow
=
Compare
CNT
CMP
Period
26.3
Clocks
The peripheral clock (CLK_PER) is required to be at least four times faster than the RTC clock (CLK_RTC) for
reading the counter value, regardless of the prescaler setting.
A 32.768 kHz crystal can be connected to the XTAL32K1 or XTAL32K2 pins, along with any required load capacitors.
Alternatively, an external digital clock can be connected to the XTAL32K1 pin.
26.4
RTC Functional Description
The RTC peripheral offers two timing functions: the Real-Time Counter (RTC) and a Periodic Interrupt Timer (PIT).
This subsection describes the RTC.
26.4.1
Initialization
Before enabling the RTC peripheral and the desired actions (interrupt requests and output events), the source clock
for the RTC counter must be configured to operate the RTC.
26.4.1.1 Configure the Clock CLK_RTC
To configure the CLK_RTC, follow these steps:
1.
2.
Configure the desired oscillator to operate as required, in the Clock Controller (CLKCTRL) peripheral.
Write the Clock Select (CLKSEL) bit field in the Clock Selection (RTC.CLKSEL) register accordingly.
The CLK_RTC clock configuration is used by both RTC and PIT functionality.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 351
AVR128DB28/32/48/64
RTC - Real-Time Counter
26.4.1.2 Configure RTC
To operate the RTC, follow these steps:
1.
2.
3.
4.
Set the compare value in the Compare (RTC.CMP) register, and/or the overflow value in the Period
(RTC.PER) register.
Enable the desired interrupts by writing to the respective interrupt enable bits (CMP, OVF) in the Interrupt
Control (RTC.INTCTRL) register.
Configure the RTC internal prescaler by writing the desired value to the Prescaler (PRESCALER) bit field in
the Control A (RTC.CTRLA) register.
Enable the RTC by writing a ‘1’ to the RTC Peripheral Enable (RTCEN) bit in the RTC.CTRLA register.
Note: The RTC peripheral is used internally during device start-up. Always check the Synchronization Busy bits in
the Status (RTC.STATUS) and Periodic Interrupt Timer Status (RTC.PITSTATUS) registers, and on the initial
configuration.
26.4.2
Operation - RTC
26.4.2.1 Enabling and Disabling
The RTC is enabled by writing the RTC Peripheral Enable (RTCEN) bit in the Control A (RTC.CTRLA) register to ‘1’.
The RTC is disabled by writing the RTC Peripheral Enable (RTCEN) bit in RTC.CTRLA to ‘0’.
26.5
PIT Functional Description
The RTC peripheral offers two timing functions: the Real-Time Counter (RTC) and a Periodic Interrupt Timer (PIT).
This subsection describes the PIT.
26.5.1
Initialization
To operate the PIT, follow these steps:
1. Configure the RTC clock CLK_RTC as described in section 26.4.1.1 Configure the Clock CLK_RTC.
2. Enable the interrupt by writing a ‘1’ to the Periodic Interrupt (PI) bit in the PIT Interrupt Control
(RTC.PITINTCTRL) register.
3. Select the period for the interrupt by writing the desired value to the Period (PERIOD) bit field in the Periodic
Interrupt Timer Control A (RTC.PITCTRLA) register.
4. Enable the PIT by writing a ‘1’ to the Periodic Interrupt Timer Enable (PITEN) bit in the RTC.PITCTRLA
register.
Note: The RTC peripheral is used internally during device start-up. Always check the Synchronization Busy bits in
the RTC.STATUS and RTC.PITSTATUS registers, and on the initial configuration.
26.5.2
Operation - PIT
26.5.2.1 Enabling and Disabling
The PIT is enabled by writing the Periodic Interrupt Timer Enable (PITEN) bit in the Periodic Interrupt Timer Control A
(RTC.PITCTRLA) register to ‘1’. The PIT is disabled by writing the Periodic Interrupt Timer Enable (PITEN) bit in
RTC.PITCTRLA to ‘0’.
26.5.2.2 PIT Interrupt Timing
Timing of the First Interrupt
The PIT function and the RTC function are running from the same counter inside the prescaler and can be configured
as described below:
• The RTC interrupt period is configured by writing the Period (RTC.PER) register
• The PIT interrupt period is configured by writing the Period (PERIOD) bit field in Periodic Interrupt Timer Control
A (RTC.PITCTRLA) register
The prescaler is OFF when both functions are OFF (RTC Peripheral Enable (RTCEN) bit in RTC.CTRLA and the
Periodic Interrupt Timer Enable (PITEN) bit in RTC.PITCTRLA are ‘0’), but it is running (that is, its internal counter is
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 352
AVR128DB28/32/48/64
RTC - Real-Time Counter
counting) when either function is enabled. For this reason, the timing of the first PIT interrupt and the first RTC count
tick will be unknown (anytime between enabling and a full period).
Continuous Operation
After the first interrupt, the PIT will continue toggling every ½ PIT period resulting in a full PIT period signal.
Example 26-1. PIT Timing Diagram for PERIOD=CYC16
For PERIOD=CYC16 in RTC.PITCTRLA, the PIT output effectively follows the state of the
prescaler counter bit 3, so the resulting interrupt output has a period of 16 CLK_RTC cycles.
The time between writing PITEN to ‘1’ and the first PIT interrupt can vary between virtually zero
and a full PIT period of 16 CLK_RTC cycles. The precise delay between enabling the PIT and its
first output depends on the prescaler’s counting phase: the first interrupt shown below is produced
by writing PITEN to ‘1’ at any time inside the leading time window.
Figure 26-1. Timing Between PIT Enable and First Interrupt
Prescaler
counter
value (LSB)
..000000
..000001
..000010
..000011
..000100
..000101
..000110
..000111
..001000
..001001
..001010
..001011
..001100
..001101
..001110
..001111
..010000
..010001
..010010
..010011
..010100
..010101
..010110
..010111
..011000
..011001
..011010
..011011
..011100
..011101
..011110
..011111
..100000
..100001
..100010
..100011
..100100
..100101
..100110
..100111
..101000
..101001
..101010
..101011
..101100
..101101
..101110
..101111
CLK_RTC
Prescaler bit 3
(CYC16)
Continuous Operation
PITENABLE=0
PIT output
Time window for writing
PITENABLE=1
First PIT output
26.6
Crystal Error Correction
The prescaler for the RTC and PIT can do internal frequency correction of the crystal clock by using the PPM error
value from the Crystal Frequency Calibration (CALIB) register when the Frequency Correction Enable (CORREN) bit
in the RTC.CTRLA register is ‘1’.
The CALIB register must be written by the user, based on the information about the frequency error. The correction
operation is performed by adding or removing a number of cycles equal to the value given in the Error Correction
Value (ERROR) bit field in the CALIB register spread throughout a million-cycle interval.
The correction of the clock will be reflected in the RTC count value available through the Count (RTC.CNT) registers
or in the PIT intervals.
If disabling the correction feature, an ongoing correction cycle will be completed before the function is disabled.
Note: If using this feature with a negative correction, the minimum prescaler configuration is DIV2.
26.7
Events
The RTC can generate the events described in the following table:
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 353
AVR128DB28/32/48/64
RTC - Real-Time Counter
Table 26-1. Event Generators in RTC
Generator Name
Description
Event
Type
Clock
Domain
Length of the Event
OVF
Overflow
Pulse
CLK_RTC
One CLK_RTC period
CMP
Compare Match
PIT_DIV8192
Prescaled RTC clock divided by 8192
PIT_DIV4096
Prescaled RTC clock divided by 4096
Given by prescaled RTC clock
divided by 4096
PIT_DIV2048
Prescaled RTC clock divided by 2048
Given by prescaled RTC clock
divided by 2048
PIT_DIV1024
Prescaled RTC clock divided by 1024
Given by prescaled RTC clock
divided by 1024
PIT_DIV512
Prescaled RTC clock divided by 512
Given by prescaled RTC clock
divided by 512
PIT_DIV256
Prescaled RTC clock divided by 256
Given by prescaled RTC clock
divided by 256
PIT_DIV128
Prescaled RTC clock divided by 128
Given by prescaled RTC clock
divided by 128
PIT_DIV64
Prescaled RTC clock divided by 64
Given by prescaled RTC clock
divided by 64
Module Event
RTC
One CLK_RTC period
Level
Given by prescaled RTC clock
divided by 8192
The conditions for generating the OVF and CMP events are identical to those that will raise the corresponding
interrupt flags in the RTC.INTFLAGS register.
Refer to the (EVSYS) Event System section for more details regarding event users and Event System configuration.
26.8
Interrupts
Table 26-2. Available Interrupt Vectors and Sources
Name Vector Description
RTC
Real-Time Counter overflow
and compare match interrupt
Conditions
•
•
PIT
Periodic Interrupt Timer
interrupt
Overflow (OVF): The counter has reached the value from the
RTC.PER register and wrapped to zero.
Compare (CMP): Match between the value from the Counter
(RTC.CNT) register and the value from the Compare (RTC.CMP)
register.
A time period has passed, as configured by the PERIOD bit field in
RTC.PITCTRLA.
When an interrupt condition occurs, the corresponding interrupt flag is set in the peripheral’s Interrupt Flags
(peripheral.INTFLAGS) register.
An interrupt source is enabled or disabled by writing to the corresponding enable bit in the peripheral’s Interrupt
Control (peripheral.INTCTRL) register.
An interrupt request is generated when the corresponding interrupt source is enabled, and the interrupt flag is set.
The interrupt request remains active until the interrupt flag is cleared. See the peripheral’s INTFLAGS register for
details on how to clear interrupt flags.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 354
AVR128DB28/32/48/64
RTC - Real-Time Counter
Note that:
• The RTC has two INTFLAGS registers: RTC.INTFLAGS and RTC.PITINTFLAGS.
• The RTC has two INTCTRL registers: RTC.INTCTRL and RTC.PITINTCTRL.
26.9
Sleep Mode Operation
The RTC will continue to operate in Idle Sleep mode. It will run in Standby Sleep mode if the Run in Standby
(RUNSTDBY) bit in RTC.CTRLA is set.
The PIT will continue to operate in any sleep mode.
26.10
Synchronization
Both the RTC and the PIT are asynchronous, operating from a different clock source (CLK_RTC) independently of
the peripheral clock (CLK_PER). For Control and Count register updates, it will take some RTC and/or peripheral
clock cycles before an updated register value is available in a register or until a configuration change affects the RTC
or PIT, respectively. This synchronization time is described for each register in the Register Description section.
For some RTC registers, a Synchronization Busy flag is available (CMPBUSY, PERBUSY, CNTBUSY, CTRLABUSY)
in the Status (RTC.STATUS) register.
For the RTC.PITCTRLA register, a Synchronization Busy flag is available (CTRLBUSY) in the Periodic Interrupt
Timer Status (RTC.PITSTATUS) register.
Check these flags before writing to the mentioned registers.
26.11
Debug Operation
If the Debug Run (DBGRUN) bit in the Debug Control (RTC.DBGCTRL) register is ‘1’, the RTC will continue normal
operation. If DBGRUN is ‘0’ and the CPU is halted, the RTC will halt the operation and ignore any incoming events.
If the Debug Run (DBGRUN) bit in the Periodic Interrupt Timer Debug Control (RTC.PITDBGCTRL) register is ‘1’, the
PIT will continue normal operation. If DBGRUN is ‘0’ in the Debug mode and the CPU is halted, the PIT output will be
low. When the PIT output is high at the time, a new positive edge occurs to set the interrupt flag when restarting from
a break. The result is an additional PIT interrupt that would not happen during normal operation. If the PIT output is
low at the break, the PIT will resume low without additional interrupt.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 355
AVR128DB28/32/48/64
RTC - Real-Time Counter
26.12
Register Summary
Offset
Name
Bit Pos.
7
0x00
0x01
0x02
0x03
0x04
0x05
0x06
0x07
CTRLA
STATUS
INTCTRL
INTFLAGS
TEMP
DBGCTRL
CALIB
CLKSEL
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
15:8
7:0
15:8
7:0
15:8
RUNSTDBY
0x08
CNT
0x0A
PER
0x0C
CMP
0x0E
...
0x0F
0x10
0x11
0x12
0x13
0x14
0x15
26.13
6
5
4
3
2
CMPBUSY
CORREN
PERBUSY
PRESCALER[3:0]
1
0
CNTBUSY
CMP
CMP
RTCEN
CTRLABUSY
OVF
OVF
TEMP[7:0]
DBGRUN
SIGN
ERROR[6:0]
CLKSEL[1:0]
CNT[7:0]
CNT[15:8]
PER[7:0]
PER[15:8]
CMP[7:0]
CMP[15:8]
Reserved
PITCTRLA
PITSTATUS
PITINTCTRL
PITINTFLAGS
Reserved
PITDBGCTRL
7:0
7:0
7:0
7:0
PERIOD[3:0]
7:0
PITEN
CTRLBUSY
PI
PI
DBGRUN
Register Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 356
AVR128DB28/32/48/64
RTC - Real-Time Counter
26.13.1 Control A
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
RUNSTDBY
R/W
0
CTRLA
0x00
0x00
-
6
R/W
0
5
4
PRESCALER[3:0]
R/W
R/W
0
0
3
R/W
0
2
CORREN
R/W
0
1
0
RTCEN
R/W
0
Bit 7 – RUNSTDBY Run in Standby
Value
Description
0
RTC disabled in Standby Sleep mode
1
RTC enabled in Standby Sleep mode
Bits 6:3 – PRESCALER[3:0] Prescaler
These bits define the prescaling of the CLK_RTC clock signal. Due to synchronization between the RTC clock and
the peripheral clock, there is a latency of two RTC clock cycles from updating the register until this has an effect.
Application software needs to check that the CTRLABUSY flag in RTC.STATUS register is cleared before writing to
this register.
Value
Name
Description
0x0
DIV1
RTC clock/1 (no prescaling)
0x1
DIV2
RTC clock/2
0x2
DIV4
RTC clock/4
0x3
DIV8
RTC clock/8
0x4
DIV16
RTC clock/16
0x5
DIV32
RTC clock/32
0x6
DIV64
RTC clock/64
0x7
DIV128
RTC clock/128
0x8
DIV256
RTC clock/256
0x9
DIV512
RTC clock/512
0xA
DIV1024
RTC clock/1024
0xB
DIV2048
RTC clock/2048
0xC
DIV4096
RTC clock/4096
0xD
DIV8192
RTC clock/8192
0xE
DIV16384
RTC clock/16384
0xF
DIV32768
RTC clock/32768
Bit 2 – CORREN Frequency Correction Enable
Value
Description
0
Frequency correction is disabled
1
Frequency correction is enabled
Bit 0 – RTCEN RTC Peripheral Enable
Value
Description
0
RTC peripheral is disabled
1
RTC peripheral is enabled
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 357
AVR128DB28/32/48/64
RTC - Real-Time Counter
26.13.2 Status
Name:
Offset:
Reset:
Property:
Bit
7
STATUS
0x01
0x00
-
6
Access
Reset
5
4
3
CMPBUSY
R
0
2
PERBUSY
R
0
1
CNTBUSY
R
0
0
CTRLABUSY
R
0
Bit 3 – CMPBUSY Compare Synchronization Busy
This bit is ‘1’ when the RTC is busy synchronizing the Compare (RTC.CMP) register in the RTC clock domain.
Bit 2 – PERBUSY Period Synchronization Busy
This bit is ‘1’ when the RTC is busy synchronizing the Period (RTC.PER) register in the RTC clock domain.
Bit 1 – CNTBUSY Counter Synchronization Busy
This bit is ‘1’ when the RTC is busy synchronizing the Count (RTC.CNT) register in the RTC clock domain.
Bit 0 – CTRLABUSY Control A Synchronization Busy
This bit is ‘1’ when the RTC is busy synchronizing the Control A (RTC.CTRLA) register in the RTC clock domain.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 358
AVR128DB28/32/48/64
RTC - Real-Time Counter
26.13.3 Interrupt Control
Name:
Offset:
Reset:
Property:
Bit
7
INTCTRL
0x02
0x00
-
6
5
4
3
Access
Reset
2
1
CMP
R/W
0
0
OVF
R/W
0
Bit 1 – CMP Compare Match Interrupt Enable
Enable interrupt-on-compare match (that is, when the value from the Count (RTC.CNT) register matches the value
from the Compare (RTC.CMP) register).
Value
Description
0
The compare match interrupt is disabled
1
The compare match interrupt is enabled
Bit 0 – OVF Overflow Interrupt Enable
Enable interrupt-on-counter overflow (that is, when the value from the Count (RTC.CNT) register matched the value
from the Period (RTC.PER) register and wraps around to zero).
Value
Description
0
The overflow interrupt is disabled
1
The overflow interrupt is enabled
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 359
AVR128DB28/32/48/64
RTC - Real-Time Counter
26.13.4 Interrupt Flag
Name:
Offset:
Reset:
Property:
Bit
7
INTFLAGS
0x03
0x00
-
6
5
4
3
Access
Reset
2
1
CMP
R/W
0
0
OVF
R/W
0
Bit 1 – CMP Compare Match Interrupt Flag
This flag is set when the value from the Count (RTC.CNT) register matches the value from the Compare (RTC.CMP)
register.
Writing a ‘1’ to this bit clears the flag.
Bit 0 – OVF Overflow Interrupt Flag
This flag is set when the value from the Count (RTC.CNT) register has reached the value from the Period (RTC.PER)
register and wrapped to zero.
Writing a ‘1’ to this bit clears the flag.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 360
AVR128DB28/32/48/64
RTC - Real-Time Counter
26.13.5 Temporary
Name:
Offset:
Reset:
Property:
TEMP
0x4
0x00
-
The Temporary register is used by the CPU for 16-bit single-cycle access to the 16-bit registers of this peripheral. The
register is common for all the 16-bit registers of this peripheral and can be read and written by software. For more
details on reading and writing 16-bit registers, refer to Accessing 16-Bit Registers in the Memories section.
Bit
7
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
TEMP[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – TEMP[7:0] Temporary
Temporary register for read/write operations in 16-bit registers.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 361
AVR128DB28/32/48/64
RTC - Real-Time Counter
26.13.6 Debug Control
Name:
Offset:
Reset:
Property:
Bit
7
DBGCTRL
0x05
0x00
-
6
5
4
3
2
Access
Reset
1
0
DBGRUN
R/W
0
Bit 0 – DBGRUN Debug Run
Value
Description
0
The peripheral is halted in Break Debug mode and ignores events.
1
The peripheral will continue to run in Break Debug mode when the CPU is halted.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 362
AVR128DB28/32/48/64
RTC - Real-Time Counter
26.13.7 Crystal Frequency Calibration
Name:
Offset:
Reset:
Property:
CALIB
0x06
0x00
-
This register stores the error value and the type of correction to be done. This register is written by software with any
error value based on external calibration and/or temperature correction/s.
Bit
Access
Reset
7
SIGN
R/W
0
6
5
4
R/W
0
R/W
0
R/W
0
3
ERROR[6:0]
R/W
0
2
1
0
R/W
0
R/W
0
R/W
0
Bit 7 – SIGN Error Correction Sign Bit
This bit shows the direction of the correction.
Value
Description
0x0
0x1
Positive correction causing the prescaler to count slower.
Negative correction causing the prescaler to count faster. This requires that the
minimum prescaler configuration is DIV2.
Bits 6:0 – ERROR[6:0] Error Correction Value
The number of correction clocks for each million RTC clock cycles interval (PPM).
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 363
AVR128DB28/32/48/64
RTC - Real-Time Counter
26.13.8 Clock Selection
Name:
Offset:
Reset:
Property:
Bit
7
CLKSEL
0x07
0x00
-
6
5
4
3
2
Access
Reset
1
0
CLKSEL[1:0]
R/W
R/W
0
0
Bits 1:0 – CLKSEL[1:0] Clock Select
Writing these bits select the source for the RTC clock (CLK_RTC).
Value
Name
Description
0x0
0x1
0x2
0x3
OSC32K
OSC1K
XOSC32K
EXTCLK
32.768 kHz from OSC32K
1.024 kHz from OSC32K
32.768 kHz from XOSC32K
External clock from the EXTCLK/
XTALHF1 pin
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 364
AVR128DB28/32/48/64
RTC - Real-Time Counter
26.13.9 Count
Name:
Offset:
Reset:
Property:
CNT
0x08
0x0000
-
The RTC.CNTL and RTC.CNTH register pair represents the 16-bit value, RTC.CNT. The low byte [7:0] (suffix L) is
accessible at the original offset. The high byte [15:8] (suffix H) can be accessed at offset + 0x01.
Due to the synchronization between the RTC clock and main clock domains, there is a latency of two RTC clock
cycles from updating the register until this has an effect. The application software needs to check that the CNTBUSY
flag in RTC.STATUS is cleared before writing to this register.
Bit
15
14
13
12
11
10
9
8
R/W
0
R/W
0
R/W
0
R/W
0
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
CNT[15:8]
Access
Reset
Bit
R/W
0
R/W
0
R/W
0
R/W
0
7
6
5
4
CNT[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 15:8 – CNT[15:8] Counter High Byte
These bits hold the MSB of the 16-bit Counter register.
Bits 7:0 – CNT[7:0] Counter Low Byte
These bits hold the LSB of the 16-bit Counter register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 365
AVR128DB28/32/48/64
RTC - Real-Time Counter
26.13.10 Period
Name:
Offset:
Reset:
Property:
PER
0x0A
0xFFFF
-
The RTC.PERL and RTC.PERH register pair represents the 16-bit value, RTC.PER. The low byte [7:0] (suffix L) is
accessible at the original offset. The high byte [15:8] (suffix H) can be accessed at offset + 0x01.
Due to the synchronization between the RTC clock and main clock domains, there is a latency of two RTC clock
cycles from updating the register until this has an effect. The application software needs to check that the PERBUSY
flag in RTC.STATUS is cleared before writing to this register.
Bit
15
14
13
12
11
10
9
8
R/W
1
R/W
1
R/W
1
R/W
1
3
2
1
0
R/W
1
R/W
1
R/W
1
R/W
1
PER[15:8]
Access
Reset
Bit
R/W
1
R/W
1
R/W
1
R/W
1
7
6
5
4
PER[7:0]
Access
Reset
R/W
1
R/W
1
R/W
1
R/W
1
Bits 15:8 – PER[15:8] Period High Byte
These bits hold the MSB of the 16-bit Period register.
Bits 7:0 – PER[7:0] Period Low Byte
These bits hold the LSB of the 16-bit Period register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 366
AVR128DB28/32/48/64
RTC - Real-Time Counter
26.13.11 Compare
Name:
Offset:
Reset:
Property:
CMP
0x0C
0x0000
-
The RTC.CMPL and RTC.CMPH register pair represents the 16-bit value, RTC.CMP. The low byte [7:0] (suffix L) is
accessible at the original offset. The high byte [15:8] (suffix H) can be accessed at offset + 0x01.
Bit
15
14
13
12
11
10
9
8
R/W
0
R/W
0
R/W
0
R/W
0
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
CMP[15:8]
Access
Reset
Bit
R/W
0
R/W
0
R/W
0
R/W
0
7
6
5
4
CMP[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 15:8 – CMP[15:8] Compare High Byte
These bits hold the MSB of the 16-bit Compare register.
Bits 7:0 – CMP[7:0] Compare Low Byte
These bits hold the LSB of the 16-bit Compare register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 367
AVR128DB28/32/48/64
RTC - Real-Time Counter
26.13.12 Periodic Interrupt Timer Control A
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
PITCTRLA
0x10
0x00
-
6
R/W
0
5
4
PERIOD[3:0]
R/W
R/W
0
0
3
2
R/W
0
1
0
PITEN
R/W
0
Bits 6:3 – PERIOD[3:0] Period
Writing this bit field selects the number of RTC clock cycles between each interrupt.
Value
Name
Description
0x0
OFF
No interrupt
0x1
CYC4
4 cycles
0x2
CYC8
8 cycles
0x3
CYC16
16 cycles
0x4
CYC32
32 cycles
0x5
CYC64
64 cycles
0x6
CYC128
128 cycles
0x7
CYC256
256 cycles
0x8
CYC512
512 cycles
0x9
CYC1024
1024 cycles
0xA
CYC2048
2048 cycles
0xB
CYC4096
4096 cycles
0xC
CYC8192
8192 cycles
0xD
CYC16384
16384 cycles
0xE
CYC32768
32768 cycles
0xF
Reserved
Bit 0 – PITEN Periodic Interrupt Timer Enable
Value
Description
0
Periodic Interrupt Timer disabled
1
Periodic Interrupt Timer enabled
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 368
AVR128DB28/32/48/64
RTC - Real-Time Counter
26.13.13 Periodic Interrupt Timer Status
Name:
Offset:
Reset:
Property:
Bit
7
PITSTATUS
0x11
0x00
-
6
5
4
3
Access
Reset
2
1
0
CTRLBUSY
R
0
Bit 0 – CTRLBUSY PITCTRLA Synchronization Busy
This bit is ‘1’ when the RTC is busy synchronizing the Periodic Interrupt Timer Control A (RTC.PITCTRLA) register in
the RTC clock domain.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 369
AVR128DB28/32/48/64
RTC - Real-Time Counter
26.13.14 PIT Interrupt Control
Name:
Offset:
Reset:
Property:
Bit
7
PITINTCTRL
0x12
0x00
-
6
5
4
3
Access
Reset
2
1
0
PI
R/W
0
Bit 0 – PI Periodic Interrupt
Value
Description
0
The periodic interrupt is disabled
1
The periodic interrupt is enabled
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 370
AVR128DB28/32/48/64
RTC - Real-Time Counter
26.13.15 PIT Interrupt Flag
Name:
Offset:
Reset:
Property:
Bit
7
PITINTFLAGS
0x13
0x00
-
6
5
4
3
Access
Reset
2
1
0
PI
R/W
0
Bit 0 – PI Periodic Interrupt Flag
This flag is set when a periodic interrupt is issued.
Writing a ‘1’ clears the flag.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 371
AVR128DB28/32/48/64
RTC - Real-Time Counter
26.13.16 Periodic Interrupt Timer Debug Control
Name:
Offset:
Reset:
Property:
Bit
7
PITDBGCTRL
0x15
0x00
-
6
5
4
3
2
Access
Reset
1
0
DBGRUN
R/W
0
Bit 0 – DBGRUN Debug Run
Value
Description
0
The peripheral is halted in Break Debug mode and ignores events.
1
The peripheral will continue to run in Break Debug mode when the CPU is halted.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 372
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
27.
USART - Universal Synchronous and Asynchronous Receiver and
Transmitter
27.1
Features
•
•
•
•
•
•
•
•
•
•
•
•
27.2
Full-Duplex Operation
Half-Duplex Operation:
– One-Wire mode
– RS-485 mode
Asynchronous or Synchronous Operation
Supports Serial Frames with Five, Six, Seven, Eight or Nine Data Bits and One or Two Stop Bits
Fractional Baud Rate Generator:
– Can generate the desired baud rate from any peripheral clock frequency
– No need for an external oscillator
Built-In Error Detection and Correction Schemes:
– Odd or even parity generation and parity check
– Buffer overflow and frame error detection
– Noise filtering including false Start bit detection and digital low-pass filter
Separate Interrupts for:
– Transmit complete
– Transmit Data register empty
– Receive complete
Master SPI Mode
Multiprocessor Communication Mode
Start-of-Frame Detection
IRCOM Module for IrDA® Compliant Pulse Modulation/Demodulation
LIN Slave Support
Overview
The Universal Synchronous and Asynchronous serial Receiver and Transmitter (USART) is a fast and flexible serial
communication peripheral. The USART supports a number of different modes of operation that can accommodate
multiple types of applications and communication devices. For example, the One-Wire Half-Duplex mode is useful
when low pin count applications are desired. The communication is frame-based, and the frame format can be
customized to support a wide range of standards.
The USART is buffered in both directions, enabling continued data transmission without any delay between frames.
Separate interrupts for receive and transmit completion allow fully interrupt-driven communication.
The transmitter consists of a single-write buffer, a Shift register, and control logic for different frame formats. The
receiver consists of a two-level receive buffer and a Shift register. The status information of the received data is
available for error checking. Data and clock recovery units ensure robust synchronization and noise filtering during
asynchronous data reception.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 373
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
27.2.1
Block Diagram
Figure 27-1. USART Block Diagram
CLOCK GENERATOR
BAUD
XCK
Baud Rate Generator
TRANSMITTER
XDIR
TX Shift Register
TXDATA
TXD
RECEIVER
RX Shift Register
RX Buffer
RXD
RXDATA
27.2.2
Signal Description
Signal
Type
Description
XCK
Output/input
Clock for synchronous operation
XDIR
Output
Transmit enable for RS-485
TxD
Output/input
Transmitting line (and receiving line in One-Wire mode)
RxD
Input
Receiving line
27.3
Functional Description
27.3.1
Initialization
Full Duplex Mode:
1.
2.
3.
4.
Set the baud rate (USARTn.BAUD).
Set the frame format and mode of operation (USARTn.CTRLC).
Configure the TXD pin as an output.
Enable the transmitter and the receiver (USARTn.CTRLB).
Notes:
• For interrupt-driven USART operation, global interrupts must be disabled during the initialization
• Before doing a reinitialization with a changed baud rate or frame format, be sure that there are no ongoing
transmissions while the registers are changed
One-Wire Half Duplex Mode:
1.
2.
Internally connect the TXD to the USART receiver (the LBME bit in the USARTn.CTRLA register).
Enable internal pull-up for the RX/TX pin (the PULLUPEN bit in the PORTx.PINnCTRL register).
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 374
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
3.
4.
5.
6.
Enable Open-Drain mode (the ODME bit in the USARTn.CTRLB register).
Set the baud rate (USARTn.BAUD).
Set the frame format and mode of operation (USARTn.CTRLC).
Enable the transmitter and the receiver (USARTn.CTRLB).
Notes:
• When Open-Drain mode is enabled, the TXD pin is automatically set to output by hardware
• For interrupt-driven USART operation, global interrupts must be disabled during the initialization
• Before doing a reinitialization with a changed baud rate or frame format, be sure that there are no ongoing
transmissions while the registers are changed
27.3.2
Operation
27.3.2.1 Frame Formats
The USART data transfer is frame-based. A frame starts with a Start bit followed by one character of data bits. If
enabled, the Parity bit is inserted after the data bits and before the first Stop bit. After the Stop bit(s) of a frame, either
the next frame can follow immediately, or the communication line can return to the Idle (high) state. The USART
accepts all combinations of the following as valid frame formats:
•
•
•
•
1 Start bit
5, 6, 7, 8, or 9 data bits
No, even, or odd Parity bit
1 or 2 Stop bits
The figure below illustrates the possible combinations of frame formats. Bits inside brackets are optional.
Figure 27-2. Frame Formats
FRAME
(IDLE)
St
0
1
2
3
4
[5]
[6]
[7]
[8]
[P]
Sp1 [Sp2]
St
Start bit, always low
(n)
Data bits (0 to 8)
P
Parity bit, may be odd or even
Sp
Stop bit, always high
IDLE
No transfer on the communication line (RxD or TxD). The Idle state is always high.
(St/IDLE)
27.3.2.2 Clock Generation
The clock used for shifting and sampling data bits is generated internally by the fractional baud rate generator or
externally from the Transfer Clock (XCK) pin.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 375
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
Figure 27-3. Clock Generation Logic Block Diagram
CLOCK GENERATOR
Sync
Register
Edge
Detector
CLK_PER
Fractional Baud Rate
Generator
BAUD
XCK
XCKO
Transmitter
TXCLK
Receiver
RXCLK
27.3.2.2.1 The Fractional Baud Rate Generator
In modes where the USART is not using the XCK input as a clock source, the fractional Baud Rate Generator is used
to generate the clock. Baud rate is given in terms of bits per second (bps) and is configured by writing the
USARTn.BAUD register. The baud rate (fBAUD) is generated by dividing the peripheral clock (fCLK_PER) by a division
factor decided by the BAUD register.
The fractional Baud Rate Generator features hardware that accommodates cases where fCLK_PER is not divisible by
fBAUD. Usually, this situation would lead to a rounding error. The fractional Baud Rate Generator expects the BAUD
register to contain the desired division factor left shifted by six bits, as implemented by the equations in Table 27-1.
The six LSbs will then hold the fractional part of the desired divisor. The fractional part of the BAUD register is used to
dynamically adjust fBAUD to achieve a closer approximation to the desired baud rate.
Since the baud rate cannot be higher than fCLK_PER, the integer part of the BAUD register needs to be at least 1.
Since the result is left shifted by six bits, the corresponding minimum value of the BAUD register is 64. The valid
range is, therefore, 64 to 65535.
In Synchronous mode, only the 10-bit integer part of the BAUD register (BAUD[15:6]) determines the baud rate, and
the fractional part (BAUD[5:0]) must, therefore, be written to zero.
The table below lists equations for translating baud rates into input values for the BAUD register. The equations take
fractional interpretation into consideration, so the BAUD values calculated with these equations can be written directly
to USARTn.BAUD without any additional scaling.
Table 27-1. Equations for Calculating Baud Rate Register Setting
Operating Mode
Asynchronous
Synchronous Master
Conditions
Baud Rate (Bits Per Seconds) USART.BAUD Register Value
Calculation
����� ≤
����_���
64 × ����_���
����� =
�
� × ����
���� =
����� ≤
����_���
����_���
����� =
�
� × ���� 15: 6
���� 15: 6 =
�����.���� ≥ 64
�����.���� ≥ 64
© 2020 Microchip Technology Inc.
Preliminary Datasheet
64 × ����_���
� × �����
����_���
� × �����
DS40002247A-page 376
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
S is the number of samples per bit
• Asynchronous Normal mode: S = 16
• Asynchronous Double-Speed mode: S = 8
• Synchronous mode: S = 2
27.3.2.3 Data Transmission
The USART transmitter sends data by periodically driving the transmission line low. The data transmission is initiated
by loading the transmit buffer (USARTn.TXDATA) with the data to be sent. The data in the transmit buffer is moved to
the Shift register once it is empty and ready to send a new frame. After the Shift register is loaded with data, the data
frame will be transmitted.
When the entire frame in the Shift register has been shifted out, and there are no new data present in the transmit
buffer, the Transmit Complete Interrupt Flag (the TXCIF bit in the USARTn.STATUS register) is set, and the interrupt
is generated if it is enabled.
TXDATA can only be written when the Data Register Empty Interrupt Flag (the DREIF bit in the USARTn.STATUS
register) is set, indicating that the register is empty and ready for new data.
When using frames with fewer than eight bits, the Most Significant bits (MSb) written to TXDATA are ignored. If 9-bit
characters are used, the DATA[8] bit in the USARTn.TXDATAH register has to be written before the DATA[7:0] bits in
the USARTn.TXDATAL register.
27.3.2.3.1 Disabling the Transmitter
When disabling the transmitter, the operation will not become effective until ongoing and pending transmissions are
completed (that is, when the Transmit Shift register and Transmit Buffer register do not contain data to be
transmitted). When the transmitter is disabled, it will no longer override the TXD pin, and the PORT module regains
control of the pin. The pin is automatically configured as an input by hardware regardless of its previous setting. The
pin can now be used as a normal I/O pin with no port override from the USART.
27.3.2.4 Data Reception
The USART receiver samples the reception line to detect and interpret the received data. The direction of the pin
must, therefore, be configured as an input by writing a ‘0’ to the corresponding bit in the Direction register
(PORTx.DIRn).
The receiver accepts data when a valid Start bit is detected. Each bit that follows the Start bit will be sampled at the
baud rate or XCK clock and shifted into the Receive Shift register until the first Stop bit of a frame is received. A
second Stop bit will be ignored by the receiver. When the first Stop bit is received, and a complete serial frame is
present in the Receive Shift register, the contents of the Shift register will be moved into the receive buffer. The
Receive Complete Interrupt Flag (the RXCIF bit in the USARTn.STATUS register) is set, and the interrupt is
generated if enabled.
The RXDATA registers are the part of the double-buffered RX buffer that can be read by the application software
when RXCIF is set. If only one frame has been received, the data and status bits for that frame are pushed to the
RXDATA registers directly. If two frames are present in the RX buffer, the RXDATA registers contain the data for the
oldest frame.
The buffer shifts out the data either when RXDATAL or RXDATAL is read, depending on the configuration. The
register which does not lead to data being shifted should be read first to be able to read both bytes before shifting.
When the Character Size (CHSIZE) bits in the Control C (USARTn.CTRLC) register is configured to 9-bit (low byte
first), a read of RXDATAH shifts the receive buffer. Otherwise, RXDATAL shifts the buffer.
27.3.2.4.1 Receiver Error Flags
The USART receiver features error detection mechanisms that uncover corruption of the transmission. These
mechanisms include the following:
• Frame Error detection - controls whether the received frame is valid
• Buffer Overflow detection - indicates data loss due to the receiver buffer being full and overwritten by the new
data
• Parity Error detection - checks the validity of the incoming frame by calculating its parity and comparing it to the
Parity bit
Each error detection mechanism controls one error flag that can be read in the RXDATAH register:
• Frame Error (FERR)
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 377
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
•
•
Buffer Overflow (BUFOVF)
Parity Error (PERR)
The error flags are located in the RX buffer together with their corresponding frame. The RXDATAH register that
contains the error flags must be read before the RXDATAL register, since reading the RXDATAL register will trigger
the RX buffer to shift out the RXDATA bytes.
Note: If the Character Size bit field (the CHSIZE bits in the USARTn.CTRLC register) is set to nine bits, low byte
first (9BITL), the RXDATAH register will, instead of the RXDATAL register, trigger the RX buffer to shift out the
RXDATA bytes. The RXDATAL register must, in that case, be read before the RXDATAH register.
27.3.2.4.2 Disabling the Receiver
When disabling the receiver, the operation is immediate. The receiver buffer will be flushed, and data from ongoing
receptions will be lost.
27.3.2.4.3 Flushing the Receive Buffer
If the RX buffer has to be flushed during normal operation, repeatedly read the DATA location (USARTn.RXDATAH
and USARTn.RXDATAL registers) until the Receive Complete Interrupt Flag (the RXCIF bit in the
USARTn.RXDATAH register) is cleared.
27.3.3
Communication Modes
The USART is a flexible peripheral that supports multiple different communication protocols. The available modes of
operation can be split into two groups: Synchronous and asynchronous communication.
The synchronous communication relies on one device on the bus to be the master, providing the rest of the devices
with a clock signal through the XCK pin. All the devices use this common clock signal for both transmission and
reception, requiring no additional synchronization mechanism.
The device can be configured to run either as a master or a slave on the synchronous bus.
The asynchronous communication does not use a common clock signal. Instead, it relies on the communicating
devices to be configured with the same baud rate. When receiving a transmission, the hardware synchronization
mechanisms are used to align the incoming transmission with the receiving device peripheral clock.
Four different modes of reception are available when communicating asynchronously. One of these modes can
receive transmissions at twice the normal speed, sampling only eight times per bit instead of the normal 16. The
other three operating modes use variations of synchronization logic, all receiving at normal speed.
27.3.3.1 Synchronous Operation
27.3.3.1.1 Clock Operation
The XCK pin direction controls whether the transmission clock is an input (Slave mode) or an output (Master mode).
The corresponding port pin direction must be set to output for Master mode or to input for Slave mode
(PORTx.DIRn). The data input (on RXD) is sampled at the XCK clock edge which is opposite the edge where data
are transmitted (on TXD) as shown in the figure below.
Figure 27-4. Synchronous Mode XCK Timing
XCK
INVEN = 0
Data transmit (TxD)
Data sample (RxD)
XCK
INVEN = 1
Data transmit (TxD)
Data sample (RxD)
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 378
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
The I/O pin can be inverted by writing a ‘1’ to the Inverted I/O Enable (INVEN) bit in the Pin n Control register of the
port peripheral (PORTx.PINnCTRL). Using the inverted I/O setting for the corresponding XCK port pin, the XCK clock
edges used for sampling RxD and transmitting on TxD can be selected. If the inverted I/O is disabled (INVEN = 0),
the rising XCK clock edge represents the start of a new data bit, and the received data will be sampled at the falling
XCK clock edge. If inverted I/O is enabled (INVEN = 1), the falling XCK clock edge represents the start of a new data
bit, and the received data will be sampled at the rising XCK clock edge.
27.3.3.1.2 External Clock Limitations
When the USART is configured in Synchronous Slave mode, the XCK signal must be provided externally by the
master device. Since the clock is provided externally, configuring the BAUD register will have no impact on the
transfer speed. Successful clock recovery requires the clock signal to be sampled at least twice for each rising and
falling edge. The maximum XCK speed in Synchronous Operation mode, fSlave_XCK, is therefore limited by:
�Slave_XCK<
����_���
4
If the XCK clock has jitter, or if the high/low period duty cycle is not 50/50, the maximum XCK clock speed must be
reduced accordingly to ensure that XCK is sampled a minimum of two times for each edge.
27.3.3.1.3 USART in Master SPI Mode
The USART may be configured to function with multiple different communication interfaces, and one of these is the
Serial Peripheral Interface (SPI) where it can function as the master device. The SPI is a four-wire interface that
enables a master device to communicate with one or multiple slaves.
Frame Formats
The serial frame for the USART in Master SPI mode always contains eight Data bits. The Data bits can be configured
to be transmitted with either the LSb or MSb first, by writing to the Data Order bit (UDORD) in the Control C register
(USARTn.CTRLC).
SPI does not use Start, Stop, or Parity bits, so the transmission frame can only consist of the Data bits.
Clock Generation
Being a master device in a synchronous communication interface, the USART in Master SPI mode must generate the
interface clock to be shared with the slave devices. The interface clock is generated using the fractional Baud Rate
Generator, which is described in 27.3.2.2.1 The Fractional Baud Rate Generator.
Each Data bit is transmitted by pulling the data line high or low for one full clock period. The receiver will sample bits
in the middle of the transmitter hold period as shown in the figure below. It also shows how the timing scheme can be
configured using the Inverted I/O Enable (INVEN) bit in the PORTx.PINnCTRL register and the USART Clock Phase
(UCPHA) bit in the USARTn.CTRLC register.
UCPHA = 1
UCPHA = 0
Figure 27-5. Data Transfer Timing Diagrams
INVEN = 0
INVEN = 1
XCK
XCK
Data transmit (TxD)
Data transmit (TxD)
Data sample (RxD)
Data sample (RxD)
XCK
XCK
Data transmit (TxD)
Data transmit (TxD)
Data sample (RxD)
Data sample (RxD)
The table below further explains the figure above.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 379
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
Table 27-2. Functionality of INVEN and UCPHA Bits
INVEN
UCPHA
Leading Edge (1)
Trailing Edge (1)
0
0
Rising, sample
Falling, transmit
0
1
Rising, transmit
Falling, sample
1
0
Falling, sample
Rising, transmit
1
1
Falling, transmit
Rising, sample
Note:
1. The leading edge is the first clock edge of a clock cycle. The trailing edge is the last clock edge of a clock
cycle.
Data Transmission
Data transmission in Master SPI mode is functionally identical to general USART operation as described in the
Operation section. The transmitter interrupt flags and corresponding USART interrupts are also identical. See
27.3.2.3 Data Transmission for further description.
Data Reception
Data reception in Master SPI mode is identical in function to general USART operation as described in the Operation
section. The receiver interrupt flags and the corresponding USART interrupts are also identical, aside from the
receiver error flags that are not in use and always read as ‘0’. See 27.3.2.4 Data Reception for further description.
USART in Master SPI Mode vs. SPI
The USART in Master SPI mode is fully compatible with a stand-alone SPI peripheral. Their data frame and timing
configurations are identical. Some SPI specific special features are, however, not supported with the USART in
Master SPI mode:
• Write Collision Flag Protection
• Double-Speed mode
• Multi-Master support
A comparison of the pins used with USART in Master SPI mode and with SPI is shown in the table below.
Table 27-3. Comparison of USART in Master SPI Mode and SPI Pins
USART
SPI
Comment
TXD
MOSI
Master out
RXD
MISO
Master in
XCK
SCK
Functionally identical
-
SS
Not supported by USART in Master SPI mode(1)
Note:
1. For the stand-alone SPI peripheral, this pin is used with the Multi-Master function or as a dedicated Slave
Select pin. The Multi-Master function is not available with the USART in Master SPI mode, and no dedicated
Slave Select pin is available.
27.3.3.2 Asynchronous Operation
27.3.3.2.1 Clock Recovery
Since there is no common clock signal when using Asynchronous mode, each communicating device generates
separate clock signals. These clock signals must be configured to run at the same baud rate for the communication
to take place. The devices, therefore, run at the same speed, but their timing is skewed in relation to each other. To
accommodate this, the USART features a hardware clock recovery unit which synchronizes the incoming
asynchronous serial frames with the internally generated baud rate clock.
The figure below illustrates the sampling process for the Start bit of an incoming frame. It shows the timing scheme
for both Normal and Double-Speed mode (the RXMODE bits in the USARTn.CTRLB register configured respectively
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 380
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
to 0x00 and 0x01). The sample rate for Normal mode is 16 times the baud rate, while the sample rate for DoubleSpeed mode is eight times the baud rate (see 27.3.3.2.4 Double-Speed Operation for more details). The horizontal
arrows show the maximum synchronization error. Note that the maximum synchronization error is larger in DoubleSpeed mode.
Figure 27-6. Start Bit Sampling
RxD
IDLE
START
BIT 0
Sample
(RXMODE = 0x0)
0
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
1
2
3
Sample
(RXMODE = 0x1)
0
1
2
3
4
5
6
7
8
1
2
When the clock recovery logic detects a falling edge from Idle (high) state to the Start bit (low), the Start bit detection
sequence is initiated. In the figure above, sample 1 denotes the first sample reading ‘0’. The clock recovery logic then
uses three subsequent samples (samples 8, 9, and 10 in Normal mode, and samples 4, 5, 6 in Double-Speed mode)
to decide if a valid Start bit is received. If two or three samples read ‘0’, the Start bit is accepted. The clock recovery
unit is synchronized, and the data recovery can begin. If less than two samples read ‘0’, the Start bit is rejected. This
process is repeated for each Start bit.
27.3.3.2.2 Data Recovery
As with clock recovery, the data recovery unit samples at a rate 8 or 16 times faster than the baud rate depending on
whether it is running in Double-Speed or Normal mode, respectively. The figure below shows the sampling process
for reading a bit in a received frame.
Figure 27-7. Sampling of Data and Parity Bits
RxD
BIT n
Sample
(CLK2X = 0)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
1
Sample
(CLK2X = 1)
1
2
3
4
5
6
7
8
1
A majority voting technique is, like with clock recovery, used on the three center samples for deciding the logic level
of the received bit. The process is repeated for each bit until a complete frame is received.
The data recovery unit will only receive the first Stop bit while ignoring the rest if there are more. If the sampled Stop
bit is read ‘0’, the Frame Error flag will be set. The figure below shows the sampling of a Stop bit. It also shows the
earliest possible beginning of the next frame's Start bit.
Figure 27-8. Stop Bit and Next Start Bit Sampling
RxD
STOP 1
(A)
(B)
(C)
Sample
(CLK2X = 0)
1
2
3
4
5
6
7
8
9
10
0/1
0/1
0/1
Sample
(CLK2X = 1)
1
2
3
4
5
6
0/1
A new high-to-low transition indicating the Start bit of a new frame can come right after the last of the bits used for
majority voting. For Normal-Speed mode, the first low-level sample can be at the point marked (A) in the figure
above. For Double-Speed mode the first low level must be delayed to point (B), being the first sample after the
majority vote samples. Point (C) marks a Stop bit of full length at the nominal baud rate.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 381
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
27.3.3.2.3 Error Tolerance
The speed of the internally generated baud rate and the externally received data rate should ideally be identical, but
due to natural clock source error, this is normally not the case. The USART is tolerant of such error, and the limits of
this tolerance make up what is sometimes known as the Operational Range.
The following tables list the operational range of the USART, being the maximum receiver baud rate error that can be
tolerated. Note that Normal-Speed mode has higher toleration of baud rate variations than Double-Speed mode.
Table 27-4. Recommended Maximum Receiver Baud Rate Error for Normal Speed Mode
D
Rslow [%]
Rfast [%]
Maximum Total Error [%]
Recommended Max. Receiver Error [%]
5
93.20
106.67
-6.80/+6.67
±3.0
6
94.12
105.79
-5.88/+5.79
±2.5
7
94.81
105.11
-5.19/+5.11
±2.0
8
95.36
104.58
-4.54/+4.58
±2.0
9
95.81
104.14
-4.19/+4.14
±1.5
10
96.17
103.78
-3.83/+3.78
±1.5
Notes:
• D: The sum of character size and parity size (D = 5 to 10 bits)
• RSLOW: The ratio of the slowest incoming data rate that can be accepted in relation to the receiver baud rate
• RFAST: The ratio of the fastest incoming data rate that can be accepted in relation to the receiver baud rate
Table 27-5. Recommended Maximum Receiver Baud Rate Error for Double Speed Mode
D
Rslow [%]
Rfast [%]
Maximum Total Error [%]
Recommended Max. Receiver Error [%]
5
94.12
105.66
-5.88/+5.66
±2.5
6
94.92
104.92
-5.08/+4.92
±2.0
7
95.52
104.35
-4.48/+4.35
±1.5
8
96.00
103.90
-4.00/+3.90
±1.5
9
96.39
103.53
-3.61/+3.53
±1.5
10
96.70
103.23
-3.30/+3.23
±1.0
Notes:
• D: The sum of character size and parity size (D = 5 to 10 bits)
• RSLOW: The ratio of the slowest incoming data rate that can be accepted in relation to the receiver baud rate
• RFAST: The ratio of the fastest incoming data rate that can be accepted in relation to the receiver baud rate
The recommendations of the maximum receiver baud rate error were made under the assumption that the receiver
and transmitter equally divide the maximum total error.
The following equations are used to calculate the maximum ratio of the incoming data rate and the internal receiver
baud rate.
����� =
•
•
•
� �+1
� � + 1 + �� − 1
����� =
� �+2
� � + 1 + ��
D: The sum of character size and parity size (D = 5 to 10 bits)
S: Samples per bit. S = 16 for Normal Speed mode and S = 8 for Double-Speed mode.
SF: First sample number used for majority voting. SF = 8 for Normal-Speed mode and SF = 4 for Double-Speed
mode.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 382
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
•
•
•
SM: Middle sample number used for majority voting. SM = 9 for Normal-Speed mode and SM = 5 for DoubleSpeed mode.
RSLOW: The ratio of the slowest incoming data rate that can be accepted in relation to the receiver baud rate
RFAST: The ratio of the fastest incoming data rate that can be accepted in relation to the receiver baud rate
27.3.3.2.4 Double-Speed Operation
Double-speed operation allows for higher baud rates under asynchronous operation with lower peripheral clock
frequencies. This operation mode is enabled by writing the RXMODE bits in the Control B (USARTn.CTRLB) register
to 0x01.
When enabled, the baud rate for a given asynchronous baud rate setting will be doubled. This is shown in the
equations in 27.3.2.2.1 The Fractional Baud Rate Generator. In this mode, the receiver will use half the number of
samples (reduced from 16 to 8) for data sampling and clock recovery. This requires a more accurate baud rate
setting and peripheral clock. See 27.3.3.2.3 Error Tolerance for more details.
27.3.3.2.5 Auto-Baud
The auto-baud feature lets the USART configure its BAUD register based on input from a communication device.
This allows the device to communicate autonomously with multiple devices communicating with different baud rates.
The USART peripheral features two auto-baud modes: Generic Auto-Baud mode and LIN Constrained Auto-Baud
mode.
Both auto-baud modes must receive an auto-baud frame as seen in the figure below.
Figure 27-9. Auto-Baud Timing
Break Field
Sync Field
Tbit
8 Tbit
The break field is detected when 12 or more consecutive low cycles are sampled and notifies the USART that it is
about to receive the synchronization field. After the break field, when the Start bit of the synchronization field is
detected, a counter running at the peripheral clock speed is started. The counter is then incremented for the next
eight Tbit of the synchronization field. When all eight bits are sampled, the counter is stopped. The resulting counter
value is in effect the new BAUD register value.
When the USART Receive mode is set to GENAUTO (the RXMODE bits in the USARTn.CTRLB register), the
Generic Auto-Baud mode is enabled. In this mode, one can set the Wait For Break (WFB) bit in the USARTn.STATUS
register to enable detection of a break field of any length (that is, also shorter than 12 cycles). This makes it possible
to set an arbitrary new baud rate without knowing the current baud rate. If the measured sync field results in a valid
BAUD value (0x0064 - 0xFFFF), the BAUD register is updated.
When USART Receive mode is set to LINAUTO mode (the RXMODE bits in the USARTn.CTRLB register), it follows
the LIN format. The WFB functionality of the Generic Auto-Baud mode is not compatible with the LIN Constrained
Auto-Baud mode. This means that the received signal must be low for 12 peripheral clock cycles or more for a break
field to be valid. When a break field has been detected, the USART expects the following synchronization field
character to be 0x55. The tolerance for the difference in baud rates between the two synchronizing devices can be
configured using the Auto-baud Window Size (ABW) bits in the Control D (USARTn.CTRLD) register. If any of these
conditions are not met, the Inconsistent Sync Field Error Flag (the ISFIF bit in the USARTn.STATUS register) is set,
and the baud rate is unchanged.
27.3.3.2.6 Half Duplex Operation
Half duplex is a type of communication where two or more devices may communicate with each other, but only one at
a time. The USART can be configured to operate in the following half duplex modes:
• One-Wire mode
• RS-485 mode
One-Wire Mode
One-Wire mode is enabled by setting the Loop-Back Mode Enable (LBME) bit in the USARTn.CTRLA register. This
will enable an internal connection between the TXD pin and the USART receiver, making the TXD pin a combined
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 383
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
TxD/RxD line. The RXD pin will be disconnected from the USART receiver and may be controlled by a different
peripheral.
In One-Wire mode, multiple devices are able to manipulate the TxD/RxD line at the same time. In the case where one
device drives the pin to a logical high level (VCC), and another device pulls the line low (GND), a short will occur. To
accommodate this, the USART features an Open-Drain mode (the ODME bit in the USARTn.CTRLB register) which
prevents the transmitter from driving a pin to a logical high level, thereby constraining it to only be able to pull it low.
Combining this function with the internal pull-up feature (the PULLUPEN bit in the PORTx.PINnCTRL register) will let
the line be held high through a pull-up resistor, allowing any device to pull it low. When the line is pulled low the
current from VCC to GND will be limited by the pull-up resistor. The TXD pin is automatically set to output by hardware
when the Open-Drain mode is enabled.
When the USART is transmitting to the TxD/RxD line, it will also receive its own transmission. This can be used to
check for overlapping transmissions by checking if the received data are the same as the transmitted data as it
should be.
RS-485 Mode
RS-485 is a communication standard supported by the USART peripheral. It is a physical interface that defines the
setup of a communication circuit. Data are transmitted using differential signaling, making communication robust
against noise. RS-485 is enabled by writing the RS485 bit (USARTn.CTRLA) to ‘1’.
The RS-485 mode supports external line driver devices that convert a single USART transmission into corresponding
differential pair signals. It implements automatic control of the XDIR pin that can be used to enable transmission or
reception for the line driver device. The USART automatically drives the XDIR pin high while the USART is
transmitting and pulls it low when the transmission is complete. An example of such a circuit is shown in the figure
below.
Figure 27-10. RS-485 Bus Connection
Line Driver
TXD
TX Driver
XDIR
Differential Bus
+
-
USART
RX Driver
RXD
The XDIR pin goes high one baud clock cycle in advance of data being shifted out to allow some guard time to
enable the external line driver. The XDIR pin will remain high for the complete frame including Stop bit(s).
Figure 27-11. XDIR Drive Timing
TxD
St
0
1
2
3
4
5
6
7
Sp1
XDIR
Guard
time
Stop
RS-485 mode is compatible with One-Wire mode. One-Wire mode enables an internal connection between the TXD
pin and the USART receiver, making the TXD pin a combined TxD/RxD line. The RXD pin will be disconnected from
the USART receiver and may be controlled by a different peripheral. An example of such a circuit is shown in the
figure below.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 384
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
Figure 27-12. RS-485 with Loop-Back Mode Connection
TXD
Line Driver
TX Driver
XDIR
Differential Bus
+
-
USART
RXD
RX Driver
27.3.3.2.7 IRCOM Mode of Operation
The USART peripheral can be configured in Infrared Communication mode (IRCOM) which is IrDA® 1.4 compatible
with baud rates up to 115.2 kbps. When enabled, the IRCOM mode enables infrared pulse encoding/decoding for the
USART.
Figure 27-13. Block Diagram
IRCOM
Event System
Events
Encoded RxD
Pulse
Decoding
Decoded RxD
USART
Pulse
Encoding
RXD
TXD
Decoded RxD
Encoded RxD
The USART is set in IRCOM mode by writing 0x02 to the CMODE bits in the USARTn.CTRLC register. The data on
the TXD/RXD pins are the inverted values of the transmitted/received infrared pulse. It is also possible to select an
event channel from the Event System as an input for the IRCOM receiver. This enables the IRCOM to receive input
from the I/O pins or sources other than the corresponding RXD pin. This will disable the RxD input from the USART
pin.
For transmission, three pulse modulation schemes are available:
•
•
•
3/16 of the baud rate period
Fixed programmable pulse time based on the peripheral clock frequency
Pulse modulation disabled
For the reception, a fixed programmable minimum high-level pulse-width for the pulse to be decoded as a logical ‘0’
is used. Shorter pulses will then be discarded, and the bit will be decoded to logical ‘1’ as if no pulse was received.
When IRCOM mode is enabled, Double-Speed mode cannot be used for the USART.
27.3.4
Additional Features
27.3.4.1 Parity
Parity bits can be used by the USART to check the validity of a data frame. The Parity bit is set by the transmitter
based on the number of bits with the value of ‘1’ in a transmission and controlled by the receiver upon reception. If
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 385
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
the Parity bit is inconsistent with the transmission frame, the receiver may assume that the data frame has been
corrupted.
Even or odd parity can be selected for error checking by writing the Parity Mode (PMODE) bits in the
USARTn.CTRLC register. If even parity is selected, the Parity bit is set to ‘1’ if the number of Data bits with value ‘1’
is odd (making the total number of bits with value ‘1’ even). If odd parity is selected, the Parity bit is set to ‘1’ if the
number of data bits with value ‘1’ is even (making the total number of bits with value ‘1’ odd).
When enabled, the parity checker calculates the parity of the data bits in incoming frames and compares the result
with the Parity bit of the corresponding frame. If a parity error is detected, the Parity Error flag (the PERR bit in the
USARTn.RXDATAH register) is set.
If LIN Constrained Auto-Baud mode is enabled (RXMODE = 0x03 in the USARTn.CTRLB register), a parity check is
only performed on the protected identifier field. A parity error is detected if one of the equations below is not true,
which sets the Parity Error flag.
�0 = ��0 XOR ��1 XOR ��2 XOR ��4
�1 = NOT ��1 XOR ��3 XOR ��4 XOR ��5
Figure 27-14. Protected Identifier Field and Mapping of Identifier and Parity Bits
Protected identifier field
St
ID0 ID1 ID2 ID3 ID4 ID5 P0
P1
Sp
27.3.4.2 Start-of-Frame Detection
The Start-of-Frame Detection feature enables the USART to wake up from Standby sleep mode upon data reception.
When a high-to-low transition is detected on the RXD pin, the oscillator is powered up, and the USART peripheral
clock is enabled. After start-up, the rest of the data frame can be received, provided that the baud rate is slow enough
in relation to the oscillator start-up time. The start-up time of the oscillators varies with supply voltage and
temperature. For details on oscillator start-up time characteristics, refer to the Electrical Characteristics section.
If a false Start bit is detected, the device will, if another wake-up source has not been triggered, go back into the
Standby sleep mode.
The Start-of-Frame detection works in Asynchronous mode only. It is enabled by writing the Start-of-Frame Detection
Enable (SFDEN) bit in the USARTn.CTRLB register. If a Start bit is detected while the device is in Standby sleep
mode, the USART Receive Start Interrupt Flag (RXSIF) bit is set.
The USART Receive Complete Interrupt Flag (RXCIF) bit and the RXSIF bit share the same interrupt line, but each
has its dedicated interrupt settings. The table below shows the USART Start Frame Detection modes, depending on
the interrupt setting.
Table 27-6. USART Start Frame Detection Modes
SFDEN RXSIF Interrupt RXCIF Interrupt Comment
0
x
x
Standard mode
1
Disabled
Disabled
Only the oscillator is powered during the frame reception. If the
interrupts are disabled and buffer overflow is ignored, all incoming
frames will be lost
1
Disabled
Enabled
System/all clocks are awakened on Receive Complete interrupt
1
Enabled
x
System/all clocks are awakened when a Start bit is detected
Note: The SLEEP instruction will not shut down the oscillator if there is ongoing communication.
27.3.4.3 Multiprocessor Communication
The Multiprocessor Communication mode (MPCM) effectively reduces the number of incoming frames that have to
be handled by the receiver in a system with multiple microcontrollers communicating via the same serial bus. This
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 386
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
mode is enabled by writing a ‘1’ to the MPCM bit in the Control B register (USARTn.CTRLB). In this mode, a
dedicated bit in the frames is used to indicate whether the frame is an address or data frame type.
If the receiver is set up to receive frames that contain five to eight data bits, the first Stop bit is used to indicate the
frame type. If the receiver is set up for frames with nine data bits, the ninth bit is used to indicate frame type. When
the frame type bit is ‘1’, the frame contains an address. When the frame type bit is ‘0’, the frame is a data frame. If 5to 8-bit character frames are used, the transmitter must be set to use two Stop bits, since the first Stop bit is used for
indicating the frame type.
If a particular slave MCU has been addressed, it will receive the following data frames as usual, while the other slave
MCUs will ignore the frames until another address frame is received.
27.3.4.3.1 Using Multiprocessor Communication
The following procedure should be used to exchange data in Multiprocessor Communication mode (MPCM):
1.
2.
3.
4.
All slave MCUs are in Multiprocessor Communication mode.
The master MCU sends an address frame, and all slaves receive and read this frame.
Each slave MCU determines if it has been selected.
The addressed MCU will disable MPCM and receive all data frames. The other slave MCUs will ignore the
data frames.
When the addressed MCU has received the last data frame, it must enable MPCM again and wait for a new
address frame from the master.
5.
The process then repeats from step 2.
27.3.5
Events
The USART can generate the events described in the table below.
Table 27-7. Event Generators in USART
Generator Name
Description
Peripheral Event
USARTn
XCK
The clock signal in SPI Master
mode and Synchronous USART
Master mode
Event Type
Generating Clock
Domain
Length of Event
Pulse
CLK_PER
One XCK period
The table below describes the event user and its associated functionality.
Table 27-8. Event Users in USART
User Name
27.3.6
Peripheral
Input
USARTn
IREI
Description
USARTn IrDA event input
Input Detection
Async/Sync
Pulse
Sync
Interrupts
Table 27-9. Available Interrupt Vectors and Sources
Name Vector Description
Conditions
RXC
Receive Complete interrupt
DRE
Data Register Empty
interrupt
The transmit buffer is empty/ready to receive new data (DREIE)
TXC
Transmit Complete
interrupt
The entire frame in the Transmit Shift register has been shifted out and there
are no new data in the transmit buffer (TXCIE)
© 2020 Microchip Technology Inc.
•
•
•
There is unread data in the receive buffer (RXCIE)
Receive of Start-of-Frame detected (RXSIE)
Auto-Baud Error/ISFIF flag set (ABEIE)
Preliminary Datasheet
DS40002247A-page 387
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
When an Interrupt condition occurs, the corresponding Interrupt flag is set in the STATUS register
(USARTn.STATUS).
An interrupt source is enabled or disabled by writing to the corresponding bit in the Control A register
(USARTn.CTRLA).
An interrupt request is generated when the corresponding interrupt source is enabled, and the Interrupt flag is set.
The interrupt request remains active until the Interrupt flag is cleared. See the USARTn.STATUS register for details
on how to clear Interrupt flags.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 388
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
27.4
Register Summary
Offset
Name
Bit Pos.
0x00
0x01
0x02
0x03
0x04
0x05
0x06
0x07
0x07
RXDATAL
RXDATAH
TXDATAL
TXDATAH
STATUS
CTRLA
CTRLB
CTRLC
CTRLC
0x08
BAUD
0x0A
0x0B
0x0C
0x0D
0x0E
CTRLD
DBGCTRL
EVCTRL
TXPLCTRL
RXPLCTRL
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
15:8
7:0
7:0
7:0
7:0
7:0
27.5
7
6
RXCIF
BUFOVF
5
4
3
2
1
0
FERR
PERR
DATA[8]
DATA[7:0]
DATA[7:0]
RXCIF
TXCIF
RXCIE
TXCIE
RXEN
TXEN
CMODE[1:0]
CMODE[1:0]
DREIF
DREIE
RXSIF
RXSIE
SFDEN
PMODE[1:0]
ISFIF
LBME
ODME
SBMODE
BDF
ABEIE
RXMODE[1:0]
CHSIZE[2:0]
UDORD
UCPHA
DATA[8]
WFB
RS485
MPCM
BAUD[7:0]
BAUD[15:8]
ABW[1:0]
DBGRUN
IREI
TXPL[7:0]
RXPL[6:0]
Register Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 389
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
27.5.1
Receiver Data Register Low Byte
Name:
Offset:
Reset:
Property:
RXDATAL
0x00
0x00
-
This register contains the eight LSbs of the data received by the USART receiver. The USART receiver is doublebuffered, and this register always represents the data for the oldest received frame. If the data for only one frame is
present in the receive buffer, this register contains that data.
The buffer shifts out the data either when RXDATAL or RXDATAH is read, depending on the configuration. The
register which does not lead to data being shifted should be read first to be to able read both bytes before shifting.
When the Character Size (CHSIZE) bits in the Control C (USARTn.CTRLC) register is configured to 9-bit (low byte
first), a read of RXDATAH shifts the receive buffer. Otherwise, RXDATAL shifts the buffer.
Bit
7
6
5
4
3
2
1
0
R
0
R
0
R
0
R
0
DATA[7:0]
Access
Reset
R
0
R
0
R
0
R
0
Bits 7:0 – DATA[7:0] Receiver Data Register
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 390
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
27.5.2
Receiver Data Register High Byte
Name:
Offset:
Reset:
Property:
RXDATAH
0x01
0x00
-
This register contains the MSb of the data received by the USART receiver, as well as status bits reflecting the status
of the received data frame. The USART receiver is double-buffered, and this register always represents the data and
status bits for the oldest received frame. If the data and status bits for only one frame is present in the receive buffer,
this register contains that data.
The buffer shifts out the data either when RXDATAL or RXDATAH is read, depending on the configuration. The
register which does not lead to data being shifted should be read first to be able to read both bytes before shifting.
When the Character Size (CHSIZE) bits in the Control C (USARTn.CTRLC) register is configured to 9-bit (low byte
first), a read of RXDATAH shifts the receive buffer. Otherwise, RXDATAL shifts the buffer.
Bit
Access
Reset
7
RXCIF
R
0
6
BUFOVF
R
0
5
4
3
2
FERR
R
0
1
PERR
R
0
0
DATA[8]
R
0
Bit 7 – RXCIF USART Receive Complete Interrupt Flag
This flag is set when there are unread data in the receive buffer and cleared when the receive buffer is empty.
Bit 6 – BUFOVF Buffer Overflow
This flag is set if a buffer overflow is detected. A buffer overflow occurs when the receive buffer is full, a new frame is
waiting in the receive shift register, and a new Start bit is detected. This flag is cleared when the Receiver Data
(USARTn.RXDATAL and USARTn.RXDATAH) registers are read.
This flag is not used in the Master SPI mode of operation.
Bit 2 – FERR Frame Error
This flag is set if the first Stop bit is ‘0’ and cleared when it is correctly read as ‘1’.
This flag is not used in the Master SPI mode of operation.
Bit 1 – PERR Parity Error
This flag is set if parity checking is enabled, and the received data has a parity error. This flag is otherwise cleared.
For details on parity calculation, refer to 27.3.4.1 Parity.
This flag is not used in the Master SPI mode of operation.
Bit 0 – DATA[8] Receiver Data Register
When using a 9-bit frame size, this bit holds the ninth bit (MSb) of the received data.
When the Receiver Mode (RXMODE) bits in the Control B (USARTn.CTRLB) register is configured to LIN
Constrained Auto-Baud (LINAUTO) mode, this bit indicates if the received data are within the response space of a
LIN frame. This bit is cleared if the received data are in the protected identifier field and is otherwise set.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 391
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
27.5.3
Transmit Data Register Low Byte
Name:
Offset:
Reset:
Property:
TXDATAL
0x02
0x00
-
The data written to this register is automatically loaded into the dedicated shift register. The shift register outputs
each of the bits serially to the TXD pin.
When using a 9-bit frame size, the ninth bit (MSb) should be written to USARTn.TXDATAH. In that case, the buffer
shifts data either when TXDATAL or TXDATAH is written, depending on the configuration. The register which does
not lead to data being shifted should be written first to be able to write both registers before shifting.
When the Character Size (CHSIZE) bits in the Control C (USARTn.CTRLC) register is configured to 9-bit (low byte
first), a write of TXDATAH shifts the transmit buffer. Otherwise, TXDATAL shifts the buffer.
This register may only be written when the Data Register Empty Interrupt Flag (DREIF) in the Status
(USARTn.STATUS) register is set.
Bit
7
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
DATA[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – DATA[7:0] Transmit Data Register
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 392
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
27.5.4
Transmit Data Register High Byte
Name:
Offset:
Reset:
Property:
TXDATAH
0x03
0x00
-
The data written to this register is automatically loaded into the dedicated shift register. The shift register outputs
each of the bits serially to the TXD pin.
When using a 9-bit frame size, the ninth bit (MSb) should be written to USARTn.TXDATAH. In that case, the buffer
shifts data either when TXDATAL or TXDATAH is written, depending on the configuration. The register which does
not lead to data being shifted should be written first to be able to write both bytes before shifting.
When the Character Size (CHSIZE) bits in the Control C (USARTn.CTRLC) register is configured to 9-bit (low byte
first), a write of TXDATAH shifts the transmit buffer. Otherwise, TXDATAL shifts the buffer.
This register may only be written when the Data Register Empty Interrupt Flag (DREIF) in the Status
(USARTn.STATUS) register is set.
Bit
7
6
5
4
3
Access
Reset
2
1
0
DATA[8]
R/W
0
Bit 0 – DATA[8] Transmit Data Register
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 393
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
27.5.5
USART Status Register
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
RXCIF
R
0
STATUS
0x04
0x00
-
6
TXCIF
R/W
0
5
DREIF
R
1
4
RXSIF
R/W
0
3
ISFIF
R/W
0
2
1
BDF
R/W
0
0
WFB
W
0
Bit 7 – RXCIF USART Receive Complete Interrupt Flag
This flag is set when there are unread data in the receive buffer and cleared when the receive buffer is empty.
Bit 6 – TXCIF USART Transmit Complete Interrupt Flag
This flag is set when the entire frame in the Transmit Shift register has been shifted out, and there are no new data in
the transmit buffer (TXDATAL and TXDATAH) registers. It is cleared by writing a ‘1’ to it.
Bit 5 – DREIF USART Data Register Empty Interrupt Flag
This flag is set when the transmit buffer (TXDATAL and TXDATAH) registers are empty and cleared when they
contain data that has not yet been moved into the transmit shift register.
Bit 4 – RXSIF USART Receive Start Interrupt Flag
This flag is set when Start-of-Frame detection is enabled, the device is in Standby sleep mode, and a valid start bit is
detected. It is cleared by writing a ‘1’ to it.
This flag is not used in the Master SPI mode operation.
Bit 3 – ISFIF Inconsistent Synchronization Field Interrupt Flag
This flag is set if an auto-baud mode is enabled, and the synchronization field is too short or too long to give a valid
baud setting. It will also be set when USART is set to LINAUTO mode, and the SYNC character differs from data
value 0x55. This flag is cleared by writing a ‘1’ to it. See section 27.3.3.2.5 Auto-Baud for more information.
Bit 1 – BDF Break Detected Flag
This flag is set if an auto-baud mode is enabled and a valid break and synchronization character is detected, and is
cleared when the next data is received. It can also be cleared by writing a ‘1’ to it. See section 27.3.3.2.5 Auto-Baud
for more information.
Bit 0 – WFB Wait For Break
This bit controls whether the Wait For Break feature is enabled or not. Refer to section 27.3.3.2.5 Auto-Baud for
more information.
Value
Description
0
Wait For Break is disabled
1
Wait For Break is enabled
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 394
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
27.5.6
Control A
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
RXCIE
R/W
0
CTRLA
0x05
0x00
-
6
TXCIE
R/W
0
5
DREIE
R/W
0
4
RXSIE
R/W
0
3
LBME
R/W
0
2
ABEIE
R/W
0
1
0
RS485
R/W
0
Bit 7 – RXCIE Receive Complete Interrupt Enable
This bit controls whether the Receive Complete Interrupt is enabled or not. When enabled, the interrupt will be
triggered when the RXCIF bit in the USARTn.STATUS register is set.
Value
Description
0
The Receive Complete Interrupt is disabled
1
The Receive Complete Interrupt is enabled
Bit 6 – TXCIE Transmit Complete Interrupt Enable
This bit controls whether the Transmit Complete Interrupt is enabled or not. When enabled, the interrupt will be
triggered when the TXCIF bit in the USARTn.STATUS register is set.
Value
Description
0
The Transmit Complete Interrupt is disabled
1
The Transmit Complete Interrupt is enabled
Bit 5 – DREIE Data Register Empty Interrupt Enable
This bit controls whether the Data Register Empty Interrupt is enabled or not. When enabled, the interrupt will be
triggered when the DREIF bit in the USARTn.STATUS register is set.
Value
Description
0
The Data Register Empty Interrupt is disabled
1
The Data Register Empty Interrupt is enabled
Bit 4 – RXSIE Receiver Start Frame Interrupt Enable
This bit controls whether the Receiver Start Frame Interrupt is enabled or not. When enabled, the interrupt will be
triggered when the RXSIF bit in the USARTn.STATUS register is set.
Value
Description
0
The Receiver Start Frame Interrupt is disabled
1
The Receiver Start Frame Interrupt is enabled
Bit 3 – LBME Loop-back Mode Enable
This bit controls whether the Loop-back mode is enabled or not. When enabled, an internal connection between the
TXD pin and the USART receiver is created, and the input from the RXD pin to the USART receiver is disconnected.
Value
Description
0
Loop-back mode is disabled
1
Loop-back mode is enabled
Bit 2 – ABEIE Auto-baud Error Interrupt Enable
This bit controls whether the Auto-baud Error Interrupt is enabled or not. When enabled, the interrupt will be triggered
when the ISFIF bit in the USARTn.STATUS register is set.
Value
Description
0
The Auto-baud Error Interrupt is disabled
1
The Auto-baud Error Interrupt is enabled
Bit 0 – RS485 RS-485 Mode
This bit controls whether the RS-485 mode is enabled or not. Refer to section RS-485 Mode for more information.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 395
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
Value
0
1
Description
RS-485 mode is disabled
RS-485 mode is enabled
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 396
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
27.5.7
Control B
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
RXEN
R/W
0
CTRLB
0x06
0x00
-
6
TXEN
R/W
0
5
4
SFDEN
R/W
0
3
ODME
R/W
0
2
1
RXMODE[1:0]
R/W
R/W
0
0
0
MPCM
R/W
0
Bit 7 – RXEN Receiver Enable
This bit controls whether the USART receiver is enabled or not. Refer to section 27.3.2.4.2 Disabling the Receiver
for more information.
Value
Description
0
The USART receiver is disabled
1
The USART receiver is enabled
Bit 6 – TXEN Transmitter Enable
This bit controls whether the USART transmitter is enabled or not. Refer to section 27.3.2.3.1 Disabling the
Transmitter for more information.
Value
Description
0
The USART transmitter is disabled
1
The USART transmitter is enabled
Bit 4 – SFDEN Start-of-Frame Detection Enable
This bit controls whether the USART Start-of-Frame Detection mode is enabled or not. Refer to section 27.3.4.2
Start-of-Frame Detection for more information.
Value
Description
0
The USART Start-of-Frame Detection mode is disabled
1
The USART Start-of-Frame Detection mode is enabled
Bit 3 – ODME Open Drain Mode Enable
This bit controls whether Open Drain mode is enabled or not. See section One-Wire Mode for more information.
Value
Description
0
Open Drain mode is disabled
1
Open Drain mode is enabled
Bits 2:1 – RXMODE[1:0] Receiver Mode
Writing these bits selects the receiver mode of the USART.
• Writing the bits to 0x00 enables Normal-Speed (NORMAL) mode. When the USART Communication Mode
(CMODE) bits in the Control C (USARTn.CTRLC) register is configured to Asynchronous USART
(ASYNCHRONOUS) or Infrared Communication (IRCOM), the RXMODE bits should always be written to 0x00.
• Writing the bits to 0x01 enables Double-Speed (CLK2X) mode. Refer to section 27.3.3.2.4 Double-Speed
Operation for more information.
• Writing the bits to 0x02 enables Generic Auto-Baud (GENAUTO) mode. Refer to section 27.3.3.2.5 Auto-Baud
for more information.
• Writing the bits to 0x03 enables Lin Constrained Auto-Baud (LINAUTO) mode. Refer to section 27.3.3.2.5 AutoBaud for more information.
Value
0x00
0x01
0x02
0x03
Name
NORMAL
CLK2X
GENAUTO
LINAUTO
© 2020 Microchip Technology Inc.
Description
Normal-Speed mode
Double-Speed mode
Generic Auto-Baud mode
LIN Constrained Auto-Baud mode
Preliminary Datasheet
DS40002247A-page 397
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
Bit 0 – MPCM Multi-Processor Communication Mode
This bit controls whether the Multi-Processor Communication mode is enabled or not. Refer to section 27.3.4.3
Multiprocessor Communication for more information.
Value
Description
0
Multi-Processor Communication mode is disabled
1
Multi-Processor Communication mode is enabled
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 398
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
27.5.8
Control C - Normal Mode
Name:
Offset:
Reset:
Property:
CTRLC
0x07
0x03
-
This register description is valid for all modes except the Master SPI mode. When the USART Communication Mode
bits (CMODE) in this register are written to ‘MSPI’, see CTRLC - Master SPI mode for the correct description.
Bit
Access
Reset
7
6
CMODE[1:0]
R/W
R/W
0
0
5
4
PMODE[1:0]
R/W
R/W
0
0
3
SBMODE
R/W
0
2
R/W
0
1
CHSIZE[2:0]
R/W
1
0
R/W
1
Bits 7:6 – CMODE[1:0] USART Communication Mode
These bits select the communication mode of the USART.
Writing a 0x03 to these bits alters the available bit fields in this register, see CTRLC - Master SPI mode.
Value
Name
Description
0x00
ASYNCHRONOUS
Asynchronous USART
0x01
SYNCHRONOUS
Synchronous USART
0x02
IRCOM
Infrared Communication
0x03
MSPI
Master SPI
Bits 5:4 – PMODE[1:0] Parity Mode
These bits enable and select the type of parity generation. See section 27.3.4.1 Parity for more information.
Value
Name
Description
0x0
DISABLED
Disabled
0x1
Reserved
0x2
EVEN
Enabled, even parity
0x3
ODD
Enabled, odd parity
Bit 3 – SBMODE Stop Bit Mode
This bit selects the number of Stop bits to be inserted by the transmitter.
The receiver ignores this setting.
Value
Description
0
1 Stop bit
1
2 Stop bits
Bits 2:0 – CHSIZE[2:0] Character Size
These bits select the number of data bits in a frame. The receiver and transmitter use the same setting. For 9BIT
character size, the order of which byte to read or write first, low or high byte of RXDATA or TXDATA, can be
configured.
Value
Name
Description
0x00
5BIT
5-bit
0x01
6BIT
6-bit
0x02
7BIT
7-bit
0x03
8BIT
8-bit
0x04
Reserved
0x05
Reserved
0x06
9BITL
9-bit (Low byte first)
0x07
9BITH
9-bit (High byte first)
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 399
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
27.5.9
Control C - Master SPI Mode
Name:
Offset:
Reset:
Property:
CTRLC
0x07
0x00
-
This register description is valid only when the USART is in Master SPI mode (CMODE written to MSPI). For other
CMODE values, see CTRLC - Normal Mode.
See 27.3.3.1.3 USART in Master SPI Mode for a full description of the Master SPI mode operation.
Bit
Access
Reset
7
6
CMODE[1:0]
R/W
R/W
0
0
5
4
3
2
UDORD
R/W
0
1
UCPHA
R/W
0
0
Bits 7:6 – CMODE[1:0] USART Communication Mode
These bits select the communication mode of the USART.
Writing a value different than 0x03 to these bits alters the available bit fields in this register, see CTRLC - Normal
Mode.
Value
Name
Description
0x00
ASYNCHRONOUS
Asynchronous USART
0x01
SYNCHRONOUS
Synchronous USART
0x02
IRCOM
Infrared Communication
0x03
MSPI
Master SPI
Bit 2 – UDORD USART Data Order
This bit controls the frame format. The receiver and transmitter use the same setting. Changing the setting of the
UDORD bit will corrupt all ongoing communication for both the receiver and the transmitter.
Value
Description
0
MSb of the data word is transmitted first
1
LSb of the data word is transmitted first
Bit 1 – UCPHA USART Clock Phase
This bit controls the phase of the interface clock. Refer to section Clock Generation for more information.
Value
Description
0
Data are sampled on the leading (first) edge
1
Data are sampled on the trailing (last) edge
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 400
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
27.5.10 Baud Register
Name:
Offset:
Reset:
Property:
BAUD
0x08
0x00
-
The USARTn.BAUDL and USARTn.BAUDH register pair represents the 16-bit value, USARTn.BAUD. The low byte
[7:0] (suffix L) is accessible at the original offset. The high byte [15:8] (suffix H) can be accessed at offset + 0x01.
Ongoing transmissions of the transmitter and receiver will be corrupted if the baud rate is changed. Writing to this
register will trigger an immediate update of the baud rate prescaler. For more information on how to set the baud rate,
see Table 27-1, Equations for Calculating Baud Rate Register Setting.
Bit
15
14
13
12
11
10
9
8
R/W
0
R/W
0
R/W
0
R/W
0
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
BAUD[15:8]
Access
Reset
Bit
R/W
0
R/W
0
R/W
0
R/W
0
7
6
5
4
BAUD[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 15:8 – BAUD[15:8] USART Baud Rate High Byte
These bits hold the MSB of the 16-bit Baud register.
Bits 7:0 – BAUD[7:0] USART Baud Rate Low Byte
These bits hold the LSB of the 16-bit Baud register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 401
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
27.5.11 Control D
Name:
Offset:
Reset:
Property:
Bit
CTRLD
0x0A
0x00
-
7
6
5
4
3
2
1
0
ABW[1:0]
Access
Reset
R/W
0
R/W
0
Bits 7:6 – ABW[1:0] Auto-baud Window Size
These bits control the tolerance for the difference between the baud rates between the two synchronizing devices
when using Lin Constrained Auto-baud mode. The tolerance is based on the number of baud samples between every
two bits. When baud rates are identical, there should be 32 baud samples between each bit pair since each bit is
sampled 16 times.
Value
Name
Description
0x00
WDW0
32±6 (18% tolerance)
0x01
WDW1
32±5 (15% tolerance)
0x02
WDW2
32±7 (21% tolerance)
0x03
WDW3
32±8 (25% tolerance)
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 402
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
27.5.12 Debug Control Register
Name:
Offset:
Reset:
Property:
Bit
7
DBGCTRL
0x0B
0x00
-
6
5
4
3
2
Access
Reset
1
0
DBGRUN
R/W
0
Bit 0 – DBGRUN Debug Run
Value
Description
0
The peripheral is halted in Break Debug mode and ignores events.
1
The peripheral will continue to run in Break Debug mode when the CPU is halted.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 403
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
27.5.13 IrDA Control Register
Name:
Offset:
Reset:
Property:
Bit
7
EVCTRL
0x0C
0x00
-
6
5
4
3
Access
Reset
2
1
0
IREI
R/W
0
Bit 0 – IREI IrDA Event Input Enable
This bit controls whether the IrDA event input is enabled or not. See section 27.3.3.2.7 IRCOM Mode of Operation
for more information.
Value
Description
0
IrDA Event input is enabled
1
IrDA Event input is disabled
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 404
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
27.5.14 IRCOM Transmitter Pulse Length Control Register
Name:
Offset:
Reset:
Property:
Bit
7
TXPLCTRL
0x0D
0x00
-
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
TXPL[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – TXPL[7:0] Transmitter Pulse Length
This 8-bit value sets the pulse modulation scheme for the transmitter. Setting this register will have effect only if
IRCOM mode is selected by the USART, and it must be configured before the USART transmitter is enabled (TXEN).
Value
Description
0x00
3/16 of the baud rate period pulse modulation is used
0x01Fixed pulse length coding is used. The 8-bit value sets the number of peripheral clock periods for the
0xFE
pulse. The start of the pulse will be synchronized with the rising edge of the baud rate clock.
0xFF
Pulse coding disabled. RX and TX signals pass through the IRCOM module unaltered. This enables
other features through the IRCOM module, such as half-duplex USART, loop-back testing, and USART
RX input from an event channel.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 405
AVR128DB28/32/48/64
USART - Universal Synchronous and Asynchrono...
27.5.15 IRCOM Receiver Pulse Length Control Register
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
RXPLCTRL
0x0E
0x00
-
6
5
4
R/W
0
R/W
0
R/W
0
3
RXPL[6:0]
R/W
0
2
1
0
R/W
0
R/W
0
R/W
0
Bits 6:0 – RXPL[6:0] Receiver Pulse Length
This 7-bit value sets the filter coefficient for the IRCOM transceiver. Setting this register will only have effect if IRCOM
mode is selected by a USART, and it must be configured before the USART receiver is enabled (RXEN).
Value
Description
0x00
Filtering disabled
0x01Filtering enabled. The value of RXPL+1 represents the number of samples required for a received
0x7F
pulse to be accepted.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 406
AVR128DB28/32/48/64
SPI - Serial Peripheral Interface
28.
SPI - Serial Peripheral Interface
28.1
Features
•
•
•
•
•
•
•
•
28.2
Full Duplex, Three-Wire Synchronous Data Transfer
Master or Slave Operation
LSb First or MSb First Data Transfer
Seven Programmable Bit Rates
End of Transmission Interrupt Flag
Write Collision Flag Protection
Wake-up from Idle Mode
Double-Speed (CK/2) Master SPI Mode
Overview
The Serial Peripheral Interface (SPI) is a high-speed synchronous data transfer interface using three or four pins. It
allows full duplex communication between an AVR® device and peripheral devices, or between several
microcontrollers. The SPI peripheral can be configured as either master or slave. The master initiates and controls all
data transactions.
The interconnection between master and slave devices with SPI is shown in the block diagram. The system consists
of two shift registers and a master clock generator. The SPI master initiates the communication cycle by pulling the
desired slave’s Slave Select (SS) signal low. The master and slave prepare the data to be sent to their respective
shift registers, and the master generates the required clock pulses on the SCK line to exchange data. Data are
always shifted from master to slave on the master output, slave input (MOSI) line, and from slave to master on the
master input, slave output (MISO) line.
28.2.1
Block Diagram
Figure 28-1. SPI Block Diagram
SLAVE
MASTER
DATA
DATA
Transmit Data
Buffer
Transmit Data
Buffer
Buffer
mode
Buffer
mode
LSb
MSb
MISO
MISO
MOSI
MOSI
8-bit Shift Register
LSb
MSb
8-bit Shift Register
SPI CLOCK
GENERATOR
SCK
SCK
SS
SS
Receive Data
Receive Data
Receive Data
Buffer
Buffer
mode
Receive Data
Buffer
Buffer
mode
DATA
DATA
The SPI is built around an 8-bit shift register that will shift data out and in at the same time. The Transmit Data
register and the Receive Data register are not physical registers but are mapped to other registers when written or
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 407
AVR128DB28/32/48/64
SPI - Serial Peripheral Interface
read: Writing the Transmit Data (SPIn.DATA) register will write the shift register in Normal mode and the Transmit
Buffer register in Buffer mode. Reading the Receive Data (SPIn.DATA) register will read the Receive Data register in
Normal mode and the Receive Data Buffer in Buffer mode.
In Master mode, the SPI has a clock generator to generate the SCK clock. In Slave mode, the received SCK clock is
synchronized and sampled to trigger the shifting of data in the shift register.
28.2.2
Signal Description
Table 28-1. Signals in Master and Slave Mode
Signal
Pin Configuration
Description
Master Mode
defined(1)
Slave Mode
MOSI
Master Out Slave In
User
MISO
Master In Slave Out
Input
SCK
Serial Clock
User defined(1)
Input
Slave Select
defined(1)
Input
SS
User
Input
User defined(1,2)
Notes:
1. If the pin data direction is configured as output, the pin level is controlled by the SPI.
2. If the SPI is in Slave mode and the MISO pin data direction is configured as output, the SS pin controls the
MISO pin output in the following way:
– If the SS pin is driven low, the MISO pin is controlled by the SPI.
– If the SS pin is driven high, the MISO pin is tri-stated.
When the SPI module is enabled, the pin data direction for the signals marked with “Input” in Table 28-1 is
overridden.
28.3
Functional Description
28.3.1
Initialization
Initialize the SPI to a basic functional state by following these steps:
1. Configure the SS pin in the port peripheral.
2. Select the SPI master/slave operation by writing the Master/Slave Select (MASTER) bit in the Control A
(SPIn.CTRLA) register.
3. In Master mode, select the clock speed by writing the Prescaler (PRESC) bits and the Clock Double (CLK2X)
bit in SPIn.CTRLA.
4. Optional: Select the Data Transfer mode by writing to the MODE bits in the Control B (SPIn.CTRLB) register.
5. Optional: Write the Data Order (DORD) bit in SPIn.CTRLA.
6. Optional: Set up the Buffer mode by writing the BUFEN and BUFWR bits in the Control B (SPIn.CTRLB)
register.
7. Optional: To disable the multi-master support in Master mode, write ‘1’ to the Slave Select Disable (SSD) bit in
SPIn.CTRLB.
8. Enable the SPI by writing a ‘1’ to the ENABLE bit in SPIn.CTRLA.
28.3.2
Operation
28.3.2.1 Master Mode Operation
When the SPI is configured in Master mode, a write to the SPIn.DATA register will start a new transfer. The SPI
master can operate in two modes, Normal and Buffer, as explained below.
28.3.2.1.1 Normal Mode
In Normal mode, the system is single-buffered in the transmit direction and double-buffered in the receive direction.
This influences the data handling in the following ways:
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 408
AVR128DB28/32/48/64
SPI - Serial Peripheral Interface
1.
New bytes to be sent cannot be written to the DATA (SPIn.DATA) register before the entire transfer has
completed. A premature write will cause corruption of the transmitted data and the Write Collision (WRCOL)
flag in SPIn.INTFLAGS will be set.
Received bytes are written to the Receive Data Buffer register immediately after the transmission is
completed.
The Receive Data Buffer register has to be read before the next transmission is completed, or the data will be
lost. This register is read by reading SPIn.DATA.
The Transmit Data Buffer and Receive Data Buffer registers are not used in Normal mode.
2.
3.
4.
After a transfer has completed, the Interrupt Flag (IF) will be set in the Interrupt Flags (SPIn.INTFLAGS) register. This
will cause the corresponding interrupt to be executed if this interrupt and the global interrupts are enabled. Setting the
Interrupt Enable (IE) bit in the Interrupt Control (SPIn.INTCTRL) register will enable the interrupt.
28.3.2.1.2 Buffer Mode
The Buffer mode is enabled by writing the BUFEN bit in the SPIn.CTRLB register to ‘1’. The BUFWR bit in
SPIn.CTRLB has no effect in Master mode. In Buffer mode, the system is double-buffered in the transmit direction
and triple-buffered in the receive direction. This influences the data handling the following ways:
1. New bytes can be written to the DATA (SPIn.DATA) register as long as the Data Register Empty Interrupt Flag
(DREIF) in the Interrupt Flag (SPIn.INTFLAGS) register is set. The first write will be transmitted right away, and
the following write will go to the Transmit Data Buffer register.
2. A received byte is placed in a two-entry Receive First-In, First-Out (RX FIFO) queue comprised of the Receive
Data register and Receive Data Buffer immediately after the transmission is completed.
3. The DATA register is used to read from the RX FIFO. The RX FIFO must be read at least every second
transfer to avoid any loss of data.
When both the shift register and the Transmit Data Buffer register become empty, the Transfer Complete Interrupt
Flag (TXCIF) in the Interrupt Flags (SPIn.INTFLAGS) register will be set. This will cause the corresponding interrupt
to be executed if this interrupt and the global interrupts are enabled. Setting the Transfer Complete Interrupt Enable
(TXCIE) in the Interrupt Control (SPIn.INTCTRL) register enables the Transfer Complete Interrupt.
28.3.2.1.3 SS Pin Functionality in Master Mode - Multi-Master Support
In Master mode, the Slave Select Disable (SSD) bit in the Control B (SPIn.CTRLB) register controls how the SPI
uses the SS pin.
•
•
•
If SSD in SPIn.CTRLB is ‘0’, the SPI can use the SS pin to transition from Master to Slave mode. This allows
multiple SPI masters on the same SPI bus.
If SSD in SPIn.CTRLB is ‘0’, and the SS pin is configured as an output pin, it can be used as a regular I/O pin or
by other peripheral modules, and will not affect the SPI system.
If SSD in SPIn.CTRLB is ‘1’, the SPI does not use the SS pin, and it can be used as a regular I/O pin, or by
other peripheral modules.
If the SSD bit in SPIn.CTRLB is ‘0’, and the SS is configured as an input pin, the SS pin must be held high to ensure
Master SPI operation. A low level will be interpreted as another Master is trying to take control of the bus. This will
switch the SPI into Slave mode, and the hardware of the SPI will perform the following actions:
1. The Master (MASTER) bit in the SPI Control A (SPIn.CTRLA) register is cleared, and the SPI system
becomes a slave. The direction of the SPI pins will be switched when the conditions in Table 28-2 are met.
2. The Interrupt Flag (IF) bit in the Interrupt Flags (SPIn.INTFLAGS) register will be set. If the interrupt is enabled
and the global interrupts are enabled, the interrupt routine will be executed.
Table 28-2. Overview of the SS Pin Functionality when the SSD Bit in SPIn.CTRLB is ‘0’
SS Configuration
Input
Output
© 2020 Microchip Technology Inc.
SS Pin-Level
Description
High
Master activated (selected)
Low
Master deactivated, switched to
Slave mode
High
Master activated (selected)
Low
Preliminary Datasheet
DS40002247A-page 409
AVR128DB28/32/48/64
SPI - Serial Peripheral Interface
Note: If the device is in Master mode and it cannot be ensured that the SS pin will stay high between two
transmissions, the status of the Master (MASTER) bit in SPIn.CTRLA has to be checked before a new byte is written.
After the Master bit has been cleared by a low level on the SS line, it must be set by the application to re-enable the
SPI Master mode.
28.3.2.2 Slave Mode
In Slave mode, the SPI peripheral receives SPI clock and Slave Select from a Master. Slave mode supports three
operational modes: One Normal mode and two configurations for the Buffered mode. In Slave mode, the control logic
will sample the incoming signal on the SCK pin. To ensure correct sampling of this clock signal, the minimum low and
high periods must each be longer than two peripheral clock cycles.
28.3.2.2.1 Normal Mode
In Normal mode, the SPI peripheral will remain Idle as long as the SS pin is driven high. In this state, the software
may update the contents of the DATA register, but the data will not be shifted out by incoming clock pulses on the
SCK pin until the SS pin is driven low. If the SS pin is driven low, the slave will start to shift out data on the first SCK
clock pulse. When one byte has been completely shifted, the SPI Interrupt Flag (IF) in SPIn.INTFLAGS is set.
The user application may continue placing new data to be sent into the DATA register before reading the incoming
data. New bytes to be sent cannot be written to the DATA register before the entire transfer has completed. A
premature write will be ignored, and the hardware will set the Write Collision (WRCOL) flag in SPIn.INTFLAGS.
When the SS pin is driven high, the SPI logic is halted, and the SPI slave will not receive any new data. Any partially
received packet in the shift register will be lost.
Figure 28-2 shows a transmission sequence in Normal mode. Notice how the value 0x45 is written to the DATA
register but never transmitted.
Figure 28-2. SPI Timing Diagram in Normal Mode (Buffer Mode Not Enabled)
SS
SCK
Write DATA
Write value
0x43
0x44
0x45
0x46
WRCOL
IF
Shift Register
Data sent
0x43
0x43
0x44
0x44
0x46
0x46
The figure above shows three transfers and one write to the DATA register while the SPI is busy with a transfer. This
write will be ignored, and the Write Collision (WRCOL) flag in SPIn.INTFLAGS is set.
28.3.2.2.2 Buffer Mode
To avoid data collisions, the SPI peripheral can be configured in Buffered mode by writing a ‘1’ to the Buffer Mode
Enable (BUFEN) bit in the Control B (SPIn.CTRLB) register. In this mode, the SPI has additional interrupt flags and
extra buffers. The extra buffers are shown in Figure 28-1. There are two different modes for the Buffer mode,
selected with the Buffer mode Wait for Receive (BUFWR) bit. The two different modes are described below with
timing diagrams.
Slave Buffer Mode with Wait for Receive Bit Written to ‘0’
In Slave mode, if the Buffer mode Wait for Receive (BUFWR) bit in SPIn.CTRLB is written to ‘0’, a dummy byte will
be sent before the transmission of user data starts. Figure 28-3 shows a transmission sequence with this
configuration. Notice how the value 0x45 is written to the Data (SPIn.DATA) register but never transmitted.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 410
AVR128DB28/32/48/64
SPI - Serial Peripheral Interface
Figure 28-3. SPI Timing Diagram in Buffer Mode with BUFWR in SPIn.CTRLB Written to ‘0’
SS
SCK
Write DATA
Write value
0x43
0x44
0x45
0x46
DREIF
TXCIF
RXCIF
Transmit Buffer
Shift Register
Data sent
0x43
Dummy
Dummy
0x46
0x44
0x44
0x43
0x43
0x44
0x46
0x46
When the Wait for Receive (BUFWR) bit in SPIn.CTRLB is written to ‘0’, all writes to the Data (SPIn.DATA) register
go to the Transmit Data Buffer register. The figure above shows that the value 0x43 is written to the Data
(SPIn.DATA) register, but it is not immediately transferred to the shift register, so the first byte sent will be a dummy
byte. The value of the dummy byte equals the values that were in the shift register at the time. After the first dummy
transfer is completed, the value 0x43 is transferred to the shift register. Then 0x44 is written to the Data (SPIn.DATA)
register and goes to the Transmit Data Buffer register. A new transfer is started, and 0x43 will be sent. The value
0x45 is written to the Data (SPIn.DATA) register, but the Transmit Data Buffer register is not updated since it is
already full containing 0x44 and the Data Register Empty Interrupt Flag (DREIF) in SPIn.INTFLAGS is low. The value
0x45 will be lost. After the transfer, the value 0x44 is moved to the shift register. During the next transfer, 0x46 is
written to the Data (SPIn.DATA) register, and 0x44 is sent out. After the transfer is complete, 0x46 is copied into the
shift register and sent out in the next transfer.
The DREIF goes low every time the Transmit Data Buffer register is written and goes high after a transfer when the
previous value in the Transmit Data Buffer register is copied into the shift register. The Receive Complete Interrupt
Flag (RXCIF) in SPIn.INTFLAGS is set one cycle after the DREIF goes high. The Transfer Complete Interrupt Flag is
set one cycle after the Receive Complete Interrupt Flag is set when both the value in the shift register and the
Transmit Data Buffer register have been sent.
Slave Buffer Mode with Wait for Receive Bit Written to ‘1’
In Slave mode, if the Buffer mode Wait for Receive (BUFWR) bit in SPIn.CRTLB is written to ‘1’, the transmission of
user data starts as soon as the SS pin is driven low. Figure 28-4 shows a transmission sequence with this
configuration. Notice how the value 0x45 is written to the Data (SPIn.DATA) register but never transmitted.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 411
AVR128DB28/32/48/64
SPI - Serial Peripheral Interface
Figure 28-4. SPI Timing Diagram in Buffer Mode with CTRLB.BUFWR Written to ‘1’
SS
SCK
Write DATA
Write value
0x43
0x44
0x45
0x46
DREIF
TXCIF
RXCIF
Transmit Buffer
Shift Register
0x46
0x44
0x43
0x43
Data sent
0x46
0x44
0x43
0x44
0x46
All writes to the Data (SPIn.DATA) register go to the Transmit Data Buffer register. The figure above shows that the
value 0x43 is written to the Data (SPIn.DATA) register, and since the SS pin is high, it is copied to the shift register in
the next cycle. Then the next write (0x44) will go to the Transmit Data Buffer register. During the first transfer, the
value 0x43 will be shifted out. In the figure above, the value 0x45 is written to the Data (SPIn.DATA) register, but the
Transmit Data Buffer register is not updated since the DREIF is low. After the transfer is completed, the value 0x44
from the Transmit Data Buffer register is copied to the shift register. The value 0x46 is written to the Transmit Data
Buffer register. During the next two transfers, 0x44 and 0x46 are shifted out. The flags behave identically to the Buffer
Mode Wait for Receive (BUFWR) bit in SPIn.CTRLB set to ‘0’.
28.3.2.2.3 SS Pin Functionality in Slave Mode
The Slave Select (SS) pin plays a central role in the operation of the SPI. Depending on the mode the SPI is in, and
the configuration of this pin, it can be used to activate or deactivate devices. The SS pin is used as a Chip Select pin.
In Slave mode, SS, MOSI, and SCK are always inputs. The behavior of the MISO pin depends on the configured data
direction of the pin in the port peripheral and the value of SS. When the SS pin is driven low, the SPI is activated and
will respond to received SCK pulses by clocking data out on MISO, if the user has configured the data direction of the
MISO pin as output. When the SS pin is driven high, the SPI is deactivated, meaning that it will not receive incoming
data. If the MISO pin data direction is configured as output, the MISO pin will be tri-stated. Table 28-3 shows an
overview of the SS pin functionality.
Table 28-3. Overview of the SS Pin Functionality
MISO Pin Mode
SS Configuration
SS Pin-Level
Description
Port Direction =
Output
Port Direction =
Input
High
Slave deactivated
(deselected)
Tri-stated
Input
Low
Slave activated
(selected)
Output
Input
Always Input
Note: In Slave mode, the SPI state machine will be reset when the SS pin is driven high. If the SS pin is driven high
during a transmission, the SPI will stop sending and receiving data immediately and both data received and data sent
must be considered lost. As the SS pin is used to signal the start and end of a transfer, it is useful for achieving
packet/byte synchronization and keeping the Slave bit counter synchronized with the master clock generator.
28.3.2.3 Data Modes
There are four combinations of SCK phase and polarity with respect to serial data. The desired combination is
selected by writing to the MODE bits in the Control B (SPIn.CTRLB) register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 412
AVR128DB28/32/48/64
SPI - Serial Peripheral Interface
The SPI data transfer formats are shown below. Data bits are shifted out and latched in on opposite edges of the
SCK signal, ensuring sufficient time for data signals to stabilize.
The leading edge is the first clock edge of a clock cycle. The trailing edge is the last clock edge of a clock cycle.
Figure 28-5. SPI Data Transfer Modes
1
2
3
4
5
6
7
8
MISO
1
2
3
4
5
6
7
8
MOSI
1
2
3
4
5
6
7
8
SPI Mode 0
Cycle #
SS
SCK
Sampling
SPI Mode 1
Cycle #
1
2
3
4
5
6
7
8
MISO
1
2
3
4
5
6
7
8
MOSI
1
2
3
4
5
6
7
8
SS
SCK
Sampling
SPI Mode 2
Cycle #
1
2
3
4
5
6
7
8
MISO
1
2
3
4
5
6
7
8
MOSI
1
2
3
4
5
6
7
8
SS
SCK
Sampling
SPI Mode 3
Cycle #
1
2
3
4
5
6
7
8
MISO
1
2
3
4
5
6
7
8
MOSI
1
2
3
4
5
6
7
8
SS
SCK
Sampling
28.3.2.4 Events
The SPI can generate the following events:
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 413
AVR128DB28/32/48/64
SPI - Serial Peripheral Interface
Table 28-4. Event Generators in SPI
Generator Name
Module
Event
SPIn
SCK
Description
Event Type
SPI Master clock
Level
Generating
Clock
Domain
CLK_PER
Length of Event
Minimum two CLK_PER periods
The SPI has no event users.
Refer to the Event System section for more details regarding event types and Event System configuration.
28.3.2.5 Interrupts
Table 28-5. Available Interrupt Vectors and Sources
Name
SPIn
Conditions
Vector Description
SPI interrupt
Normal Mode
•
•
IF: Interrupt Flag interrupt
WRCOL: Write Collision interrupt
Buffer Mode
•
•
•
•
SSI: Slave Select Trigger Interrupt
DRE: Data Register Empty interrupt
TXC: Transfer Complete interrupt
RXC: Receive Complete interrupt
When an interrupt condition occurs, the corresponding interrupt flag is set in the peripheral’s Interrupt Flags
(peripheral.INTFLAGS) register.
An interrupt source is enabled or disabled by writing to the corresponding enable bit in the peripheral’s Interrupt
Control (peripheral.INTCTRL) register.
An interrupt request is generated when the corresponding interrupt source is enabled, and the interrupt flag is set.
The interrupt request remains active until the interrupt flag is cleared. See the peripheral’s INTFLAGS register for
details on how to clear interrupt flags.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 414
AVR128DB28/32/48/64
SPI - Serial Peripheral Interface
28.4
Register Summary
Offset
Name
Bit Pos.
7
6
5
4
0x00
0x01
0x02
0x03
0x03
0x04
CTRLA
CTRLB
INTCTRL
INTFLAGS
INTFLAGS
DATA
7:0
7:0
7:0
7:0
7:0
7:0
DORD
BUFWR
TXCIE
WRCOL
TXCIF
MASTER
CLK2X
BUFEN
RXCIE
IF
RXCIF
DREIE
SSIE
DREIF
SSIF
DATA[7:0]
28.5
3
2
1
PRESC[1:0]
SSD
0
ENABLE
MODE[1:0]
IE
BUFOVF
Register Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 415
AVR128DB28/32/48/64
SPI - Serial Peripheral Interface
28.5.1
Control A
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
CTRLA
0x00
0x00
-
6
DORD
R/W
0
5
MASTER
R/W
0
4
CLK2X
R/W
0
3
2
1
PRESC[1:0]
R/W
R/W
0
0
0
ENABLE
R/W
0
Bit 6 – DORD Data Order
Value
Description
0
The MSb of the data word is transmitted first
1
The LSb of the data word is transmitted first
Bit 5 – MASTER Master/Slave Select
This bit selects the desired SPI mode.
If SS is configured as input and driven low while this bit is ‘1’, then this bit is cleared and the IF in SPIn.INTFLAGS is
set. The user has to write MASTER = 1 again to re-enable SPI Master mode.
This behavior is controlled by the Slave Select Disable (SSD) bit in SPIn.CTRLB.
Value
Description
0
SPI Slave mode selected
1
SPI Master mode selected
Bit 4 – CLK2X Clock Double
When this bit is written to ‘1’ the SPI speed (SCK frequency, after internal prescaler) is doubled in Master mode.
Value
Description
0
SPI speed (SCK frequency) is not doubled
1
SPI speed (SCK frequency) is doubled in Master mode
Bits 2:1 – PRESC[1:0] Prescaler
This bit field controls the SPI clock rate configured in Master mode. These bits have no effect in Slave mode. The
relationship between SCK and the peripheral clock frequency (fCLK_PER) is shown below.
The output of the SPI prescaler can be doubled by writing the CLK2X bit to ‘1’.
Value
Name
Description
0x0
DIV4
CLK_PER/4
0x1
DIV16
CLK_PER/16
0x2
DIV64
CLK_PER/64
0x3
DIV128
CLK_PER/128
Bit 0 – ENABLE SPI Enable
Value
Description
0
SPI is disabled
1
SPI is enabled
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 416
AVR128DB28/32/48/64
SPI - Serial Peripheral Interface
28.5.2
Control B
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
BUFEN
R/W
0
CTRLB
0x01
0x00
-
6
BUFWR
R/W
0
5
4
3
2
SSD
R/W
0
1
0
MODE[1:0]
R/W
0
R/W
0
Bit 7 – BUFEN Buffer Mode Enable
Writing this bit to ‘1’ enables Buffer mode. This will enable two receive buffers and one transmit buffer. Both will have
separate interrupt flags, transmit complete and receive complete.
Bit 6 – BUFWR Buffer Mode Wait for Receive
When writing this bit to ‘0’ the first data transferred will be a dummy sample.
Value
Description
0
One SPI transfer must be completed before the data are copied into the shift register.
1
If writing to the Data register when the SPI is enabled and SS is high, the first write will go directly to
the shift register.
Bit 2 – SSD Slave Select Disable
If this bit is set when operating as SPI Master (MASTER = 1 in SPIn.CTRLA), SS does not disable Master mode.
Value
Description
0
Enable the Slave Select line when operating as SPI master
1
Disable the Slave Select line when operating as SPI master
Bits 1:0 – MODE[1:0] Mode
These bits select the Transfer mode. The four combinations of SCK phase and polarity with respect to the serial data
are shown below. These bits decide whether the first edge of a clock cycle (leading edge) is rising or falling and
whether data setup and sample occur on the leading or trailing edge. When the leading edge is rising, the SCK signal
is low when idle, and when the leading edge is falling, the SCK signal is high when idle.
Value
Name
Description
0x0
0
Leading edge: Rising, sample
Trailing edge: Falling, setup
0x1
1
Leading edge: Rising, setup
Trailing edge: Falling, sample
0x2
2
Leading edge: Falling, sample
Trailing edge: Rising, setup
0x3
3
Leading edge: Falling, setup
Trailing edge: Rising, sample
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 417
AVR128DB28/32/48/64
SPI - Serial Peripheral Interface
28.5.3
Interrupt Control
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
RXCIE
R/W
0
INTCTRL
0x02
0x00
-
6
TXCIE
R/W
0
5
DREIE
R/W
0
4
SSIE
R/W
0
3
2
1
0
IE
R/W
0
Bit 7 – RXCIE Receive Complete Interrupt Enable
In Buffer mode, this bit enables the Receive Complete interrupt. The enabled interrupt will be triggered when the
RXCIF in the SPIn.INTFLAGS register is set. In the Non-Buffer mode, this bit is ‘0’.
Bit 6 – TXCIE Transfer Complete Interrupt Enable
In Buffer mode, this bit enables the Transfer Complete interrupt. The enabled interrupt will be triggered when the
TXCIF in the SPIn.INTFLAGS register is set. In the Non-Buffer mode, this bit is ‘0’.
Bit 5 – DREIE Data Register Empty Interrupt Enable
In Buffer mode, this bit enables the Data Register Empty interrupt. The enabled interrupt will be triggered when the
DREIF in the SPIn.INTFLAGS register is set. In the Non-Buffer mode, this bit is ‘0’.
Bit 4 – SSIE Slave Select Trigger Interrupt Enable
In Buffer mode, this bit enables the Slave Select interrupt. The enabled interrupt will be triggered when the SSIF in
the SPIn.INTFLAGS register is set. In the Non-Buffer mode, this bit is ‘0’.
Bit 0 – IE Interrupt Enable
This bit enables the SPI interrupt when the SPI is not in Buffer mode. The enabled interrupt will be triggered when
RXCIF/IF is set in the SPIn.INTFLAGS register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 418
AVR128DB28/32/48/64
SPI - Serial Peripheral Interface
28.5.4
Interrupt Flags - Normal Mode
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
IF
R/W
0
INTFLAGS
0x03
0x00
-
6
WRCOL
R/W
0
5
4
3
2
1
0
Bit 7 – IF Interrupt Flag
This flag is set when a serial transfer is complete, and one byte is completely shifted in/out of the SPIn.DATA register.
If SS is configured as input and is driven low when the SPI is in Master mode, this will also set this flag. The IF is
cleared by hardware when executing the corresponding interrupt vector. Alternatively, the IF can be cleared by first
reading the SPIn.INTFLAGS register when IF is set, and then accessing the SPIn.DATA register.
Bit 6 – WRCOL Write Collision
The WRCOL flag is set if the SPIn.DATA register is written before a complete byte has been shifted out. This flag is
cleared by first reading the SPIn.INTFLAGS register when WRCOL is set, and then accessing the SPIn.DATA
register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 419
AVR128DB28/32/48/64
SPI - Serial Peripheral Interface
28.5.5
Interrupt Flags - Buffer Mode
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
RXCIF
R/W
0
INTFLAGS
0x03
0x00
-
6
TXCIF
R/W
0
5
DREIF
R/W
0
4
SSIF
R/W
0
3
2
1
0
BUFOVF
R/W
0
Bit 7 – RXCIF Receive Complete Interrupt Flag
This flag is set when there are unread data in the Receive Data Buffer register and cleared when the Receive Data
Buffer register is empty (that is, it does not contain any unread data).
When interrupt-driven data reception is used, the Receive Complete Interrupt routine must read the received data
from the DATA register in order to clear RXCIF. If not, a new interrupt will occur directly after the return from the
current interrupt. This flag can also be cleared by writing a ‘1’ to its bit location.
Bit 6 – TXCIF Transfer Complete Interrupt Flag
This flag is set when all the data in the Transmit shift register has been shifted out, and there is no new data in the
transmit buffer (SPIn.DATA). The flag is cleared by writing a ‘1’ to its bit location.
Bit 5 – DREIF Data Register Empty Interrupt Flag
This flag indicates whether the Transmit Data Buffer register is ready to receive new data. The flag is ‘1’ when the
transmit buffer is empty and ‘0’ when the transmit buffer contains data to be transmitted that has not yet been moved
into the shift register. The DREIF is cleared after a Reset to indicate that the transmitter is ready.
The DREIF is cleared by writing to DATA. When interrupt-driven data transmission is used, the Data Register Empty
Interrupt routine must either write new data to DATA in order to clear DREIF or disable the Data Register Empty
interrupt. If not, a new interrupt will occur directly after the return from the current interrupt.
Bit 4 – SSIF Slave Select Trigger Interrupt Flag
This flag indicates that the SPI has been in Master mode and the SS pin has been pulled low externally, so the SPI is
now working in Slave mode. The flag will only be set if the Slave Select Disable (SSD) bit is not ‘1’. The flag is
cleared by writing a ‘1’ to its bit location.
Bit 0 – BUFOVF Buffer Overflow
This flag indicates data loss due to a Receive Data Buffer full condition. This flag is set if a Buffer Overflow condition
is detected. A Buffer Overflow occurs when the receive buffer is full (two bytes), and a third byte has been received in
the shift register. If there is no transmit data, the Buffer Overflow will not be set before the start of a new serial
transfer. This flag is cleared when the DATA register is read, or by writing a ‘1’ to its bit location.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 420
AVR128DB28/32/48/64
SPI - Serial Peripheral Interface
28.5.6
Data
Name:
Offset:
Reset:
Property:
Bit
7
DATA
0x04
0x00
-
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
DATA[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – DATA[7:0] SPI Data
The DATA register is used for sending and receiving data. Writing to the register initiates the data transmission when
in Master mode, while preparing data for sending in Slave mode. The byte written to the register shifts out on the SPI
output line when a transaction is initiated.
The SPIn.DATA register is not a physical register. Depending on what mode is configured, it is mapped to other
registers as described below.
• Normal mode:
– Writing the DATA register will write the shift register
– Reading from DATA will read from the Receive Data register
• Buffer mode:
– Writing the DATA register will write to the Transmit Data Buffer register.
– Reading from DATA will read from the Receive Data Buffer register. The contents of the Receive Data
register will then be moved to the Receive Data Buffer register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 421
AVR128DB28/32/48/64
TWI - Two-Wire Interface
29.
TWI - Two-Wire Interface
29.1
Features
•
•
•
•
•
•
•
29.2
Two-Wire Communication Interface
Philips I2C Compatible
– Standard mode
– Fast mode
– Fast mode Plus
System Management Bus (SMBus) 2.0 Compatible
– Support arbitration between Start/repeated Start and data bit
– Slave arbitration allows support for the Address Resolution Protocol (ARP)
– Configurable SMBus Layer 1 time-outs in hardware
– Independent time-outs for Dual mode
Independent Master and Slave Operation
– Combined (same pins) or Dual mode (separate pins)
– Single or multi-master bus operation with full arbitration support
Hardware Support for Slave Address Match
– Operates in all Sleep modes
– 7-bit address recognition
– General call address recognition
– Support for address range masking or secondary address match
Input Filter for Bus Noise Suppression
Smart Mode Support
Overview
The Two-Wire Interface (TWI) is a bidirectional, two-wire communication interface (bus) with a Serial Data Line (SDA)
and a Serial Clock Line (SCL).
The TWI bus connects one or several slave devices to one or several master devices. Any device connected to the
bus can act as a master, a slave, or both. The master generates the SCL by using a Baud Rate Generator (BRG) and
initiates data transactions by addressing one slave and telling whether it wants to transmit or receive data. The BRG
is capable of generating the Standard mode (Sm) and Fast mode (Fm, Fm+) bus frequencies from 100 kHz up to 1
MHz.
The TWI will detect Start and Stop conditions, bus collisions and bus errors. Arbitration lost, errors, collision, and
clock hold are also detected and indicated in separate status flags available in both Master and Slave modes.
The TWI supports multi-master bus operation and arbitration. An arbitration scheme handles the case where more
than one master tries to transmit data at the same time. The TWI also supports Smart mode, which can auto-trigger
operations and thus reduce software complexity. The TWI supports Dual mode with simultaneous master and slave
operations, which are implemented as independent units with separate enabling and configuration. The TWI supports
Quick Command mode where the master can address a slave without exchanging data.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 422
AVR128DB28/32/48/64
TWI - Two-Wire Interface
29.2.1
Block Diagram
Figure 29-1. TWI Block Diagram
Master
BAUD
Slave
TxDATA
TxDATA
SCL
0
Baud Rate Generator
SCL Hold Low
0
SCL Hold Low
shift register
shift register
SDA
0
0
RxDATA
29.2.2
ADDR/ADDRMASK
RxDATA
==
Signal Description
Signal
Description
Type
SCL
Serial Clock Line
Digital I/O
SDA
Serial Data Line
Digital I/O
29.3
Functional Description
29.3.1
General TWI Bus Concepts
The TWI provides a simple, bidirectional, two-wire communication bus consisting of:
• Serial Data Line (SDA) for packet transfer
• Serial Clock Line (SCL) for the bus clock
The two lines are open-collector lines (wired-AND).
The TWI bus topology is a simple and efficient method of interconnecting multiple devices on a serial bus. A device
connected to the bus can be a master or a slave. Only master devices can control the bus and the bus
communication.
A unique address is assigned to each slave device connected to the bus, and the master will use it to control the
slave and initiate a transaction. Several masters can be connected to the same bus. This is called a multi-master
environment. An arbitration mechanism is provided for resolving bus ownership among masters, since only one
master device may own the bus at any given time.
A master indicates the start of a transaction by issuing a Start condition (S) on the bus. The master provides the clock
signal for the transaction. An address packet with a 7-bit slave address (ADDRESS) and a direction bit, representing
whether the master wishes to read or write data (R/W), are then sent.
The addressed I2C slave will then acknowledge (ACK) the address, and data packet transactions can begin. Every 9bit data packet consists of eight data bits followed by a 1-bit reply indicating whether the data was acknowledged or
not by the receiver.
After all the data packets (DATA) are transferred, the master issues a Stop condition (P) on the bus to end the
transaction.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 423
AVR128DB28/32/48/64
TWI - Two-Wire Interface
Figure 29-2. Basic TWI Transaction Diagram Topology for a 7-bit Address Bus
SDA
SCL
6 ... 0
S
7 ... 0
ADDRESS
S
ADDRESS
R/W
R/W
ACK
7 ... 0
DATA
ACK
DATA
A
A
DATA
P
ACK/NACK
A/A
DATA
P
Direction
Address Packet
Data Packet #0
Data Packet #1
Transaction
Bus Driver
Master driving bus
S
START condition
Slave driving bus
Sr
repeated START condition
Either Master or Slave driving bus
P
STOP condition
Data Package Direction
R
Master Read
W
'0'
Acknowledge
A
Acknowledge (ACK)
'0'
'1'
29.3.2
Special Bus Conditions
Master Write
A
Not Acknowledge (NACK)
'1'
TWI Basic Operation
29.3.2.1 Initialization
If used, the following bits must be configured before enabling the TWI device:
• The SDA Setup Time (SDASETUP) bit from the Control A (TWIn.CTRLA) register
• The SDA Hold Time (SDAHOLD) bit field from the Control A (TWIn.CTRLA) register
• The FM Plus Enable (FMPEN) bit from the Control A (TWIn.CTRLA) register
29.3.2.1.1 Master Initialization
The Master Baud Rate (TWIn.MBAUD) register must be written to a value that will result in a valid TWI bus clock
frequency. Writing a ‘1’ to the Enable TWI Master (ENABLE) bit in the Master Control A (TWIn.MCTRLA) register will
enable the TWI master. The Bus State (BUSSTATE) bit field from the Master Status (TWIn.MSTATUS) register must
be set to 0x1, to force the bus state to Idle.
29.3.2.1.2 Slave Initialization
The address of the slave must be written in the Slave Address (TWIn.SADDR) register. Writing a ‘1’ to the Enable
TWI Slave (ENABLE) bit in the Slave Control A (TWIn.SCTRLA) register will enable the TWI slave. The slave device
will wait for a master device to issue a Start condition and the matching slave address.
29.3.2.2 TWI Master Operation
The TWI master is byte-oriented, with an optional interrupt after each byte. There are separate interrupt flags for the
master write and read operation. Interrupt flags can also be used for polled operation. There are dedicated status
flags for indicating ACK/NACK received, bus error, arbitration lost, clock hold, and bus state.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 424
AVR128DB28/32/48/64
TWI - Two-Wire Interface
When an interrupt flag is set to ‘1’, the SCL is forced low. This will give the master time to respond or handle any
data, and will, in most cases, require software interaction. Clearing the interrupt flags releases the SCL. The number
of interrupts generated is kept to a minimum by an automatic handling of most conditions.
29.3.2.2.1 Clock Generation
The TWI supports several transmission modes with different frequency limitations:
• Standard mode (Sm) up to 100 kHz
• Fast mode (Fm) up to 400 kHz
• Fast mode Plus (Fm+) up to 1 MHz
The Master Baud Rate (TWIn.MBAUD) register must be written to a value that will result in a TWI bus clock
frequency equal to or less than those frequency limits, depending on the transmission mode.
The low (TLOW) and high (THIGH) times are determined by the Master Baud Rate (TWIn.MBAUD) register, while the
rise (TR) and fall (TOF) times are determined by the bus topology.
Figure 29-3. SCL Timing
SCL
THD;STA
TSU;STA
TLOW
THIGH
TOF
TSP
THD;DAT
TSU;DAT
TR
TBUF
TSU;STO
SDA
S
•
•
•
•
P
S
TLOW is the low period of SCL clock
THIGH is the high period of SCL clock
TR is determined by the bus impedance; for internal pull-ups. Refer to Electrical Characteristics for details.
TOF is the output fall time and is determined by the open-drain current limit and bus impedance. Refer to
Electrical Characteristics for details.
Properties of the SCL Clock
The SCL frequency is given by:
�SCL =
1
[Hz]
�LOW + �HIGH + �OF + �R
���� =
����_���
10 + 2 × ���� + ����_��� × ��
The SCL clock is designed to have a 50/50 duty cycle, where the low period of the duty cycle comprises of TOF and
TLOW. THIGH will not start until a high state of SCL has been detected. The BAUD bit field in the TWIn.MBAUD
register and the SCL frequency are related by the following formula:
(1)
Equation 1 can be transformed to express BAUD:
���� =
����_���
����_��� × ��
− 5+
2 × ����
2
(2)
Calculation of the BAUD Value
To ensure operation within the specifications of the desired speed mode (Sm, Fm, Fm+), follow these steps:
1. Calculate a value for the BAUD bit field using equation 2
2. Calculate TLOW using the BAUD value from step 1:
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 425
AVR128DB28/32/48/64
TWI - Two-Wire Interface
3.
���� =
���� + 5
− ���
����_���
(3)
Check if your TLOW from equation 3 is above the specified minimum of the desired mode (TLOW_Sm= 4700 ns,
TLOW_Fm= 1300 ns, TLOW_Fm+= 500 ns)
– If the calculated TLOW is above the limit, use the BAUD value from equation 2
– If the limit is not met, calculate a new BAUD value using equation 4 below, where TLOW_mode is either
TLOW_Sm, TLOW_Fm, or TLOW_Fm+ from the mode specifications:
���� = ����_��� × (����_���� + ���) − 5
(4)
29.3.2.2.2 TWI Bus State Logic
The bus state logic continuously monitors the activity on the TWI bus when the master is enabled. It continues to
operate in all Sleep modes, including Power-Down.
The bus state logic includes Start and Stop condition detectors, collision detection, inactive bus time-out detection,
and a bit counter. These are used to determine the bus state. The software can get the current bus state by reading
the Bus State (BUSSTATE) bit field in the Master Status (TWIn.MSTATUS) register.
The bus state can be Unknown, Idle, Busy or Owner, and it is determined according to the state diagram shown
below.
Figure 29-4. Bus State Diagram
RESET
UNKNOWN
(0b00)
Time-out or Stop Condition
External Start Condition
IDLE
(0b01)
Time-out or Stop Condition
BUSY
(0b11)
Write ADDR to generate
Start Condition
OWNER
(0b10)
Lost Arbitration
External Repeated
Start Condition
Stop Condition
Write ADDR to generate
Repeated Start Condition
1.
2.
3.
4.
Unknown: The bus state machine is active when the TWI master is enabled. After the TWI master has been
enabled, the bus state is Unknown. The bus state will also be set to Unknown after a System Reset is
performed or after the TWI master is disabled.
Idle: The bus state machine can be forced to enter the Idle state by writing 0x1 to the Bus State (BUSSTATE)
bit field. The bus state logic cannot be forced into any other state. If no state is set by the application software,
the bus state will become Idle when the first Stop condition is detected. If the Inactive Bus Time-Out
(TIMEOUT) bit field from the Master Control A (TWIn.MCTRLA) register is configured to a nonzero value, the
bus state will change to Idle on the occurrence of a time-out. When the bus is Idle, it is ready for a new
transaction.
Busy: If a Start condition, generated externally, is detected when the bus is Idle, the bus state becomes Busy.
The bus state changes back to Idle when a Stop condition is detected or when a time-out, if configured, is set.
Owner: If a Start condition is generated internally when the bus is Idle, the bus state becomes Owner. If the
complete transaction is performed without interference, the master issues a Stop condition and the bus state
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 426
AVR128DB28/32/48/64
TWI - Two-Wire Interface
changes back to Idle. If a collision is detected, the arbitration is lost and the bus state becomes Busy until a
Stop condition is detected.
29.3.2.2.3 Transmitting Address Packets
The master starts performing a bus transaction when the Master Address (TWIn.MADDR) register is written with the
slave address and the R/W direction bit. The value of the MADDR register is then copied in the Master Data
(TWIn.MDATA) register. If the bus state is Busy, the TWI master will wait until the bus state becomes Idle before
issuing the Start condition. The TWI will issue a Start condition, and the shift register performs a byte transmit
operation on the bus.
Depending on the arbitration and the R/W direction bit, one of four cases (M1 to M4) arises after the transmission of
the address packet.
Figure 29-5. TWI Master Operation
M4
MASTER DATA INTERRUPT
P
P
A
Sr
BUSY
P
IDLE
S
ADDRESS
W
OWNER
M3
A
Sr
A
IF
M1
DATA
Sr
A
M4
M4
IF
A
The master provides data
on the bus
Mn
P
P
R
BUSY
BUSY
Interrupt flag raised
Addressed slave provides
data on the bus
BUSY
OWNER
A
M3
A
Sr
DATA
IF
M2
Sr
A
Diagram cases
Case M1: Address Packet Transmit Complete - Direction Bit Set to ‘0’
If a slave device responds to the address packet with an ACK, the Write Interrupt Flag (WIF) is set to ‘1’, the
Received Acknowledge (RXACK) flag is set to ‘0’, and the Clock Hold (CLKHOLD) flag is set to ‘1’. The WIF, RXACK
and CLKHOLD flags are located in the Master Status (TWIn.MSTATUS) register.
The clock hold is active at this point, forcing the SCL low. This will stretch the low period of the clock to slow down the
overall clock frequency, forcing delays required to process the data and preventing further activity on the bus.
The software can prepare to:
• Transmit data packets to the slave
Case M2: Address Packet Transmit Complete - Direction Bit Set to ‘1’
If a slave device responds to the address packet with an ACK, the RXACK flag is set to ‘0’, and the slave can start
sending data to the master without any delays because the slave owns the bus at this moment. The clock hold is
active at this point, forcing the SCL low.
The software can prepare to:
• Read the received data packet from the slave
Case M3: Address Packet Transmit Complete - Address not Acknowledged by Slave
If no slave device responds to the address packet, the WIF and the RXACK flags will be set to ‘1’. The clock hold is
active at this point, forcing the SCL low.
The missing ACK response can indicate that the I2C slave is busy with other tasks, or it is in a Sleep mode, and it is
not able to respond.
The software can prepare to take one of the following actions:
• Retransmit the address packet
• Complete the transaction by issuing a Stop condition in the Command (MCMD) bit field from the Master Control
B (TWIn.MCTRLB) register, which is the recommended action
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 427
AVR128DB28/32/48/64
TWI - Two-Wire Interface
Case M4: Arbitration Lost or Bus Error
If arbitration is lost, both the WIF and the Arbitration Lost (ARBLOST) flags in the Master Status (TWIn.MSTATUS)
register are set to ‘1’. The SDA is disabled and the SCL is released. The bus state changes to Busy, and the master
is no longer allowed to perform any operation on the bus until the bus state is changed back to Idle.
A bus error will behave similarly to the arbitration lost condition. In this case, the Bus Error (BUSERR) flag in the
Master Status (TWIn.MSTATUS) register is set to ‘1’, in addition to the WIF and ARBLOST flags.
The software can prepare to:
• Abort the operation and wait until the bus state changes to Idle by reading the Bus State (BUSSTATE) bit field in
the Master Status (TWIn.MSTATUS) register
29.3.2.2.4 Transmitting Data Packets
Assuming the above M1 case, the TWI master can start transmitting data by writing to the Master Data
(TWIn.MDATA) register, which will also clear the Write Interrupt Flag (WIF). During the data transfer, the master is
continuously monitoring the bus for collisions and errors. The WIF flag will be set to ‘1’ after the data packet transfer
has been completed.
If the transmission is successful and the master receives an ACK bit from the slave, the Received Acknowledge
(RXACK) flag will be set to ‘0’, meaning that the slave is ready to receive new data packets.
The software can prepare to take one of the following actions:
• Transmit a new data packet
• Transmit a new address packet
• Complete the transaction by issuing a Stop condition in the Command (MCMD) bit field from the Master Control
B (TWIn.MCTRLB) register
If the transmission is successful and the master receives a NACK bit from the slave, the RXACK flag will be set to ‘1’,
meaning that the slave is not able to or does not need to receive more data.
The software can prepare to take one of the following actions:
• Transmit a new address packet
• Complete the transaction by issuing a Stop condition in the Command (MCMD) bit field from the Master Control
B (TWIn.MCTRLB) register
The RXACK status is valid only if the WIF flag is set to ‘1’ and the Arbitration Lost (ARBLOST) and Bus Error
(BUSERR) flags are set to ‘0’.
The transmission can be unsuccessful if a collision is detected. Then, the master will lose arbitration, the Arbitration
Lost (ARBLOST) flag will be set to ‘1’, and the bus state changes to Busy. An arbitration lost during the sending of
the data packet is treated the same way as the above M4 case.
The WIF, ARBLOST, BUSERR and RXACK flags are all located in the Master Status (TWIn.MSTATUS) register.
29.3.2.2.5 Receiving Data Packets
Assuming the M2 case above, the clock is released for one byte, allowing the slave to put one byte of data on the
bus. The master will receive one byte of data from the slave, and the Read Interrupt Flag (RIF) will be set to ‘1’
together with the Clock Hold (CLKHOLD) flag. The action selected by the Acknowledge Action (ACKACT) bit in the
Master Control B (TWIn.MCTRLB) register is automatically sent on the bus when a command is written to the
Command (MCMD) bit field in the TWIn.MCTRLB register.
The software can prepare to take one of the following actions:
• Respond with an ACK by writing ‘0’ to the ACKACT bit in the TWIn.MCTRLB register and prepare to receive a
new data packet
• Respond with a NACK by writing ‘1’ to the ACKACT bit and then transmit a new address packet
•
Respond with a NACK by writing ‘1’ to the ACKACT bit and then complete the transaction by issuing a Stop
condition in the MCMD bit field from the TWIn.MCTRLB register
A NACK response might not be successfully executed, as arbitration can be lost during the transmission. If a collision
is detected, the master loses arbitration, and the Arbitration Lost (ARBLOST) flag is set to ‘1’ and the bus state
changes to Busy. The Master Write Interrupt Flag (WIF) is set if the arbitration was lost when sending a NACK or a
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 428
AVR128DB28/32/48/64
TWI - Two-Wire Interface
bus error occurred during the procedure. An arbitration lost during the sending of the data packet is treated in the
same way as the above M4 case.
The RIF, CLKHOLD, ARBLOST and WIF flags are all located in the Master Status (TWIn.MSTATUS) register.
Note: The RIF and WIF flags are mutually exclusive and cannot be set simultaneously.
29.3.2.3 TWI Slave Operation
The TWI slave is byte-oriented with optional interrupts after each byte. There are separate interrupt flags for the slave
data and for address/Stop recognition. Interrupt flags can also be used for polled operation. There are dedicated
status flags for indicating ACK/NACK received, clock hold, collision, bus error, and R/W direction bit.
When an interrupt flag is set to ‘1’, the SCL is forced low. This will give the slave time to respond or handle any data,
and will, in most cases, require software interaction. The number of interrupts generated is kept to a minimum by
automatic handling of most conditions.
The Address Recognition Mode (PMEN) bit in the Slave Control A (TWIn.SCTRLA) register can be configured to
allow the slave to respond to all received addresses.
29.3.2.3.1 Receiving Address Packets
When the TWI is configured as a slave, it will wait for a Start condition to be detected. When this happens, the
successive address packet will be received and checked by the address match logic. The slave will ACK a correct
address and store the address in the Slave Data (TWIn.SDATA) register. If the received address is not a match, the
slave will not acknowledge or store the address, but wait for a new Start condition.
The Address or Stop Interrupt Flag (APIF) in the Slave Status (TWIn.SSTATUS) register is set to ‘1’ when a Start
condition is succeeded by one of the following:
•
•
A valid address match with the address stored in the Address (ADDR[7:1]) bit field in the Slave Address
(TWIn.SADDR) register
The General Call Address 0x00, and the Address (ADDR[0]) bit in the Slave Address (TWIn.SADDR) register is
set to ‘1’
•
A valid address match with the secondary address stored in the Address Mask (ADDRMASK) bit field and the
Address Mask Enable (ADDREN) bit is set to ‘1’ in the Slave Address Mask (TWIn.SADDRMASK) register
•
Any address if the Address Recognition Mode (PMEN) bit in the Slave Control A (TWIn.SCTRLA) register is set
to ‘1’
A Start condition immediately followed by a Stop condition is an illegal operation, and the Bus Error (BUSERR) flag in
the Slave Status (TWIn.SSTATUS) register is set.
Depending on the Read/Write Direction (DIR) bit in the Slave Status (TWIn.SSTATUS) register and the bus condition,
one of four distinct cases (S1 to S4) arises after the reception of the address packet.
Figure 29-6. TWI Slave Operation
SLAVE ADDRESS INTERRUPT
S4
S3
IF
Sn
S
SLAVE DATA INTERRUPT
A
ADDRESS
R
IF
Interrupt flag raised
W
IF
Addressed slave provides
data on the bus
Interrupton STOP
Condition Enabled
IF
S3
Sr
S4
A
S3
Sr
S4
S3
Sr
S4
P
S3
A
Sr
S4
A
Sr
S4
A
IF
P
P
S3
P
A
A
The master provides data
on the bus
P
S2
DATA
A
S3
P
DATA
IF
S1
Diagram cases
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 429
AVR128DB28/32/48/64
TWI - Two-Wire Interface
Case S1: Address Packet Accepted - Direction Bit Set to ‘0’
If an ACK is sent by the slave after the address packet is received and the Read/Write Direction (DIR) bit in the Slave
Status (TWIn.SSTATUS) register is set to ‘0’, the master indicates a write operation.
The clock hold is active at this point, forcing the SCL low. This will stretch the low period of the clock to slow down the
overall clock frequency, forcing delays required to process the data and preventing further activity on the bus.
The software can prepare to:
• Read the received data packet from the master
Case S2: Address Packet Accepted - Direction Bit Set to ‘1’
If an ACK is sent by the slave after the address packet is received and the DIR bit is set to ‘1’, the master indicates a
read operation, and the Data Interrupt Flag (DIF) in the Slave Status (TWIn.SSTATUS) register will be set to ‘1’.
The clock hold is active at this point, forcing the SCL low.
The software can prepare to:
• Transmit data packets to the master
Case S3: Stop Condition Received
When the Stop condition is received, the Address or Stop (AP) flag will be set to ‘0’, indicating that a Stop condition,
and not an address match, activated the Address or Stop Interrupt Flag (APIF).
The AP and APIF flags are located in the Slave Status (TWIn.SSTATUS) register.
The software can prepare to:
• Wait until a new address packet will be addressed to it
Case S4: Collision
If the slave is not able to send a high-level data bit or a NACK, the Collision (COLL) bit in the Slave Status
(TWIn.SSTATUS) register is set to ‘1’. The slave will commence its operation as normal, except no low values will be
shifted out on the SDA. The data and acknowledge output from the slave logic will be disabled. The clock hold is
released. A Start or repeated Start condition will be accepted.
The COLL bit is intended for systems where the Address Resolution Protocol (ARP) is employed. A detected collision
in non-ARP situations indicates that there has been a protocol violation and must be treated as a bus error.
29.3.2.3.2 Receiving Data Packets
Assuming the above S1 case, the slave must be ready to receive data. When a data packet is received, the Data
Interrupt Flag (DIF) in the Slave Status (TWIn.SSTATUS) register is set to ‘1’. The action selected by the
Acknowledge Action (ACKACT) bit in the Slave Control B (TWIn.SCTRLB) register is automatically sent on the bus
when a command is written to the Command (SCMD) bit field in the TWIn.SCTRLB register.
The software can prepare to take one of the following actions:
• Respond with an ACK by writing ‘0’ to the ACKACT bit in the TWIn.SCTRLB register, indicating that the slave is
ready to receive more data
• Respond with a NACK by writing ‘1’ to the ACKACT bit, indicating that the slave cannot receive any more data
and the master must issue a Stop or repeated Start condition
29.3.2.3.3 Transmitting Data Packets
Assuming the above S2 case, the slave can start transmitting data by writing to the Slave Data (TWIn.SDATA)
register. When a data packet transmission is completed, the Data Interrupt Flag (DIF) in the Slave Status
(TWIn.SSTATUS) register is set to ‘1’.
The software can prepare to take one of the following actions:
• Check if the master responded with an ACK by reading the Received Acknowledge (RXACK) bit from the Slave
Status (TWIn.SSTATUS) register and start transmitting new data packets
• Check if the master responded with a NACK by reading the RXACK and stop transmitting data packets. The
master must send a Stop or repeated Start condition after the NACK.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 430
AVR128DB28/32/48/64
TWI - Two-Wire Interface
29.3.3
Additional Features
29.3.3.1 SMBus
If the TWI is used in an SMBus environment, the Inactive Bus Time-Out (TIMEOUT) bit field from the Master Control
A (TWIn.MCTRLA) register must be configured. It is recommended to write to the Master Baud Rate (TWIn.MBAUD)
register before setting the time-out because it is dependent on the baud rate setting.
A frequency of 100 kHz can be used for the SMBus environment. For the Standard mode (Sm) and Fast mode (Fm),
the operating frequency has slew rate limited output, while for the Fast mode Plus (Fm+), it has x10 output drive
strength.
The TWI also allows for an SMBus compatible SDA hold time configured in the SDA Hold Time (SDAHOLD) bit field
from the Control A (TWIn.CTRLA) register.
29.3.3.2 Multi Master
A master can start a bus transaction only if it has detected that the bus is in the Idle state. As the TWI bus is a multimaster bus, more devices may try to initiate a transaction at the same time. This results in multiple masters owning
the bus simultaneously. The TWI solves this problem by using an arbitration scheme where the master loses control
of the bus if it is not able to transmit a high-level data bit on the SDA and the Bus State (BUSSTATE) bit field from the
Master Status (TWIn.MSTATUS) register will be changed to Busy. The masters that lose the arbitration must wait until
the bus becomes Idle before attempting to reacquire the bus ownership.
Both devices can issue a Start condition, but DEVICE1 loses arbitration when attempting to transmit a high level (bit
5) while DEVICE2 is transmitting a low level.
Figure 29-7. TWI Arbitration
DEVICE1 Loses arbitration
DEVICE1_SDA
DEVICE2_SDA
SDA
(wired-AND)
bit 7
bit 6
bit 5
bit 4
SCL
S
29.3.3.3 Smart Mode
The TWI interface has a Smart mode that simplifies the application code and minimizes the user interaction needed
to adhere to the I2C protocol.
For the TWI Master, the Smart mode will automatically send the ACK action as soon as the Master Data
(TWIn.MDATA) register is read. This feature is only active when the Acknowledge Action (ACKACT) bit in the Master
Control B (TWIn.MCTRLB) register is set to ACK. If the ACKACT bit is set to NACK, the TWI Master will not generate
a NACK after the MDATA register is read. This feature is enabled when the Smart Mode Enable (SMEN) bit in the
Master Control A (TWIn.MCTRLA) register is set to ‘1’.
For the TWI Slave, the Smart mode will automatically send the ACK action as soon as the Slave Data (TWIn.SDATA)
register is read. The Smart mode will automatically set the Data Interrupt Flag (DIF) to ‘0’ in the Slave Status
(TWIn.SSTATUS) register if the TWIn.SDATA register is read or written. This feature is enabled when the Smart
Mode Enable (SMEN) bit in the Slave Control A (TWIn.SCTRLA) register is set to ‘1’.
29.3.3.4 Dual Mode
The TWI supports Dual mode operation where the master and the slave will operate simultaneously and
independently. In this case, the Control A (TWIn.CTRLA) register will configure the master device, and the Dual Mode
Control (TWIn.DUALCTRL) register will configure the slave device. See the 29.3.2.1 Initialization section for more
details about the master configuration.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 431
AVR128DB28/32/48/64
TWI - Two-Wire Interface
If used, the following bits must be configured before enabling the TWI Dual mode:
• The SDA Hold Time (SDAHOLD) bit field
• The FM Plus Enable (FMPEN) bit from the DUALCTRL register
The Dual mode can be enabled by writing a ‘1’ to the Dual Control Enable (ENABLE) bit in the DUALCTRL register.
29.3.3.5 Quick Command Mode
With Quick Command mode, the R/W bit from the address packet denotes the command. This mode is enabled by
writing ‘1’ to the Quick Command Enable (QCEN) bit in the Master Control A (TWIn.MCTRLA) register. There are no
data sent or received.
The Quick Command mode is SMBus specific, where the R/W bit can be used to turn a device function on/off or to
enable/disable a low-power Standby mode. This mode can be enabled to auto-trigger operations and reduce the
software complexity.
After the master receives an ACK from the slave, either the Read Interrupt Flag (RIF) or Write Interrupt Flag (WIF)
will be set, depending on the value of the R/W bit. When either the RIF or WIF flag is set after issuing a Quick
Command, the TWI will accept a Stop command by writing the Command (MCMD) bit field in the Master Control B
(TWIn.MCTRLB) register.
The RIF and WIF flags, together with the value of the last Received Acknowledge (RXACK) flag are all located in the
Master Status (TWIn.MSTATUS) register.
Figure 29-8. Quick Command Frame Format
BUSY
P
BUSY
IDLE
S
ADDRESS
R/W
OWNER
A
P
The master provides data
on the bus
Addressed slave provides
data on the bus
29.3.3.6 10-bit Address
Regardless of whether the transaction is a read or write, the master must start by sending the 10-bit address with the
R/W direction bit set to ‘0’.
The slave address match logic supports recognition of 7-bit addresses and general call address. The Slave Address
(TWIn.SADDR) register is used by the slave address match logic to determine if a master device has addressed the
TWI slave.
The TWI slave address match logic only supports recognition of the first byte of a 10-bit address and the second byte
must be handled in software. The first byte of the 10-bit address will be recognized if the upper five bits of the Slave
Address (TWIn.SADDR) register are 0b11110. Thus, the first byte will consist of five indication bits, the two Most
Significant bits (MSb) of the 10-bits address, and the R/W direction bit. The Least Significant Byte (LSB) of the
address that follows from the master will come in the form of a data packet.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 432
AVR128DB28/32/48/64
TWI - Two-Wire Interface
Figure 29-9. 10-bit Address Transmission
S
SW
1
1
1
1
0
A9
A8
W
A
A7 A6
A5
A4
A3
A2
A1
A0
S
W
A
Software interaction
The master provides data
on the bus
Addressed slave provides
data on the bus
29.3.4
Interrupts
Table 29-1. Available Interrupt Vectors and Sources
Name
Vector Description
Slave
TWI Slave interrupt
Master
TWI Master interrupt
Conditions
•
DIF: Data Interrupt Flag in TWIn.SSTATUS is set to ‘1’
•
APIF: Address or Stop Interrupt Flag in TWIn.SSTATUS is set to ‘1’
•
RIF: Read Interrupt Flag in TWIn.MSTATUS is set to ‘1’
•
WIF: Write Interrupt Flag in TWIn.MSTATUS is set to ‘1’
When an interrupt condition occurs, the corresponding interrupt flag is set in the Master Status (TWIn.MSTATUS)
register or the Slave Status (TWIn.SSTATUS) register.
When several interrupt request conditions are supported by an interrupt vector, the interrupt requests are ORed
together into one combined interrupt request to the interrupt controller. The user must read the Interrupt flags from
the TWIn.MSTATUS register or the TWIn.SSTATUS register, to determine which of the interrupt conditions are
present.
29.3.5
Sleep Mode Operation
The bus state logic and the address recognition hardware continue to operate in all sleep modes. If a slave device is
in Sleep mode and a Start condition followed by the address of the slave is detected, clock stretching is active during
the wake-up period until the main clock is available. The master will stop operation in all sleep modes. When the Dual
mode is active, the device will wake up only when the Start condition is sent on the bus of the slave.
29.3.6
Debug Operation
During run-time debugging, the TWI will continue normal operation. Halting the CPU in Debugging mode will halt the
normal operation of the TWI. The TWI can be forced to operate with halted CPU by writing a ‘1’ to the Debug Run
(DBGRUN) bit in the Debug Control (TWIn.DBGCTRL) register. When the CPU is halted in Debug mode and the
DBGRUN bit is ‘1’, reading or writing the Master Data (TWIn.MDATA) register or the Slave Data (TWIn.SDATA)
register will neither trigger a bus operation, nor cause transmit and clear flags. If the TWI is configured to require
periodical service by the CPU through interrupts or similar, improper operation or data loss may result during halted
debugging.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 433
AVR128DB28/32/48/64
TWI - Two-Wire Interface
29.4
Register Summary
Offset
Name
Bit Pos.
0x00
0x01
0x02
0x03
0x04
0x05
0x06
0x07
0x08
0x09
0x0A
0x0B
0x0C
0x0D
0x0E
CTRLA
DUALCTRL
DBGCTRL
MCTRLA
MCTRLB
MSTATUS
MBAUD
MADDR
MDATA
SCTRLA
SCTRLB
SSTATUS
SADDR
SDATA
SADDRMASK
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
29.5
7
6
5
4
INPUTLVL
INPUTLVL
SDASETUP
RIEN
WIEN
QCEN
RIF
WIF
DIEN
APIEN
DIF
APIF
3
2
SDAHOLD[1:0]
SDAHOLD[1:0]
TIMEOUT[1:0]
FLUSH
ACKACT
CLKHOLD
RXACK
ARBLOST
BUSERR
BAUD[7:0]
ADDR[7:0]
DATA[7:0]
PIEN
PMEN
ACKACT
CLKHOLD
RXACK
COLL
BUSERR
ADDR[7:0]
DATA[7:0]
ADDRMASK[6:0]
1
0
FMPEN
FMPEN
ENABLE
DBGRUN
SMEN
ENABLE
MCMD[1:0]
BUSSTATE[1:0]
SMEN
ENABLE
SCMD[1:0]
DIR
AP
ADDREN
Register Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 434
AVR128DB28/32/48/64
TWI - Two-Wire Interface
29.5.1
Control A
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
CTRLA
0x00
0x00
-
6
INPUTLVL
R/W
0
5
4
SDASETUP
R/W
0
3
2
SDAHOLD[1:0]
R/W
R/W
0
0
1
FMPEN
R/W
0
0
Bit 6 – INPUTLVL Input Voltage Transition Level
This bit is used to select between I2C and SMBUS.
Value
Name
Description
0
I2C
I2C input voltage transition level
1
SMBUS
SMBus 3.0 input voltage transition level
Bit 4 – SDASETUP SDA Setup Time
By default, there are four clock cycles of setup time on the SDA out signal while reading from the slave part of the
TWI module.
Value
Name
Description
0
4CYC
SDA setup time is four clock cycles
1
8CYC
SDA setup time is eight clock cycles
Bits 3:2 – SDAHOLD[1:0] SDA Hold Time
This bit field selects the SDA hold time for the TWI. See the Electrical Characteristics section for details.
Value
Name
Description
0x0
OFF
Hold time OFF
0x1
50NS
Short hold time
0x2
300NS
Meets the SMBus 2.0 specifications under typical conditions
0x3
500NS
Meets the SMBus 2.0 across all corners
Bit 1 – FMPEN FM Plus Enable
Writing a ‘1’ to this bit selects the 1 MHz bus speed for the TWI in default configuration or for the TWI Master in Dual
mode configuration.
Value
Name
Description
0
OFF
Operating in Standard mode or Fast mode
1
ON
Operating in Fast mode Plus
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 435
AVR128DB28/32/48/64
TWI - Two-Wire Interface
29.5.2
Dual Mode Control Configuration
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
DUALCTRL
0x01
0x00
-
6
INPUTLVL
R/W
0
5
4
3
2
SDAHOLD[1:0]
R/W
R/W
0
0
1
FMPEN
R/W
0
0
ENABLE
R/W
0
Bit 6 – INPUTLVL Input Voltage Transition Level
This bit is used to select between I2C and SMBUS.
Value
Name
Description
0
I2C
I2C input voltage transition level
1
SMBUS
SMBus 3.0 input voltage transition level
Bits 3:2 – SDAHOLD[1:0] SDA Hold Time
This bit field selects the SDA hold time for the TWI Slave. This bit field is ignored if the Dual mode is not enabled.
Value
Name
Description
0x0
OFF
Hold time OFF
0x1
50NS
Short hold time
0x2
300NS
Meets the SMBus 2.0 specifications under typical conditions
0x3
500NS
Meets the SMBus 2.0 across all corners
Bit 1 – FMPEN FM Plus Enable
Writing a ‘1’ to this bit selects the 1 MHz bus speed for the TWI Slave.
Value
Name
Description
0
OFF
Operating in Standard mode or Fast mode
1
ON
Operating in Fast mode Plus
Bit 0 – ENABLE Dual Control Enable
Writing a ‘1’ to this bit will enable the Dual mode configuration.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 436
AVR128DB28/32/48/64
TWI - Two-Wire Interface
29.5.3
Debug Control
Name:
Offset:
Reset:
Property:
Bit
7
DBGCTRL
0x02
0x00
-
6
5
4
3
2
Access
Reset
1
0
DBGRUN
R/W
0
Bit 0 – DBGRUN Debug Run
See the 29.3.6 Debug Operation section for more details.
Value
Description
0
The TWI is halted in Break Debug mode and ignores events
1
The TWI will continue to run in Break Debug mode when the CPU is halted
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 437
AVR128DB28/32/48/64
TWI - Two-Wire Interface
29.5.4
Master Control A
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
RIEN
R/W
0
MCTRLA
0x03
0x00
-
6
WIEN
R/W
0
5
4
QCEN
R/W
0
3
2
TIMEOUT[1:0]
R/W
R/W
0
0
1
SMEN
R/W
0
0
ENABLE
R/W
0
Bit 7 – RIEN Read Interrupt Enable
A TWI master read interrupt will be generated only if this bit and the Global Interrupt Enable (I) bit in the Status
(CPU.SREG) register are set to ‘1’.
Writing a ‘1’ to this bit enables the interrupt on the Read Interrupt Flag (RIF) in the Master Status (TWIn.MSTATUS)
register. When the master read interrupt occurs, the RIF flag is set to ‘1’.
Bit 6 – WIEN Write Interrupt Enable
A TWI master write interrupt will be generated only if this bit and the Global Interrupt Enable (I) bit in the Status
(CPU.SREG) register are set to ‘1’.
Writing a ‘1’ to this bit enables the interrupt on the Write Interrupt Flag (WIF) in the Master Status (TWIn.MSTATUS)
register. When the master write interrupt occurs, the WIF flag is set to ‘1’.
Bit 4 – QCEN Quick Command Enable
Writing a ‘1’ to this bit enables the Quick Command mode. If the Quick Command mode is enabled and a slave
acknowledges the address, the corresponding Read Interrupt Flag (RIF) or Write Interrupt Flag (WIF) will be set
depending on the value of R/W bit.
The software must issue a Stop command by writing to the Command (MCMD) bit field in the Master Control B
(TWIn.MCTRLB) register.
Bits 3:2 – TIMEOUT[1:0] Inactive Bus Time-Out
Setting this bit field to a nonzero value will enable the inactive bus time-out supervisor. If the bus is inactive for longer
than the TIMEOUT setting, the bus state logic will enter the Idle state.
Value
Name
Description
0x0
DISABLED
Bus time-out disabled - I2C
0x1
50US
50 µs - SMBus (assume the baud rate is set to 100 kHz)
0x2
100US
100 µs (assume the baud rate is set to 100 kHz)
0x3
200US
200 µs (assume the baud rate is set to 100 kHz)
Bit 1 – SMEN Smart Mode Enable
Writing a ‘1’ to this bit enables the Master Smart mode. When the Smart mode is enabled, the existing value in the
Acknowledge Action (ACKACT) bit from the Master Control B (TWIn.MCTRLB) register is sent immediately after
reading the Master Data (TWIn.MDATA) register.
Bit 0 – ENABLE Enable TWI Master
Writing a ‘1’ to this bit enables the TWI as master.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 438
AVR128DB28/32/48/64
TWI - Two-Wire Interface
29.5.5
Master Control B
Name:
Offset:
Reset:
Property:
Bit
MCTRLB
0x04
0x00
-
7
6
5
4
Access
Reset
3
FLUSH
R/W
0
2
ACKACT
R/W
0
1
0
MCMD[1:0]
R/W
0
R/W
0
Bit 3 – FLUSH Flush
This bit clears the internal state of the master and the bus states changes to Idle. The TWI will transmit invalid data if
the Master Data (TWIn.MDATA) register is written before the Master Address (TWIn.MADDR) register.
Writing a ‘1’ to this bit generates a strobe for one clock cycle, disabling the master, and then re-enabling the master.
Writing a ‘0’ to this bit has no effect.
Bit 2 – ACKACT Acknowledge Action
The ACKACT(1) bit represents the behavior in the Master mode under certain conditions defined by the bus state and
the software interaction. If the Smart Mode Enable (SMEN) bit in the Master Control A (TWIn.MCTRLA) register is set
to ‘1’, the acknowledge action is performed when the Master Data (TWIn.MDATA) register is read, else a command
must be written to the Command (MCDM) bit field in the Master Control B (TWIn.MCTRLB) register.
The acknowledge action is not performed when the Master Data (TWIn.MDATA) register is written, since the master
is sending data.
Value
Name
Description
0
ACK
Send ACK
1
NACK
Send NACK
Bits 1:0 – MCMD[1:0] Command
The MCMD(1) bit field is a strobe. This bit field is always read as ‘0’.
Writing to this bit field triggers a master operation as defined by the table below.
Table 29-2. Command Settings
MCMD[1:0]
Group
Configuration
DIR
Description
0x0
NOACT
X
Reserved
0x1
REPSTART
0x2
RECVTRANS
0x3
STOP
X
Execute Acknowledge Action followed by repeated Start condition
W
Execute Acknowledge Action (no action) followed by a byte write operation(2)
R
Execute Acknowledge Action followed by a byte read operation
X
Execute Acknowledge Action followed by issuing a Stop condition
Notes:
1. The ACKACT bit and the MCMD bit field can be written at the same time.
2. For a master write operation, the TWI will wait for new data to be written to the Master Data (TWIn.MDATA)
register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 439
AVR128DB28/32/48/64
TWI - Two-Wire Interface
29.5.6
Master Status
Name:
Offset:
Reset:
Property:
Bit
7
RIF
R/W
0
Access
Reset
MSTATUS
0x05
0x00
-
6
WIF
R/W
0
5
CLKHOLD
R/W
0
4
RXACK
R
0
3
ARBLOST
R/W
0
2
BUSERR
R/W
0
1
0
BUSSTATE[1:0]
R/W
R/W
0
0
Bit 7 – RIF Read Interrupt Flag
This flag is set to ‘1’ when the master byte read operation is successfully completed.
The RIF flag can be used for a master read interrupt. More information can be found in the Read Interrupt Enable
(RIEN) bit from the Master Control A (TWIn.MCTRLA) register.
This flag is automatically cleared when accessing several other TWI registers. The RIF flag can be cleared by
choosing one of the following methods:
1. Writing a ‘1’ to it.
2.
3.
4.
Writing to the Master Address (TWIn.MADDR) register.
Writing/Reading the Master Data (TWIn.MDATA) register.
Writing to the Command (MCMD) bit field from the Master Control B (TWIn.MCTRLB) register.
Bit 6 – WIF Write Interrupt Flag
This flag is set to ‘1’ when a master transmit address or byte write operation is completed, regardless of the
occurrence of a bus error or arbitration lost condition.
The WIF flag can be used for a master write interrupt. More information can be found from the Write Interrupt Enable
(WIEN) bit in the Master Control A (TWIn.MCTRLA) register.
This flag can be cleared by choosing one of the methods described for the RIF flag.
Bit 5 – CLKHOLD Clock Hold
When this bit is read as ‘1’, it indicates that the master is currently holding the SCL low, stretching the TWI clock
period.
This bit can be cleared by choosing one of the methods described for the RIF flag.
Bit 4 – RXACK Received Acknowledge
When this flag is read as ‘0’, it indicates that the most recent Acknowledge bit from the slave was ACK and the slave
is ready for more data.
When this flag is read as ‘1’, it indicates that the most recent Acknowledge bit from the slave was NACK and the
slave is not able to or does not need to receive more data.
Bit 3 – ARBLOST Arbitration Lost
When this bit is read as ‘1’, it indicates that the master has lost arbitration. This can happen in one of the following
cases:
1. While transmitting a high data bit.
2. While transmitting a NACK bit.
3. While issuing a Start condition (S).
4. While issuing a repeated Start (Sr).
This flag can be cleared by choosing one of the methods described for the RIF flag.
Bit 2 – BUSERR Bus Error
The BUSERR flag indicates that an illegal bus operation has occurred. An illegal bus operation is detected if a
protocol violating the Start (S), repeated Start (Sr), or Stop (P) conditions is detected on the TWI bus lines. A Start
condition directly followed by a Stop condition is one example of a protocol violation.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 440
AVR128DB28/32/48/64
TWI - Two-Wire Interface
The BUSERR flag can be cleared by choosing one of the following methods:
1. Writing a ‘1’ to it.
2.
Writing to the Master Address (TWIn.MADDR) register.
The TWI bus error detector is part of the TWI Master circuitry. For the bus errors to be detected, the TWI Master must
be enabled (ENABLE bit in TWIn.MCTRLA is ‘1’), and the main clock frequency must be at least four times the SCL
frequency.
Bits 1:0 – BUSSTATE[1:0] Bus State
This bit field indicates the current TWI bus state.
Value
Name
Description
0x0
UNKNOWN
Unknown bus state
0x1
IDLE
Idle bus state
0x2
OWNER
This TWI controls the bus
0x3
BUSY
Busy bus state
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 441
AVR128DB28/32/48/64
TWI - Two-Wire Interface
29.5.7
Master Baud Rate
Name:
Offset:
Reset:
Property:
Bit
7
MBAUD
0x06
0x00
-
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
BAUD[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – BAUD[7:0] Baud Rate
This bit field is used to derive the SCL high and low time. It must be written while the master is disabled. The master
can be disabled by writing ‘0’ to the Enable TWI Master (ENABLE) bit from the Master Control A (TWIn.MCTRLA)
register.
Refer to the 29.3.2.2.1 Clock Generation section for more information on how to calculate the frequency of the SCL.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 442
AVR128DB28/32/48/64
TWI - Two-Wire Interface
29.5.8
Master Address
Name:
Offset:
Reset:
Property:
Bit
7
MADDR
0x07
0x00
-
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
ADDR[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – ADDR[7:0] Address
This register contains the address of the external slave device. When this bit field is written, the TWI will issue a Start
condition, and the shift register performs a byte transmit operation on the bus depending on the bus state.
This register can be read at any time without interfering with the ongoing bus activity since a read access does not
trigger the master logic to perform any bus protocol related operations.
The master control logic uses the bit 0 of this register as the R/W direction bit.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 443
AVR128DB28/32/48/64
TWI - Two-Wire Interface
29.5.9
Master Data
Name:
Offset:
Reset:
Property:
Bit
7
MDATA
0x08
0x00
-
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
DATA[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – DATA[7:0] Data
This bit field provides direct access to the master’s physical shift register, which is used to shift out data on the bus
(transmit) and to shift in data received from the bus (receive). The direct access implies that the MDATA register
cannot be accessed during byte transmissions.
Reading valid data or writing data to be transmitted can only be successful when the CLKHOLD bit is read as ‘1’ or
when an interrupt occurs.
A write access to the MDATA register will command the master to perform a byte transmit operation on the bus,
directly followed by receiving the Acknowledge bit from the slave. This is independent of the Acknowledge Action
(ACKACT) bit from the Master Control B (TWIn.MCTRLB) register. The write operation is performed regardless of
winning or losing arbitration before the Write Interrupt Flag (WIF) is set to ‘1’.
If the Smart Mode Enable (SMEN) bit in the Master Control A (TWIn.MCTRLA) register is set to ‘1’, a read access to
the MDATA register will command the master to perform an acknowledge action. This is dependent on the setting of
the Acknowledge Action (ACKACT) bit from the Master Control B (TWIn.MCTRLB) register.
Notes:
1. The WIF and RIF interrupt flags are cleared automatically if the MDATA register is read while ACKACT is set
to ‘1’.
2.
3.
The ARBLOST and BUSEER flags are left unchanged.
The WIF, RIF, ARBLOST, and BUSERR flags together with the Clock Hold (CLKHOLD) bit are all located in
the Master Status (TWIn.MSTATUS) register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 444
AVR128DB28/32/48/64
TWI - Two-Wire Interface
29.5.10 Slave Control A
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
DIEN
R/W
0
SCTRLA
0x09
0x00
-
6
APIEN
R/W
0
5
PIEN
R/W
0
4
3
2
PMEN
R/W
0
1
SMEN
R/W
0
0
ENABLE
R/W
0
Bit 7 – DIEN Data Interrupt Enable
Writing this bit to ‘1’ enables an interrupt on the Data Interrupt Flag (DIF) from the Slave Status (TWIn.SSTATUS)
register.
A TWI slave data interrupt will be generated only if this bit, the DIF flag, and the Global Interrupt Enable (I) bit in
Status (CPU.SREG) register are all ‘1’.
Bit 6 – APIEN Address or Stop Interrupt Enable
Writing this bit to ‘1’ enables an interrupt on the Address or Stop Interrupt Flag (APIF) from the Slave Status
(TWIn.SSTATUS) register.
A TWI slave address or stop interrupt will be generated only if this bit, the APIF flag, and the Global Interrupt Enable
(I) bit in the Status (CPU.SREG) register are all ‘1’.
Notes:
1. The slave stop interrupt shares the interrupt flag and vector with the slave address interrupt.
2. The Stop Interrupt Enable (PIEN) bit in the Slave Control A (TWIn.SCTRLA) register must be written to ‘1’ for
the APIF to be set on a Stop condition.
3. When the interrupt occurs, the Address or Stop (AP) bit in the Slave Status (TWIn.SSTATUS) register will
determine whether an address match or a Stop condition caused the interrupt.
Bit 5 – PIEN Stop Interrupt Enable
Writing this bit to ‘1’ allows the Address or Stop Interrupt Flag (APIF) in the Slave Status (TWIn.SSTATUS) register to
be set when a Stop condition occurs. To use this feature, the main clock frequency must be at least four times the
SCL frequency.
Bit 2 – PMEN Address Recognition Mode
If this bit is written to ‘1’, the slave address match logic responds to all received addresses.
If this bit is written to ‘0’, the address match logic uses the Slave Address (TWIn.SADDR) register to determine which
address to recognize as the slave’s address.
Bit 1 – SMEN Smart Mode Enable
Writing this bit to ‘1’ enables the slave Smart mode. When the Smart mode is enabled, issuing a command by writing
to the Command (SCMD) bit field in the Slave Control B (TWIn.SCTRLB) register or accessing the Slave Data
(TWIn.SDATA) register resets the interrupt, and the operation continues. If the Smart mode is disabled, the slave
always waits for a new slave command before continuing.
Bit 0 – ENABLE Enable TWI Slave
Writing this bit to ‘1’ enables the TWI slave.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 445
AVR128DB28/32/48/64
TWI - Two-Wire Interface
29.5.11 Slave Control B
Name:
Offset:
Reset:
Property:
Bit
7
SCTRLB
0x0A
0x00
-
6
5
4
3
Access
Reset
2
ACKACT
R/W
0
1
0
SCMD[1:0]
R/W
0
R/W
0
Bit 2 – ACKACT Acknowledge Action
The ACKACT(1) bit represents the behavior of the slave device under certain conditions defined by the bus protocol
state and the software interaction. If the Smart Mode Enable (SMEN) bit in the Slave Control A (TWIn.SCTRLA)
register is set to ‘1’, the acknowledge action is performed when the Slave Data (TWIn.SDATA) register is read, else a
command must be written to the Command (SCMD) bit field in the Slave Control B (TWIn.SCTRLB) register.
The acknowledge action is not performed when the Slave Data (TWIn.SDATA) register is written, since the slave is
sending data.
Value
Name
Description
0
ACK
Send ACK
1
NACK
Send NACK
Bits 1:0 – SCMD[1:0] Command
The SCMD(1) bit field is a strobe. This bit field is always read as ‘0’.
Writing to this bit field triggers a slave operation as defined by the table below.
Table 29-3. Command Settings
SCMD[1:0]
Group
Configuration
DIR Description
0x0
0x1
NOACT
—
0x2
COMPTRANS
0x3
RESPONSE
X
No action
X
Reserved
Used to complete a transaction
W
Execute Acknowledge Action succeeded by waiting for any Start (S/Sr)
condition
R
Wait for any Start (S/Sr) condition
Used in response to an address interrupt (APIF)
W
Execute Acknowledge Action succeeded by reception of next byte
R
Execute Acknowledge Action succeeded by slave data interrupt
Used in response to a data interrupt (DIF)
W
Execute Acknowledge Action succeeded by reception of next byte
R
Execute a byte read operation followed by Acknowledge Action
Note: 1. The ACKACT bit and the SCMD bit field can be written at the same time. The ACKACT will be updated
before the command is triggered.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 446
AVR128DB28/32/48/64
TWI - Two-Wire Interface
29.5.12 Slave Status
Name:
Offset:
Reset:
Property:
Bit
7
DIF
R/W
0
Access
Reset
SSTATUS
0x0B
0x00
-
6
APIF
R/W
0
5
CLKHOLD
R
0
4
RXACK
R
0
3
COLL
R/W
0
2
BUSERR
R/W
0
1
DIR
R
0
0
AP
R
0
Bit 7 – DIF Data Interrupt Flag
This flag is set to ‘1’ when the slave byte transmit or receive operation is successfully completed without any bus
errors. This flag can be set to ‘1’ with an unsuccessful transaction in case of a collision detection. More information
can be found in the Collision (COLL) bit description.
The DIF flag can generate a slave data interrupt. More information can be found in Data Interrupt Enable (DIEN) bit
from the Slave Control A (TWIn.SCTRLA) register.
This flag is automatically cleared when accessing several other TWI registers. The DIF flag can be cleared by
choosing one of the following methods:
1. Writing/Reading the Slave Data (TWIn.SDATA) register.
2. Writing to the Command (SCMD) bit field from the Slave Control B (TWIn.SCTRLB) register.
Bit 6 – APIF Address or Stop Interrupt Flag
This flag is set to ‘1’ when the slave address has been received or by a Stop condition.
The APIF flag can generate a slave address or stop interrupt. More information can be found in the Address or Stop
Interrupt Enable (APIEN) bit from the Slave Control A (TWIn.SCTRLA) register.
This flag can be cleared by choosing one of the methods described for the DIF flag.
Bit 5 – CLKHOLD Clock Hold
When this bit is read as ‘1’, it indicates that the slave is currently holding the SCL low, stretching the TWI clock
period.
This bit is set to ‘1’ when an address or data interrupt occurs. Resetting the corresponding interrupt will indirectly set
this bit to ‘0’.
Bit 4 – RXACK Received Acknowledge
When this flag is read as ‘0’, it indicates that the most recent Acknowledge bit from the master was ACK.
When this flag is read as ‘1’, it indicates that the most recent Acknowledge bit from the master was NACK.
Bit 3 – COLL Collision
When this bit is read as ‘1’, it indicates that the slave has not been able to do one of the following:
1.
2.
Transmit high bits on the SDA. The Data Interrupt Flag (DIF) will be set to ‘1’ at the end as a result of the
internal completion of an unsuccessful transaction.
Transmit the NACK bit. The collision occurs because the slave address match already took place, and the
APIF flag is set to ‘1’ as a result.
Writing a ‘1’ to this bit will clear the COLL flag. The flag is automatically cleared if any Start condition (S/Sr) is
detected.
Note: The APIF and DIF flags can only generate interrupts whose handlers can be used to check for the collision.
Bit 2 – BUSERR Bus Error
The BUSERR flag indicates that an illegal bus operation has occurred. Illegal bus operation is detected if a protocol
violating the Start (S), repeated Start (Sr), or Stop (P) conditions is detected on the TWI bus lines. A Start condition
directly followed by a Stop condition is one example of a protocol violation. Writing a ‘1’ to this bit will clear the
BUSERR flag.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 447
AVR128DB28/32/48/64
TWI - Two-Wire Interface
The TWI bus error detector is part of the TWI Master circuitry. For the bus errors to be detected by the slave, the TWI
Dual mode or the TWI Master must be enabled, and the main clock frequency must be at least four times the SCL
frequency. The TWI Dual mode can be enabled by writing ‘1’ to the ENABLE bit in the TWIn.DUALCTRL register.
The TWI Master can be enabled by writing ‘1’ to the ENABLE bit in the TWIn.MCTRLA register.
Bit 1 – DIR Read/Write Direction
This bit indicates the current TWI bus direction. The DIR bit reflects the direction bit value from the last address
packet received from a master TWI device.
When this bit is read as ‘1’, it indicates that a master read operation is in progress.
When this bit is read as ‘0’, it indicates that a master write operation is in progress.
Bit 0 – AP Address or Stop
When the TWI slave Address or Stop Interrupt Flag (APIF) is set ‘1’, this bit determines whether the interrupt is due
to an address detection or a Stop condition.
Value
Name
Description
0
STOP
A Stop condition generated the interrupt on the APIF flag
1
ADR
Address detection generated the interrupt on the APIF flag
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 448
AVR128DB28/32/48/64
TWI - Two-Wire Interface
29.5.13 Slave Address
Name:
Offset:
Reset:
Property:
Bit
7
SADDR
0x0C
0x00
-
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
ADDR[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – ADDR[7:0] Address
The Slave Address (TWIn.SADDR) register is used by the slave address match logic to determine if a master device
has addressed the TWI slave. The Address or Stop Interrupt Flag (APIF) and the Address or Stop (AP) bit in the
Slave Status (TWIn.SSTATUS) register are set to ‘1’ if an address packet is received.
The upper seven bits (ADDR[7:1]) of the SADDR register represent the main slave address.
The Least Significant bit (ADDR[0]) of the SADDR register is used for the recognition of the General Call Address
(0x00) of the I2C protocol. This feature is enabled when this bit is set to ‘1’.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 449
AVR128DB28/32/48/64
TWI - Two-Wire Interface
29.5.14 Slave Data
Name:
Offset:
Reset:
Property:
Bit
7
SDATA
0x0D
0x00
-
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
DATA[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – DATA[7:0] Data
This bit field provides access to the slave data register.
Reading valid data or writing data to be transmitted can only be successfully achieved when the SCL is held low by
the slave (i.e., when the slave CLKHOLD bit is set to ‘1’). It is not necessary to check the Clock Hold (CLKHOLD) bit
from the Slave Status (TWIn.SSTATUS) register in software before accessing the SDATA register if the software
keeps track of the present protocol state by using interrupts or observing the interrupt flags.
If the Smart Mode Enable (SMEN) bit in the Slave Control A (TWIn.SCTRLA) register is set to ‘1’, a read access to
the SDATA register, when the clock hold is active, auto-triggers bus operations and will command the slave to
perform an acknowledge action. This is dependent on the setting of the Acknowledge Action (ACKACT) bit from the
Slave Control B (TWIn.SCTRLB) register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 450
AVR128DB28/32/48/64
TWI - Two-Wire Interface
29.5.15 Slave Address Mask
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
SADDRMASK
0x0E
0x00
-
7
6
5
R/W
0
R/W
0
R/W
0
4
ADDRMASK[6:0]
R/W
0
3
2
1
R/W
0
R/W
0
R/W
0
0
ADDREN
R/W
0
Bits 7:1 – ADDRMASK[6:0] Address Mask
The ADDRMASK bit field acts as a second address match or an address mask register depending on the ADDREN
bit.
If the ADDREN bit is written to ‘0’, the ADDRMASK bit field can be loaded with a 7-bit Slave Address mask. Each of
the bits in the Slave Address Mask (TWIn.SADDRMASK) register can mask (disable) the corresponding address bits
in the TWI Slave Address (TWIn.SADDR) register. When a bit from the mask is written to ‘1’, the address match logic
ignores the comparison between the incoming address bit and the corresponding bit in the Slave Address
(TWIn.SADDR) register. In other words, masked bits will always match, making it possible to recognize ranges of
addresses.
If the ADDREN bit is written to ‘1’, the Slave Address Mask (TWIn.SADDRMASK) register can be loaded with a
second slave address in addition to the Slave Address (TWIn.SADDR) register. In this mode, the slave will have two
unique addresses, one in the Slave Address (TWIn.SADDR) register and the other one in the Slave Address Mask
(TWIn.SADDRMASK) register.
Bit 0 – ADDREN Address Mask Enable
If this bit is written to ‘0’, the TWIn.SADDRMASK register acts as a mask to the TWIn.SADDR register.
If this bit is written to ‘1’, the slave address match logic responds to the two unique addresses in slave TWIn.SADDR
and TWIn.SADDRMASK.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 451
AVR128DB28/32/48/64
CRCSCAN - Cyclic Redundancy Check Memory Sca...
30.
CRCSCAN - Cyclic Redundancy Check Memory Scan
30.1
Features
•
•
•
•
30.2
CRC-16-CCITT or CRC-32 (IEEE 802.3)
Check of the Entire Flash Section, Application Code, and/or Boot Section
Selectable NMI Trigger on Failure
User-Configurable Check During Internal Reset Initialization
Overview
The Cyclic Redundancy Check (CRC) is an important safety feature. It scans the Nonvolatile Memory (NVM) making
sure the code is correct.
The device will not execute code if Flash fault has occurred. By ensuring no code corruption has occurred, a
potentially unintended behavior in the application that can cause a dangerous situation can be avoided. The CRC
scan can be set up to scan the entire Flash, only the boot section, or both the boot and application code sections.
The CRC generates a checksum that is compared to a pre-calculated one. If the two checksums match, the Flash is
OK, and the application code can start running.
The BUSY bit in the Status (CRCSCAN.STATUS) register indicates if a CRC scan is ongoing or not, while the OK bit
in the Status (CRCSCAN.STATUS) register indicates if the checksum comparison matches or not.
The CRCSCAN can be set up to generate a Non-Maskable Interrupt (NMI) if the checksums do not match.
30.2.1
Block Diagram
Figure 30-1. Cyclic Redundancy Check Block Diagram
Memory
(Boot, App,
Flash)
CTRLB
CTRLA
Source
Enable,
Reset
CRC
calculation
BUSY
STATUS
OK
CHECKSUM
30.3
Functional Description
30.3.1
Initialization
NMI Req
To enable a CRC in software (or via the debugger):
1. Write the Source (SRC) bit field of the Control B (CRCSCAN.CTRLB) register to select the desired source
settings.
2. Enable the CRCSCAN by writing a ‘1’ to the ENABLE bit in the Control A (CRCSCAN.CTRLA) register.
3.
The CRC will start after three cycles. The CPU will continue executing during these three cycles.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 452
AVR128DB28/32/48/64
CRCSCAN - Cyclic Redundancy Check Memory Sca...
Selection between CRC32 and CRC16 is done through fuse settings. The CRCSCAN can be configured to perform a
code memory scan before the device leaves Reset. If this check fails, the CPU is not allowed to start normal code
execution. This feature is enabled and controlled by the CRCSRC field in FUSE.SYSCFG0 (see the Fuses section for
more information).
If the CRCSCAN is enabled, a successful CRC check will have the following outcome:
• Normal code execution starts
• The ENABLE bit in CRCSCAN.CTRLA will be ‘1’
•
•
The SRC bit field in CRCSCAN.CTRLB will reflect the checked section(s)
The OK flag in CRCSCAN.STATUS will be ‘1’
If the CRCSCAN is enabled, a non-successful CRC check will have the following outcome:
• Normal code execution does not start. The CPU will hang executing no code.
• The ENABLE bit in CRCSCAN.CTRLA will be ‘1’
30.3.2
•
•
The SRC bit field in CRCSCAN.CTRLB will reflect the checked section(s)
The OK flag in CRCSCAN.STATUS will be ‘0’
•
This condition may be observed using the debug interface
Operation
When operating, the CRCSCAN has priority access to the Flash and will stall the CPU until completed.
The CRC will use three clock cycles for each 16-bit fetch. The CRCSCAN can be configured to do a scan from startup.
An n-bit CRC applied to a data block of arbitrary length will detect any single alteration (error burst) up to n bits in
length. For longer error bursts a fraction 1-2-n will be detected.
The CRC generator supports CRC-16-CCITT and CRC-32 (IEEE 802.3).
The polynomial options are:
•
•
CRC-16-CCITT: x16 + x12 + x5 + 1
CRC-32: x32 + x26 + x23 + x22 + x16 + x12 + x11 + x10 + x8 + x7 + x5 + x4 + x2 + x + 1
The CRC reads byte-by-byte the content of the section(s) it is set up to check, starting with byte 0, and generates a
new checksum per byte. The byte is sent through a shift register as depicted below, starting with the Most Significant
bit. If the last bytes in the section contain the correct checksum, the CRC will pass. See 30.3.2.1 Checksum for how
to place the checksum. The initial value of the Checksum register is 0xFFFF.
30.3.2.1 Checksum
The pre-calculated checksum must be present in the last location of the section to be checked. If the BOOT section is
to be checked, the checksum must be saved in the last bytes of the BOOT section. The same is done for
APPLICATION and the entire Flash. Table 30-1 shows explicitly how the checksum must be stored for the different
sections. Refer to the CRCSCAN.CTRLB register description for how to configure the sections to be checked.
Table 30-1. Placement of the Pre-Calculated Checksum for CRC16 in Flash
Section to Check
CHECKSUM[15:8]
CHECKSUM[7:0]
BOOT
BOOTEND-1
BOOTEND
BOOT and APPLICATION
APPEND-1
APPEND
Full Flash
FLASHEND-1
FLASHEND
Table 30-2. Placement of the Pre-Calculated Checksum for CRC32 in Flash
Section to Check
BOOT
BOOT and
APPLICATION
Full Flash
CHECKSUM[31:24]
BOOTEND
APPEND
CHECKSUM[23:16]
BOOTEND-1
APPEND-1
CHECKSUM[15:8]
BOOTEND-2
APPEND-2
CHECKSUM[7:0]
BOOTEND-3
APPEND-3
FLASHEND
FLASHEND-1
FLASHEND-2
FLASHEND-3
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 453
AVR128DB28/32/48/64
CRCSCAN - Cyclic Redundancy Check Memory Sca...
30.3.3
Interrupts
Table 30-3. Available Interrupt Vectors and Sources
Name
Vector Description
Conditions
NMI
Non-Maskable Interrupt
CRC failure
When the interrupt condition occurs the OK flag in the Status (CRCSCAN.STATUS) register is cleared to ‘0’.
A Non-Maskable Interrupt (NMI) is enabled by writing a ‘1’ to the respective Enable (NMIEN) bit in the Control A
(CRCSCAN.CTRLA) register, but can only be disabled with a System Reset. An NMI is generated when the OK flag
in the CRCSCAN.STATUS register is cleared, and the NMIEN bit is ‘1’. The NMI request remains active until a
System Reset and cannot be disabled.
An NMI can be triggered even if interrupts are not globally enabled.
30.3.4
Sleep Mode Operation
In all CPU Sleep modes, the CRCSCAN is halted and will resume operation when the CPU wakes up.
The CRCSCAN starts operation three cycles after writing the Enable (ENABLE) bit in the Control A
(CRCSCAN.CTRLA) register. During these three cycles, it is possible to enter Sleep mode. In this case:
1. The CRCSCAN will not start until the CPU is woken up.
2. Any interrupt handler will execute after CRCSCAN has finished.
30.3.5
Debug Operation
Whenever the debugger reads or writes a peripheral or memory location, the CRCSCAN will be disabled.
If the CRCSCAN is busy when the debugger accesses the device, the CRCSCAN will restart the ongoing operation
when the debugger accesses an internal register or when the debugger disconnects.
The BUSY bit in the Status (CRCSCAN.STATUS) register will read ‘1’ if the CRCSCAN was busy when the debugger
caused it to disable, but it will not actively check any section as long as the debugger keeps it disabled. There are
synchronized CRC status bits in the debugger's internal register space, which can be read by the debugger without
disabling the CRCSCAN. Reading the debugger's internal CRC status bits will make sure that the CRCSCAN is
enabled.
It is possible to write the CRCSCAN.STATUS register directly from the debugger:
• BUSY bit in CRCSCAN.STATUS:
– Writing the BUSY bit to ‘0’ will stop the ongoing CRC operation (so that the CRCSCAN does not restart its
operation when the debugger allows it).
– Writing the BUSY bit to ‘1’ will make the CRC start a single check with the settings in the Control B
(CRCSCAN.CTRLB) register, but not until the debugger allows it.
•
As long as the BUSY bit in CRCSCAN.STATUS is ‘1’, CRCSCAN.CTRLB and the Non-Maskable Interrupt
Enable (NMIEN) bit in the Control A (CRCSCAN.CTRLA) register cannot be altered.
OK bit in CRCSCAN.STATUS:
– Writing the OK bit to ‘0’ can trigger a Non-Maskable Interrupt (NMI) if the NMIEN bit in CRCSCAN.CTRLA
is ‘1’. If an NMI has been triggered, no writes to the CRCSCAN are allowed.
– Writing the OK bit to ‘1’ will make the OK bit read as ‘1’ when the BUSY bit in CRCSCAN.STATUS is ‘0’.
Writes to CRCSCAN.CTRLA and CRCSCAN.CTRLB from the debugger are treated in the same way as writes from
the CPU.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 454
AVR128DB28/32/48/64
CRCSCAN - Cyclic Redundancy Check Memory Sca...
30.4
Register Summary
Offset
Name
Bit Pos.
7
0x00
0x01
0x02
CTRLA
CTRLB
STATUS
7:0
7:0
7:0
RESET
30.5
6
5
4
3
2
1
0
NMIEN
ENABLE
SRC[1:0]
OK
BUSY
Register Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 455
AVR128DB28/32/48/64
CRCSCAN - Cyclic Redundancy Check Memory Sca...
30.5.1
Control A
Name:
Offset:
Reset:
Property:
CTRLA
0x00
0x00
-
If an NMI has been triggered this register is not writable.
Bit
Access
Reset
7
RESET
R/W
0
6
5
4
3
2
1
NMIEN
R/W
0
0
ENABLE
R/W
0
Bit 7 – RESET Reset CRCSCAN
Writing this bit to ‘1’ resets the CRCSCAN. The CRCSCAN Control and Status (CRCSCAN.CTRLA,
CRCSCAN.CTRLB, CRCSCAN.STATUS) register will be cleared one clock cycle after the RESET bit is written to ‘1’.
If NMIEN is ‘0’, this bit is writable both when the CRCSCAN is busy (the BUSY bit in CRCSCAN.STATUS is ‘1’) and
not busy (the BUSY bit is ‘0’), and will take effect immediately.
If NMIEN is ‘1’, this bit is only writable when the CRCSCAN is not busy (the BUSY bit in CRCSCAN.STATUS is ‘0’).
The RESET bit is a strobe bit.
Bit 1 – NMIEN Enable NMI Trigger
When this bit is written to ‘1’, any CRC failure will trigger an NMI.
This bit can only be cleared by a System Reset. It is not cleared by a write to the RESET bit.
This bit can only be written to ‘1’ when the CRCSCAN is not busy (the BUSY bit in CRCSCAN.STATUS is ‘0’).
Bit 0 – ENABLE Enable CRCSCAN
Writing this bit to ‘1’ enables the CRCSCAN with the current settings. It will stay ‘1’ even after a CRC check has
completed, but writing it to ‘1’ again will start a new check.
Writing the bit to ‘0’ has no effect.
The CRCSCAN can be configured to run a scan during the microcontroller (MCU) start-up sequence to verify the
Flash sections before letting the CPU start normal code execution (see the 30.3.1 Initialization section). If this feature
is enabled, the ENABLE bit will read as ‘1’ when normal code execution starts.
To see whether the CRCSCAN is busy with an ongoing check, poll the BUSY bit in the Status (CRCSCAN.STATUS)
register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 456
AVR128DB28/32/48/64
CRCSCAN - Cyclic Redundancy Check Memory Sca...
30.5.2
Control B
Name:
Offset:
Reset:
Property:
CTRLB
0x01
0x00
-
The Control B register contains the source settings for the CRC. It is not writable when the CRCSCAN is busy, or
when an NMI has been triggered.
Bit
7
6
5
4
3
2
1
0
SRC[1:0]
Access
Reset
R/W
0
R/W
0
Bits 1:0 – SRC[1:0] CRC Source
The SRC bit field selects which section of the Flash will be checked by the CRCSCAN. To set up section sizes, refer
to the Fuses section.
The CRCSCAN can be enabled during internal Reset initialization to verify Flash sections before letting the CPU start
(see the Fuses section). If the CRCSCAN is enabled during internal Reset initialization, the SRC bit field will read out
as FLASH, BOOTAPP, or BOOT when normal code execution starts (depending on the configuration).
Value
Name
Description
0x0
FLASH
The CRC is performed on the entire Flash (boot, application code, and application data
sections).
0x1
BOOTAPP The CRC is performed on the boot and application code sections of Flash.
0x2
BOOT
The CRC is performed on the boot section of Flash.
0x3
Reserved.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 457
AVR128DB28/32/48/64
CRCSCAN - Cyclic Redundancy Check Memory Sca...
30.5.3
Status
Name:
Offset:
Reset:
Property:
Bit
7
STATUS
0x02
0x02
-
6
5
4
3
Access
Reset
2
1
OK
R
1
0
BUSY
R
0
Bit 1 – OK CRC OK
When this bit is read as ‘1’, the previous CRC completed successfully. The bit is set to ‘1’ by default before a CRC
scan is run. The bit is not valid unless BUSY is ‘0’.
Bit 0 – BUSY CRC Busy
When this bit is read as ‘1’, the CRCSCAN is busy. As long as the module is busy, the access to the control registers
is limited.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 458
AVR128DB28/32/48/64
CCL – Configurable Custom Logic
31.
CCL – Configurable Custom Logic
31.1
Features
•
•
•
•
•
•
•
•
•
31.2
Glue Logic for General Purpose PCB Design
6 Programmable Look-Up Tables (LUTs)
Combinatorial Logic Functions: Any Logic Expression which is a Function of up to Three Inputs.
Sequencer Logic Functions:
– Gated D flip-flop
– JK flip-flop
– Gated D latch
– RS latch
Flexible LUT Input Selection:
– I/Os
– Events
– Subsequent LUT output
– Internal peripherals such as:
• Analog comparator
• Timers/Counters
• USART
• SPI
Clocked by a System Clock or other Peripherals
Output can be Connected to I/O Pins or an Event System
Optional Synchronizer, Filter, or Edge Detector Available on Each LUT Output
Optional Interrupt Generation from Each LUT Output:
– Rising edge
– Falling edge
– Both edges
Overview
The Configurable Custom Logic (CCL) is a programmable logic peripheral which can be connected to the device
pins, to events, or to other internal peripherals. The CCL can serve as ‘glue logic’ between the device peripherals and
external devices. The CCL can eliminate the need for external logic components, and can also help the designer to
overcome real-time constraints by combining Core Independent Peripherals (CIPs) to handle the most time-critical
parts of the application independent of the CPU.
The CCL peripheral provides a number of Look-up Tables (LUTs). Each LUT consists of three inputs, a truth table, a
synchronizer/filter, and an edge detector. Each LUT can generate an output as a user programmable logic expression
with three inputs. The output is generated from the inputs using the combinatorial logic and can be filtered to remove
spikes. The CCL can be configured to generate an interrupt request on changes in the LUT outputs.
Neighboring LUTs can be combined to perform specific operations. A sequencer can be used for generating complex
waveforms.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 459
AVR128DB28/32/48/64
CCL – Configurable Custom Logic
31.2.1
Block Diagram
Figure 31-1. Configurable Custom Logic
Even LUT n
INSEL
Internal
Events
I/O
Peripherals
FILTSEL
LUTn-TRUTHSEL[2:0]
Filter/
Synch
TRUTH
CLKSRC
Edge
Detector
LUTn-OUT
CLK_LUTn
Clock Sources
LUTn-TRUTHSEL[2]
Sequencer
Odd LUT n+1
INSEL
Internal
Events
I/O
Peripherals
FILTSEL
LUTn+1-TRUTHSEL[2:0]
Filter/
Synch
TRUTH
CLKSRC
SEQSEL
EDGEDET
EDGEDET
LUTn+1-OUT
Edge
Detector
CLK_LUTn+1
Clock Sources
LUTn+1-TRUTHSEL[2]
Table 31-2. Sequencer and LUT Connection
31.2.2
Sequencer
Even and Odd LUT
SEQ0
LUT0 and LUT1
SEQ1
LUT2 and LUT3
SEQ2
LUT4 and LUT5
Signal Description
Name
Type
Description
LUTn-OUT
Digital output
Output from the look-up table
LUTn-IN[2:0]
Digital input
Input to the look-up table. LUTn-IN[2] can serve as CLK_LUTn.
Refer to I/O Multiplexing and Considerations for details on the pin mapping for this peripheral. One signal can be
mapped to several pins.
31.2.2.1 CCL Input Selection MUX
The following peripherals outputs are available as inputs into the CCL LUT.
Value
Input Source
0x00
MASK
0x01
FEEDBACK
0x02
LINK
LUT[n+1]
0x03
EVENTA
EVENTA
0x04
EVENTB
EVENTB
0x05
INn
LUTn-IN0
LUTn-IN1
LUTn-IN2
0x06
ACn
AC0 OUT
AC1 OUT
AC2 OUT
© 2020 Microchip Technology Inc.
INSEL0[3:0]
INSEL1[3:0]
INSEL2[3:0]
Masked input
LUTn
Preliminary Datasheet
DS40002247A-page 460
AVR128DB28/32/48/64
CCL – Configurable Custom Logic
...........continued
Value
Input Source
INSEL0[3:0]
INSEL1[3:0]
INSEL2[3:0]
0x07
ZCDn
0x08
USARTn(1)
ZCD0 OUT
ZCD1 OUT
ZCD2 OUT
USART0 TXD
USART1 TXD
USART2 TXD
0x09
SPI0(2)
SPI0 MOSI
SPI0 MOSI
SPI0 SCK
0x0A
TCA0
WO0
WO1
WO2
0x0B
TCA1
WO0
WO1
WO2
0x0C
TCBn
TCB0 WO
TCB1 WO
TCB2 WO
0x0D
TCD0
WOA
WOB
WOC
Notes:
1. USART connections to the CCL work only in asynchronous/synchronous USART master mode.
2. SPI connections to the CCL work only in master SPI mode.
31.3
Functional Description
31.3.1
Operation
31.3.1.1 Enable-Protected Configuration
The configuration of the LUTs and sequencers is enable-protected, meaning that they can only be configured when
the corresponding even LUT is disabled (ENABLE=0 in the LUT n Control A register, CCL.LUTnCTRLA). This is a
mechanism to suppress the undesired output from the CCL under (re-)configuration.
The following bits and registers are enable-protected:
•
•
Sequencer Selection (SEQSEL) in the Sequencer Control n register (CCL.SEQCTRLn)
LUT n Control x registers (CCL.LUTnCTRLx), except the ENABLE bit in CCL.LUTnCTRLA
The enable-protected bits in the CCL.LUTnCTRLx registers can be written at the same time as ENABLE in
CCL.LUTnCTRLA is written to ‘1’, but not at the same time as ENABLE is written to ‘0’.
The enable protection is denoted by the enable-protected property in the register description.
31.3.1.2 Enabling, Disabling, and Resetting
The CCL is enabled by writing a ‘1’ to the ENABLE bit in the Control register (CCL.CTRLA). The CCL is disabled by
writing a ‘0’ to that ENABLE bit.
Each LUT is enabled by writing a ‘1’ to the LUT Enable bit (ENABLE) in the LUT n Control A register
(CCL.LUTnCTRLA). Each LUT is disabled by writing a ‘0’ to the ENABLE bit in CCL.LUTnCTRLA.
31.3.1.3 Truth Table Logic
The truth table in each LUT unit can generate a combinational logic output as a function of up to three inputs (LUTnTRUTHSEL[2:0]). The unused inputs can be turned off (tied low). The truth table for the combinational logic
expression is defined by the bits in the CCL.TRUTHn registers. Each combination of the input bits (LUTnTRUTHSEL[2:0]) corresponds to one bit in the TRUTHn register, as shown in the table below.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 461
AVR128DB28/32/48/64
CCL – Configurable Custom Logic
Figure 31-2. Truth Table Output Value Selection of a LUT
TRUTH[0]
TRUTH[1]
TRUTH[2]
TRUTH[3]
TRUTH[4]
TRUTH[5]
TRUTH[6]
TRUTH[7]
OUT
LUTn-TRUTHSEL[2:0]
Table 31-3. Truth Table of a LUT
LUTn-TRUTHSEL[2]
LUTn-TRUTHSEL[1]
LUTn-TRUTHSEL[0]
OUT
0
0
0
TRUTH[0]
0
0
1
TRUTH[1]
0
1
0
TRUTH[2]
0
1
1
TRUTH[3]
1
0
0
TRUTH[4]
1
0
1
TRUTH[5]
1
1
0
TRUTH[6]
1
1
1
TRUTH[7]
31.3.1.4 Truth Table Inputs Selection
Input Overview
The inputs can be individually:
•
•
•
•
•
OFF
Driven by peripherals
Driven by internal events from the Event System
Driven by I/O pin inputs
Driven by other LUTs
The input for each LUT is configured by writing the Input Source Selection bits in the LUT Control registers:
• INSEL0 in CCL.LUTnCTRLB
• INSEL1 in CCL.LUTnCTRLB
• INSEL2 in CCL.LUTnCTRLC
Internal Feedback Inputs (FEEDBACK)
The output from a sequencer can be used as an input source for the two LUTs it is connected to.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 462
AVR128DB28/32/48/64
CCL – Configurable Custom Logic
Figure 31-3. Feedback Input Selection
Even LUT
Sequencer
Odd LUT
When selected (INSELy=FEEDBACK in LUTnCTRLx), the sequencer (SEQ) output is used as input for the
corresponding LUTs.
Linked LUT (LINK)
When selecting the LINK input option, the next LUT’s direct output is used as LUT input. In general, LUT[n+1] is
linked to the input of LUT[n]. LUT0 is linked to the input of the last LUT.
Example 31-1. Linking all LUTs on a Device with Four LUTs
•
•
•
•
LUT1 is the input for LUT0
LUT2 is the input for LUT1
LUT3 is the input for LUT2
LUT0 is the input for LUT3 (wrap-around)
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 463
AVR128DB28/32/48/64
CCL – Configurable Custom Logic
Figure 31-4. Linked LUT Input Selection
LUT0
SEQ0
LUT1
LUT2
SEQ1
LUT3
Event Input Selection (EVENTx)
Events from the Event System can be used as inputs to the LUTs by writing to the INSELn bit groups in the LUT n
Control B and C registers.
I/O Pin Inputs (IO)
When selecting the IO option, the LUT input will be connected to its corresponding I/O pin. Refer to the I/O
Multiplexing section in the data sheet for more details about where the LUTn-INy pins are located.
Peripherals
The different peripherals on the three input lines of each LUT are selected by writing to the Input Select (INSEL) bits
in the LUT Control registers (LUTnCTRLB and LUTnCTRLC).
31.3.1.5 Filter
By default, the LUT output is a combinational function of the LUT inputs. This may cause some short glitches when
the inputs change the value. These glitches can be removed by clocking through filters if demanded by application
needs.
The Filter Selection bits (FILTSEL) in the LUT n Control A registers (CCL.LUTnCTRLA) define the digital filter
options.
When FILTSEL=SYNCH, the output is synchronized with CLK_LUTn. The output will be delayed by two positive
CLK_LUTn edges.
When FILTSEL=FILTER, only the input that is persistent for more than two positive CLK_LUTn edges will pass
through the gated flip-flop to the output. The output will be delayed by four positive CLK_LUTn edges.
One clock cycle later, after the corresponding LUT is disabled, all internal filter logic is cleared.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 464
AVR128DB28/32/48/64
CCL – Configurable Custom Logic
Figure 31-5. Filter
FILTSEL
DISABLE
Input
SYNCH
OUT
Q
D
R
Q
D
R
Q
D
R
D
EN
Q
FILTER
R
CLK_LUTn
CLR
31.3.1.6 Edge Detector
The edge detector can be used to generate a pulse when detecting a rising edge on its input. To detect a falling edge,
the TRUTH table can be programmed to provide inverted output.
The edge detector is enabled by writing ‘1’ to the Edge Detection bit (EDGEDET) in the LUT n Control A register
(CCL.LUTnCTRLA). In order to avoid unpredictable behavior, a valid filter option must be enabled.
The edge detection is disabled by writing a ‘0’ to EDGEDET in CCL.LUTnCTRLA. After disabling a LUT, the
corresponding internal edge detector logic is cleared one clock cycle later.
Figure 31-6. Edge Detector
EDGEDET
CLK_LUTn
31.3.1.7 Sequencer Logic
Each LUT pair can be connected to a sequencer. The sequencer can function as either D flip-flop, JK flip-flop, gated
D latch, or RS latch. The function is selected by writing the Sequencer Selection (SEQSEL) bit group in the
Sequencer Control register (CCL.SEQCTRLn).
The sequencer receives its input from either the LUT, filter or edge detector, depending on the configuration.
A sequencer is clocked by the same clock as the corresponding even LUT. The clock source is selected by the Clock
Source (CLKSRC) bit group in the LUT n Control A register (CCL.LUTnCTRLA).
The flip-flop output (OUT) is refreshed on the rising edge of the clock. When the even LUT is disabled, the latch is
cleared asynchronously. The flip-flop Reset signal (R) is kept enabled for one clock cycle.
Gated D Flip-Flop (DFF)
The D input is driven by the even LUT output, and the G input is driven by the odd LUT output.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 465
AVR128DB28/32/48/64
CCL – Configurable Custom Logic
Figure 31-7. D Flip-Flop
Even LUT
CLK_LUTn
Odd LUT
Table 31-4. DFF Characteristics
R
G
D
OUT
1
X
X
Clear
0
1
1
Set
0
1
0
Clear
0
0
X
Hold state (no change)
JK Flip-Flop (JK)
The J input is driven by the even LUT output, and the K input is driven by the odd LUT output.
Figure 31-8. JK Flip-Flop
Even LUT
CLK_LUTn
Odd LUT
Table 31-5. JK Characteristics
R
J
K
OUT
1
X
X
Clear
0
0
0
Hold state (no change)
0
0
1
Clear
0
1
0
Set
0
1
1
Toggle
Gated D Latch (DLATCH)
The D input is driven by the even LUT output, and the G input is driven by the odd LUT output.
Figure 31-9. D Latch
Even LUT
D
Odd LUT
G
Q
OUT
Table 31-6. D Latch Characteristics
G
D
OUT
0
X
Hold state (no change)
1
0
Clear
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 466
AVR128DB28/32/48/64
CCL – Configurable Custom Logic
...........continued
G
D
OUT
1
1
Set
RS Latch (RS)
The S input is driven by the even LUT output, and the R input is driven by the odd LUT output.
Figure 31-10. RS Latch
Even LUT
S
Odd LUT
R
Q
OUT
Table 31-7. RS Latch Characteristics
S
R
OUT
0
0
Hold state (no change)
0
1
Clear
1
0
Set
1
1
Forbidden state
31.3.1.8 Clock Source Settings
The filter, edge detector, and sequencer are, by default, clocked by the peripheral clock (CLK_PER). It is also
possible to use other clock inputs (CLK_LUTn) to clock these blocks. This is configured by writing the Clock Source
(CLKSRC) bits in the LUT Control A register.
Figure 31-11. Clock Source Settings
Edge
detector
CLK_PER
IN[2]
OSCHF
OSC32K
OSC1K
Filter
Sequential
logic
CLKSRC
LUTn
When the Clock Source (CLKSRC) bit is written to 0x1, LUTn-TRUTHSEL[2] is used to clock the corresponding filter
and edge detector (CLK_LUTn). The sequencer is clocked by the CLK_LUTn of the even LUT in the pair. When
CLKSRC is written to 0x1, LUTn-TRUTHSEL[2] is treated as OFF (low) in the TRUTH table.
The CCL peripheral must be disabled while changing the clock source to avoid undefined outputs from the peripheral.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 467
AVR128DB28/32/48/64
CCL – Configurable Custom Logic
31.3.2
Interrupts
Table 31-8. Available Interrupt Vectors and Sources
Name Vector Description Conditions
CCL
CCL interrupt
INTn in INTFLAG is raised as configured by the INTMODEn bits in the
CCL.INTCTRLn register
When an interrupt condition occurs, the corresponding interrupt flag is set in the peripheral’s Interrupt Flags
(peripheral.INTFLAGS) register.
An interrupt source is enabled or disabled by writing to the corresponding enable bit in the peripheral’s Interrupt
Control (peripheral.INTCTRL) register.
An interrupt request is generated when the corresponding interrupt source is enabled, and the interrupt flag is set.
The interrupt request remains active until the interrupt flag is cleared. See the peripheral’s INTFLAGS register for
details on how to clear interrupt flags.
When several interrupt request conditions are supported by an interrupt vector, the interrupt requests are ORed
together into one combined interrupt request to the interrupt controller. The user must read the peripheral’s
INTFLAGS register to determine which of the interrupt conditions are present.
31.3.3
Events
The CCL can generate the events shown in the table below.
Table 31-9. Event Generators in the CCL
Generator Name
Description
Event Type Generating Clock Domain Length of Event
Peripheral Event
CCL
LUTn
LUT output level Level
Asynchronous
Depends on the CCL
configuration
The CCL has the event users below for detecting and acting upon input events.
Table 31-10. Event Users in the CCL
User Name
Peripheral
Input
CCL
LUTnx
Description
Input Detection
Async/Sync
LUTn input x or clock signal
No detection
Async
The event signals are passed directly to the LUTs without synchronization or input detection logic.
Two event users are available for each LUT. They can be selected as LUTn inputs by writing to the INSELn bit groups
in the LUT n Control B and Control C registers (CCL.LUTnCTRLB or LUTnCTRLC).
Refer to the Event System (EVSYS) section for more details regarding the event types and the EVSYS configuration.
31.3.4
Sleep Mode Operation
Writing the Run In Standby bit (RUNSTDBY) in the Control A register (CCL.CTRLA) to ‘1’ will allow the selected clock
source to be enabled in Standby Sleep mode.
If RUNSTDBY is ‘0’ the peripheral clock will be disabled in Standby Sleep mode. If the filter, edge detector, and/or
sequencer are enabled, the LUT output will be forced to ‘0’ in Standby Sleep mode. In Idle Sleep mode, the TRUTH
table decoder will continue the operation and the LUT output will be refreshed accordingly, regardless of the
RUNSTDBY bit.
If the Clock Source bit (CLKSRC) in the LUT n Control A register (CCL.LUTnCTRLA) is written to ‘1’, the LUTnTRUTHSEL[2] will always clock the filter, edge detector, and sequencer. The availability of the LUTn-TRUTHSEL[2]
clock in sleep modes will depend on the sleep settings of the peripheral used.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 468
AVR128DB28/32/48/64
CCL – Configurable Custom Logic
31.4
Register Summary
Offset
Name
Bit Pos.
0x00
0x01
0x02
0x03
0x04
0x05
0x06
0x07
0x08
0x09
0x0A
0x0B
0x0C
0x0D
0x0E
0x0F
0x10
0x11
0x12
0x13
0x14
0x15
0x16
0x17
0x18
0x19
0x1A
0x1B
0x1C
0x1D
0x1E
0x1F
CTRLA
SEQCTRL0
SEQCTRL1
SEQCTRL2
Reserved
INTCTRL0
INTCTRL1
INTFLAGS
LUT0CTRLA
LUT0CTRLB
LUT0CTRLC
TRUTH0
LUT1CTRLA
LUT1CTRLB
LUT1CTRLC
TRUTH1
LUT2CTRLA
LUT2CTRLB
LUT2CTRLC
TRUTH2
LUT3CTRLA
LUT3CTRLB
LUT3CTRLC
TRUTH3
LUT4CTRLA
LUT4CTRLB
LUT4CTRLC
TRUTH4
LUT5CTRLA
LUT5CTRLB
LUT5CTRLC
TRUTH5
7:0
7:0
7:0
7:0
31.5
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7
6
5
4
3
2
1
RUNSTDBY
0
ENABLE
SEQSEL0[3:0]
SEQSEL1[3:0]
SEQSEL2[3:0]
INTMODE3[1:0]
EDGEDET
EDGEDET
EDGEDET
EDGEDET
EDGEDET
EDGEDET
INTMODE2[1:0]
INTMODE1[1:0]
INTMODE0[1:0]
INTMODE5[1:0]
INTMODE4[1:0]
INT5
INT4
INT3
INT2
INT1
INT0
OUTEN
FILTSEL[1:0]
CLKSRC[2:0]
ENABLE
INSEL1[3:0]
INSEL0[3:0]
INSEL2[3:0]
TRUTH[7:0]
OUTEN
FILTSEL[1:0]
CLKSRC[2:0]
ENABLE
INSEL1[3:0]
INSEL0[3:0]
INSEL2[3:0]
TRUTH[7:0]
OUTEN
FILTSEL[1:0]
CLKSRC[2:0]
ENABLE
INSEL1[3:0]
INSEL0[3:0]
INSEL2[3:0]
TRUTH[7:0]
OUTEN
FILTSEL[1:0]
CLKSRC[2:0]
ENABLE
INSEL1[3:0]
INSEL0[3:0]
INSEL2[3:0]
TRUTH[7:0]
OUTEN
FILTSEL[1:0]
CLKSRC[2:0]
ENABLE
INSEL1[3:0]
INSEL0[3:0]
INSEL2[3:0]
TRUTH[7:0]
OUTEN
FILTSEL[1:0]
CLKSRC[2:0]
ENABLE
INSEL1[3:0]
INSEL0[3:0]
INSEL2[3:0]
TRUTH[7:0]
Register Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 469
AVR128DB28/32/48/64
CCL – Configurable Custom Logic
31.5.1
Control A
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
CTRLA
0x00
0x00
-
6
RUNSTDBY
R/W
0
5
4
3
2
1
0
ENABLE
R/W
0
Bit 6 – RUNSTDBY Run in Standby
Writing this bit to ‘1’ will enable the peripheral to run in Standby Sleep mode.
Value
Description
0
The CCL will not run in Standby Sleep mode
1
The CCL will run in Standby Sleep mode
Bit 0 – ENABLE Enable
Value
Description
0
The peripheral is disabled
1
The peripheral is enabled
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 470
AVR128DB28/32/48/64
CCL – Configurable Custom Logic
31.5.2
Sequencer Control 0
Name:
Offset:
Reset:
Property:
Bit
7
SEQCTRL0
0x01
0x00
Enable-Protected
6
Access
Reset
5
4
3
R/W
0
2
1
SEQSEL0[3:0]
R/W
R/W
0
0
0
R/W
0
Bits 3:0 – SEQSEL0[3:0] Sequencer Selection
This bit group selects the sequencer configuration for LUT0 and LUT1.
Value
Name
Description
0x0
DISABLE
The sequencer is disabled
0x1
DFF
D flip-flop
0x2
JK
JK flip-flop
0x3
LATCH
D latch
0x4
RS
RS latch
Other
Reserved
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 471
AVR128DB28/32/48/64
CCL – Configurable Custom Logic
31.5.3
Sequencer Control 1
Name:
Offset:
Reset:
Property:
Bit
7
SEQCTRL1
0x02
0x00
Enable-Protected
6
Access
Reset
5
4
3
R/W
0
2
1
SEQSEL1[3:0]
R/W
R/W
0
0
0
R/W
0
Bits 3:0 – SEQSEL1[3:0] Sequencer Selection
This bit group selects the sequencer configuration for LUT2 and LUT3.
Value
Name
Description
0x0
DISABLE
The sequencer is disabled
0x1
DFF
D flip-flop
0x2
JK
JK flip-flop
0x3
LATCH
D latch
0x4
RS
RS latch
Other
Reserved
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 472
AVR128DB28/32/48/64
CCL – Configurable Custom Logic
31.5.4
Sequencer Control 2
Name:
Offset:
Reset:
Property:
Bit
7
SEQCTRL2
0x03
0x00
Enable-Protected
6
Access
Reset
5
4
3
R/W
0
2
1
SEQSEL2[3:0]
R/W
R/W
0
0
0
R/W
0
Bits 3:0 – SEQSEL2[3:0] Sequencer Selection
This bit group selects the sequencer configuration for LUT4 and LUT5.
Value
Name
Description
0x0
DISABLE
The sequencer is disabled
0x1
DFF
D flip-flop
0x2
JK
JK flip-flop
0x3
LATCH
D latch
0x4
RS
RS latch
Other
Reserved
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 473
AVR128DB28/32/48/64
CCL – Configurable Custom Logic
31.5.5
Interrupt Control 0
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
INTCTRL0
0x05
0x00
-
7
6
INTMODE3[1:0]
R/W
R/W
0
0
5
4
INTMODE2[1:0]
R/W
R/W
0
0
3
2
INTMODE1[1:0]
R/W
R/W
0
0
1
0
INTMODE0[1:0]
R/W
R/W
0
0
Bits 0:1, 2:3, 4:5, 6:7 – INTMODE
The bits in INTMODEn select the interrupt sense configuration for LUTn-OUT.
Value
Name
Description
0x0
INTDISABLE
Interrupt disabled
0x1
RISING
Sense rising edge
0x2
FALLING
Sense falling edge
0x3
BOTH
Sense both edges
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 474
AVR128DB28/32/48/64
CCL – Configurable Custom Logic
31.5.6
Interrupt Control 1
Name:
Offset:
Reset:
Property:
Bit
7
INTCTRL1
0x06
0x00
6
Access
Reset
5
4
3
2
INTMODE5[1:0]
R/W
R/W
0
0
1
0
INTMODE4[1:0]
R/W
R/W
0
0
Bits 0:1, 2:3 – INTMODE
The bits in INTMODEn select the interrupt sense configuration for LUTn-OUT.
Value
Name
Description
0x0
INTDISABLE
Interrupt disabled
0x1
RISING
Sense rising edge
0x2
FALLING
Sense falling edge
0x3
BOTH
Sense both edges
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 475
AVR128DB28/32/48/64
CCL – Configurable Custom Logic
31.5.7
Interrupt Flag
Name:
Offset:
Reset:
Property:
Bit
7
INTFLAGS
0x07
0x00
-
6
Access
Reset
5
INT5
R/W
0
4
INT4
R/W
0
3
INT3
R/W
0
2
INT2
R/W
0
1
INT1
R/W
0
0
INT0
R/W
0
Bits 0, 1, 2, 3, 4, 5 – INT Interrupt Flag
The INTn flag is set when the LUTn output change matches the Interrupt Sense mode as defined in CCL.INTCTRLn.
Writing a ‘1’ to this flag’s bit location will clear the flag.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 476
AVR128DB28/32/48/64
CCL – Configurable Custom Logic
31.5.8
LUT n Control A
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
EDGEDET
R/W
0
LUTnCTRLA
0x08 + n*0x04 [n=0..5]
0x00
Enable-Protected
6
OUTEN
R/W
0
5
4
FILTSEL[1:0]
R/W
R/W
0
0
3
R/W
0
2
CLKSRC[2:0]
R/W
0
1
R/W
0
0
ENABLE
R/W
0
Bit 7 – EDGEDET Edge Detection
Value
Description
0
Edge detector is disabled
1
Edge detector is enabled
Bit 6 – OUTEN Output Enable
This bit enables the LUT output to the LUTn OUT pin. When written to ‘1’, the pin configuration of the PORT I/OController is overridden.
Value
Description
0
Output to pin disabled
1
Output to pin enabled
Bits 5:4 – FILTSEL[1:0] Filter Selection
These bits select the LUT output filter options.
Value
Name
0x0
DISABLE
0x1
SYNCH
0x2
FILTER
0x3
-
Description
Filter disabled
Synchronizer enabled
Filter enabled
Reserved
Bits 3:1 – CLKSRC[2:0] Clock Source Selection
This bit selects between various clock sources to be used as the clock (CLK_LUTn) for a LUT.
The CLK_LUTn of the even LUT is used for clocking the sequencer of a LUT pair.
Value
Input Source
Description
0x0
0x1
0x2
0x3
0x4
0x5
0x6
0x07
CLKPER
IN2
OSCHF
OSC32K
OSC1K
-
CLK_PER is clocking the LUT
IN2 is clocking the LUT
Reserved
Reserved
Internal high-frequency oscillator before prescaler is clocking LUT
Internal 32.786 kHz oscillator
Internal 32.768 kHz oscillator divided by 32
Reserved
Bit 0 – ENABLE LUT Enable
Value
Description
0
The LUT is disabled
1
The LUT is enabled
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 477
AVR128DB28/32/48/64
CCL – Configurable Custom Logic
31.5.9
LUT n Control B
Name:
Offset:
Reset:
Property:
LUTnCTRLB
0x09 + n*0x04 [n=0..5]
0x00
Enable-Protected
Notes:
1. SPI connections to the CCL work in master SPI mode only.
2. USART connections to the CCL work only when the USART is in one of the following modes:
– Asynchronous USART
– Synchronous USART master
Bit
7
Access
Reset
R/W
0
6
5
INSEL1[3:0]
R/W
R/W
0
0
4
3
R/W
0
R/W
0
2
1
INSEL0[3:0]
R/W
R/W
0
0
0
R/W
0
Bits 7:4 – INSEL1[3:0] LUT n Input 1 Source Selection
These bits select the source for input 1 of LUT n.
Value
Name
Description
0x0
0x1
0x2
0x3
0x4
0x5
0x6
0x7
0x8
0x9
0xA
0xB
0xC
0xD
Other
MASK
FEEDBACK
LINK
EVENTA
EVENTB
IN1
AC1
ZCD1
USART1
SPI0
TCA0
TCA1
TCB1
TCD0
-
Masked input
Feedback input
Output from LUT[n+1] as input source
Event A as input source
Event B as input source
LUTn-IN1 as input source
AC1 OUT as input source
ZCD1 OUT as input source
USART1 TXD as input source
SPI0 MOSI as input source
TCA0 WO1 as input source
TCA1 WO1 as input source
TCB1 WO as input source
TCD0 WOB as input source
Reserved
Bits 3:0 – INSEL0[3:0] LUT n Input 0 Source Selection
These bits select the source for input 0 of LUT n.
Value
Name
Description
0x0
0x1
0x2
0x3
0x4
0x5
0x6
0x7
0x8
0x9
0xA
0xB
0xC
0xD
MASK
FEEDBACK
LINK
EVENTA
EVENTB
IN0
AC0
ZCD0
USART0
SPI0
TCA0
TCA1
TCB0
TCD0
Masked input
Feedback input
Output from LUT[n+1] as input source
Event A as input source
Event B as input source
LUTn-IN0 as input source
AC0 OUT as input source
ZCD0 OUT as input source
USART0 TXD as input source
SPI0 MOSI as input source
TCA0 WO0 as input source
TCA1 WO0 as input source
TCB0 WO as input source
TCD0 WOA as input source
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 478
AVR128DB28/32/48/64
CCL – Configurable Custom Logic
...........continued
Value
Name
Description
Other
-
Reserved
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 479
AVR128DB28/32/48/64
CCL – Configurable Custom Logic
31.5.10 LUT n Control C
Name:
Offset:
Reset:
Property:
Bit
7
LUTnCTRLC
0x0A + n*0x04 [n=0..5]
0x00
Enable-Protected
6
5
4
Access
Reset
3
R/W
0
2
1
INSEL2[3:0]
R/W
R/W
0
0
0
R/W
0
Bits 3:0 – INSEL2[3:0] LUT n Input 2 Source Selection
These bits select the source for input 2 of LUT n.
Value
Name
Description
0x0
0x1
0x2
0x3
0x4
0x5
0x6
0x7
0x8
0x9
0xA
0xB
0xC
0xD
Other
MASK
FEEDBACK
LINK
EVENTA
EVENTB
IN2
AC2
ZCD2
USART2
SPI0
TCA0
TCA1
TCB2
TCD0
-
Masked input
Feedback input
Output from LUT[n+1] as input source
Event A as input source
Event B as input source
LUTn-IN2 as input source
AC2 OUT as input source
ZCD2 OUT as input source
USART2 TXD as input source
SPI0 SCK as input source
TCA0 WO2 as input source
TCA1 WO2 as input source
TCB2 WO as input source
TCD0 WOC as input source
Reserved
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 480
AVR128DB28/32/48/64
CCL – Configurable Custom Logic
31.5.11 TRUTHn
Name:
Offset:
Reset:
Property:
Bit
7
TRUTHn
0x0B + n*0x04 [n=0..5]
0x00
Enable-Protected
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
TRUTH[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – TRUTH[7:0] Truth Table
These bits define the value of truth logic as a function of inputs LUTn-TRUTHSEL[2:0]. See also section Truth Table
Logic.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 481
AVR128DB28/32/48/64
AC - Analog Comparator
32.
AC - Analog Comparator
32.1
Features
•
•
•
•
•
•
•
•
32.2
Selectable Response Time
Selectable Hysteresis
Analog Comparator Output Available on Pin
Comparator Output Inversion Available
Flexible Input Selection:
– Four Positive pins
– Three Negative pins
– Internal reference voltage generator (DACREF)
Interrupt Generation on:
– Rising edge
– Falling edge
– Both edges
Window Function Interrupt Generation on:
– Signal above window
– Signal inside window
– Signal below window
– Signal outside window
Event Generation:
– Comparator output
– Window function
Overview
The analog comparator (AC) compares the voltage levels on two inputs and gives a digital output based on this
comparison. The AC can be configured to generate interrupt requests and/or events based on several different
combinations of input change.
The input selection includes analog port pins and internally generated inputs. The AC digital output goes through
controller logic, enabling customization of the signal for use internally with the Event System or externally on the pin.
The dynamic behavior of the AC can be adjusted by a hysteresis feature. The hysteresis can be customized to
optimize the operation for each application.
The individual comparators can be used independently (Normal mode) or paired to form a window comparison
(Window mode).
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 482
AVR128DB28/32/48/64
AC - Analog Comparator
32.2.1
Block Diagram
Figure 32-1. Analog Comparator
AINP0
From ACn
.
.
.
+
AINPn
-
Voltage
divider
Event out
CMP
(Int. req.)
Invert
VREF
.
..
Controller
logic
Hysteresis
AINNn
AC
Enable
AINN0
OUT
CTRLA
DACREF
32.2.2
MUXCTRL
Signal Description
Signal
Description
Type
AINNn
Negative input n
Analog
AINPn
Positive input n
Analog
OUT
Comparator output of AC
Digital
32.3
Functional Description
32.3.1
Initialization
For basic operation, follow these steps:
1. Configure the desired input pins in the port peripheral as analog inputs.
2. Select the positive and negative input sources by writing to the Positive and Negative Input MUX Selection
(MUXPOS and MUXNEG) bit fields in the MUX Control (ACn.MUXCTRL) register.
3. Optional: Enable the output to pin by writing a ‘1’ to the Output Pad Enable (OUTEN) bit in the Control A
(ACn.CTRLA) register.
4. Enable the AC by writing a ‘1’ to the ENABLE bit in ACn.CTRLA.
During the start-up time after enabling the AC, the INITVAL bit in the CTRLB register can be used to set the AC
output before the AC is ready. If VREF is used as a reference source, the respective start-up time of the reference
source must be added. For details about the start-up time of the AC and VREF peripherals, refer to the Electrical
Characteristics section.
To avoid the pin being tri-stated when the AC is disabled, the OUT pin must be configured as output.
32.3.2
Operation
32.3.2.1 Input Hysteresis
Applying an input hysteresis helps to prevent constant toggling of the output when the noise-afflicted input signals are
close to each other.
The input hysteresis can either be disabled or have one of three levels. The hysteresis is configured by writing to the
Hysteresis Mode Select (HYSMODE) bit field in the Control A (ACn.CTRLA) register. For details about typical values
of hysteresis levels, refer to the Electrical Characteristics section.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 483
AVR128DB28/32/48/64
AC - Analog Comparator
32.3.2.2 Input and Reference Selection
The input selection to the ACn is controlled by the Positive and Negative Multiplexers (MUXPOS and MUXNEG) bit
fields in the MUX Control (ACn.MUXCTRL) register. For positive input of ACn, an analog pin can be selected, while
for negative input, the selection can be made between analog pins and internal DAC reference voltage (DACREF).
For details about the possible selections, refer to the MUX Control (ACn.MUXCTRL) register description.
The generated voltage depends on the DACREF register value and the reference voltage selected in the VREF
module, and is calculated as:
�DACREF =
DACREF
× �REF
256
The internal reference voltages (VREF), except for VREFA and VDD, are generated from an internal band gap
reference.
After switching inputs to I/O pins or setting a new voltage reference, the ACn requires time to settle. Refer to the
Electrical Characteristics section for more details.
32.3.2.3 Normal Mode
The AC has one positive input and one negative input. The output of the comparator is ‘1’ when the difference
between the positive and the negative input voltage is positive, and ‘0’ otherwise. This output is available on the
output pin (OUT) through a logic XOR gate. This allows the inversion of the OUT pin when the INVERT bit in the
MUX Control (ACn.MUXCTRL) register is ‘1’.
To avoid random output and set a specific level on the OUT pin during the ACn initialization, the INITVAL bit in the
same register is used.
32.3.2.4 Power Modes
For power sensitive applications, the AC provides multiple power modes with balance power consumption and
response time. A mode is selected by writing to the Power Profile (POWER) bit field in the Control A (ACn.CTRLA)
register.
32.3.2.5 Window Mode
Each AC (i.e., ACx) can be configured to work together with another comparator (i.e., ACy) in Window mode. In this
mode, a voltage range (the window) is defined, and the selected comparator indicates whether an input signal is
within this range or not.
The WINSEL bit field in the Control B (ACn.CTRLB) register selects which ACy instance is connected to the current
comparator (ACx) to create the window comparator. The user is responsible for configuring the MUXPOS and
MUXNEG bit fields in the MUX Control (ACn.MUXCTRL) register for ACx and ACy, so they match the setup in the
figure below. Note that the MUXPOS bit field in the ACn.MUXCTRL register of both ACs must be configured to the
same pin.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 484
AVR128DB28/32/48/64
AC - Analog Comparator
Figure 32-2. Analog Comparators in Window Mode
ACx
Common input
+
-
Window limit 1
WINSTATE[1:0]
CMPSTATE
CMPIF (IRQ)
Controller
logic
ACy
+
Controller
logic
-
Window limit 2
The status of the input signal is reported by the Window State (WINSTATE) flags in the Status (ACn.STATUS)
register. The status can be:
•
•
•
Above window - the input signal is above the upper limit.
Inside window - the input signal is between the lower and upper limit.
Below window - the input signal is below the lower limit.
Writing to the INTMODE bit field in the Interrupt Control (INTCTRL) register selects one of these window modes for
triggering an event or requesting an interrupt:
• Above window - the interrupt/event is issued when the input signal is above the upper limit.
• Inside window - the interrupt/event is issued when the input signal is between the lower and upper limit.
• Below window - the interrupt/event is issued when the input signal is below the lower limit.
• Outside window - the interrupt/event is issued when the input signal is not between the lower and upper limit.
The CMPSTATE bit is ‘1’ when the Window state matches the selected Interrupt Mode (INTMODE) bit field, and ‘0’
otherwise.
The window interrupt is enabled by writing a ‘1’ to the Analog Comparator Interrupt Enable (CMP) bit in the Interrupt
Control (ACn.INTCTRL) register.
32.3.3
Events
The AC can generate the following events:
Table 32-1. Event Generators in AC
Generator Name
Module
Event
ACn
OUT
Description
Comparator
output level
Event Type
Level
Generating
Clock Domain
Asynchronous
Length of Event
Given by AC output level
The AC has no event users.
Refer to the Event System (EVSYS) section for more details regarding event types and Event System configuration.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 485
AVR128DB28/32/48/64
AC - Analog Comparator
32.3.4
Interrupts
Table 32-2. Available Interrupt Vectors and Sources
Name
Vector Description
Conditions
CMP
Analog comparator interrupt
AC output is toggling as configured by INTMODE in ACn.INTCTRL
When an interrupt condition occurs, the corresponding interrupt flag is set in the Status (ACn.STATUS) register.
An interrupt source is enabled or disabled by writing to the corresponding bit in the peripheral’s Interrupt Control
(ACn.INTCTRL) register.
The AC can generate a comparator interrupt, CMP, and can request this interrupt on either rising, falling, or both
edges of the toggling comparator output. This is configured by writing to the Interrupt Mode (INTMODE) bit field in the
Interrupt Control (ACn.INTCTRL) register. The interrupt is enabled by writing a ‘1’ to the Analog Comparator Interrupt
Enable (CMP) bit in the Interrupt Control (ACn.INTCTRL) register. The interrupt request remains active until the
interrupt flag is cleared. Refer to the Status (ACn.STATUS) register description for details on how to clear the
interrupt flags.
32.3.5
Sleep Mode Operation
In Idle Sleep mode the AC will continue to operate as normal.
In Standby Sleep mode the AC is disabled by default. If the Run in Standby Mode (RUNSTDBY) bit in the Control A
(ACn.CTRLA) register is written to ‘1’, the AC will continue to operate as normal with an event, interrupt and AC
output on the pin even if the CLK_PER is not running in Standby Sleep mode.
In Power-Down Sleep mode the AC and the output to the pad are disabled.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 486
AVR128DB28/32/48/64
AC - Analog Comparator
32.4
Register Summary
Offset
Name
Bit Pos.
7
6
0x00
0x01
0x02
0x03
...
0x04
0x05
0x06
0x07
CTRLA
CTRLB
MUXCTRL
7:0
7:0
7:0
RUNSTDBY
OUTEN
INVERT
INITVAL
32.5
5
4
3
POWER[1:0]
MUXPOS[2:0]
2
1
0
HYSMODE[1:0]
ENABLE
WINSEL[1:0]
MUXNEG[2:0]
Reserved
DACREF
INTCTRL
STATUS
7:0
7:0
7:0
WINSTATE[1:0]
DACREF[7:0]
INTMODE[1:0]
CMPSTATE
CMP
CMPIF
Register Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 487
AVR128DB28/32/48/64
AC - Analog Comparator
32.5.1
Control A
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
RUNSTDBY
R/W
0
CTRLA
0x00
0x00
-
6
OUTEN
R/W
0
5
4
3
POWER[1:0]
R/W
R/W
0
0
2
1
HYSMODE[1:0]
R/W
R/W
0
0
0
ENABLE
R/W
0
Bit 7 – RUNSTDBY Run in Standby Mode
Writing this bit to ‘1’ allows the AC to continue operation in Standby Sleep mode. Since the clock is stopped,
interrupts and status flags are not updated.
Value
Description
0
In Standby Sleep mode, the peripheral is halted
1
In Standby Sleep mode, the peripheral continues operation
Bit 6 – OUTEN Output Pad Enable
Writing this bit to ‘1’ makes the OUT signal available on the pin.
Bits 4:3 – POWER[1:0] Power Profile
This setting controls the current through the comparator, which allows the AC to trade power consumption for the
response time. Refer to the Electrical Characteristics section for power consumption and response time.
Value
Name
Description
0x0
0x1
0x2
0x3
PROFILE0
PROFILE1
PROFILE2
-
Power profile 0. Shortest response time and highest consumption.
Power profile 1
Power profile 2
Reserved
Bits 2:1 – HYSMODE[1:0] Hysteresis Mode Select
Writing to this bit field selects the Hysteresis mode for the AC input. For details about typical values of hysteresis
levels, refer to the Electrical Characteristics section.
Value
Name
Description
0x0
NONE
No hysteresis
0x1
SMALL
Small hysteresis
0x2
MEDIUM
Medium hysteresis
0x3
LARGE
Large hysteresis
Bit 0 – ENABLE Enable AC
Writing this bit to ‘1’ enables the AC.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 488
AVR128DB28/32/48/64
AC - Analog Comparator
32.5.2
Control B
Name:
Offset:
Reset:
Property:
Bit
7
CTRLB
0x01
0x00
-
6
5
4
3
2
Access
Reset
1
0
WINSEL[1:0]
R/W
R/W
0
0
Bits 1:0 – WINSEL[1:0] Window Selection Mode
This bit field selects the AC connected to the current comparator in Window mode.
Value
Name
Description
0x0
DISABLED
Window function disabled
0x1
UPSEL1
Windows enabled, with ACn+1 connected
0x2
UPSEL2
Windows enabled, with ACn+2 connected
0x3
Reserved
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 489
AVR128DB28/32/48/64
AC - Analog Comparator
32.5.3
MUX Control
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
INVERT
R/W
0
MUXCTRL
0x02
0x00
-
6
INITVAL
R/W
0
5
R/W
0
4
MUXPOS[2:0]
R/W
0
3
2
R/W
0
R/W
0
1
MUXNEG[2:0]
R/W
0
0
R/W
0
Bit 7 – INVERT Invert AC Output
Writing this bit to ‘1’ enables inversion of the output of the AC. This inversion has to be taken into account when using
the AC output signal as an input signal to other peripherals or parts of the system.
Bit 6 – INITVAL AC Output Initial Value
To avoid that the AC output toggles before the comparator is ready, the INITVAL can be used to set the initial state of
the comparator output.
Value
Name
Description
0x0
LOW
Output initialized to ‘0’
0x1
HIGH
Output initialized to ‘1’
Bits 5:3 – MUXPOS[2:0] Positive Input MUX Selection
Writing to this bit field selects the input signal to the positive input of the AC.
Value
Name
Description
0x0
0x1
0x2
0x3
Other
AINP0
AINP1
AINP2
AINP3
-
Positive pin 0
Positive pin 1
Positive pin 2
Positive pin 3
Reserved
Bits 2:0 – MUXNEG[2:0] Negative Input MUX Selection
Writing to this bit field selects the input signal to the negative input of the AC.
Value
Name
Description
0x0
0x1
0x2
0x3
Other
AINN0
AINN1
AINN2
DACREF
-
Negative pin 0
Negative pin 1
Negative pin 2
DAC Reference
Reserved
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 490
AVR128DB28/32/48/64
AC - Analog Comparator
32.5.4
DAC Voltage Reference
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
DACREF
0x05
0xFF
R/W
7
6
5
R/W
1
R/W
1
R/W
1
4
3
DACREF[7:0]
R/W
R/W
1
1
2
1
0
R/W
1
R/W
1
R/W
1
Bits 7:0 – DACREF[7:0] DACREF Data Value
This bit field defines the output voltage from the internal voltage divider. The DAC voltage reference depends on the
DACREF value and the reference voltage selected in the VREF module, and is calculated as:
DACREF 7: 0
�DACREF =
× �REF
256
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 491
AVR128DB28/32/48/64
AC - Analog Comparator
32.5.5
Interrupt Control
Name:
Offset:
Reset:
Property:
Bit
INTCTRL
0x06
0x00
-
7
6
Access
Reset
5
4
INTMODE[1:0]
R/W
R/W
0
0
3
2
1
0
CMP
R/W
0
Bits 5:4 – INTMODE[1:0] Interrupt Mode
Writing to this bit field selects which edge(s) of the AC output or when entering a window state triggers an interrupt
request.
Table 32-3. Interrupt Generation in Window Mode
Value
Name
Description
0x0
0x1
0x2
0x3
ABOVE
INSIDE
BELOW
OUTSIDE
Enables Window mode above interrupt
Enables Window mode inside interrupt
Enables Window mode below interrupt
Enables Window mode outside interrupt
Table 32-4. Interrupt Generation with Single Comparator
Value
Name
Description
0x0
0x1
0x2
0x3
BOTHEDGE
NEGEDGE
POSEDGE
Positive and negative inputs crosses
Reserved
Positive input goes above negative input
Positive input goes below negative input
Bit 0 – CMP AC Interrupt Enable
This bit enables the AC interrupt. The enabled interrupt will be triggered when the CMPIF bit in the ACn.STATUS
register is set.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 492
AVR128DB28/32/48/64
AC - Analog Comparator
32.5.6
Status
Name:
Offset:
Reset:
Property:
Bit
STATUS
0x07
0x00
-
7
6
WINSTATE[1:0]
R
R
0
0
Access
Reset
5
4
CMPSTATE
R
0
3
2
1
0
CMPIF
R/W
0
Bits 7:6 – WINSTATE[1:0] Window State
When the window function is enabled, these flags indicate the current status of the input signal with respect to the
window.
Not valid when the Window mode is disabled.
Table 32-5. Window State Settings
Value
Name
Description
0x0
0x1
0x2
Other
ABOVE
INSIDE
BELOW
-
Above window
Inside window
Below window
Reserved
Bit 4 – CMPSTATE AC State
If this bit is ‘1’, the OUT signal is high. If this bit is ‘0’, the OUT signal is low. In Window mode, if this bit is ‘1’, the
Window state matches the selected Interrupt mode (INTMODE) bit field. If INTMODE is ‘OUTSIDE’, both ‘ABOVE’ and
‘BELOW’ are valid matches. It will have a synchronizer delay to get updated in the I/O register (three cycles).
Bit 0 – CMPIF AC Interrupt Flag
This bit is ‘1’ when the OUT signal matches the Interrupt Mode (INTMODE) bit field as defined in the ACn.INTCTRL
register. Writing a ‘1’ to this flag bit location will clear the flag.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 493
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
33.
ADC - Analog-to-Digital Converter
33.1
Features
•
•
•
•
•
•
•
•
•
•
•
•
•
33.2
12-Bit Resolution
Up to 130 ksps at 12-Bit Resolution
Differential and Single-Ended Conversion
Up to 22 Inputs
Rail-to-Rail Input Voltage Range
Free-Running and Single Conversion
Accumulation of Up to 128 Samples per Conversion
Multiple Voltage Reference Options
Temperature Sensor Input Channel
Programmable Input Sampling Duration
Configurable Threshold and Window Comparator
Event Triggered Conversion
Interrupt and Event on Conversion Complete
Overview
The Analog-to-Digital Converter (ADC) is a 12-bit Successive Approximation Register (SAR) ADC, with a sampling
rate up to 130 ksps at 12-bit resolution. The ADC is connected to an analog input multiplexer for selection between
multiple single-ended or differential inputs. In single-ended conversions, the ADC measures the voltage between the
selected input and 0V (GND). In differential conversions, the ADC measures the voltage between two selected input
channels. The selected ADC input channels can either be internal (e.g., a voltage reference) or external analog input
pins.
An ADC conversion can be started by software, or by using the Event System (EVSYS) to route an event from other
peripherals. This makes it possible to do a periodic sampling of input signals, trigger an ADC conversion on a special
condition or trigger an ADC conversion in Standby sleep mode.
A digital window compare feature is available for monitoring the input signal and can be configured only to trigger an
interrupt if the sample is below or above a user-defined threshold, or inside or outside a user-defined window, with
minimum software intervention required.
The ADC input signal is fed through a sample-and-hold circuit which ensures that the input voltage to the ADC is held
at a constant level during sampling.
The ADC supports sampling in bursts where a configurable number of conversions are accumulated into a single
ADC result (Sample Accumulation). Furthermore, a sample delay can be configured to tune the ADC burst sampling
frequency away from any harmonic noise aliased from the sampled signal.
The ADC voltage reference is configured in the Voltage Reference (VREF) peripheral and can use one of the
following sources as voltage reference:
• Multiple Internally Generated Voltages
• AVDD Supply Voltage
• External VREF Pin (VREFA)
This device has one instance of the ADC peripheral: ADC0.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 494
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
33.2.1
Block Diagram
Figure 33-1. Block Diagram
MUXPOS
VREF
AVDD
VREFA
Internal Reference
AIN0
AIN1
VADCREF
.
.
.
AINn
VAINP
Internal
Inputs
ADC
AIN0
AIN1
VAINN
.
.
.
Result
formatting
AINn
Internal
Inputs
MUXNEG
33.2.2
RES
Control Logic
>
<
CTRLA
EVCTRL
COMMAND
WINLT
WINHT
Result ready
(IRQ )
Window compare
(IRQ )
Signal Description
Pin Name
Type
Description
AIN[n:0]
Analog input
Analog input to be converted
VREFA
Analog input
External voltage reference pin
33.3
Functional Description
33.3.1
Definitions
•
•
•
33.3.2
ACC
Conversion: The operation in which analog values on the selected ADC inputs are transformed into a digital
representation.
Sample: The output of a single ADC conversion.
Result: The value placed in the Result (ADCn.RES) register. Depending on the ADC configuration, this value is
a single sample or the sum of multiple accumulated samples.
Initialization
The following steps are recommended to initialize ADC operation:
1. Configure the ADC voltage reference in the Voltage Reference (VREF) peripheral.
2. Optional: Select between Single-Ended or Differential mode by writing to the Conversion Mode (CONVMODE)
bit in the Control A (ADCn.CTRLA) register.
3. Configure the resolution by writing to the Resolution Selection (RESSEL) bit field in the ADCn.CTRLA register.
4. Optional: Configure to left adjust by writing a ‘1’ to the Left Adjust Result (LEFTADJ) bit in the ADCn.CTRLA
register.
5. Optional: Select the Free-Running mode by writing a ‘1’ to the Free-Running (FREERUN) bit in the
ADCn.CTRLA register.
6. Optional: Configure the number of samples to be accumulated per conversion by writing to the Sample
Accumulation Number Select (SAMPNUM) bit field in the Control B (ADCn.CTRLB) register.
7. Configure the ADC clock (CLK_ADC) by writing to the Prescaler (PRESC) bit field in the Control C
(ADCn.CTRLC) register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 495
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
8. Select the positive ADC input by writing to the MUXPOS bit field in the ADCn.MUXPOS register.
9. Optional: Select the negative ADC input by writing to the MUXNEG bit field in the ADCn.MUXNEG register.
10. Optional: Enable Start Event input by writing a ‘1’ to the Start Event Input (STARTEI) bit in the Event Control
(ADCn.EVCTRL) register, and configure the Event System accordingly.
11. Enable the ADC by writing a ‘1’ to the ADC Enable (ENABLE) bit in the ADCn.CTRLA register.
Following these steps will initialize the ADC for basic measurements.
For details about the start-up time of the VREF peripheral, refer to the Electrical Characteristics section.
The ADC does not consume power when the ENABLE bit is ‘0’. The ADC generates a 10- or 12-bit result which can
be read from the Result (ADCn.RES) register.
Notes: Changing the following registers during a conversion will give unpredictable results:
• In ADCn.CTRLA:
– Conversion Mode (CONVMODE) bit
– Left Adjust Result (LEFTADJ) bit
– Resolution Selection (RESSEL) bit field
• In ADCn.CTRLB:
– Sample Accumulation Number Select (SAMPNUM) bit field
• In ADCn.CTRLC:
– Prescaler (PRESC) bit field
33.3.3
Operation
33.3.3.1 Operation Modes
The ADC supports differential and single-ended conversions. This is configured in the CONVMODE bit in the
ADCn.CTRLA register.
The operation modes can be split into two groups:
• Single conversion of one sample per trigger
• Accumulated conversion of n conventions per trigger, the result is accumulated
The accumulated conversion utilizes 12-bit conversions and can be configured with or without truncation of the
accumulated result. The accumulator is always reset to zero when a new accumulated conversion is started.
33.3.3.2 Starting a Conversion
Once the initialization is finished, a conversion is started by writing a ‘1’ to the ADC Start Conversion (STCONV) bit in
the Command (ADCn.COMMAND) register. This bit is ‘1’ as long as the conversion is in progress. The STCONV bit
will be set during a conversion and cleared once the conversion is complete.
If a different input channel is selected while a conversion is in progress, the ADC will finish the current conversion
before changing the channel.
Depending on the accumulator setting, the conversion result is a single sample, or an accumulation of samples. Once
the triggered operation is finished, the Result Ready (RESRDY) flag in the Interrupt Flags (ADCn.INTFLAGS) register
is set. The corresponding interrupt vector is executed if the Result Ready Interrupt Enable (RESRDY) bit in the
Interrupt Control (ADCn.INTCTRL) register is ‘1’ and the Global Interrupt Enable bit is ‘1’.
The RESRDY interrupt flag in the ADCn.INTFLAGS register will be set even if the specific interrupt is disabled,
allowing software to check for any finished conversion by polling the flag. A conversion can thus be triggered without
causing an interrupt upon completion.
Alternatively, a conversion can be triggered by an event. This is enabled by writing a ‘1’ to the Start Event Input
(STARTEI) bit in the Event Control (ADCn.EVCTRL) register. Any incoming event routed to the ADC through the
Event System (EVSYS) will trigger an ADC conversion. This provides a method to start conversions with predictable
intervals or at specific conditions.
The ADC will trigger a conversion on the rising edge of an event signal. When an event occurs, the STCONV bit in
the ADCn.COMMAND register is set and it will be cleared when the conversion is complete. Refer to Figure 33-2.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 496
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
Figure 33-2. ADC Event Trigger Logic
PRESCALER
CLK_ADC
START
STARTEI
CONVERSION
LOGIC
PERIPHERAL EVENT
STCONV
In Free-Running mode, the first conversion is started by writing a ‘1’ to the STCONV bit in the ADCn.COMMAND
register. A new conversion cycle is started immediately after the previous conversion cycle has completed. A
completed conversion will set the RESRDY flag in the ADCn.INTFLAGS register.
33.3.3.3 Clock Generation
The ADC peripheral contains a prescaler which generates the ADC clock (CLK_ADC) from the peripheral clock
(CLK_PER). The minimum ADC_CLK frequency is 125 kHz. The prescaling is selected by writing to the Prescaler
(PRESC) bit field in the Control C (ADCn.CTRLC) register. The prescaler begins counting from the moment the ADC
conversion starts and is reset for every new conversion. Refer to Figure 33-3.
Figure 33-3. ADC Prescaler
CTRLC
Prescaler
CLK_PER/256
CLK_PER/2
CLK_PER
Reset
"Start"
CLK_PER/4
ENABLE
PRESC
ADC clock source
(CLK_ADC)
When initiating a conversion by writing a ‘1’ to the Start Conversion (STCONV) bit in the ADCn.COMMAND register
or from event, the conversion starts after one CLK_PER cycle. The prescaler is kept in Reset, as long as there is no
ongoing conversion. This assures a fixed delay from the trigger to the actual start of a conversion of maximum 2
CLK_PER cycles.
33.3.3.4 Conversion Timing
A normal conversion takes place in the following order:
1. Write a ‘1’ to the STCONV bit in the Command (ADCn.COMMAND) register.
2.
3.
4.
5.
Start-up for maximum 2 CLK_PER cycles.
Sample-and-hold for 2 CLK_ADC cycles.
Conversion for 13.5 CLK_ADC cycles.
Result formatting for 2 CLK_PER cycles.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 497
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
When a conversion is complete, the result is available in the Result (ADCn.RES) register, and the Result Ready
(RESRDY) interrupt flag is set in the Interrupt Flags (ADCn.INTFLAGS) register.
33.3.3.4.1 Single Conversion
The figure below shows the timing diagram for a single 12-bit ADC conversion.
Figure 33-4. Timing Diagram - Single Conversion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
CLK_ADC
STCONV
State
Idle
Converting
Sampling
F
Result
RESRDY
(Event)
RESRDY
(Int Req)
Start-up
(2 CLK_PER
cycles)
Sampling
(2 CLK_ADC
cycles)
Conversion
(13.5 CLK_ADC
cycles)
Result formatting
(2 CLK_PER
cycles)
For a single conversion, the total conversion time is calculated by:
Total Conversion Time (12-bit) =
13.5 + 2
4
+
�CLK_ADC �CLK_PER
33.3.3.4.2 Accumulated Conversion
The figure below shows the timing diagram for the ADC when accumulating two samples in Accumulation mode.
Figure 33-5. Timing Diagram - Accumulated Conversion
CLK_ADC
STCONV
State
Idle
Sampling
Converting
F
Sampling
Converting
F
Result
RESRDY
(Event)
RESRDY
(Int Req)
Start-up
(2 CLK_PER cycles)
Second conversion in accumulation
First conversion in accumulation
The number of samples to accumulate is configured with the Sample Number (SAMPNUM) bit field in the Control B
(ADCn.CTRLB) register. The STCONV bit is set for the entire conversion. The total conversion time for n samples is
given by:
Total Conversion Time (12-bit) =
2
13.5 + 2
2
+�
+
�CLK_PER
�CLK_ADC �CLK_PER
33.3.3.4.3 Free-Running Conversion
In Free-Running mode, a new conversion is started as soon as the previous conversion has completed. This is
signaled by the RESRDY bit in the Interrupt Flags (ADCn.INTFLAGS) register.
The figure below shows the timing diagram for the ADC in Free-Running mode with single conversion.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 498
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
Figure 33-6. Timing Diagram - Free-Running Conversion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
1
15
2
3
CLK_ADC
State
Converting
F
Sampling
Converting
F
Sampling
Converting
RESRDY
(Event)
RESRDY
(Int Req)
Software clear
The Result Ready event and the interrupt flag are set after each conversion. It is possible to combine accumulated
conversion and Free-Running mode.
To safely change any of these settings when using Free-Running mode, disable Free-Running mode, and wait for the
conversion to complete before doing any changes. Enable Free-Running mode again before starting the next
conversion.
33.3.3.4.4 Adjusting Conversion Time
Both sampling time and sampling length can be adjusted using the Sampling Delay Selection (SAMPDLY) bit field in
the Control D (ADCn.CTRLD) register and Sample Length (SAMPLEN) bit field in the Sample Control
(ADCn.SAMPCTRL) register. Both of these control the ADC sampling time and sampling length in a number of
CLK_ADC cycles. Increasing SAMPLEN allows sampling high-impedance sources without reducing CLK_ADC
frequency. Adjusting SAMPDLY is intended for tuning the sampling frequency away from harmonic noise in the
analog signal. Total sampling time is given by:
SampleTime =
2 + SAMPDLY + SAMPLEN
�CLK_ADC
The equation above implies that the total conversion time for n samples is now:
Total Conversion Time (12-bit) =
2
13.5 + 2 + ������� + �������
2
+�
+
�CLK_PER
�CLK_ADC
�CLK_PER
Some of the analog resources used by the ADC require time to initialize before a conversion can start. The
Initialization Delay (INITDLY) bit field in the Control D (ADCn.CTRLD) register can be used to prevent starting a
conversion prematurely by halting sampling for the configured delay duration.
The figure below shows the timing diagram for the ADC and the usage of the INITDLY, SAMPDLY and SAMPLEN bit
fields:
Figure 33-7. Timing Diagram - Conversion with Delays and Custom Sampling Length
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
CLK_ADC
ENABLE
STCONV
State
Idle
Initialization
Sampling
Converting
F
Result
RESRDY
(Event)
RESRDY
(Int Req)
Start
- up
(2 CLK_PER
cycles)
INITDLY
(0 – 256
CLK_ADC cycles)
SAMPDLY
(0 – 15
CLK_ADC cycles)
SAMPLEN
(0 – 255
CLK_ADC cycles)
Sampling
(2 CLK_ADC
cycles)
Conversion
(13.5 CLK_ADC
cycles)
Result formatting
(2 CLK_PER
cycles)
33.3.3.5 Conversion Result (Output Formats)
The result of an analog-to-digital conversion is written to the 16-bit Result (ADCn.RES) register and is given by the
following equations:
Single-ended 12-bit conversion: ��� =
© 2020 Microchip Technology Inc.
�AINP
× 4096 ∈ 0, 4095
�ADCREF
Preliminary Datasheet
DS40002247A-page 499
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
Single-ended 10-bit conversion: ��� =
Differential 12-bit conversion: ��� =
Differential 10-bit conversion: ��� =
�AINP
× 1024 ∈ 0, 1023
�ADCREF
��INP − �AINN
× 2048 ∈ −2048, 2047
�ADCREF
�AINP − �AINN
× 512 ∈ −512, 511
�ADCREF
where VAINP and VAINN are the positive and negative ADC inputs and VADCREF is the selected ADC voltage reference.
The data format used for single-ended conversions is unsigned one’s complement, while two's complement with sign
extension is used for differential conversions. Consequently, for differential conversions the sign bit is padded to the
higher bits in the Result register, if needed.
By default, conversion results are stored in the Result register as right-adjusted 16-bit values. The eight Least
Significant bits (LSbs) are then located in the low byte of the Result register. By writing a ‘1’ to the Left Adjust Result
(LEFTADJ) bit in the Control A (ADCn.CTRLA) register, the values will be left-adjusted by placing the eight Most
Significant bits (MSbs) in the high byte of the Result register.
The two figures below illustrate the relationship between the analog input and the corresponding ADC output.
Figure 33-8. Unsigned Single-Ended, Input Range, and Result Representation
VADCREF
VAINP
0V
Dec
4095
4094
4093
...
2049
2048
2047
2046
...
2
1
0
Hex
FFF
FFE
FFD
...
801
800
7FF
7FE
...
2
1
0
16-bit result
0x0FFF
0x07FF
0x0FFE
0x0FFD
...
0x0801
0x0800
0x07FF
0x07FE
...
0x0002
0x0001
0x0000
Where VAINP is the single-ended or internal input.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 500
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
Figure 33-9. Signed Differential Input, Input Range, and Result Representation
VADCREF
0V
VAINP-VAINN
-VADCREF
If a single-ended analog input is above the ADC voltage reference level, the 12-bit ADC result will be 0xFFF
(decimal 4095). Likewise, if the input is below 0V, the ADC result will be 0x000.
If the voltage difference between VAINP and VAINN for a 12-bit differential conversion is above the ADC voltage
reference level, the ADC result will be 0x7FF (decimal 2047). If the voltage difference is larger than the voltage
reference level in the negative direction, the ADC result will be 0x800 (decimal -2048).
33.3.3.6 Accumulation
By default, conversion results are stored in the Result register as right-adjusted 16-bit values. The eight Least
Significant bits (LSbs) are then located in the low byte of the Result register. By writing a ‘1’ to the Left Adjust Result
(LEFTADJ) bit in the Control A (ADCn.CTRLA) register, the values will be left-adjusted by placing the eight Most
Significant bits (MSbs) in the high byte of the Result register.
The result from multiple consecutive conversions can be accumulated. The number of samples to be accumulated is
specified by the Sample Accumulation Number Select (SAMPNUM) bit field in the Control B (ADCn.CTRLB) register.
When accumulating more than 16 samples, the result might be too large to match the 16-bit Result register size. To
avoid overflow, the LSbs of the result are truncated to fit within the available register size.
The two following tables show how the Result (ADCn.RES) register value is stored for single-ended and differential
conversions.
Table 33-1. Result Format in Single-Ended Mode
RES[15:8]
Accumulations
1
2
4
8
LEFTADJ
Bit
15
Bit
14
Bit
13
Bit
12
0
0
0
0
0
1
0
0
Bit
9
Bit
8
Bit
7
Bit
6
Bit
5
0
0
0
0
1
© 2020 Microchip Technology Inc.
Bit
3
Bit
2
Bit
1
Bit
0
0
0
0
0
0
0
0
0
0
Accumulation [12:0]
Accumulation [13:0]
Accumulation [13:0]
0
Bit
4
Conversion [11:0]
Accumulation [12:0]
1
0
Bit
10
Conversion [11:0]
1
0
Bit
11
RES[7:0]
Accumulation [14:0]
0
Accumulation [14:0]
Preliminary Datasheet
DS40002247A-page 501
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
...........continued
RES[15:8]
Accumulations
LEFTADJ
Bit
15
Bit
14
Bit
13
Bit
12
Bit
11
RES[7:0]
Bit
10
Bit
9
0
16
Bit
7
Bit
6
Bit
5
Bit
4
Bit
3
Bit
2
Bit
1
Bit
0
Accumulation [15:0]
1
0
32, 64, 128
Bit
8
Truncated Accumulation [15:0]
1
Table 33-2. Result Format in Differential Mode
RES[15:8]
Accumulations
LEFTADJ
Bit 15
0
1
Bit
14
0
Sign extension
Sign extension
0
1
32, 64, 128
Bit
8
Bit
7
Bit
6
Bit
5
Bit
4
Bit
3
Bit
2
Bit
1
Bit
0
0
0
0
0
0
0
0
0
0
Signed conversion [11:0]
Signed accumulation [12:0]
Signed accumulation [13:0]
Signed accumulation [13:0]
1
16
Bit
9
Signed accumulation [12:0]
1
8
Bit
10
Sign extension
1
0
Bit
11
Signed conversion [11:0]
0
4
Bit
12
Sign extension
1
2
Bit
13
RES[7:0]
0
1
Signed accumulation [14:0]
0
Signed accumulation [14:0]
Signed accumulation [15:0]
Signed truncated accumulation [15:0]
33.3.3.7 Channel Selection
The input selection for the ADC is controlled by the MUXPOS and MUXNEG bit fields in the ADCn.MUXPOS and
ADCn.MUXNEG registers, respectively. If the ADC is running single-ended conversions, only MUXPOS is used, while
both are used in differential conversions.
The MUXPOS bit field of the ADCn.MUXPOS register and the MUXNEG bit field of the ADCn.MUXNEG register are
buffered through a temporary register. This ensures that the input selection only comes into effect at a safe point
during the conversion. The channel selections are continuously updated until a conversion is started.
Once the conversion starts, the channel selections are locked to ensure sufficient sampling time for the ADC. The
continuous updating of input channel selection resumes in the last CLK_ADC clock cycle before the conversion
completes. The next conversion starts on the following rising CLK_ADC clock edge after the STCONV bit is written to
‘1’.
33.3.3.8 Temperature Measurement
An on-chip temperature sensor is available. Follow the steps below to do a temperature measurement. The resulting
value will be right-adjusted.
1.
2.
In the Voltage Reference (VREF) peripheral, select the internal 2.048V reference as the ADC reference
voltage.
Select the temperature sensor as input in the ADCn.MUXPOS register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 502
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
3.
4.
Acquire the temperature sensor output voltage by running a 12-bit, right-adjusted, single-ended conversion.
Process the measurement result as described below.
The measured voltage has an almost linear relationship with the temperature. Due to process variations, the
temperature sensor output voltage varies between individual devices at the same temperature. The individual
compensation factors determined during production test are stored in the Signature Row. These compensations
factors are generated for the internal 2.048V reference.
•
•
SIGROW.TEMPSENSE0 contains the slope of the temperature sensor characteristics
SIGROW.TEMPSENSE1 contains the offset of the temperature sensor characteristics
In order to achieve more accurate results, the result of the temperature sensor measurement must be processed in
the application software using compensation values from device production or user calibration. The temperature (in
Kelvin) is calculated by the following equation:
�=
Offset − ADC Result × Slope
4096
It is recommended to follow these steps in the application code when using the compensation values from the
Signature Row:
uint16_t sigrow_offset = SIGROW.TEMPSENSE1; // Read unsigned value from signature row
uint16_t sigrow_slope = SIGROW.TEMPSENSE0; // Read unsigned value from signature row
uint16_t adc_reading = ADCn.RES;
// ADC conversion result
uint32_t temp = sigrow_offset - adc_reading;
temp *= sigrow_slope;
// Result will overflow 16-bit variable
temp += 0x0800;
// Add 4096/2 to get correct rounding on division below
temp >>= 12;
// Round off to nearest degree in Kelvin, by dividing with 2^12 (4096)
uint16_t temperature_in_K = temp;
To increase the precision of the measurement to less than 1 Kelvin it is possible to adjust the last two steps to round
off to a fraction of one degree. Add 4096/4 and right shift by 11 for a precision of ½ Kelvin, or add 4096/8 and right
shift by 10 for a ¼ Kelvin precision.
If accumulation is used to reduce noise in the temperature measurement, the ADC result needs to be adjusted to a
12-bit value before the calculation is performed.
If another reference (VADCREF) than 2.048V is required, the offset and slope values need to be adjusted according to
the following equations:
Slope = TEMPSENSE0 ×
Offset = TEMPSENSE1 ×
33.3.3.9 Window Comparator
VADCREF
2.048V
2.048V
VADCREF
The ADC can raise the Window Comparator Interrupt (WCMP) flag in the Interrupt Flags (ADCn.INTFLAGS) register
and request an interrupt (WCMP) when the output of a conversion or accumulation is above and/or below certain
thresholds. The available modes are:
• The result is below a threshold
• The result is above a threshold
• The result is inside a window (above the lower threshold and below the upper threshold)
• The result is outside a window (either under the lower threshold or above the upper threshold)
The thresholds are defined by writing to the Window Comparator Low and High Threshold (ADCn.WINLT and
ADCn.WINHT) registers. Writing to the Window Comparator Mode (WINCM) bit field in the Control E (ADCn.CTRLE)
register selects the Window mode to use.
When accumulating multiple samples, the comparison between the result and the threshold will happen after the last
sample was acquired. Consequently, the flag is raised only once, after taking the last sample of the accumulation.
Assuming the ADC is already configured to run, follow these steps to use the Window Comparator:
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 503
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
1.
2.
3.
Set the required threshold(s) by writing to the Window Comparator Low and High Threshold (ADCn.WINLT
and ADCn.WINHT) registers.
Optional: Enable the interrupt request by writing a ‘1’ to the Window Comparator Interrupt Enable (WCMP) bit
in the Interrupt Control (ADCn.INTCTRL) register.
Enable the Window Comparator and select a mode by writing a valid non-zero value to the Window
Comparator Mode (WINCM) bit field in the Control E (ADCn.CTRLE) register.
When accumulating samples, the window comparator thresholds are applied to the accumulated value and not to
each sample. Using left adjustment of the result will make the comparator values independent of number of samples.
33.3.4
I/O Lines and Connections
The analog input pins and the VREF pin (AINx and VREFA) are configured in the I/O Pin Controller (PORT).
To reduce power consumption, the digital input buffer has to be disabled on the pins used as inputs for ADC. This is
configured by the I/O Pin Controller (PORT).
33.3.5
Events
The ADC can generate the following events:
Table 33-3. Event Generators in ADC
Generator Name
Peripheral
ADCn
Description
Event Type
Event
RESRDY
Result ready
Pulse
Generating
Clock Domain
CLK_PER
Length of Event
One clock period
The conditions for generating an event are identical to those that will raise the corresponding flag in the Interrupt
Flags (ADCn.INTFLAGS) register.
The ADC has one event user for detecting and acting upon input events. The table below describes the event user
and the associated functionality.
Table 33-4. Event Users and Available Event Actions in ADC
User Name
Peripheral
Input
ADCn
START
Description
Input Detection
Async/Sync
ADC start conversion
Edge
Async
The ADC can be configured to start a conversion on the rising edge of an event signal by writing a ‘1’ to the STARTEI
bit field in the Event Control (ADCn.EVCTRL) register. Refer to the Event System (EVSYS) chapter for more details
regarding event types and Event System configuration.
When an input event trigger occurs, the positive edge will be detected, the Start Conversion (STCONV) bit in the
Command (ADCn.COMMAND) register will be set, and the conversion will start. When the conversion is completed,
the Result Ready (RESRDY) flag in the Interrupt Flags (ADCn.INTFLAGS) register is set and the STCONV bit in
ADCn.COMMAND is cleared.
33.3.6
Interrupts
Table 33-5. Available Interrupt Vectors and Sources
Name
Vector Description
Conditions
RESRDY
Result Ready interrupt
The conversion result is available in ADCn.RES.
WCMP
Window Comparator interrupt
As defined by WINCM in ADCn.CTRLE.
When an interrupt condition occurs, the corresponding interrupt flag is set in the Interrupt Flags (ADCn.INTFLAGS)
register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 504
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
An interrupt source is enabled or disabled by writing to the corresponding enable bit in the Interrupt Control
(ADCn.INTCTRL) register.
An interrupt request is generated when the corresponding interrupt source is enabled and the interrupt flag is set. The
interrupt request remains active until the interrupt flag is cleared. Refer to the ADCn.INTFLAGS register for details on
how to clear interrupt flags.
33.3.7
Debug Operation
By default, halting the CPU in Debugging mode will halt the normal operation of the peripheral.
This peripheral can be forced to operate while the CPU is halted by writing a ‘1’ to the Debug Run (DBGRUN) bit in
the Debug Control (ADCn.DBGCTRL) register.
33.3.8
Sleep Mode Operation
By default, the ADC is disabled in Standby sleep mode.
The ADC can stay fully operational in Standby Sleep mode if the Run in Standby (RUNSTDBY) bit in the Control A
(ADCn.CTRLA) register is written to ‘1’.
In this case, the ADC will stay active, any ongoing conversions will be completed, and interrupts will be executed as
configured.
In Standby seep mode, an ADC conversion can be triggered only via the Event System (EVSYS), or the ADC must
be in Free-Running mode with the first conversion triggered by software before entering sleep. The peripheral clock is
requested if needed and is turned off after the conversion is completed.
The reference source and supply infrastructure need time to stabilize when activated in Standby sleep mode.
Configure a delay for the start of the first conversion by writing a non-zero value to the Initialization Delay (INITDLY)
bit field in the Control D (ADCn.CTRLD) register.
In Power-Down sleep mode, no conversions are possible. Any ongoing conversions are halted and will be resumed
when going out of sleep. At the end of the conversion, the Result Ready (RESRDY) flag will be set, but the content of
the Result (ADCn.RES) registers will be invalid since the ADC was halted during a conversion. It is recommended to
make sure conversions have completed before entering Power-Down sleep mode.
33.3.9
Synchronization
Not applicable.
33.3.10 Configuration Change Protection
Not applicable.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 505
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
33.4
Register Summary
Offset
Name
Bit Pos.
7
0x00
0x01
0x02
0x03
0x04
0x05
0x06
...
0x07
0x08
0x09
0x0A
0x0B
0x0C
0x0D
0x0E
0x0F
CTRLA
CTRLB
CTRLC
CTRLD
CTRLE
SAMPCTRL
7:0
7:0
7:0
7:0
7:0
7:0
RUNSTDBY
0x10
RES
0x12
WINLT
0x14
WINHT
33.5
6
5
4
CONVMODE
LEFTADJ
3
2
RESSEL[1:0]
INITDLY[2:0]
1
FREERUN
SAMPNUM[2:0]
PRESC[3:0]
SAMPDLY[3:0]
WINCM[2:0]
0
ENABLE
SAMPLEN[7:0]
Reserved
MUXPOS
MUXNEG
COMMAND
EVCTRL
INTCTRL
INTFLAGS
DBGCTRL
TEMP
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
7:0
15:8
7:0
15:8
7:0
15:8
MUXPOS[6:0]
MUXNEG[6:0]
SPCONV
WCMP
WCMP
STCONV
STARTEI
RESRDY
RESRDY
DBGRUN
TEMP[7:0]
RES[7:0]
RES[15:8]
WINLT[7:0]
WINLT[15:8]
WINHT[7:0]
WINHT[15:8]
Register Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 506
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
33.5.1
Control A
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
RUNSTDBY
R/W
0
CTRLA
0x00
0x00
-
6
5
CONVMODE
R/W
0
4
LEFTADJ
R/W
0
3
2
RESSEL[1:0]
R/W
R/W
0
0
1
FREERUN
R/W
0
0
ENABLE
R/W
0
Bit 7 – RUNSTDBY Run in Standby
This bit determines whether the ADC still runs during Standby.
Value
Description
0
ADC will not run in Standby sleep mode. An ongoing conversion will finish before the ADC enters sleep
mode
1
ADC will run in Standby sleep mode
Bit 5 – CONVMODE Conversion Mode
This bit defines if the ADC is working in Single-Ended or Differential mode.
Value
Name
Description
0x0
SINGLEENDED The ADC is operating in Single-Ended mode where only the positive input is used.
The ADC result is presented as an unsigned value.
0x1
DIFF
The ADC is operating in Differential mode where both positive and negative inputs
are used. The ADC result is presented as a signed value.
Bit 4 – LEFTADJ Left Adjust Result
Writing a ‘1’ to this bit will enable left adjustment of the ADC result.
Bits 3:2 – RESSEL[1:0] Resolution Selection
This bit field selects the ADC resolution. When changing the resolution from 12-bit to 10-bit, the conversion time is
reduced from 13.5 CLK_ADC cycles to 11.5 CLK_ADC cycles.
Value
Description
0x00
12-bit resolution
0x01
10-bit resolution
Other
Reserved
Bit 1 – FREERUN Free-Running
Writing a ‘1’ to this bit will enable the Free-Running mode for the ADC. The first conversion is started by writing a ‘1’
to the Start Conversion (STCONV) bit in the Command (ADCn.COMMAND) register.
Bit 0 – ENABLE ADC Enable
Value
Description
0
ADC is disabled
1
ADC is enabled
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 507
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
33.5.2
Control B
Name:
Offset:
Reset:
Property:
Bit
7
CTRLB
0x01
0x00
-
6
5
4
3
Access
Reset
2
R/W
0
1
SAMPNUM[2:0]
R/W
0
0
R/W
0
Bits 2:0 – SAMPNUM[2:0] Sample Accumulation Number Select
This bit field selects how many consecutive ADC sampling results are accumulated automatically. When this bit field
is written to a value greater than 0x0, the according number of consecutive ADC sampling results are accumulated
into the ADC Result (ADCn.RES) register.
Value
Name
Description
0x0
NONE
No accumulation
0x1
ACC2
2 results accumulated
0x2
ACC4
4 results accumulated
0x3
ACC8
8 results accumulated
0x4
ACC16
16 results accumulated
0x5
ACC32
32 results accumulated
0x6
ACC64
64 results accumulated
0x7
ACC128
128 results accumulated
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 508
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
33.5.3
Control C
Name:
Offset:
Reset:
Property:
Bit
7
CTRLC
0x02
0x00
-
6
Access
Reset
5
4
3
R/W
0
2
1
PRESC[3:0]
R/W
R/W
0
0
0
R/W
0
Bits 3:0 – PRESC[3:0] Prescaler
This bit field defines the division factor from the peripheral clock (CLK_PER) to the ADC clock (CLK_ADC).
Value
Name
Description
0x0
DIV2
CLK_PER divided by 2
0x1
DIV4
CLK_PER divided by 4
0x2
DIV8
CLK_PER divided by 8
0x3
DIV12
CLK_PER divided by 12
0x4
DIV16
CLK_PER divided by 16
0x5
DIV20
CLK_PER divided by 20
0x6
DIV24
CLK_PER divided by 24
0x7
DIV28
CLK_PER divided by 28
0x8
DIV32
CLK_PER divided by 32
0x9
DIV48
CLK_PER divided by 48
0xA
DIV64
CLK_PER divided by 64
0xB
DIV96
CLK_PER divided by 96
0xC
DIV128
CLK_PER divided by 128
0xD
DIV256
CLK_PER divided by 256
Other
Reserved
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 509
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
33.5.4
Control D
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
R/W
0
CTRLD
0x03
0x00
-
6
INITDLY[2:0]
R/W
0
5
R/W
0
4
3
R/W
0
2
1
SAMPDLY[3:0]
R/W
R/W
0
0
0
R/W
0
Bits 7:5 – INITDLY[2:0] Initialization Delay
This bit field defines the initialization delay before the first sample when enabling the ADC or changing to an internal
reference voltage. Setting this delay will ensure that the components of ADC are ready before starting the first
conversion. The initialization delay will also be applied when waking up from deep Sleep to do a measurement.
The delay is expressed as a number of CLK_ADC cycles.
Value
Name
Description
0x0
DLY0
Delay 0 CLK_ADC cycles
0x1
DLY16
Delay 16 CLK_ADC cycles
0x2
DLY32
Delay 32 CLK_ADC cycles
0x3
DLY64
Delay 64 CLK_ADC cycles
0x4
DLY128
Delay 128 CLK_ADC cycles
0x5
DLY256
Delay 256 CLK_ADC cycles
Other
Reserved
Bits 3:0 – SAMPDLY[3:0] Sampling Delay
This bit field defines the delay between consecutive ADC samples. This allows modifying the sampling frequency
used during hardware accumulation, to suppress periodic noise that may otherwise disturb the sampling. The delay is
expressed as CLK_ADC cycles and is given directly by the bit field setting.
Value
Name
Description
0x0
DLY0
Delay 0 CLK_ADC cycles
0x1
DLY1
Delay 1 CLK_ADC cycles
0x2
DLY2
Delay 2 CLK_ADC cycles
...
...
0xF
DLY15
Delay 15 CLK_ADC cycles
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 510
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
33.5.5
Control E
Name:
Offset:
Reset:
Property:
Bit
7
CTRLE
0x04
0x00
-
6
5
4
3
Access
Reset
2
R/W
0
1
WINCM[2:0]
R/W
0
0
R/W
0
Bits 2:0 – WINCM[2:0] Window Comparator Mode
This bit field enables the Window Comparator and defines when the Window Comparator Interrupt Flag (WCMP) in
the Interrupt Flags (ADCn.INTFLAGS) register is set. In the table below, RESULT is the accumulated 16-bit result.
WINLT and WINHT are the 16-bit lower threshold value and the 16-bit upper threshold value given by the
ADCn.WINLT and ADCn.WINHT registers, respectively.
Value
Name
Description
0x0
NONE
No Window Comparison (default)
0x1
BELOW
RESULT < WINLT
0x2
ABOVE
RESULT > WINHT
0x3
INSIDE
WINLT ≤ RESULT ≤ WINHT
0x4
OUTSIDE
RESULT < WINLT or RESULT >WINHT
Other
Reserved
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 511
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
33.5.6
Sample Control
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
SAMPCTRL
0x05
0x00
-
7
6
5
R/W
0
R/W
0
R/W
0
4
3
SAMPLEN[7:0]
R/W
R/W
0
0
2
1
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – SAMPLEN[7:0] Sample Length
This bit field extends the ADC sampling time with the number of CLK_ADC cycles given by the bit field value.
Increasing the sampling time allows sampling sources with higher impedance. By default, the sampling time is two
CLK_ADC cycles. The total conversion time increases with the selected sampling length.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 512
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
33.5.7
MUX Selection for Positive ADC Input
Name:
Offset:
Reset:
Property:
Bit
7
Access
Reset
MUXPOS
0x08
0x00
-
6
5
4
R/W
0
R/W
0
R/W
0
3
MUXPOS[6:0]
R/W
0
2
1
0
R/W
0
R/W
0
R/W
0
Bits 6:0 – MUXPOS[6:0] MUX Selection for Positive ADC Input
This bit field selects which analog input is connected to the positive input of the ADC. If this bit field is changed during
a conversion, the change will not take effect until the conversion is complete.
Value
Name
Description
0x00-0x15
0x16-0x3F
0x40
0x41
0x42
0x43
0x44
0x45
0x46-0x47
0x48
0x49
0x4A
0x4B
Other
AIN0-AIN21
GND
TEMPSENSE
VDDDIV10
VDDIO2DIV10
DAC0
DACREF0
DACREF1
DACREF2
-
ADC input pin 0-21
Reserved
Ground
Reserved
Temperature sensor
Reserved
VDD divided by 10
VDDIO2 divided by 10
Reserved
DAC0
DACREF0
DACREF1
DACREF2
Reserved
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 513
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
33.5.8
MUX Selection for Negative ADC Input
Name:
Offset:
Reset:
Property:
Bit
7
Access
Reset
MUXNEG
0x09
0x00
-
6
5
4
R/W
0
R/W
0
R/W
0
3
MUXNEG[6:0]
R/W
0
2
1
0
R/W
0
R/W
0
R/W
0
Bits 6:0 – MUXNEG[6:0] MUX Selection for Negative ADC Input
This bit field selects which analog input is connected to the negative input of the ADC. If this bit field is changed
during a conversion, the change will not take effect until the conversion is complete.
Value
Name
Description
0x00-0x0F
0x10-0x3F
0x40
0x41-0x47
0x48
Other
AIN0-AIN15
GND
DAC0
-
ADC input pin 0-15
Reserved
Ground
Reserved
DAC0
Reserved
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 514
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
33.5.9
Command
Name:
Offset:
Reset:
Property:
Bit
7
COMMAND
0x0A
0x00
-
6
5
4
3
Access
Reset
2
1
SPCONV
R/W
0
0
STCONV
R/W
0
Bit 1 – SPCONV Stop Conversion
Writing a ‘1’ to this bit will end the current measurement. This bit will take precedence over the Start Conversion
(STCONV) bit. Writing a ‘0’ to this bit has no effect.
Bit 0 – STCONV Start Conversion
Writing a ‘1’ to this bit will start a conversion as soon as any ongoing conversions are completed. If in Free-Running
mode, this will start the first conversion. STCONV will read as ‘1’ as long as a conversion is in progress. When the
conversion is complete, this bit is automatically cleared. Writing a ‘0’ to this bit has no effect.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 515
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
33.5.10 Event Control
Name:
Offset:
Reset:
Property:
Bit
7
EVCTRL
0x0B
0x00
-
6
5
4
3
Access
Reset
2
1
0
STARTEI
R/W
0
Bit 0 – STARTEI Start Event Input
This bit enables the event input as trigger for starting a conversion. When a ‘1’ is written to this bit, a rising event
edge will trigger an ADC conversion.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 516
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
33.5.11 Interrupt Control
Name:
Offset:
Reset:
Property:
Bit
7
INTCTRL
0x0C
0x00
-
6
5
4
3
Access
Reset
2
1
WCMP
R/W
0
0
RESRDY
R/W
0
Bit 1 – WCMP Window Comparator Interrupt Enable
Writing a ‘1’ to this bit enables the window comparator interrupt.
Bit 0 – RESRDY Result Ready Interrupt Enable
Writing a ‘1’ to this bit enables the Result Ready interrupt.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 517
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
33.5.12 Interrupt Flags
Name:
Offset:
Reset:
Property:
Bit
7
INTFLAGS
0x0D
0x00
-
6
5
4
3
Access
Reset
2
1
WCMP
R/W
0
0
RESRDY
R/W
0
Bit 1 – WCMP Window Comparator Interrupt Flag
This window comparator flag is set when the measurement is complete and if the result matches the selected
Window Comparator mode defined by the WINCM bit field in the Control E (ADCn.CTRLE) register. The comparison
is done at the end of the conversion. The flag is cleared by either writing a ‘1’ to the bit position or by reading the
Result (ADCn.RES) register. Writing a ‘0’ to this bit has no effect.
Bit 0 – RESRDY Result Ready Interrupt Flag
The Result Ready interrupt flag is set when a measurement is complete and a new result is ready. The flag is cleared
by either writing a ‘1’ to the bit location or by reading the Result (ADCn.RES) register. Writing a ‘0’ to this bit has no
effect.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 518
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
33.5.13 Debug Control
Name:
Offset:
Reset:
Property:
Bit
7
DBGCTRL
0x0E
0x00
-
6
5
4
3
2
1
Access
Reset
0
DBGRUN
R/W
0
Bit 0 – DBGRUN Run in Debug Mode
When written to ‘1’, the peripheral will continue operating in Debug mode when the CPU is halted.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 519
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
33.5.14 Temporary
Name:
Offset:
Reset:
Property:
TEMP
0x0F
0x00
-
The Temporary register is used by the CPU for 16-bit single-cycle access to the 16-bit registers of this peripheral. The
register is common for all the 16-bit registers of this peripheral and can be read and written by software. For more
details on reading and writing 16-bit registers, refer to Accessing 16-Bit Registers in the Memories section.
Bit
7
6
5
4
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
TEMP[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – TEMP[7:0] Temporary
Temporary register for read and write operations to and from 16-bit registers.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 520
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
33.5.15 Result
Name:
Offset:
Reset:
Property:
RES
0x10
0x00
-
The ADCn.RESL and ADCn.RESH register pair represents the 16-bit value, ADCn.RES. The low byte [7:0] (suffix L)
is accessible at the original offset. The high byte [15:8] (suffix H) can be accessed at offset + 0x01.
Refer to the 33.3.3.5 Conversion Result (Output Formats) section for details on the output from this register.
Bit
15
14
13
12
11
10
9
8
R
0
R
0
R
0
R
0
3
2
1
0
R
0
R
0
R
0
R
0
RES[15:8]
Access
Reset
R
0
R
0
R
0
R
0
Bit
7
6
5
4
RES[7:0]
Access
Reset
R
0
R
0
R
0
R
0
Bits 15:8 – RES[15:8] Result High Byte
This bit field constitutes the high byte of the ADCn.RES register, where the MSb is RES[15].
Bits 7:0 – RES[7:0] Result Low Byte
This bit field constitutes the low byte of the ADCn.RES register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 521
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
33.5.16 Window Comparator Low Threshold
Name:
Offset:
Reset:
Property:
WINLT
0x12
0x00
-
This register is the 16-bit low threshold for the digital comparator monitoring the Result (ADCn.RES) register. The
data format must be according to the Conversion mode and left/right adjustment setting.
The ADCn.WINLTH and ADCn.WINLTL register pair represents the 16-bit value, ADCn.WINLT. The low byte [7:0]
(suffix L) is accessible at the original offset. The high byte [15:8] (suffix H) can be accessed at offset + 0x01.
Bit
Access
Reset
Bit
15
14
13
R/W
0
R/W
0
R/W
0
7
6
5
12
11
WINLT[15:8]
R/W
R/W
0
0
4
10
9
8
R/W
0
R/W
0
R/W
0
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
WINLT[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 15:8 – WINLT[15:8] Window Comparator Low Threshold High Byte
This bit field holds the MSB of the 16-bit register.
Bits 7:0 – WINLT[7:0] Window Comparator Low Threshold Low Byte
This bit field holds the LSB of the 16-bit register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 522
AVR128DB28/32/48/64
ADC - Analog-to-Digital Converter
33.5.17 Window Comparator High Threshold
Name:
Offset:
Reset:
Property:
WINHT
0x14
0x00
-
This register is the 16-bit high threshold for the digital comparator monitoring the Result (ADCn.RES) register. The
data format must be according to the Conversion mode and left/right adjustment setting.
The ADCn.WINHTH and ADCn.WINHTL register pair represents the 16-bit value, ADCn.WINHT. The low byte [7:0]
(suffix L) is accessible at the original offset. The high byte [15:8] (suffix H) can be accessed at offset + 0x01.
Bit
Access
Reset
Bit
15
14
13
R/W
0
R/W
0
R/W
0
7
6
5
12
11
WINHT[15:8]
R/W
R/W
0
0
4
10
9
8
R/W
0
R/W
0
R/W
0
3
2
1
0
R/W
0
R/W
0
R/W
0
R/W
0
WINHT[7:0]
Access
Reset
R/W
0
R/W
0
R/W
0
R/W
0
Bits 15:8 – WINHT[15:8] Window Comparator High Threshold High Byte
This bit field holds the MSB of the 16-bit register.
Bits 7:0 – WINHT[7:0] Window Comparator High Threshold Low Byte
This bit field holds the LSB of the 16-bit register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 523
AVR128DB28/32/48/64
DAC - Digital-to-Analog Converter
34.
DAC - Digital-to-Analog Converter
34.1
Features
•
•
•
•
34.2
10-bit Resolution
Up to 140 ksps Conversion Rate
High Drive Capabilities
The DAC Output Can Be Used as Input to the ADC Positive Input
Overview
The Digital-to-Analog Converter (DAC) converts a digital value written to the Data (DACn.DATA) register to an analog
voltage. The conversion range is between GND and the selected voltage reference in the Voltage Reference (VREF)
peripheral. The DAC has one continuous time output with high drive capabilities. The DAC conversion can be started
from the application by writing to the Data (DACn.DATA) register.
34.2.1
Block Diagram
Figure 34-1. DAC Block Diagram
Other
Peripherals
DAC
DATA
OUT
Output
Buffer
VREF
ENABLE
CTRLA
34.2.2
OUTEN
Signal Description
Signal
OUT
Description
DAC output
34.3
Functional Description
34.3.1
Initialization
Type
Analog
To operate the DAC, the following steps are required:
1.
2.
3.
Select the DAC reference voltage in the Voltage Reference (VREF) peripheral by writing the appropriate
Reference Selection bits.
Configure the further usage of the DAC output:
– Configure an internal peripheral to use the DAC output. Refer to the documentation of the respective
peripherals.
– Enable the output to a pin by writing a ‘1’ to the Output Buffer Enable (OUTEN) bit. The input for the DAC
pin must be disabled in the Port peripheral (ISC = INPUT_DISABLE in PORTx.PINCTRLn).
Write an initial digital value to the Data (DACn.DATA) register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 524
AVR128DB28/32/48/64
DAC - Digital-to-Analog Converter
4.
34.3.2
Enable the DAC by writing a ‘1’ to the ENABLE bit in the Control A (DACn.CTRLA) register.
Operation
34.3.2.1 Enabling, Disabling and Resetting
The DAC is enabled by writing a ‘1’ to the ENABLE bit in the Control A (DACn.CTRLA) register, and disabled by
writing a ‘0’ to this bit.
34.3.2.2 Starting a Conversion
When the ENABLE bit in the Control A (DACn.CTRLA) register is written to ‘1’, a conversion starts as soon as the
Data (DACn.DATA) register is written.
When the ENABLE bit in DACn.CTRLA is written to ‘0’, writing to the Data register does not trigger a conversion.
Instead, the conversion starts when the ENABLE bit in DACn.CTRLA is written to ‘1’.
34.3.2.3 DAC as Source For Internal Peripherals
The analog output of the DAC can be internally connected to other peripherals when the ENABLE bit in the Control A
(DACn.CTRLA) register is written to ‘1’. When the DAC analog output is only being used internally, the Output Buffer
Enable (OUTEN) bit in DACn.CTRLA can be ‘0’.
34.3.2.4 DAC Output on Pin
The analog output of the DAC can be connected to a pin by writing a ‘1’ to the Output Buffer Enable (OUTEN) bit in
the Control A (DACn.CTRLA) register. The pin used by the DAC must have the input disabled from the Port
peripheral. There is an output buffer between the DAC output and the pin, which ensures the analog value does not
depend on the load of the pin. The output buffer can only source current, it has very limited sinking capability.
34.3.3
Sleep Mode Operation
If the Run in Standby (RUNSTDBY) bit in the Control A (DACn.CTRLA) register is written to ‘1’, the DAC will continue
to operate in Standby sleep mode. If the RUNSTDBY bit is zero, the DAC will stop the conversion in Standby sleep
mode.
If the conversion is stopped in Standby sleep mode, the DAC and the output buffer are disabled to reduce power
consumption. When the device is exiting Standby sleep mode, the DAC and the output buffer (if the OUTEN bit in the
Control A (DACn.CTRLA) register is written to ‘1’) are enabled again. Therefore, a start-up time is required before a
new conversion is initiated.
In Power-Down sleep mode, the DAC and the output buffer are disabled to reduce power consumption.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 525
AVR128DB28/32/48/64
DAC - Digital-to-Analog Converter
34.4
Register Summary
Offset
Name
Bit Pos.
7
6
0x00
0x01
CTRLA
Reserved
7:0
RUNSTDBY
OUTEN
0x02
DATA
34.5
7:0
15:8
5
4
3
2
1
0
ENABLE
DATA[1:0]
DATA[9:2]
Register Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 526
AVR128DB28/32/48/64
DAC - Digital-to-Analog Converter
34.5.1
Control A
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
RUNSTDBY
R/W
0
CTRLA
0x00
0x00
-
6
OUTEN
R/W
0
5
4
3
2
1
0
ENABLE
R/W
0
Bit 7 – RUNSTDBY Run in Standby Mode
If this bit is written to ‘1’, the DAC or the output buffer will not automatically be disabled when the device is entering
Standby sleep mode.
Bit 6 – OUTEN Output Buffer Enable
Writing a ‘1’ to this bit enables the output buffer and sends the OUT signal to a pin.
Bit 0 – ENABLE DAC Enable
Writing a ‘1’ to this bit enables the DAC.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 527
AVR128DB28/32/48/64
DAC - Digital-to-Analog Converter
34.5.2
DATA
Name:
Offset:
Reset:
Property:
DATA
0x02
0x00
-
The DACn.DATAL and DACn.DATAH register pair represents the 10-bit value, DACn.DATA. The two LSbs [1:0] are
accessible at the original offset. The eight MSbs [9:2] can be accessed at offset + 0x01.
The output will be updated after DACn.DATAH is written.
Bit
15
14
13
12
11
10
9
8
DATA[9:2]
Access
Reset
Bit
R/W
0
7
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
R/W
0
6
5
4
3
2
1
0
DATA[1:0]
Access
Reset
R/W
0
R/W
0
Bits 15:6 – DATA[9:0]
These bits contain the digital data, which will be converted to an analog voltage.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 528
AVR128DB28/32/48/64
OPAMP - Analog Signal Conditioning
35.
OPAMP - Analog Signal Conditioning
35.1
Features
•
•
•
•
•
•
•
•
35.2
Internal Resistor Ladder Facilitates Analog Signal Conditioning with Zero External Components
Selectable Configurations:
– Standalone general-purpose operational amplifier (op amp)
– Unity-gain buffer
– Non-inverting/inverting Programmable Gain Amplifier (PGA)
– Cascaded PGAs
– Instrumentation amplifier
Input Selection:
– I/O pins
– DAC
– Ground
– VDD/2 reference
– Output from another op amp
– Internal resistor ladder
Output Selection:
– On I/O pins
– As input for ADC
– As input for AC
– As input for another op amp
Internal Timer Generates READY Event When Settling is Complete
Low-Power Support:
– Optional event-triggered operation
Event-System-Controlled Dump Mode to Support Signal Integration
Offset and Gain Calibration Using the ADC
Overview
The Analog Signal Conditioning (OPAMP) peripheral features three operational amplifiers (op amps), designated
OPn where n is zero, one or two. These op amps are implemented with a flexible connection scheme using analog
multiplexers and resistor ladders. This allows a large number of analog signal conditioning configurations to be
achieved, many of which require no external components. A multiplexer at the non-inverting (+) input of each op amp
allows connection to either an external pin, a wiper position from a resistor ladder, a DAC output, ground, VDD/2, or
an output from another op amp. A second multiplexer at the inverting (-) input of each op amp allows connection to
either an external pin, a wiper position from a resistor ladder, the output of the op amp, or DAC output. Three more
multiplexers connected to each resistor ladder provide additional configuration flexibility. Two of these multiplexers
select the top and bottom connections to the resistor ladder, and the third controls the wiper position.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 529
AVR128DB28/32/48/64
OPAMP - Analog Signal Conditioning
35.2.1
Block Diagram
Figure 35-1. Op Amp n (OPn) Block Diagram
OPnINMUX.MUXPOS
OPnINP
DAC
GND
VDD/2
CTRLA.ENABLE
OPnCTRLA.OUTMODE
*
*
+
Output
Driver
OPn
OPnINMUX.MUXNEG
OPnOUT
-
OPnINN
OPnOUT
VDD
DAC
OPnRESMUX.MUXTOP
OPnRESMUX.MUXWIP
R2
OPnWIP
R1
OPnRESMUX.MUXBOT
OPnINP
OPnINN
DAC
GND
* Additional internal analog signals -- see OPnINMUX
and OPnRESMUX register descriptions for details
*
35.2.2
Signal Description
Signal Name
Type
OPnINP
Analog input
Non-inverting (+) input pin for OPn
OPnINN
Analog input
Inverting (-) input pin for OPn
OPnOUT
Analog output
35.3
Functional Description
35.3.1
Initialization
Description
Output from OPn
To initialize the OPAMP peripheral for basic, always-on operation, the following steps are recommended:
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 530
AVR128DB28/32/48/64
OPAMP - Analog Signal Conditioning
1.
2.
Configure the timebase of the OPAMP peripheral by writing to the Timebase (TIMEBASE) bit field in the
Timebase (OPAMP.TIMEBASE) register.
For each op amp that will be used:
2.1.
Configure the Op Amp n Input Multiplexer (OPAMP.OPnINMUX):
• Select the non-inverting (+) input of the op amp by writing to the Multiplexer for Positive Input
(MUXPOS) bit field
• Select the inverting (-) input of the op amp by writing to the Multiplexer for Negative Input
(MUXNEG) bit field
2.2.
Configure the Op Amp n Resistor Ladder Multiplexer (OPAMP.OPnRESMUX):
• Select the connection to the top resistor in the resistor ladder by writing to the Multiplexer for Top
(MUXTOP) bit field
• Select the connection to the bottom resistor in the resistor ladder by writing to the Multiplexer for
Bottom (MUXBOT) bit field
• Select the wiper position in the resistor ladder by writing to the Multiplexer for Wiper (MUXWIP)
bit field
2.3.
Configure the Op Amp n Control A (OPAMP.OPnCTRLA) register:
• Configure the op amp to be always on by writing ‘1’ to the Always On (ALWAYSON) bit
•
Disable events for the op amp by writing ‘0’ to the Event Enable (EVENTEN) bit
•
3.
4.
35.3.2
Select the normal output mode by writing the NORMAL setting to the Output Mode (OUTMODE)
bit field
2.4.
Optional: Configure the allowed settling time by writing to the Settle Timer (SETTLE) bit field in the
Op Amp n Settle Timer (OPAMP.OPnSETTLE) register.
Enable the OPAMP peripheral by writing a ‘1’ to the OPAMP Enable (ENABLE) bit in the Control A
(OPAMP.CTRLA) register.
For each op amp whose settling time was configured in step 2.4 above, wait for the SETTLED bit in the Op
Amp n Status (OPAMP.OPnSTATUS) register to become ‘1’. This indicates that the op amp start-up and
settling have completed, and the OPAMP peripheral is ready for use.
Operation
35.3.2.1 MUXPOS - Non-Inverting (+) Input Selection
As shown in Figure 35-1, the non-inverting (+) input of an op amp is connected to an analog multiplexer that allows
one input source to be selected out of a variety of input sources. The source selection is configured using the
MUXPOS bit field of the Op Amp n Input Multiplexer (OPAMP.OPnINMUX) register. If the OPnINMUX register is
changed while the OPn output is enabled, a glitch will occur on the output signal due to the opening and closing of
analog switches, and some time will elapse before the output settles to the new value.
The following non-inverting input options are available for OPn:
• INP (OPnINP) - Positive input pin for OPn
• WIP (OPnWIP) - Wiper from OPn’s resistor ladder
• DAC - DAC output (DAC and DAC output buffer must be enabled)
• GND - Ground
• VDDDIV2 - VDD/2
OP1 has the following additional input available:
• LINKOUT (OP0OUT) - OP0 output
OP2 has the following additional inputs available:
• LINKOUT (OP1OUT) - OP1 output
• LINKWIP (OP0WIP) - Wiper from OP0’s resistor ladder
35.3.2.2 MUXNEG - Inverting (-) Input Selection
As shown in Figure 35-1, the inverting (-) input of an op amp is also connected to an analog multiplexer that allows
one of a variety of input sources to be chosen. The selection is configured using the MUXNEG bit field of the Op Amp
n Input Multiplexer (OPAMP.OPnINMUX) register. If the OPnINMUX register is changed while the OPn output is
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 531
AVR128DB28/32/48/64
OPAMP - Analog Signal Conditioning
enabled, a glitch will occur on the output signal due to the opening and closing of analog switches, and some time will
elapse before the output settles to the new value.
The following inverting input options are available for OPn:
• INN (OPnINN) - Negative input pin for OPn
• WIP (OPnWIP) - Wiper from OPn’s resistor ladder
• OUT (OPnOUT) - OPn output
• DAC - DAC output (DAC and DAC output buffer must be enabled)
35.3.2.3 MUXTOP, MUXBOT, MUXWIP - Resistor Ladder Configuration
A resistor ladder and three connected multiplexers are associated with each op amp, as shown in Figure 35-1. These
multiplexers are configured using the MUXTOP, MUXBOT and MUXWIP bit fields of the Op Amp n Resistor Ladder
Multiplexer (OPAMP.OPnRESMUX) register. If the OPnRESMUX register is changed while the OPn output is
enabled, a glitch will occur on the output signal due to the opening and closing of analog switches, and some time will
elapse before the output settles to the new value.
The top of the resistor ladder is connected to a multiplexer that is controlled by the MUXTOP bit field. It can connect
to the following signals:
• OFF - Multiplexer off, no signal connected
• OUT (OPnOUT) - OPn output
• VDD - VDD
The bottom of the resistor ladder is also connected to a multiplexer. This multiplexer is controlled by the MUXBOT bit
field, and it can connect to the following signals:
• OFF - Multiplexer off, no signal connected
• INP (OPnINP) - Positive input pin for OPn
• INN (OPnINN) - Negative input pin for OPn
• DAC - DAC output (DAC and DAC output buffer must be enabled)
• GND - Ground
OP0 has the following additional connection available via the MUXBOT bit field:
• LINKOUT (OP2OUT) - OP2 output
OP1 has the following additional connection available via the MUXBOT bit field:
• LINKOUT (OP0OUT) - OP0 output
OP2 has the following additional connection available via the MUXBOT bit field:
• LINKOUT (OP1OUT) - OP1 output
A third multiplexer is connected to eight different wiper positions on the resistor ladder. This multiplexer is controlled
by the MUXWIP bit field and allows the OPnWIP signal to be connected to any of eight different ratios of the upper
(R2) and lower (R1) resistors:
• WIP0 - R2/R1 = 1/15
• WIP1 - R2/R1 = 1/7
• WIP2 - R2/R1 = 1/3
• WIP3 - R2/R1 = 1
• WIP4 - R2/R1 = 5/3
• WIP5 - R2/R1 = 3
• WIP6 - R2/R1 = 7
• WIP7 - R2/R1 = 15
The tolerances of the resistor ratios, as well as the absolute values and tolerances of the resistors, are specified in
the Electrical Characteristics section.
35.3.2.4 Output Modes
As shown in Figure 35-1, an op amp has an output driver that is directly connected to an OPnOUT pin. When the
output driver for OPn is enabled, the OPnOUT pin cannot be driven by other peripherals. The output driver can
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 532
AVR128DB28/32/48/64
OPAMP - Analog Signal Conditioning
operate in different modes controlled by the OUTMODE bit field of the OPnCTRLA register. The following output
modes are available:
• OFF - The output driver is disabled, but can be overridden by a DRIVEn event
• NORMAL - The output driver is enabled and operates normally
35.3.2.5 Input Voltage Range
In applications where a rail-to-rail input voltage range is not needed, the OPAMP peripheral may be configured to
save power. Writing ‘1’ to the Input Range Select (IRSEL) bit in the Power Control (PWRCTRL) register reduces the
voltage range of the op amp inputs, which also reduces power consumption. See the Electrical Characteristics
section for more information about the expected voltage range and power consumption.
35.3.2.6 Internal Timer
When an op amp is enabled, it takes some time for the op amp to start up and begin functioning correctly. There are
two components to this start-up time, warmup time and settling time:
1. The warmup time is the time required to elapse before the internal circuitry in the op amp stabilizes. The
warmup time is documented in the Electrical Characteristics section. It does not depend on external circuitry.
2. The settling time is the additional time allowed for the op amp output to settle and become stable after the
warmup time is completed. It depends on a variety of factors, including the strength of the output driver and
the load on the op amp output.
Each op amp has an internal timer that is used to determine the delay between op amp enable and the indication that
it is ready to use. For the internal timer to function correctly, two bit fields must be configured:
1. The timebase of the OPAMP peripheral must be configured by writing to the Timebase (TIMEBASE) bit field in
the Timebase (OPAMP.TIMEBASE) register. Determine how many cycles of the peripheral clock (CLK_PER)
are equal to 1 μs. If that number is an integer, subtract one from it and write it to the TIMEBASE bit field. If that
number is not an integer, round it down to an integer and write it to the TIMEBASE bit field. For example, if the
peripheral clock period is 260 ns, then 3.85 cycles are equal to 1 μs. Since 3.85 is not an integer, it should be
rounded down to 3 and then written to the TIMEBASE bit field.
2. The settling time of OPn must be configured by writing to the Settle Timer (SETTLE) bit field in the Op Amp n
Settle Timer (OPAMP.OPnSETTLE) register. Since the settling time depends on a variety of factors, including
the load on the op amp, it may not be known until the later stages of design and development. If the settling
time is unknown, the maximum value of ‘0x7F’ (127 μs) should be written to the SETTLE bit field.
After the user-programmed settling time has elapsed, the SETTLED bit is changed from ‘0’ to ‘1’ in OPnSTATUS. If
the op amp is in EVENT_ENABLED mode, the READYn event is also issued.
If the OPnINMUX or OPnRESMUX register is changed while the OPn output is enabled, a glitch will occur on the
output signal due to the opening and closing of analog switches, and some time will elapse before the output settles
to the new value. For this reason, the settle timer restarts whenever there is a write to the OPnINMUX or
OPnRESMUX registers.
The settle timer also restarts whenever there is a write to the OPnCTRLA register.
35.3.2.7 Enable and Disable
An op amp can be configured to be in one of three possible Enable/Disable modes:
• Enabled/disabled by software with all events disabled (SW_ENABLED_WITHOUT_EVENTS mode)
• Enabled/disabled by the event system (EVENT_ENABLED mode)
• Enabled/disabled by software with other events enabled (SW_ENABLED_WITH_EVENTS mode)
To select SW_ENABLED_WITHOUT_EVENTS mode, a ‘0’ must be written to the EVENTEN bit in OPnCTRLA. In
SW_ENABLED_WITHOUT_EVENTS mode, an op amp is enabled by writing ‘1’ to the ALWAYSON bit in
OPnCTRLA. The op amp stays enabled as long as the ALWAYSON bit is ‘1’. Writing a ‘0’ to the ALWAYSON bit in
OPnCTRLA will disable the op amp. The op amp will not respond to incoming events and will not generate events.
To select EVENT_ENABLED mode, the ALWAYSON bit in OPnCTRLA must be written to ‘0’, and the EVENTEN bit
must be written to ‘1’. In this mode, the op amp is enabled by the ENABLEn event, and it is disabled by the
DISABLEn event. In EVENT_ENABLED mode, the op amp will issue a READYn event when settling is complete. The
op amp will also respond to incoming DUMPn and DRIVEn events. Special cases are handled as follows:
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 533
AVR128DB28/32/48/64
OPAMP - Analog Signal Conditioning
•
•
•
•
If an ENABLEn event is received while the op amp is in the process of starting up as a result of a prior
ENABLEn event, the most recent ENABLEn event will be ignored, and the READYn event will be issued when
the settling time from the first ENABLEn event has elapsed
If an ENABLEn event is received while the op amp is already enabled and has completed settling, the op amp
will stay enabled with no change in state. The READYn event will also be generated immediately after the
ENABLEn event in this case.
If a DISABLEn event is received while the op amp is already disabled, the DISABLEn event is ignored, and the
op amp remains disabled
If both an ENABLEn and DISABLEn event are received simultaneously, the DISABLEn event is ignored, and the
ENABLEn event is handled as described in the cases above
To select SW_ENABLED_WITH_EVENTS mode, a ‘1’ must be written to the EVENTEN bit in OPnCTRLA. In
SW_ENABLED_WITH_EVENTS mode, an op amp is enabled by writing ‘1’ to the ALWAYSON bit in OPnCTRLA.
The op amp stays enabled as long as the ALWAYSON bit is ‘1’. While the op amp is enabled, it will respond to
incoming DUMPn and DRIVEn events but will be unaffected by incoming ENABLEn and DISABLEn events. Writing a
‘0’ to the ALWAYSON bit in OPnCTRLA will disable the op amp.
35.3.2.8 Offset Calibration
The input offset voltage of an op amp can be calibrated by using the DAC or an external constant-voltage source in
conjunction with the ADC.
To calibrate an op amp, perform the following steps:
• Use the DAC or the external I/O pin connected to the op amp non-inverting (+) input as a constant-voltage
reference. If the DAC is used, it and its output buffer must be enabled.
• Configure the op amp with:
– MUXPOS in the OPnINMUX register set to the selected calibration source
– MUXNEG in the OPnINMUX register set to OPnOUT, so the op amp will function as a voltage follower
• Use the ADC to measure the voltage of the selected calibration source. This value becomes the calibration
target.
• Configure the ADC input multiplexer to select the output from OPn, then use the ADC to measure this voltage
• Use the difference between the two ADC measurements to determine how much the calibration register
(OPnCAL) should be adjusted. The calibration step size is found in the Electrical Characteristics section.
35.3.3
Events
An op amp must be in EVENT_ENABLED or SW_ENABLED_WITH_EVENTS mode to generate events and respond
to incoming events. To select EVENT_ENABLED mode for OPn, the EVENTEN bit in OPnCTRLA must be written to
‘1’, and the ALWAYSON bit must be written to ‘0’. To select SW_ENABLED_WITH_EVENTS mode for OPn, the
EVENTEN bit in OPnCTRLA must be written to ‘1’, and the ALWAYSON bit must be written to ‘1’.
OPn can generate the event as described in the table below.
Table 35-1. Event Generators in OPAMP
Generator Name
Peripheral
Event
OPAMP
READYn
Description
OPn is ready
Event Type
Generating Clock Domain
Pulse
CLK_PER
Length of Event
One CLK_PER period
The table below describes the event users for OPn and their associated functionality.
Table 35-2. Event Users in OPAMP
User Name
Peripheral
Input
OPAMP
ENABLEn
OPAMP
DISABLEn
Description
Input Detection
Async/Sync
Enable OPn
Edge
Async
Disable OPn
Edge
Sync
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 534
AVR128DB28/32/48/64
OPAMP - Analog Signal Conditioning
...........continued
User Name
Peripheral
Input
OPAMP
DUMPn
OPAMP
DRIVEn
Description
Input Detection
Async/Sync
Dump OPn VOUT to VINN
Level
Sync
Enable OPn output driver in NORMAL mode
Level
Sync
Refer to the Event System (EVSYS) section for more details regarding event types and Event System configuration.
35.3.4
Interrupts
The OPAMP peripheral does not generate interrupts.
35.3.5
Sleep Mode Operation
If the RUNSTDBY bit is ‘0’ in OPnCTRLA, OPn is shut off in Standby Sleep mode, and its output driver is disabled.
The op amp will release its override of the I/O pin, and the behavior of the I/O pin can be controlled using standard
GPIO control.
If the RUNSTDBY bit is ‘1’ in OPnCTRLA, OPn remains operational in Standby Sleep mode with its enable/disable
behavior determined by the ALWAYSON and EVENTEN bits in OPnCTRLA.
35.3.6
Debug Operation
When the CPU is halted in Debug mode, the analog portions of the OPAMP peripheral will continue to operate as
before the CPU halt. If the DBGRUN bit is ‘1’ in the DBGCTRL register, the digital interface of the OPAMP peripheral
will also continue operating normally.
35.3.7
Application Usage
The OPAMP peripheral is highly flexible and can be used in various analog signal conditioning applications. This
section describes a broad range of configurations and the multiplexer settings required to achieve them. Many of
these configurations require no external components.
Figure 35-2. Op Amp Connected Directly to Pins
+
OPnINP
OPn
OPnOUT
-
OPnINN
The figure above displays an op amp connected directly to the pins of the device without being connected to any of
the internal resistors. This is useful for situations where the user desires to make all connections to other components
externally. The multiplexer settings required to achieve this configuration are:
Table 35-3. Op Amp Connected Directly to Pins
OPn
MUXPOS
MUXNEG
MUXBOT
MUXWIP
MUXTOP
INP
INN
OFF
WIP0
OFF
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 535
AVR128DB28/32/48/64
OPAMP - Analog Signal Conditioning
Figure 35-3. Voltage Follower
VIN
+
VOUT
OPn
-
The figure above displays a voltage follower, also known as a unity-gain buffer. The non-inverting (+) input is
connected to a pin, and the output is connected to the inverting (-) input. The multiplexer settings required for this
configuration are:
Table 35-4. Voltage Follower
OPn
MUXPOS
MUXNEG
MUXBOT
MUXWIP
MUXTOP
INP
OUT
OFF
WIP0
OFF
Figure 35-4. Non-Inverting PGA
VOUT = VIN(1+R2/R1)
VIN
VOUT
OPn
R2
R1
The figure above displays a non-inverting Programmable Gain Amplifier. The required multiplexer settings are as
follows:
Table 35-5. Non-Inverting PGA
OPn
MUXPOS
MUXNEG
MUXBOT
MUXWIP
MUXTOP
INP
WIP
GND
Setting
determines gain
OUT
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 536
AVR128DB28/32/48/64
OPAMP - Analog Signal Conditioning
Figure 35-5. Inverting PGA
VOUT = -(R2/R1)(VIN-VDD/2)
+ VDD/2
+
VDD/2
VOUT
OPn
R1
-
VIN
R2
The figure above shows the inverting PGA configuration. The multiplexer settings required are:
Table 35-6. Inverting PGA
OPn
MUXPOS
MUXNEG
MUXBOT
MUXWIP
MUXTOP
VDDDIV2
WIP
INN
Setting
determines gain
OUT
Figure 35-6. Integrator
C (external)
DUMPn
VIN
-
R (external)
VOUT
OPn
VDD/2
+
The figure above shows how an op amp can be configured as an integrator. An external capacitor and external
resistor are required. The DUMPn event closes a switch to discharge the capacitor and reset the integrator. The
multiplexer settings required are:
Table 35-7. Integrator
OPn
MUXPOS
MUXNEG
MUXBOT
MUXWIP
MUXTOP
VDDDIV2
INN
OFF
WIP0
OFF
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 537
AVR128DB28/32/48/64
OPAMP - Analog Signal Conditioning
Figure 35-7. Differential Amplifier using Two Op Amps
VDIFF = OP1OUT - V2
+
V2
= (V2-V1)R2/R1
OP1
OP1OUT = V2 - (V1-V2)R2/R1
V1
-
+
R2(OP1)
R1(OP1)
OP0
-
The figure above shows how a differential amplifier can be constructed from two op amps. The differential amplifier
accepts two input signals, V1 and V2. The output also consists of two signals, OP1OUT and V2. The difference of the
output signals, VDIFF, is proportional to the difference (V2-V1) of the two input signals. The multiplexer settings
required to implement this configuration are as follows:
Table 35-8. Differential Amplifier using Two Op Amps
MUXPOS
MUXNEG
MUXBOT
MUXWIP
MUXTOP
OP0
INP
OUT
OFF
WIP0
OFF
OP1
INP
WIP
LINKOUT
(OP0OUT)
Setting
determines gain
OUT
Figure 35-8. Cascaded (Two) Non-Inverting PGA
VIN
OP0OUT = VIN(1+R2/R1)
OP0
OP1
OP1OUT = OP0OUT(1+R2/R1)
R2
R1
R2
R1
The figure above shows a cascaded non-inverting PGA constructed from two op amps. The following multiplexer
settings are required:
Table 35-9. Cascaded (Two) Non-Inverting PGA
OP0
MUXPOS
MUXNEG
MUXBOT
MUXWIP
MUXTOP
INP
WIP
GND
Determines gain
OUT
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 538
AVR128DB28/32/48/64
OPAMP - Analog Signal Conditioning
...........continued
OP1
MUXPOS
MUXNEG
MUXBOT
MUXWIP
MUXTOP
LINKOUT
(OP0OUT)
WIP
GND
Determines gain
OUT
Figure 35-9. Cascaded (Two) Inverting PGA
VDD/2
+
OP1OUT = -(R2/R1)(OP0OUT-VDD/2) + VDD/2
OP1
VDD/2
+
OP0
R2
R1
OP0OUT = -(R2/R1)(VIN-VDD/2) + VDD/2
VIN
R2
R1
The figure above shows a cascaded inverting PGA constructed from two op amps. The required multiplexer settings
are:
Table 35-10. Cascaded (Two) Inverting PGA
MUXPOS
MUXNEG
MUXBOT
MUXWIP
MUXTOP
OP0
VDDDIV2
WIP
INN
Determines gain
OUT
OP1
VDDDIV2
WIP
LINKOUT
(OP0OUT)
Determines gain
OUT
Figure 35-10. Cascaded (Three) Non-Inverting PGA
VIN
OP0OUT = VIN(1+R2/R1)
OP1OUT = OP0OUT(1+R2/R1)
OP0
OP2OUT = OP1OUT(1+R2/R1)
OP1
VOUT
OP2
R2
R1
R1
R2
R2
R1
The figure above shows a cascaded non-inverting PGA constructed from three op amps. The following multiplexer
settings are required:
Table 35-11. Cascaded (Three) Non-Inverting PGA
OP0
MUXPOS
MUXNEG
MUXBOT
MUXWIP
MUXTOP
INP
WIP
GND
Determines gain
OUT
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 539
AVR128DB28/32/48/64
OPAMP - Analog Signal Conditioning
...........continued
MUXPOS
MUXNEG
MUXBOT
MUXWIP
MUXTOP
OP1
LINKOUT
(OP0OUT)
WIP
GND
Determines gain
OUT
OP2
LINKOUT
(OP1OUT)
WIP
GND
Determines gain
OUT
Figure 35-11. Cascaded (Three) Inverting PGA
OP2OUT = -(R2/R1)(OP1OUT-VDD/2)
VDD/2
+ VDD/2
+
VOUT
OP2
VDD/2
-
+
OP1
R1
VDD/2
+
R2
OP1OUT = -(R2/R1)(OP0OUT-VDD/2) + VDD/2
OP0
R1
R2
OP0OUT = -(R2/R1)(VIN-VDD/2) + VDD/2
VIN
R1
R2
The figure above shows a cascaded inverting PGA constructed from three op amps. The required multiplexer settings
are:
Table 35-12. Cascaded (Three) Inverting PGA
MUXPOS
MUXNEG
MUXBOT
MUXWIP
MUXTOP
OP0
VDDDIV2
WIP
INN
Determines gain
OUT
OP1
VDDDIV2
WIP
LINKOUT
(OP0OUT)
Determines gain
OUT
OP2
VDDDIV2
WIP
LINKOUT
(OP1OUT)
Determines gain
OUT
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 540
AVR128DB28/32/48/64
OPAMP - Analog Signal Conditioning
Figure 35-12. Instrumentation Amplifier
V2
R1(OP0)
R2(OP0)
OP0
VOUT = (V2-V1)*Gain
OP2
V1
R2(OP2)
R1(OP2)
OP1
The figure above displays three op amps configured as an instrumentation amplifier. The following multiplexer
settings are required:
Table 35-13. Instrumentation Amplifier
MUXPOS
MUXNEG
MUXBOT
MUXWIP
MUXTOP
OP0
INP
OUT
GND
See the table
below
OUT
OP1
INP
OUT
OFF
WIP0
OFF
OP2
LINKWIP
(OP0WIP)
WIP
LINKOUT
(OP1OUT)
See the table
below
OUT
The resistor ladders associated with OP0 and OP2 must be configured as follows to select the desired gain:
Table 35-14. Gain Selection for Instrumentation Amplifier
GAIN
OP0RESMUX.MUXWIP
OP2RESMUX.MUXWIP
1/15
0x7
0x0
1/7
0x6
0x1
1/3
0x5
0x2
1
0x3
0x3
3
0x2
0x5
7
0x1
0x6
15
0x0
0x7
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 541
AVR128DB28/32/48/64
OPAMP - Analog Signal Conditioning
35.4
Register Summary
Offset
Name
Bit Pos.
0x00
0x01
0x02
0x03
...
0x0E
0x0F
0x10
0x11
0x12
0x13
0x14
0x15
0x16
...
0x17
0x18
0x19
0x1A
0x1B
0x1C
0x1D
0x1E
...
0x1F
0x20
0x21
0x22
0x23
0x24
0x25
CTRLA
DBGCTRL
TIMEBASE
7:0
7:0
7:0
35.5
7
6
5
4
3
2
1
0
ENABLE
DBGRUN
TIMEBASE[6:0]
Reserved
PWRCTRL
OP0CTRLA
OP0STATUS
OP0RESMUX
OP0INMUX
OP0SETTLE
OP0CAL
7:0
7:0
7:0
7:0
7:0
7:0
7:0
RUNSTDBY
OUTMODE[1:0]
MUXWIP[2:0]
MUXBOT[2:0]
MUXNEG[2:0]
IRSEL
ALWAYSON
SETTLED
MUXTOP[1:0]
MUXPOS[2:0]
EVENTEN
SETTLE[6:0]
CAL[7:0]
Reserved
OP1CTRLA
OP1STATUS
OP1RESMUX
OP1INMUX
OP1SETTLE
OP1CAL
7:0
7:0
7:0
7:0
7:0
7:0
RUNSTDBY
7:0
7:0
7:0
7:0
7:0
7:0
RUNSTDBY
OUTMODE[1:0]
MUXWIP[2:0]
MUXBOT[2:0]
MUXNEG[2:0]
EVENTEN
ALWAYSON
SETTLED
MUXTOP[1:0]
MUXPOS[2:0]
SETTLE[6:0]
CAL[7:0]
Reserved
OP2CTRLA
OP2STATUS
OP2RESMUX
OP2INMUX
OP2SETTLE
OP2CAL
OUTMODE[1:0]
MUXWIP[2:0]
MUXBOT[2:0]
MUXNEG[2:0]
EVENTEN
ALWAYSON
SETTLED
MUXTOP[1:0]
MUXPOS[2:0]
SETTLE[6:0]
CAL[7:0]
Register Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 542
AVR128DB28/32/48/64
OPAMP - Analog Signal Conditioning
35.5.1
Control A
Name:
Offset:
Reset:
Property:
Bit
7
CTRLA
0x00
0x00
-
6
5
4
3
Access
Reset
2
1
0
ENABLE
R/W
0
Bit 0 – ENABLE Enable OPAMP Peripheral
This bit controls whether the OPAMP peripheral is enabled or not.
Value
Description
0
The OPAMP peripheral is disabled
1
The OPAMP peripheral is enabled
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 543
AVR128DB28/32/48/64
OPAMP - Analog Signal Conditioning
35.5.2
Debug Control
Name:
Offset:
Reset:
Property:
Bit
7
DBGCTRL
0x01
0x00
-
6
5
4
3
Access
Reset
2
1
0
DBGRUN
R/W
0
Bit 0 – DBGRUN Run in Debug Mode
This bit controls whether or not the digital interface of the OPAMP peripheral will continue operation in Debug mode
when the CPU is halted. The analog portions of the OPAMP peripheral will continue to operate.
Value
Description
0
The digital interface of the OPAMP peripheral will not continue operating in Debug mode when the
CPU is halted
1
The digital interface of the OPAMP peripheral will continue operating in Debug mode when the CPU is
halted
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 544
AVR128DB28/32/48/64
OPAMP - Analog Signal Conditioning
35.5.3
Timebase
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
TIMEBASE
0x02
0x01
-
6
5
4
R/W
0
R/W
0
R/W
0
3
TIMEBASE[6:0]
R/W
0
2
1
0
R/W
0
R/W
0
R/W
1
Bits 6:0 – TIMEBASE[6:0] Timebase
This bit field controls the maximum value of a counter that counts CLK_PER cycles to achieve a time interval equal to
or larger than 1 μs. It should be written with one less than the number of CLK_PER cycles that are equal to or larger
than 1 μs. This is used for the internal timing of the warmup and settling times.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 545
AVR128DB28/32/48/64
OPAMP - Analog Signal Conditioning
35.5.4
Power Control
Name:
Offset:
Reset:
Property:
Bit
7
PWRCTRL
0x0F
0x00
-
6
5
4
3
2
1
Access
Reset
0
IRSEL
R/W
0
Bit 0 – IRSEL Input Range Select
This bit selects the op amp input voltage range.
Value
Description
0
The op amp input voltage range is rail-to-rail
1
The op amp input voltage range and power consumption are reduced. See the Electrical
Characteristics section for more information.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 546
AVR128DB28/32/48/64
OPAMP - Analog Signal Conditioning
35.5.5
Op Amp n Control A
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
RUNSTDBY
R/W
0
OPnCTRLA
0x10 + n*0x08 [n=0..2]
0x00
-
6
5
4
3
2
OUTMODE[1:0]
R/W
R/W
0
0
1
EVENTEN
R/W
0
0
ALWAYSON
R/W
0
Bit 7 – RUNSTDBY Run in Standby Mode
This bit controls whether or not the op amp functions in Standby sleep mode.
Value
Description
0
OPn is disabled when in Standby sleep mode, and its output driver is disabled
1
OPn will continue operating as configured in Standby sleep mode
Bits 3:2 – OUTMODE[1:0] Output Mode
This bit field selects the output mode of the output driver.
Value
Name
Description
0x0
OFF
The output driver for OPn is disabled, but this can be overridden by the DRIVEn event
0x1
NORMAL The output driver for OPn is enabled in Normal mode
0x2 Reserved
0x3
Bit 1 – EVENTEN Event Enable
This bit enables event reception and generation.
Value
Description
0
No events are enabled for OPn
1
All events are enabled for OPn
Bit 0 – ALWAYSON Always On
This bit controls whether the op amp is always on or not.
Value
Description
0
OPn is not always on but can be enabled by the ENABLEn event and disabled by the DISABLEn event
1
OPn is always on
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 547
AVR128DB28/32/48/64
OPAMP - Analog Signal Conditioning
35.5.6
Op Amp n Status
Name:
Offset:
Reset:
Property:
Bit
7
OPnSTATUS
0x11 + n*0x08 [n=0..2]
0x00
-
6
5
4
3
2
1
Access
Reset
0
SETTLED
R
0
Bit 0 – SETTLED Op Amp has Settled
This bit is cleared when the op amp is waiting for settling related to enabling or configuration changes.
This bit is set when the allowed settling time is finished.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 548
AVR128DB28/32/48/64
OPAMP - Analog Signal Conditioning
35.5.7
Op Amp n Resistor Ladder Multiplexer
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
R/W
0
OPnRESMUX
0x12 + n*0x08 [n=0..2]
0x00
-
6
MUXWIP[2:0]
R/W
0
5
4
R/W
0
R/W
0
3
MUXBOT[2:0]
R/W
0
2
R/W
0
1
0
MUXTOP[1:0]
R/W
R/W
0
0
Bits 7:5 – MUXWIP[2:0] Multiplexer for Wiper
This bit field selects the resistor ladder wiper (potentiometer) position.
Value
Name
Description
0x0
WIP0
R1 = 15R, R2 = 1R
0x1
WIP1
R1 = 14R, R2 = 2R
0x2
WIP2
R1 = 12R, R2 = 4R
0x3
WIP3
R1 = 8R, R2 = 8R
0x4
WIP4
R1 = 6R, R2 = 10R
0x5
WIP5
R1 = 4R, R2 = 12R
0x6
WIP6
R1 = 2R, R2 = 14R
0x7
WIP7
R1 = 1R, R2 = 15R
Bits 4:2 – MUXBOT[2:0] Multiplexer for Bottom
This bit field selects the analog signal connected to the bottom resistor in the resistor ladder.
Value
Name
Description
0x0
OFF
Multiplexer off
0x1
INP
Positive input pin for OPn
0x2
INN
Negative input pin for OPn
0x3
DAC
DAC output (DAC and DAC output buffer must be enabled)
0x4
LINKOUT
OP[n-1] output(1)
0x5
GND
Ground
Other
Reserved
Note: When selecting LINKOUT for OP0, MUXBOT is connected to the output of OP2.
Bits 1:0 – MUXTOP[1:0] Multiplexer for Top
This bit field selects the analog signal connected to the top resistor in the resistor ladder.
Value
Name
Description
0x0
OFF
Multiplexer off
0x1
OUT
OPn output
0x2
VDD
VDD
Other
Reserved
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 549
AVR128DB28/32/48/64
OPAMP - Analog Signal Conditioning
35.5.8
Op Amp n Input Multiplexer
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
OPnINMUX
0x13 + n*0x08 [n=0..2]
0x00
-
6
R/W
0
5
MUXNEG[2:0]
R/W
0
4
3
R/W
0
2
R/W
0
1
MUXPOS[2:0]
R/W
0
0
R/W
0
Bits 6:4 – MUXNEG[2:0] Multiplexer for Negative Input
This bit field selects which analog signal is connected to the inverting (-) input of OPn.
Value
Name
Description
0x0
INN
Negative input pin for OPn
0x1
WIP
Wiper from OPn’s resistor ladder
0x2
OUT
OPn output (unity gain)
0x3
DAC
DAC output (DAC and DAC output buffer must be enabled)
Other
Reserved
Bits 2:0 – MUXPOS[2:0] Multiplexer for Positive Input
This bit field selects which analog signal is connected to the non-inverting (+) input of OPn.
Value
Name
Description
0x0
INP
Positive input pin for OPn
0x1
WIP
Wiper from OPn’s resistor ladder
0x2
DAC
DAC output (DAC and DAC output buffer must be enabled)
0x3
GND
Ground
0x4
VDDDIV2
VDD/2
0x5
LINKOUT
OP[n-1] output (Setting only available for OP1 and OP2)
0x6
LINKWIP
Wiper from OP0’s resistor ladder (Setting only available for OP2)
Other
Reserved
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 550
AVR128DB28/32/48/64
OPAMP - Analog Signal Conditioning
35.5.9
Op Amp n Settle Timer
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
OPnSETTLE
0x14 + n*0x08 [n=0..2]
0x00
-
6
5
4
R/W
0
R/W
0
R/W
0
3
SETTLE[6:0]
R/W
0
2
1
0
R/W
0
R/W
0
R/W
0
Bits 6:0 – SETTLE[6:0] Settle Timer
This bit field specifies the number of microseconds allowed for the op amp output to settle. This value, together with
the value in the TIMEBASE register, is used by an internal timer to determine when to generate the READYn event
and set the SETTLED flag in the OPnSTATUS register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 551
AVR128DB28/32/48/64
OPAMP - Analog Signal Conditioning
35.5.10 Op Amp n Calibration
Name:
Offset:
Reset:
Property:
Bit
7
OPnCAL
0x15 + n*0x08 [n=0..2]
Fuse readout
-
6
5
4
3
2
1
0
R/W
x
R/W
x
R/W
x
R/W
x
CAL[7:0]
Access
Reset
R/W
x
R/W
x
R/W
x
R/W
x
Bits 7:0 – CAL[7:0] Calibration of Input Offset Voltage
This bit field is a calibration value that adjusts the input offset voltage of the op amp. ‘0x00’ provides the most
negative value of offset adjustment, ‘0x80’ provides no offset adjustment, and ‘0xFF’ provides the most positive
value of offset adjustment.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 552
AVR128DB28/32/48/64
ZCD - Zero-Cross Detector
36.
ZCD - Zero-Cross Detector
36.1
Features
•
•
•
•
•
•
36.2
Detect Zero-Crossings on High-Voltage Alternating Signals
Only One External Resistor Required
The Detector Output is Available on a Pin
The Polarity of the Detector Output can be Inverted
Interrupt Generation on:
– Rising edge
– Falling edge
– Both edges
Event Generation:
– Detector output
Overview
The Zero-Cross Detector (ZCD) detects when an alternating voltage crosses through a threshold voltage near the
ground potential. The threshold is the zero-cross reference voltage, ZCPINV, and the typical value can be found in the
Electrical Specifications section of the peripheral.
The connection from the ZCD input pin (ZCIN) to the alternating voltage must be made through a series currentlimiting resistor (RSERIES). The ZCD applies either a current source or sink to the ZCD input pin to maintain a constant
voltage on the pin, thereby preventing the pin voltage from forward biasing the ESD protection diodes in the device.
When the applied voltage is greater than the reference voltage, the ZCD sinks current. When the applied voltage is
less than the reference voltage, the ZCD sources current.
The ZCD can be used when monitoring an alternating waveform for, but not limited to, the following purposes:
•
•
•
•
Period Measurement
Accurate Long-Term Time Measurement
Dimmer Phase-Delayed Drive
Low-EMI Cycle Switching
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 553
AVR128DB28/32/48/64
ZCD - Zero-Cross Detector
36.2.1
Block Diagram
Figure 36-1. Zero-Cross Detector
Zero-Cross Detector
VPULLUP
Optional
RPULLUP
VDD
ZCIN
RSERIES
External
voltage
source
ZCPINV
RPULLDOWN
+
Optional
INVERT
36.2.2
Controller
logic
CROSS
(INT Req.)
OUT
Signal Description
Signal
Description
Type
ZCIN
Input
Analog
OUT
Output
Digital
36.3
Functional Description
36.3.1
Initialization
For basic operation, follow these steps:
• Configure the desired input pin in the PORT peripheral as an analog pin with digital input buffer disabled.
Internal pull-up and pull-down resistors must also be disabled.
• Optional: Enable the output pin by writing a ‘1’ to the Output Enable (OUTEN) bit in the Control A
(ZCDn.CTRLA) register.
• Enable the ZCD by writing a ‘1’ to the ENABLE bit in ZCDn.CTRLA
After the ZCD is enabled, there is a start-up time during which the output of the ZCD may be invalid. The start-up
time can be determined by referring to the ZCD electrical characteristics for the device.
36.3.2
Operation
36.3.2.1 External Resistor Selection
The ZCD requires a current-limiting resistor in series (RSERIES) with the external voltage source. If the peak amplitude
(VPEAK) of the external voltage source is expected to be stable, the resistor value must be chosen such that an
IZCD_MAX/2 resistor current results in a voltage drop equal to the expected peak voltage. The power rating of the
resistor should be at least the mean square voltage divided by the resistor value. (How to handle a peak voltage that
varies between a minimum (VMINPEAK) and maximum (VMAXPEAK) value is described in the section below on
Handling VPEAK Variations.)
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 554
AVR128DB28/32/48/64
ZCD - Zero-Cross Detector
Equation 36-1. External Resistor
�����
������� =
3 × 10−4
Figure 36-2. External Voltage Source
Rev. 30-000001A
7/18/2017
VMAXPEAK
VMINPEAK
VPEAK
Z CPINV
36.3.2.2 ZCD Logic Output
The STATE flag in the ZCDn.STATUS register indicates whether the input signal is above or below the reference
voltage, ZCPINV. By default, the STATE flag is ‘1’ when the input signal is above the reference voltage and ‘0’ when
the input signal is below the reference voltage. The polarity of the STATE flag can be reversed by writing the INVERT
bit to ‘1’ in the ZCDn.CTRLA register. The INVERT bit will also affect ZCD interrupt polarity.
36.3.2.3 Correction for ZCPINV Offset
The actual voltage at which the ZCD switches is the zero-cross reference voltage. Because this reference voltage is
slightly offset from the ground, the zero-cross event generated by the ZCD will occur either early or late for the true
zero-crossing.
36.3.2.3.1 Correction By Offset Current
When the alternating waveform is referenced to the ground, as shown in the figure below, the zero-cross is detected
too late as the waveform rises and too early as the waveform falls.
Figure 36-3. Sine Wave Referenced to Ground
ZERO-CROSS
DETECTED
TRUE ZERO-CROSS
LATE
ZCPINV
GROUND
TRUE
ZERO-CROSS
ZERO-CROSS
DETECTED
EARLY
When the waveform is referenced to VDD, as shown in the figure below, the zero-cross is detected too late as the
waveform falls and too early as the waveform rises.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 555
AVR128DB28/32/48/64
ZCD - Zero-Cross Detector
Figure 36-4. Sine Wave Referenced to VDD
TRUE
ZERO-CROSSES
VDD
ZCPINV
GROUND
ZERO-CROSS
DETECTED
LATE
ZERO-CROSS
DETECTED
EARLY
The actual offset time can be determined for sinusoidal waveforms of a known frequency � using the equations
shown below.
Equation 36-2. ZCD Event Offset
When the External Voltage source is referenced to ground
������
sin−1 �
����
������� =
2��
When the External Voltage source is referenced to VDD
������� =
��� − ������
�����
sin−1
2��
This offset time can be compensated by adding a pull-up or pull-down biasing resistor to the ZCD input pin. A pull-up
resistor is used when the external voltage source is referenced to ground, as shown in the figure below.
Figure 36-5. External Voltage Source Referenced to Ground
Zero-Cross Detector
VPULLUP
RPULLUP
VDD
ZCIN
RSERIES
External
voltage
source
ZCPINV
+
INVERT
Controller
logic
CROSS
(INT Req.)
OUT
A pull-down resistor is used when the voltage is referenced to VDD, as shown in the figure below.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 556
AVR128DB28/32/48/64
ZCD - Zero-Cross Detector
Figure 36-6. External Voltage Source Referenced to VDD
Zero-Cross Detector
VDD
External
voltage
source
VDD
R
ZCIN
SERIES
-
RPULLDOWN
ZCPINV
+
INVERT
Controller
logic
CROSS
(INT Req.)
OUT
The resistor adds a bias to the ZCD input pin so that the external voltage source must go to zero to pull the pin
voltage to the ZCPINV switching voltage. The pull-up or pull-down value can be determined with the equations shown
below.
Equation 36-3. ZCD Pull-up/Pull-down Resistor
When the External Voltage source is referenced to ground
������� =
������� ������� − ������
������
When the External Voltage source is referenced to VDD
��������� =
������� ������
��� − ������
36.3.2.3.2 Correction by AC Coupling
When the external voltage source is sinusoidal, the effects of the ZCPINV offset can be eliminated by isolating the
external voltage source from the ZCD input pin with a capacitor in series with the current-limiting resistor, as shown in
the figure below.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 557
AVR128DB28/32/48/64
ZCD - Zero-Cross Detector
Figure 36-7. AC Coupling the ZCD
Zero-Cross Detector
VDD
RSERIES
C
ZCIN
External
voltage
source
+
ZCPINV
INVERT
Controller
logic
CROSS
(INT Req.)
OUT
The phase shift resulting from the capacitor will cause the ZCD output to switch in advance of the actual zerocrossing event. The phase shift will be the same for both rising and falling zero-crossings, which can be compensated
for by either delaying the CPU response to the ZCD switch by a timer or other means or selecting a capacitor value
large enough that the phase shift is negligible.
To determine the series resistor and capacitor values for this configuration, start by computing the impedance, Z, to
obtain a peak current of IZCD_MAX/2. Next, select a suitably large non-polarized capacitor and compute its reactance,
XC, at the external voltage source frequency. Finally, compute the series resistor (RSERIES), capacitor peak voltage,
and phase shift by using the formulas shown below.
When this technique is used, and the input signal is not present, the ZCD may oscillate. Oscillation can be prevented
by connecting the ZCD input pin to ground with a high-value resistor such as 200 kΩ, but this resistor will introduce
an offset in the detection of the zero-cross event.
Equation 36-4. R-C Equations
VPEAK = External voltage source peak voltage
f = External voltage source frequency
C = Series capacitor
R = Series resistor
VC = Peak capacitor voltage
Φ = Capacitor-induced zero-crossing phase advance in radians
TΦ = Time zero-cross event occurs before actual zero-crossing
�=
3 × 10−4
�� =
�=
�����
1
2���
�2 − ��2
�� = �� 3 × 10−4
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 558
AVR128DB28/32/48/64
ZCD - Zero-Cross Detector
Φ = tan −1�
�Φ =
Φ
2��
��
�
Equation 36-5. R-C Calculation Example
���� = 120
����� = ���� × 2 = 169.7
� = 60 ��
� = 0.1 ��
�=
�����
−4
3 × 10
�� =
�=
=
169.7
3 × 10−4
= 565.7 �Ω
1
1
= 26.53 �Ω
=
2���
2� × 60 × 10−7
�2 − ��2 = 565.1 �Ω ��������
�� = 560 �٠����
�� =
��2 + ��2 = 560.6 �Ω
����� =
�����
= 302.7 × 10−6�
��
�� = �� × ����� = 8.0 �
Φ = tan −1�
�Φ =
��
= 0.047�������
�
Φ
= 125.6��
2��
36.3.2.4 Handling VPEAK Variations
If the peak amplitude of the external voltage is expected to vary, the series resistor (RSERIES) must be selected to
keep the ZCD source and sink currents below the absolute maximum rating of ±IZCD_MAX and above a reasonable
minimum range. A general rule of thumb for the ZCD is that the maximum peak voltage should be no more than six
times the minimum peak voltage. To ensure that the maximum current does not exceed ±IZCD_MAX and the minimum
is at least IZCD_MAX/6, compute the series resistance, as shown in the equation below. The compensating pull-up or
pull-down for this series resistance can be determined using the ZCD Pull-up/Pull-down Resistor equations shown
earlier, as the pull-up/pull-down resistor value is independent of the peak voltage.
Equation 36-6. Series Resistor for External Voltage Range
�������� + ��������
������� =
7 × 10−4
36.3.3
Events
The ZCD can generate the following events:
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 559
AVR128DB28/32/48/64
ZCD - Zero-Cross Detector
Table 36-1. ZCD Event Generator
Generator Name
Peripheral
Event
ZCDn
OUT
Description
Event Type
Generating
Clock Domain
Length of Event
ZCD output level
Level
Asynchronous
Determined by
the ZCD output
level
The ZCD has no event inputs. Refer to the Event System (EVSYS) section for more details regarding event types
and Event System configuration.
36.3.4
Interrupts
Table 36-2. Available Interrupt Vectors and Sources
Name
Vector Description Conditions
CROSS ZCD interrupt
Zero-cross detection as configured by INTMODE in ZCDn.INTCTRL and INVERT in
ZCDn.CTRLA
When a ZCD interrupt condition occurs, the CROSSIF flag is set in the Status (ZCDn.STATUS) register.
ZCD interrupts are enabled or disabled by writing to the INTMODE field in the Interrupt Control (ZCDn.INTCTRL)
register.
A ZCD interrupt request is generated when the interrupt source is enabled, and the CROSSIF flag is set. The
interrupt request remains active until the interrupt flag is cleared. See the ZCDn.STATUS register description for
details on how to clear interrupt flags.
36.3.5
Sleep Mode Operation
In Idle sleep mode, the ZCD will continue to operate as normal.
In Standby sleep mode, the ZCD is disabled by default. If the Run in Standby (RUNSTDBY) bit in the Control A
(ZCDn.CTRLA) register is written to ‘1’, the ZCD will continue to operate as normal with interrupt generation, event
generation, and ZCD output on pin even if CLK_PER is not running in Standby sleep mode.
In Power Down sleep mode, the ZCD is disabled, including its output to pin.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 560
AVR128DB28/32/48/64
ZCD - Zero-Cross Detector
36.4
Register Summary
Offset
Name
Bit Pos.
7
6
0x00
0x01
0x02
0x03
CTRLA
Reserved
INTCTRL
STATUS
7:0
RUNSTDBY
OUTEN
36.5
7:0
7:0
5
4
3
INVERT
STATE
2
1
0
ENABLE
INTMODE[1:0]
CROSSIF
Register Description
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 561
AVR128DB28/32/48/64
ZCD - Zero-Cross Detector
36.5.1
Control A
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
RUNSTDBY
R/W
0
CTRLA
0x00
0x00
-
6
OUTEN
R/W
0
5
4
3
INVERT
R/W
0
2
1
0
ENABLE
R/W
0
Bit 7 – RUNSTDBY Run in Standby
Writing this bit to '1' will cause the ZCD to remain active when the device enters Standby sleep mode.
Bit 6 – OUTEN Output Pin Enable
Writing this bit to ‘1’ connects the OUT signal to a supported pin.
Bit 3 – INVERT Invert Enable
Writing this bit to ‘1’ inverts the ZCD output.
Bit 0 – ENABLE ZCD Enable
Writing this bit to '1' enables the ZCD.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 562
AVR128DB28/32/48/64
ZCD - Zero-Cross Detector
36.5.2
Interrupt Control
Name:
Offset:
Reset:
Property:
Bit
7
INTCTRL
0x02
0x00
-
6
5
4
3
Access
Reset
2
1
0
INTMODE[1:0]
R/W
R/W
0
0
Bits 1:0 – INTMODE[1:0] Interrupt Mode
Writing to these bits selects which edge(s) of the ZCD OUT signal will trigger the ZCD interrupt request.
Value
Name
Description
0x0
NONE
No interrupt
0x1
RISING
Interrupt on rising OUT signal
0x2
FALLING
Interrupt on falling OUT signal
0x3
BOTH
Interrupt on both rising and falling OUT signal
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 563
AVR128DB28/32/48/64
ZCD - Zero-Cross Detector
36.5.3
Status
Name:
Offset:
Reset:
Property:
Bit
7
STATUS
0x03
0x00
-
6
Access
Reset
5
4
STATE
R
0
3
2
1
0
CROSSIF
R/W
0
Bit 4 – STATE ZCD State
This bit indicates the current status of the OUT signal from the ZCD. This includes a three-cycle synchronizer delay.
Bit 0 – CROSSIF Cross Interrupt Flag
This is the zero-cross interrupt flag. Writing this bit to ‘1’ will clear the interrupt flag. Writing this bit to ‘0’ will have no
effect.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 564
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
37.
UPDI - Unified Program and Debug Interface
37.1
Features
•
•
•
37.2
UPDI One-Wire Interface for External Programming and On-Chip-Debugging (OCD)
– Uses a dedicated pin of the device for programming
– No GPIO pins occupied during the operation
– Asynchronous half-duplex UART protocol towards the programmer
Programming:
– Built-in error detection and error signature generation
– Override of response generation for faster programming
Debugging:
– Memory-mapped access to device address space (NVM, RAM, I/O)
– No limitation on the device clock frequency
– Unlimited number of user program breakpoints
– Two hardware breakpoints
– Support for advanced OCD features
• Run-time readout of the CPU Program Counter (PC), Stack Pointer (SP) and Status Register (SREG)
for code profiling
• Detection and signalization of the Break/Stop condition in the CPU
• Program flow control for Run, Stop and Reset debug instructions
– Nonintrusive run-time chip monitoring without accessing the system registers
– Interface for reading the result of the CRC check of the Flash on a locked device
Overview
The Unified Program and Debug Interface (UPDI) is a proprietary interface for external programming and OCD of a
device.
The UPDI supports programming of Nonvolatile Memory (NVM) space, Flash, EEPROM, fuses, lock bits, and the
user row. Some memory-mapped registers are accessible only with the correct access privilege enabled (key, lock
bits) and only in the OCD Stopped mode or certain Programming modes. These modes are unlocked by sending the
correct key to the UPDI. See the NVMCTRL - Nonvolatile Memory Controller section for programming via the NVM
controller and executing NVM controller commands.
The UPDI is partitioned into three separate protocol layers: the UPDI Physical (PHY) Layer, the UPDI Data Link (DL)
Layer and the UPDI Access (ACC) Layer. The default PHY layer handles bidirectional UART communication over the
UPDI pin line towards a connected programmer/debugger and provides data recovery and clock recovery on an
incoming data frame in the One-Wire Communication mode. Received instructions and corresponding data are
handled by the DL layer, which sets up the communication with the ACC layer based on the decoded instruction.
Access to the system bus and memory-mapped registers is granted through the ACC layer.
Programming and debugging are done through the PHY layer, which is a one-wire UART based on a half-duplex
interface using a dedicated pin for data reception and transmission. The clocking of the PHY layer is done by a
dedicated internal oscillator.
The ACC layer is the interface between the UPDI and the connected bus matrix. This layer grants access via the
UPDI interface to the bus matrix with memory-mapped access to system blocks such as memories, NVM, and
peripherals.
The Asynchronous System Interface (ASI) provides direct interface access to select features in the OCD, NVM, and
System Management systems. This gives the debugger direct access to system information without requesting bus
access.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 565
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
37.2.1
Block Diagram
Figure 37-1. UPDI Block Diagram
ASI
Memories
UPDI Pin
(RX/TX Data)
UPDI
Physical
layer
UPDI
Access
layer
Bus Matrix
UPDI Controller
NVM
Peripherals
ASI Access
37.2.2
OCD
NVM
Controller
System
Management
ASI Internal Interfaces
Clocks
The PHY layer and the ACC layer can operate on different clock domains. The PHY layer clock is derived from the
dedicated internal oscillator, and the ACC layer clock is the same as the peripheral clock. There is a synchronization
boundary between the PHY and the ACC layer, which ensures correct operation between the clock domains. The
UPDI clock output frequency is selected through the ASI, and the default UPDI clock start-up frequency is 4 MHz
after enabling or resetting the UPDI. The UPDI clock frequency can be changed by writing to the UPDI Clock Divider
Select (UPDICLKDIV) bit field in the ASI Control A (UPDI.ASI_CTRLA) register.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 566
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
Figure 37-2. UPDI Clock Domains
ASI
SYNCH
UPDI Controller
UPDI
Physical
layer
Clock
Controller
37.2.3
Clock
Controller
CLK_UPDI
CLK_UPDI
source
~
UPDI
Access
layer
CLK_PER
CLK_PER
UPDICLKDIV
~
Physical Layer
The PHY layer is the communication interface between a connected programmer/debugger and the device. The main
features of the PHY layer can be summarized as follows:
• Dedicated pin on the device with no other function
• Support for UPDI One-Wire Asynchronous mode, using half-duplex UART communication on the UPDI pin
• Internal baud detection, clock and data recovery on the UART frame
• Error detection (parity, clock recovery, frame, system errors)
• Transmission response generation (ACK)
• Generation of error signatures during operation
• Guard time control
37.2.4
Pinout Description
The following table shows the functionality of the pin used by the UPDI. See the I/O Multiplexing section in the Device
Data Sheet for more information about the UPDI physical pin.
37.3
37.3.1
Function
Pin Name
UPDI
UPDI
Functional Description
Principle of Operation
The communication through the UPDI is based on standard UART communication, using a fixed frame format, and
automatic baud rate detection for clock and data recovery. In addition to the data frame, several control frames are
important to the communication: DATA, IDLE, BREAK, SYNCH, ACK.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 567
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
Figure 37-3. Supported UPDI Frame Formats
DATA
St
0
1
2
3
4
5
6
7
P
S1 S2
P
S1 S2
P
S1 S2
IDLE
BREAK
SYNCH (0x55)
St
Synch Part
End_synch
ACK (0x40)
St
Frame
Description
DATA
A DATA frame consists of one Start (St) bit which is always low, eight Data bits, one Parity (P) bit for even
parity and two Stop (S1 and S2) bits which are always high. If the Parity bit or Stop bits have an incorrect
value, an error will be detected and signalized by the UPDI. The parity bit-check in the UPDI can be
disabled by writing to the Parity Disable (PARD) bit in the Control A (UPDI.CTRLA) register, in which case
the parity generation from the debugger is ignored.
IDLE
This is a special frame that consists of 12 high bits. This is the same as keeping the transmission line in
an Idle state.
BREAK
This is a special frame that consists of 12 low bits. It is used to reset the UPDI back to its default state
and is typically used for error recovery.
SYNCH The SYNCH frame is used by the Baud Rate Generator to set the baud rate for the coming transmission.
A SYNCH character is always expected by the UPDI in front of every new instruction, and after a
successful BREAK has been transmitted.
ACK
The ACK frame is transmitted from the UPDI whenever an ST or an STS instruction has successfully
crossed the synchronization boundary and gained bus access. When an ACK is received by the
debugger, the next transmission can start.
37.3.1.1 UPDI UART
The communication is initiated from the master debugger/programmer side, and every transmission must start with a
SYNCH character, which the UPDI can use to recover the transmission baud rate and store this setting for the
incoming data. The baud rate set by the SYNCH character will be used for both reception and transmission of the
subsequent instruction and data bytes. See the 37.3.3 UPDI Instruction Set section for details on when the next
SYNCH character is expected in the instruction stream.
There is no writable Baud Rate register in the UPDI, so the baud rate sampled from the SYNCH character is used for
data recovery when sampling the data byte.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 568
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
The transmission baud rate of the PHY layer is related to the selected UPDI clock, which can be adjusted by writing
to the UPDI Clock Divider Select (UPDICLKDIV) bit field in the ASI Control A (UPDI.ASI_CTRLA) register. The
receive and transmit baud rates are always the same within the accuracy of the auto-baud.
Table 37-1. Recommended UART Baud Rate Based on UPDICLKDIV Setting
UPDICLKDIV[1:0]
Max. Recommended Baud Rate
Min. Recommended Baud Rate
0x0 (32 MHz)
1.6 Mbps
0.600 kbps
0x1 (16 MHz)
0.9 Mbps
0.300 kbps
0x2 (8 MHz)
450 kbps
0.150 kbps
0x3 (4 MHz) - Default
225 kbps
0.075 kbps
The UPDI Baud Rate Generator utilizes fractional baud counting to minimize the transmission error. With the fixed
frame format used by the UPDI, the maximum and recommended receiver transmission error limits can be seen in
the following table:
Table 37-2. Receiver Baud Rate Error
Data + Parity Bits
Rslow
Rfast
Max. Total Error [%]
Recommended Max. RX Error [%]
9
96.39
104.76
+4.76/-3.61
+1.5/-1.5
37.3.1.2 BREAK Character
The BREAK character is used to reset the internal state of the UPDI to the default setting. This is useful if the UPDI
enters an Error state due to a communication error or when the synchronization between the debugger and the UPDI
is lost.
To ensure that a BREAK is successfully received by the UPDI in all cases, the debugger must send two consecutive
BREAK characters. The first BREAK will be detected if the UPDI is in Idle state and will not be detected if it is sent
while the UPDI is receiving or transmitting (at a very low baud rate). However, this will cause a frame error for the
reception (RX) or a contention error for the transmission (TX), and abort the ongoing operation. The UPDI will then
detect the next BREAK successfully.
Upon receiving a BREAK, the UPDI oscillator setting in the ASI Control A (UPDI.ASI_CTRLA) register is reset to the
4 MHz default UPDI clock selection. This changes the baud rate range of the UPDI, according to Table 37-1.
37.3.1.2.1 BREAK in One-Wire Mode
In One-Wire mode, the programmer/debugger and UPDI can be totally out of synch, requiring a worst-case length for
the BREAK character to be sure that the UPDI can detect it. Assuming the slowest UPDI clock speed of 4 MHz (250
ns), the maximum length of the 8-bit SYNCH pattern value that can be contained in 16 bits is
65535 × 250 ns = 16.4 ms/byte = 16.4 ms/8 bits = 2.05 ms/bit .
This gives a worst-case BREAK frame duration of 2.05 ms*12 bits ≈ 24.6 ms for the slowest prescaler setting. When
the prescaler setting is known, the time of the BREAK frame can be relaxed according to the values from the next
table:
Table 37-3. Recommended BREAK Character Duration
UPDICLKDIV[1:0]
Recommended BREAK Character Duration
0x0 (32 MHz)
3.075 ms
0x1 (16 MHz)
6.15 ms
0x2 (8 MHz)
12.30 ms
0x3 (4 MHz)
24.60 ms
37.3.1.3 SYNCH Character
The SYNCH character has eight bits and follows the regular UPDI frame format. It has a fixed data bit value of
‘0x55’. The SYNCH character has two main purposes:
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 569
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
1.
2.
It acts as the enabling character for the UPDI after a disable.
It is used by the Baud Rate Generator to set the baud rate for the subsequent transmission. If an invalid
SYNCH character is sent, the next transmission will not be sampled correctly.
37.3.1.3.1 SYNCH in One-Wire Mode
The SYNCH character is used before each new instruction. When using the REPEAT instruction, the SYNCH
character is expected only before the first instruction after REPEAT.
The SYNCH is a known character which, through its property of toggling for each bit, allows the UPDI to measure
how many UPDI clock cycles are needed to sample the 8-bit SYNCH pattern. The information obtained through the
sampling is used to provide Asynchronous Clock Recovery and Asynchronous Data Recovery on reception, and to
keep the baud rate of the connected programmer when doing transmit operations.
37.3.2
Operation
The UPDI must be enabled before the UART communication can start.
37.3.2.1 UPDI Enabling
The devices have a dedicated UPDI pin with no other function. The enable sequence for the UPDI is device
independent and is described in the following paragraphs.
37.3.2.1.1 One-Wire Enable
The UPDI pin has a constant pull-up enable, and by driving the UPDI pin low for more than 200 ns, a connected
programmer will initiate the start-up sequence.
The negative edge transition will cause an edge detector (located in the high-voltage domain if it is in a Multi-Voltage
System) to start driving the UPDI pin low, so when the programmer releases the line, it will stay low until the
requested UPDI oscillator is ready. The expected arrival time for the clock will depend on the oscillator
implementation regarding the accuracy, overshoot and readout of the oscillator calibration. For a Multi-Voltage
System, the line will be driven low until the regulator is at the correct level, and the system is powered up with the
selected oscillator ready and stable. The programmer must poll the UPDI pin after releasing it the first time to detect
when the pin transitions to high again. This transition means that the edge detector has released the pin (pull-up),
and the UPDI can receive a SYNCH character. Upon successful detection of the SYNCH character, the UPDI is
enabled and will prepare for the reception of the first instruction.
The enable transmission sequence is shown in the next figure, where the active driving periods for the programmer
and edge detector are included. The “UPDI pin” waveform shows the pin value at any given time.
The delay given for the edge detector active drive period is a typical start-up time waiting for 256 cycles on a 32 MHz
oscillator + the calibration readout. Refer to the Electrical Characteristics section for details on the expected start-up
times.
Note: The first instruction issued after the initial enable SYNCH does not need an extra SYNCH to be sent because
the enable sequence SYNCH sets up the Baud Rate Generator for the first instruction.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 570
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
Figure 37-4. UPDI Enable Sequence
1
Drive low from the debugger to request the UPDI clock.
2
UPDI clock ready; Communication channel ready.
1
UPDI Pin
St
D0
D1
D2
UPDI.txd
D4
D5
D6
D7
Sp
SYNC (0x55)
(Auto-baud)
Handshake / BREAK
TRES
UPDI.rxd
D3
(Ignore)
2
Hi-Z
Hi-Z
UPDI.txd = 0
TUPDI
debugger.
UPDI.txd
Hi-Z
Hi-Z
Debugger.txd = 0
TDeb0
Debugger.txd = z.
TDebZ
To avoid the UPDI from staying enabled if an accidental trigger of the edge detector happens, the UPDI will
automatically disable itself and lower its clock request. See the Disable During Start-up section for more details.
37.3.2.2 UPDI Disabling
37.3.2.2.1 Disable During Start-up
During the enable sequence, the UPDI can disable itself in case of an invalid enable sequence. There are two
mechanisms implemented to reset any requests the UPDI has given to the Power Management and set the UPDI to
the disabled state. A new enable sequence must then be initiated to enable the UPDI.
Time-Out Disable
When the start-up negative edge detector releases the pin after the UPDI has received its clock, or when the
regulator is stable and the system has power in a Multi-Voltage system, the default pull-up drives the UPDI pin high. If
the programmer does not detect that the pin is high, and does not initiate a transmission of the SYNCH character
within 16.4 ms at 4 MHz UPDI clock after the UPDI has released the pin, the UPDI will disable itself.
Note: Start-up oscillator frequency is device-dependent. The UPDI will count for 65536 cycles on the UPDI clock
before issuing the time-out.
Incorrect SYNCH pattern
An incorrect SYNCH pattern is detected if the length of the SYNCH character is longer than the number of samples
that can be contained in the UPDI Baud Rate register (overflow), or shorter than the minimum fractional count that
can be handled for the sampling length of each bit. If any of these errors are detected, the UPDI will disable itself.
37.3.2.2.2 UPDI Regular Disable
Any programming or debugging session that does not require any specific operation from the UPDI after
disconnecting the programmer has to be terminated by writing the UPDI Disable (UPDIDIS) bit in the Control B
(UPDI.CTRLB) register, upon which the UPDI will issue a System Reset and disable itself. The Reset will restore the
CPU to the Run state, independent of the previous state. It will also lower the UPDI clock request to the system, and
reset any UPDI KEYs and settings.
If the disable operation is not performed, the UPDI and the oscillator’s request will remain enabled. This causes
increased power consumption for the application.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 571
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
37.3.2.3 UPDI Communication Error Handling
The UPDI contains a comprehensive error detection system that provides information to the debugger when
recovering from an error scenario. The error detection consists of detecting physical transmission errors like parity
error, contention error, and frame error, to more high-level errors like access time-out error. See the UPDI Error
Signature (PESIG) bit field in the Status B (UPDI.STATUSB) register for an overview of the available error signatures.
Whenever the UPDI detects an error, it will immediately enter an internal Error state to avoid unwanted system
communication. In the Error state, the UPDI will ignore all incoming data requests, except when a BREAK character
is received. The following procedure must always be applied when recovering from an Error condition.
1. Send a BREAK character. See the 37.3.1.2 BREAK Character section for recommended BREAK character
handling.
2. Send a SYNCH character at the desired baud rate for the next data transfer.
3. Execute a Load Control Status (LDCS) instruction to read the UPDI Error Signature (PESIG) bit field in the
Status B (UPDI.STATUSB) register and get the information about the occurred error.
4. The UPDI has now recovered from the Error state and is ready to receive the next SYNCH character and
instruction.
37.3.2.4 Direction Change
To ensure correct timing for a half-duplex UART operation, the UPDI has a built-in guard time mechanism to relax the
timing when changing direction from RX to TX mode. The guard time is represented by Idle bits inserted before the
next Start bit of the first response byte is transmitted. The number of Idle bits can be configured through the Guard
Time Value (GTVAL) bit field in the Control A (UPDI.CTRLA) register. The duration of each Idle bit is given by the
baud rate used by the current transmission.
Figure 37-5. UPDI Direction Change by Inserting Idle Bits
RX Data Frame
St
RX Data Frame
Dir Change
P
Data from
debugger to UPDI
S1
S2
IDLE bits
TX Data Frame
St
G uard Tim e #
IDLE bits inserted
TX Data Frame
P
S1
S2
Data from UPDI to
debugger
The UPDI guard time is the minimum Idle time that the connected debugger will experience when waiting for data
from the UPDI. The maximum Idle time is the same as time-out. The Idle time before a transmission will be more than
the expected guard time when the synchronization time plus the data bus accessing time is longer than the guard
time.
It is recommended to always use the insertion of minimum two Guard Time bits on the UPDI side, and one guard time
cycle insertion from the debugger side.
37.3.3
UPDI Instruction Set
The communication through the UPDI is based on a small instruction set. These instructions are part of the UPDI
Data Link (DL) layer. The instructions are used to access the UPDI registers, since they are mapped into an internal
memory space called “ASI Control and Status (CS) space”, as well as the memory-mapped system space. All
instructions are byte instructions and must be preceded by a SYNCH character to determine the baud rate for the
communication. See 37.3.1.1 UPDI UART for information about setting the baud rate for the transmission. The
following figure gives an overview of the UPDI instruction set.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 572
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
Figure 37-6. UPDI Instruction Set Overview
Opcode
LDS
STS
0
0
0
1
Size A
0
0
0
0
Opcode
LD
ST
0
0
0
1
Ptr
1
1
Size B
Size A/B
0
STCS
1
1
0
1
0
0
0
0
0
LDS
0
0
1
LD
0
1
0
STS
0
1
1
ST
1
0
0
LDCS (LDS Control/Status)
1
0
1
REPEAT
1
1
0
STCS (STS Control/Status)
1
1
1
KEY
Size A - Address Size
0
CS Address
LDCS
OPCODE
0
0
0
0
Byte - can address 0-255 B
0
1
Word (2 Bytes) - for memories up to 64 KB in size
1
0
3 Bytes - for memories above 64 KB in size
1
1
Reserved
Ptr - Pointer Access
0
0
* (ptr)
0
1
* (ptr++)
1
0
ptr
1
1
Reserved
Size B - Data Size
Size B
REPEAT
1
0
1
0
0
0
SIB
KEY
1
1
1
0
Size C
0
0
0
Byte
0
1
Word (2 Bytes)
1
0
Reserved
1
1
Reserved
CS Address (CS - Control/Status reg.)
Size C - Key Size
0
0
64 bits (8 Bytes)
0
1
128 bits (16 Bytes)
1
0
Reserved
1
1
Reserved
SIB - System Information Block Sel.
0
Receive KEY
1
Send SIB
37.3.3.1 LDS - Load Data from Data Space Using Direct Addressing
The LDS instruction is used to load data from the system bus into the PHY layer shift register for serial readout. The
LDS instruction is based on direct addressing, and the address must be given as an operand to the instruction for the
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 573
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
data transfer to start. The maximum supported size for the address and data is 32 bits. The LDS instruction supports
repeated memory access when combined with the REPEAT instruction.
After issuing the LDS instruction, the number of desired address bytes, as indicated by the Size A field followed by
the output data size, which is selected by the Size B field, must be transmitted. The output data is issued after the
specified Guard Time (GT). When combined with the REPEAT instruction, the address must be sent in for each
iteration of the repeat, meaning after each time the output data sampling is done. There is no automatic address
increment when using REPEAT with LDS, as it uses a direct addressing protocol.
Figure 37-7. LDS Instruction Operation
OPCODE
Size A
Size B
Size A - Address Size
0
L DS
0
0
0
0
0
0
1
Word (2 Bytes) - for memories up to 64 KB in size
1
0
3 Bytes - for memories above 64 KB in size
1
1
Reserved
Byte - can address 0-255 B
Size B - Data Size
0
0
Byte
0
1
Word (2 Bytes)
1
0
Reserved
1
1
Reserved
ADDRESS_SIZE
Synch
(0x55)
LDS
Adr_0
RX
Adr_n
Data_0
Data_n
TX
ΔGT
When the instruction is decoded, and the address byte(s) are received as dictated by the decoded instruction, the DL
layer will synchronize all required information to the ACC layer, which will handle the bus request and synchronize
data buffered from the bus back again to the DL layer. This will create a synchronization delay that must be taken into
consideration upon receiving the data from the UPDI.
37.3.3.2 STS - Store Data to Data Space Using Direct Addressing
The STS instruction is used to store data that are shifted serially into the PHY layer shift register to the system bus
address space. The STS instruction is based on direct addressing, and the address must be given as an operand to
the instruction for the data transfer to start. The address is the first set of operands, and data are the second set. The
size of the address and data operands are given by the size fields presented in the figure below. The maximum size
for both address and data is 32 bits.
The STS supports repeated memory access when combined with the REPEAT instruction.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 574
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
Figure 37-8. STS Instruction Operation
OPCODE
0
STS
1
Size A
0
Size B
Size A - Address Size
0
0
0
Byte - can address 0-255 B
0
1
Word (2 Bytes) - for memories up to 64 KB in size
1
0
3 Bytes - for memories above 64 KB in size
1
1
Reserved
Size B - Data Size
0
0
Byte
0
1
Word (2 Bytes)
1
0
Reserved
1
1
Reserved
ADDRESS_SIZE
Synch
(0x55)
STS
Adr_0
DATA_SIZE
Adr_n
Data_0
RX
Data_n
ACK
ΔGT
ACK
TX
ΔGT
The transfer protocol for an STS instruction is depicted in the above figure, following this sequence:
1.
2.
3.
The address is sent.
An Acknowledge (ACK) is sent back from the UPDI if the transfer was successful.
The number of bytes, as specified in the STS instruction, is sent.
4.
A new ACK is received after the data have been successfully transferred.
37.3.3.3 LD - Load Data from Data Space Using Indirect Addressing
The LD instruction is used to load data from the data space and into the PHY layer shift register for serial readout.
The LD instruction is based on indirect addressing, which means that the Address Pointer in the UPDI needs to be
written before the data space read access. Automatic pointer post-increment operation is supported and is useful
when the LD instruction is utilized with the REPEAT instruction. It is also possible to do an LD from the UPDI Pointer
register. The maximum supported size for address and data load is 32 bits.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 575
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
Figure 37-9. LD Instruction Operation
OPCODE
LD
0
0
Ptr
1
Size A/B
Ptr - Pointer Access
0
Size A - Address Size
Synch
(0x55)
0
0
* ( ptr)
0
1
* (ptr++)
1
0
ptr
1
1
Reserved
Size B - Data Size
0
0
Byte - can address 0-255 B
0
0
Byte
0
1
Word (2 Bytes) - for memories up to 64 KB in size
0
1
Word (2 Bytes)
1
0
3 Bytes - for memories above 64 KB in size
1
0
Reserved
1
1
Reserved
1
1
Reserved
LD
DATA_SIZE
Data_0
RX
Data_n
TX
ΔGT
The figure above shows an example of a typical LD sequence, where the data are received after the Guard Time
(GT) period. Loading data from the UPDI Pointer register follows the same transmission protocol.
For the LD instruction from the data space, the pointer register must be set up by using an ST instruction to the UPDI
Pointer register. After the ACK has been received on a successful Pointer register write, the LD instruction must be
set up with the desired DATA SIZE operands. An LD to the UPDI Pointer register is done directly with the LD
instruction.
37.3.3.4 ST - Store Data from UPDI to Data Space Using Indirect Addressing
The ST instruction is used to store data from the UPDI PHY shift register to the data space. The ST instruction is
used to store data that are shifted serially into the PHY layer. The ST instruction is based on indirect addressing,
which means that the Address Pointer in the UPDI needs to be written before the data space. The automatic pointer
post-increment operation is supported and is useful when the ST instruction is utilized with the REPEAT instruction.
The ST instruction is also used to store the UPDI Address Pointer into the Pointer register. The maximum supported
size for storing address and data is 32 bits.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 576
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
Figure 37-10. ST Instruction Operation
OPCODE
ST
0
1
Ptr
1
Size A/B
Ptr - Pointer Access
0
0
0
* ( ptr)
0
1
* ( ptr++)
1
0
ptr
1
1
Reserved
Size A - Address Size
Size B - Data Size
0
0
Byte - can address 0-255 B
0
0
1 Byte
0
1
Word (2 Bytes) - for memories up to 64 KB in size
0
1
Word (2 Bytes)
1
0
3 Bytes - for memories above 64 KB in size
1
0
Reserved
1
1
Reserved
1
1
Reserved
ADDRESS_SIZE
Synch
(0x55)
ST
ADR_0
ADR_n
RX
ACK
TX
ΔGT
BLOCK_SIZE
Synch
(0x55)
ST
Data_0
RX
Data_n
ACK
TX
ΔGT
The figure above gives an example of an ST instruction to the UPDI Pointer register and the storage of regular data.
A SYNCH character is sent before each instruction. In both cases, an Acknowledge (ACK) is sent back by the UPDI if
the ST instruction was successful.
To write the UPDI Pointer register, the following procedure has to be followed:
1. Set the PTR field in the ST instruction to signature 0x2.
2.
3.
Set the address size (Size A) field to the desired address size.
After issuing the ST instruction, send Size A bytes of address data.
4.
Wait for the ACK character, which signifies a successful write to the Address register.
After the Address register is written, sending data is done in a similarly:
1. Set the PTR field in the ST instruction to signature 0x0 to write to the address specified by the UPDI Pointer
register. If the PTR field is set to 0x1, the UPDI pointer is automatically updated to the next address according
to the data size Size B field of the instruction after the write is executed.
2. Set the Size B field in the instruction to the desired data size.
3. After sending the ST instruction, send Size B bytes of data.
4.
Wait for the ACK character, which signifies a successful write to the bus matrix.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 577
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
When used with the REPEAT instruction, it is recommended to set up the Address register with the start address for
the block to be written and use the Pointer Post Increment register to automatically increase the address for each
repeat cycle. When using the REPEAT instruction, the data frame of Size B data bytes can be sent after each
received ACK.
37.3.3.5 LDCS - Load Data from Control and Status Register Space
The LDCS instruction is used to load serial readout data from the UPDI Control and the Status register space located
in the DL layer into the PHY layer shift register. The LDCS instruction is based on direct addressing, where the
address is part of the instruction operands. The LDCS instruction can access only the UPDI CS register space. This
instruction supports only byte access, and the data size is not configurable.
Figure 37-11. LDCS Instruction Operation
OPCODE
LDCS
1
0
CS Address
0
0
CS Address (CS - Control/Status reg.)
Synch
(0x55)
LDCS
RX
Data
TX
Δgt
The figure above shows a typical example of LDCS data transmission. A data byte from the LDCS is transmitted from
the UPDI after the guard time is completed.
37.3.3.6 STCS (Store Data to Control and Status Register Space)
The STCS instruction is used to store data to the UPDI Control and Status register space. Data are shifted in serially
into the PHY layer shift register and written as a whole byte to a selected CS register. The STCS instruction is based
on direct addressing, where the address is part of the instruction operand. The STCS instruction can access only the
internal UPDI register space. This instruction supports only byte access, and the data size is not configurable.
Figure 37-12. STCS Instruction Operation
OPCODE
STCS
1
1
CS Address
0
CS Address ( CS - Control/Status reg.)
0
Synch
(0x55)
STCS
Data
RX
TX
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 578
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
The figure above shows the data frame transmitted after the SYNCH character and the instruction frames. The STCS
instruction byte can be immediately followed by the data byte. There is no response generated from the STCS
instruction, as is the case for the ST and STS instructions.
37.3.3.7 REPEAT - Set Instruction Repeat Counter
The REPEAT instruction is used to store the repeat count value into the UPDI Repeat Counter register on the DL
layer. When instructions are used with REPEAT, the protocol overhead for SYNCH and instruction frame can be
omitted on all instructions except the first instruction after the REPEAT is issued. REPEAT is most useful for memory
instructions (LD, ST, LDS, STS), but all instructions can be repeated, except for the REPEAT instruction itself.
The DATA_SIZE operand field refers to the size of the repeat value. Only up to 255 repeats are supported. The
instruction loaded directly after the REPEAT instruction will be issued for RPT_0 + 1 times. If the Repeat Counter
register is ‘0’, the instruction will run just once. An ongoing repeat can be aborted only by sending a BREAK
character.
Figure 37-13. REPEAT Instruction Operation used with ST Instruction
OPCODE
REPEAT
1
0
Size B
1
0
0
0
Size B - Data Size
0
0
1 Byte
0
1
Word (2 Bytes)
1
0
Reserved
1
1
Reserved
REPEAT_SIZE
Synch
(0x55)
REPEAT
RPT_0
Repeat Number of Blocks of Data_SIZE
DATA_SIZE
Synch
(0x55)
ST
(ptr++)
Data_0
Data_n
DATA_SIZE
DATA_SIZE
DataB_1
DataB_n
RX
ACK
Δd
ACK
Δd
Δd
Δd
TX
Δd
The figure above gives an example of repeat operation with an ST instruction using pointer post-increment operation.
After the REPEAT instruction is sent with RPT_0 = n, the first ST instruction is issued with SYNCH and instruction
frame, while the next n ST instructions are executed by only sending data bytes according to the ST operand
DATA_SIZE, and maintaining the Acknowledge (ACK) handshake protocol.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 579
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
Figure 37-14. REPEAT used with LD Instruction
REPEAT_SIZE
Synch
(0x55)
Synch
(0x55)
RPT_0
REPEAT
RPT_1
LD
(ptr++)
RX
Repeat Number of Blocks of DATA_SIZE
DATA_SIZE
DATA_SIZE
DataB_1
DataB_n
TX
ΔGT
For LD, data will come out continuously after the LD instruction. Note the guard time on the first data block.
If using indirect addressing instructions (LD/ST), it is recommended to always use the pointer post-increment option
when combined with REPEAT. The ST/LD instruction is necessary only before the first data block (number of data
bytes determined by DATA_SIZE). Otherwise, the same address will be accessed in all repeated access operations.
For direct addressing instructions (LDS/STS), the address must always be transmitted as specified in the instruction
protocol, before data can be received (LDS) or sent (STS).
37.3.3.8 KEY - Set Activation Key or Send System Information Block
The KEY instruction is used for communicating key bytes to the UPDI or for providing the programmer with a System
Information Block (SIB), opening up for executing protected features on the device. See Table 37-4 for an overview of
functions that are activated by keys. For the KEY instruction, only a 64-bit key size is supported. The maximum
supported size for SIB is 128 bits.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 580
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
Figure 37-15. KEY Instruction Operation
SIB
KEY
1
1
1
0
Size C
0
Size C - Key Size
0
0
64 bits (8 Bytes)
0
1
128 bits (16 Bytes) (SIB only)
1
0
Reserved
1
1
Reserved
SIB - System Information Block Sel.
0
Send KEY
1
Receive SIB
KEY_SIZE
Synch
(0x55)
KEY
KEY_0
KEY_n
RX
TX
Synch
(0x55)
RX
KEY
SIB_0
Δgt
SIB_n
TX
SIB_SIZE
The figure above shows the transmission of a key and the reception of a SIB. In both cases, the Size C (SIZE_C)
field in the operand determines the number of frames being sent or received. There is no response after sending a
KEY to the UPDI. When requesting the SIB, data will be transmitted from the UPDI according to the current guard
time setting.
37.3.4
CRC Checking of Flash During Boot
Some devices support running a CRC check of the Flash contents as part of the boot process. This check can be
performed even when the device is locked. The result of this CRC check can be read from the ASI_CRC_STATUS
register. Refer to the CRCSCAN section in the device data sheet for more information on this feature.
37.3.5
Inter-Byte Delay
When performing a multi-byte transfer (LD combined with REPEAT), or reading out the System Information Block
(SIB), the output data will come out in a continuous stream. Depending on the application, on the receiver side, the
data might come out too fast, and there might not be enough time for the data to be processed before the next Start
bit arrives.
The inter-byte delay works by inserting a fixed number of Idle bits for multi-byte transfers. The reason for adding an
inter-byte delay is that there is no guard time inserted when all data is going in the same direction.
The inter-byte delay feature can be enabled by writing a ‘1’ to the Inter-Byte Delay Enable (IBDLY) bit in the Control A
(UPDI.CTRLA) register. As a result, two extra Idle bits will be inserted between each byte to relax the sampling time
for the debugger.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 581
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
Figure 37-16. Inter-Byte Delay Example with LD and RPT
Too Fast Transmission, no Inter-Byte Delay
RX
Debugger
Data
TX
RPT
CNT
LD*(ptr)
GT
Debugger
Processing
D0
SB D1 SB
D0
D2 SB
D1 lost
D3
SB
D3 lost
D2
D4 SB
D5
SB
D4
Data Sampling OK with Inter-Byte Delay
RX
Debugger
Data
TX
RPT
CNT
LD*(ptr)
GT
Debugger
Processing
D0 SB IB
D1
D0
SB IB
D1
D2
SB IB
D2
D3
SB
D3
Notes:
1. GT denotes the guard time insertion.
2. SB is for Stop bit.
3. IB is the inserted inter-byte delay.
4. The rest of the frames are data and instructions.
37.3.6
System Information Block
The System Information Block (SIB) can be read out at any time by setting the SIB bit according to the KEY
instruction from 37.3.3.8 KEY - Set Activation Key or Send System Information Block. The SIB is always accessible
to the debugger, regardless of lock bit settings, and provides a compact form of supplying information about the
device and system parameters for the debugger. The information is vital in identifying and setting up the proper
communication channel with the device. The output of the SIB is interpreted as ASCII symbols. The key size field
must be set to 16 bytes when reading out the complete SIB, and an 8-byte size can be used to read out only the
Family_ID. See the figure below for SIB format description and which data are available at different readout sizes.
Figure 37-17. System Information Block Format
16 8
37.3.7
[Byte][Bits]
[6:0] [55:0]
[7][7:0]
[10:8][23:0]
[13:11][23:0]
[14][7:0]
[15][7:0]
Field Name
Family_ID
Reserved
NVM_VERSION
OCD_VERSION
RESERVED
DBG_OSC_FREQ
Enabling of Key Protected Interfaces
The access to some internal interfaces and features is protected by the UPDI key mechanism. To activate a key, the
correct key data must be transmitted by using the KEY instruction, as described in 37.3.3.8 KEY - Set Activation Key
or Send System Information Block. The table below describes the available keys and the condition required when
doing the operation with the key active.
Table 37-4. Key Activation Overview
Key Name
Description
Chip Erase
Start NVM chip erase.
Clear lock bits
© 2020 Microchip Technology Inc.
Requirements for Operation
-
Preliminary Datasheet
Conditions for Key
Invalidation
UPDI Disable/UPDI Reset
DS40002247A-page 582
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
...........continued
Key Name
Description
Requirements for Operation
Conditions for Key
Invalidation
NVMPROG
Activate NVM
Programming
Lock bits cleared.
ASI_SYS_STATUS.NVMPROG set
Programming done/UPDI
Reset
USERROW-Write
Program the user row
on the locked device
Lock bits set.
Write to key Status bit/
ASI_SYS_STATUS.UROWPROG set UPDI Reset
The table below gives an overview of the available key signatures that must be shifted in to activate the interfaces.
Table 37-5. Key Activation Signatures
Key Name
Key Signature (LSB Written First)
Size
Chip Erase
0x4E564D4572617365
64 bits
NVMPROG
0x4E564D50726F6720
64 bits
USERROW-Write
0x4E564D5573267465
64 bits
37.3.7.1 Chip Erase
The following steps must be followed to issue a chip erase:
1. Enter the Chip Erase key by using the KEY instruction. See Table 37-5 for the CHIPERASE signature.
2.
3.
4.
5.
6.
7.
8.
Optional: Read the Chip Erase (CHIPERASE) bit in the ASI Key Status (UPDI.ASI_KEY_STATUS) register to
see that the key is successfully activated.
Write the signature to the Reset Request (RSTREQ) bit in the ASI Reset Request (UPDI.ASI_RESET_REQ)
register. This will issue a System Reset.
Write 0x00 to the ASI Reset Request (UPDI.ASI_RESET_REQ) register to clear the System Reset.
Read the NVM Lock Status (LOCKSTATUS) bit from the ASI System Status (UPDI.ASI_SYS_STATUS)
register.
The chip erase is done when LOCKSTATUS bit is ‘0’. If the LOCKSTATUS bit is ‘1’, return to step 5.
Check the Chip Erase Key Failed (ERASE_FAILED) bit in the ASI System Status (UPDI.ASI_SYS_STATUS)
register to verify if the chip erase was successful.
If the ERASE_FAILED bit is ‘0’, the chip erase was successful.
After a successful chip erase, the lock bits will be cleared, and the UPDI will have full access to the system. Until the
lock bits are cleared, the UPDI cannot access the system bus, and only CS-space operations can be performed.
CAUTION
During chip erase, the BOD is forced in ON state by writing to the Active (ACTIVE) bit field from the
Control A (BOD.CTRLA) register and uses the BOD Level (LVL) bit field from the BOD Configuration
(FUSE.BODCFG) fuse and the BOD Level (LVL) bit field from the Control B (BOD.CTRLB) register. If the
supply voltage VDD is below that threshold level, the device is unavailable until VDD is increased
adequately. See the BOD section for more details.
37.3.7.2 NVM Programming
If the device is unlocked, it is possible to write directly to the NVM Controller or to the Flash memory using the UPDI.
This will lead to unpredictable code execution if the CPU is active during the NVM programming. To avoid this, the
following NVM Programming sequence has to be executed.
1.
2.
3.
Follow the chip erase procedure, as described in 37.3.7.1 Chip Erase. If the part is already unlocked, this
point can be skipped.
Enter the NVMPROG key by using the KEY instruction. See Table 37-5 for the NVMPROG signature.
Optional: Read the NVM Programming Key Status (NVMPROG) bit from the ASI Key Status
(UPDI.KEY_STATUS) register to see if the key has been activated.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 583
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
4.
5.
6.
7.
Write the signature to the Reset Request (RSTREQ) bit in the ASI Reset Request (UPDI.ASI_RESET_REQ)
register. This will issue a System Reset.
Write 0x00 to the ASI Reset Request (UPDI.ASI_RESET_REQ) register to clear the System Reset.
Read the NVM Programming Key Status (NVMPROG) bit from the ASI System Status
(UPDI.ASI_SYS_STATUS) register.
NVM Programming can start when the NVMPROG bit is ‘1’. If the NVMPROG bit is ‘0’, return to step 6.
8.
9.
Write data to NVM through the UPDI.
Write the signature to the Reset Request (RSTREQ) bit in the ASI Reset Request (UPDI.ASI_RESET_REQ)
register. This will issue a System Reset.
10. Write 0x00 to the ASI Reset Request (UPDI.ASI_RESET_REQ) register to clear the System Reset.
11. Programming is complete.
37.3.7.3 User Row Programming
The User Row Programming feature allows programming new values to the user row (USERROW) on a locked
device. To program with this functionality enabled, the following sequence must be followed:
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
Enter the USERROW-Write key located in Table 37-5 by using the KEY instruction. See Table 37-5 for the
USERROW-Write signature.
Optional: Read the User Row Write Key Status (UROWWRITE) bit from the ASI Key Status
(UPDI.ASI_KEY_STATUS) register to see if the key has been activated.
Write the signature to the Reset Request (RSTREQ) bit in the ASI Reset Request (UPDI.ASI_RESET_REQ)
register. This will issue a System Reset.
Write 0x00 to the ASI Reset Request (UPDI.ASI_RESET_REQ) register to clear the System Reset.
Read the Start User Row Programming (UROWPROG) bit from the ASI System Status
(UPDI.ASI_SYS_STATUS) register.
User Row Programming can start when the UROWPROG bit is ‘1’. If UROWPROG is ‘0’, return to step 5.
The data to be written to the User Row must first be written to a buffer in the RAM. The writable area in the
RAM has a size of 32 bytes, and it is only possible to write user row data to the first 32 byte addresses of the
RAM. Addressing outside this memory range will result in a nonexecuted write. The data will map 1:1 with the
user row space when the data is copied into the user row upon completion of the Programming sequence.
When all user row data has been written to the RAM, write the User Row Programming Done (UROWDONE)
bit in the ASI System Control A (UPDI.ASI_SYS_CTRLA) register.
Read the Start User Row Programming (UROWPROG) bit from the ASI System Status
(UPDI.ASI_SYS_STATUS) register.
The User Row Programming is completed when UROWPROG bit is ‘0’. If UROWPROG bit is ‘1’, return to step
9.
Write to the User Row Write Key Status (UROWWRITE) bit in the ASI Key Status (UPDI.ASI_KEY_STATUS)
register.
Write the signature to the Reset Request (RSTREQ) bit in the ASI Reset Request (UPDI.ASI_RESET_REQ)
register. This will issue a System Reset.
Write 0x00 to the ASI Reset Request (UPDI.ASI_RESET_REQ) register to clear the System Reset.
14. The User Row Programming is complete.
It is not possible to read back data from the RAM in this mode. Only writes to the first 32 bytes of the RAM are
allowed.
37.3.8
Events
The UPDI can generate the following events:
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 584
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
Table 37-6. Event Generators in UPDI
Generator Name
Module
Event
UPDI
SYNCH
Description
SYNCH
character
Event Type
Level
Generating
Clock
Domain
CLK_UPDI
Length of Event
SYNCH char on UPDI pin
synchronized to CLK_UPDI
This event is set on the UPDI clock for each detected positive edge in the SYNCH character, and it is not possible to
disable this event from the UPDI.
The UPDI has no event users.
Refer to the Event System section for more details regarding event types and Event System configuration.
37.3.9
Sleep Mode Operation
The UPDI PHY layer runs independently of all sleep modes, and the UPDI is always accessible for a connected
debugger independent of the device’s sleep state. If the system enters a sleep mode that turns the system clock off,
the UPDI will not be able to access the system bus and read memories and peripherals. When enabled, the UPDI will
request the system clock so that the UPDI always has contact with the rest of the device. Thus, the UPDI PHY layer
clock is unaffected by the sleep mode’s settings. By reading the System Domain in Sleep (INSLEEP) bit in the ASI
System Status (UPDI.ASI_SYS_STATUS) register, it is possible to monitor if the system domain is in a sleep mode.
It is possible to prevent the system clock from stopping when going into a sleep mode, by writing to the Request
System Clock (CLKREQ) bit in the ASI System Control A (UPDI.ASI_SYS_CTRLA) register. If this bit is set, the
system sleep mode state is emulated, and the UPDI can access the system bus and read the peripheral registers
even in the deepest sleep modes.
The CLKREQ bit is by default ‘1’ when the UPDI is enabled, which means that the default operation is keeping the
system clock in ON state during the sleep modes.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 585
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
37.4
Register Summary
Offset
Name
Bit Pos.
0x00
0x01
0x02
0x03
0x04
...
0x06
0x07
0x08
0x09
0x0A
STATUSA
STATUSB
CTRLA
CTRLB
7:0
7:0
7:0
7:0
ASI_KEY_STATUS
ASI_RESET_REQ
ASI_CTRLA
ASI_SYS_CTRLA
7:0
7:0
7:0
7:0
0x0B
ASI_SYS_STATUS
7:0
0x0C
ASI_CRC_STATUS
7:0
37.5
7
6
5
4
3
DTD
NACKDIS
RSD
CCDETDIS
2
1
0
UPDIREV[3:0]
IBDLY
PARD
PESIG[2:0]
GTVAL[2:0]
UPDIDIS
Reserved
UROWWRITE NVMPROG CHIPERASE
RSTREQ[7:0]
UPDICLKDIV[1:0]
UROWDONE
CLKREQ
ERASE_FAIL
ED
SYSRST
INSLEEP
NVMPROG UROWPROG
LOCKSTATUS
CRC_STATUS[2:0]
Register Description
These registers are readable only through the UPDI with special instructions and are not readable through the CPU.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 586
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
37.5.1
Status A
Name:
Offset:
Reset:
Property:
Bit
7
STATUSA
0x00
0x10
-
6
5
4
R
0
R
1
3
2
1
0
UPDIREV[3:0]
Access
Reset
R
0
R
0
Bits 7:4 – UPDIREV[3:0] UPDI Revision
This bit field contains the revision of the current UPDI implementation.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 587
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
37.5.2
Status B
Name:
Offset:
Reset:
Property:
Bit
7
STATUSB
0x01
0x00
-
6
5
4
3
Access
Reset
2
R
0
1
PESIG[2:0]
R
0
0
R
0
Bits 2:0 – PESIG[2:0] UPDI Error Signature
This bit field describes the UPDI error signature and is set when an internal UPDI Error condition occurs. The PESIG
bit field is cleared on a read from the debugger.
Table 37-7. Valid Error Signatures
PESIG[2:0]
Error Type
Error Description
0x0
0x1
0x2
0x3
0x4
0x5
0x6
0x7
No error
Parity error
Frame error
Access Layer Time-Out Error
Clock Recovery error
Bus error
Contention error
No error detected (Default)
Wrong sampling of the Parity bit
Wrong sampling of the Stop bits
UPDI can get no data or response from the Access layer
Wrong sampling of the Start bit
Reserved
Address error or access privilege error
Signalize Driving Contention on the UPDI pin
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 588
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
37.5.3
Control A
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
IBDLY
R/W
0
CTRLA
0x02
0x00
-
6
5
PARD
R/W
0
4
DTD
R/W
0
3
RSD
R/W
0
2
R/W
0
1
GTVAL[2:0]
R/W
0
0
R/W
0
Bit 7 – IBDLY Inter-Byte Delay Enable
Writing a ‘1’ to this bit enables a fixed-length inter-byte delay between each data byte transmitted from the UPDI
when doing multi-byte LD(S). The fixed length is two IDLE bits.
Bit 5 – PARD Parity Disable
Writing a ‘1’ to this bit will disable the parity detection in the UPDI by ignoring the Parity bit. This feature is
recommended to be used only during testing.
Bit 4 – DTD Disable Time-Out Detection
Writing a ‘1’ to this bit will disable the time-out detection on the PHY layer, which requests a response from the ACC
layer within a specified time (65536 UPDI clock cycles).
Bit 3 – RSD Response Signature Disable
Writing a ‘1’ to this bit will disable any response signatures generated by the UPDI. This reduces the protocol
overhead to a minimum when writing large blocks of data to the NVM space. When accessing the system bus, the
UPDI may experience delays. If the delay is predictable, the response signature may be disabled, otherwise loss of
data may occur.
Bits 2:0 – GTVAL[2:0] Guard Time Value
This bit field selects the guard time value that will be used by the UPDI when the transmission direction switches from
RX to TX.
Value
Description
0x0
UPDI guard time: 128 cycles (default)
0x1
UPDI guard time: 64 cycles
0x2
UPDI guard time: 32 cycles
0x3
UPDI guard time: 16 cycles
0x4
UPDI guard time: 8 cycles
0x5
UPDI guard time: 4 cycles
0x6
UPDI guard time: 2 cycles
0x7
Reserved
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 589
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
37.5.4
Control B
Name:
Offset:
Reset:
Property:
Bit
7
CTRLB
0x03
0x00
-
6
Access
Reset
5
4
NACKDIS
R/W
0
3
CCDETDIS
R/W
0
2
UPDIDIS
R/W
0
1
0
Bit 4 – NACKDIS Disable NACK Response
Writing a ‘1’ to this bit disables the NACK signature sent by the UPDI when a System Reset is issued during ongoing
LD(S) and ST(S) operations.
Bit 3 – CCDETDIS Collision and Contention Detection Disable
Writing a ‘1’ to this bit disables the contention detection. Writing a ‘0’ to this bit enables the contention detection.
Bit 2 – UPDIDIS UPDI Disable
Writing a ‘1’ to this bit disables the UPDI PHY interface. The clock request from the UPDI is lowered, and the UPDI is
reset. All the UPDI PHY configurations and keys will be reset when the UPDI is disabled.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 590
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
37.5.5
ASI Key Status
Name:
Offset:
Reset:
Property:
Bit
7
ASI_KEY_STATUS
0x07
0x00
-
6
Access
Reset
5
UROWWRITE
R/W
0
4
NVMPROG
R
0
3
CHIPERASE
R
0
2
1
0
Bit 5 – UROWWRITE User Row Write Key Status
This bit is set to ‘1’ if the UROWWRITE key is successfully decoded. This bit must be written as the final part of the
user row write procedure to correctly reset the programming session.
Bit 4 – NVMPROG NVM Programming Key Status
This bit is set to ‘1’ if the NVMPROG key is successfully decoded. The bit is cleared when the NVM Programming
sequence is initiated, and the NVMPROG bit in ASI_SYS_STATUS is set.
Bit 3 – CHIPERASE Chip Erase Key Status
This bit is set to ‘1’ if the Chip Erase key is successfully decoded. The bit is cleared by the Reset Request issued as
part of the Chip Erase sequence described in the 37.3.7.1 Chip Erase section.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 591
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
37.5.6
ASI Reset Request
Name:
Offset:
Reset:
Property:
ASI_RESET_REQ
0x08
0x00
-
A Reset is signalized to the System when writing the Reset signature to this register.
Bit
Access
Reset
7
6
5
R/W
0
R/W
0
R/W
0
4
3
RSTREQ[7:0]
R/W
R/W
0
0
2
1
0
R/W
0
R/W
0
R/W
0
Bits 7:0 – RSTREQ[7:0] Reset Request
The UPDI will not be reset when issuing a System Reset from this register.
Value
Name
Description
0x00
RUN
Clear Reset condition
0x59
RESET
Normal Reset
Other
Reset condition is cleared
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 592
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
37.5.7
ASI Control A
Name:
Offset:
Reset:
Property:
Bit
7
ASI_CTRLA
0x09
0x03
-
6
5
4
3
Access
Reset
2
1
0
UPDICLKDIV[1:0]
R/W
R/W
1
1
Bits 1:0 – UPDICLKDIV[1:0] UPDI Clock Divider Select
This bit field selects the prescaler setting for the UPDI internal oscillator. The default setting after Reset and enable is
4 MHz.
Value
Description
0x0
32 MHz UPDI clock
0x1
16 MHz UPDI clock
0x2
8 MHz UPDI clock
0x3
4 MHz UPDI clock
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 593
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
37.5.8
ASI System Control A
Name:
Offset:
Property:
Bit
7
ASI_SYS_CTRLA
0x0A
-
6
5
4
3
Access
Reset
2
1
UROWDONE
R/W
0
0
CLKREQ
R/W
0
Bit 1 – UROWDONE User Row Programming Done
This bit must be written when the user row data have been written to the RAM. Writing a ‘1’ to this bit will start the
process of programming the user row data to the Flash.
If this bit is written before the user row data is written to the RAM by the UPDI, the CPU will proceed without the
written data.
This bit is writable only if the USERROW-Write key is successfully decoded.
Bit 0 – CLKREQ Request System Clock
If this bit is written to ‘1’, the ASI is requesting the system clock, independent of the system sleep modes. This makes
it possible for the UPDI to access the ACC layer even if the system is in a sleep mode.
Writing a ‘0’ to this bit will lower the clock request.
This bit is set by default when the UPDI is enabled.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 594
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
37.5.9
ASI System Status
Name:
Offset:
Reset:
Property:
Bit
Access
Reset
7
ASI_SYS_STATUS
0x0B
0x01
-
6
ERASE_FAILE
D
R
0
5
SYSRST
4
INSLEEP
3
NVMPROG
2
UROWPROG
R
0
R
0
R
0
R
0
1
0
LOCKSTATUS
R
1
Bit 6 – ERASE_FAILED Chip Erase Key Failed
This bit is set to ‘1’ if the chip erase has failed. This bit is set to ‘0’ on Reset. A Reset held from the
ASI_RESET_REQ register will also affect this bit.
Bit 5 – SYSRST System Reset Active
When this bit is set to ‘1’, there is an active Reset on the system domain. When this bit is set to ‘0’, the system is not
in the Reset state.
This bit is set to ‘0’ on read.
A Reset held from the ASI_RESET_REQ register will also affect this bit.
Bit 4 – INSLEEP System Domain in Sleep
When this bit is set to ‘1’, the system domain is in Idle or deeper Sleep mode. When this bit is set to ‘0’, the system is
not in any sleep mode.
Bit 3 – NVMPROG Start NVM Programming
When this bit is set to ‘1’, NVM Programming can start from the UPDI.
When the UPDI is done, the system must be reset through the UPDI Reset register.
Bit 2 – UROWPROG Start User Row Programming
When this bit is set to ‘1’, User Row Programming can start from the UPDI.
When the User Row data have been written to the RAM, the UROWDONE bit in the ASI_SYS_CTRLA register must
be written.
Bit 0 – LOCKSTATUS NVM Lock Status
When this bit is set to ‘1’, the device is locked. If a chip erase is done, and the lock bits are set to ‘0’, this bit will be
read as ‘0’.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 595
AVR128DB28/32/48/64
UPDI - Unified Program and Debug Interface
37.5.10 ASI CRC Status
Name:
Offset:
Reset:
Property:
Bit
7
ASI_CRC_STATUS
0x0C
0x00
-
6
5
4
3
Access
Reset
2
R
0
1
CRC_STATUS[2:0]
R
0
0
R
0
Bits 2:0 – CRC_STATUS[2:0] CRC Execution Status
This bit field signalizes the status of the CRC conversion. This bit field is one-hot encoded.
Value
Name
Description
0x0
DISABLED
Not enabled
0x1
BUSY
CRC enabled, busy
0x2
PASS
CRC enabled, done with PASS signature
0x4
FAIL
CRC enabled, done with FAILED signature
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 596
AVR128DB28/32/48/64
Instruction Set Summary
38.
Instruction Set Summary
The instruction set summary can be found as part of the AVR Instruction Set Manual, located at www.microchip.com/
DS40002198. Refer to the CPU version called AVRxt, for details regarding the devices documented in this data
sheet.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 597
AVR128DB28/32/48/64
Electrical Characteristics
39.
Electrical Characteristics
39.1
Disclaimer
All typical values are measured at T = 25°C and VDD = VDDIO2 = 3.0V unless otherwise specified. All minimum and
maximum values are valid across operating temperature and voltage unless otherwise specified.
Typical values given should be considered for design guidance only, and actual part variation around these values is
expected.
39.2
Absolute Maximum Ratings
Stresses beyond those listed in this section may cause permanent damage to the device. This is a stress rating only
and functional operation of the device at these or other conditions beyond those indicated in the operational sections
of this specification is not implied. Exposure to absolute maximum rating conditions for extended periods may affect
device reliability.
Table 39-1. Absolute Maximum Ratings
Parameter
Ambient temperature under bias
Storage temperature
Voltage on Pins With Respect to GND
Condition
Rating
-40 to +125
-65 to +150
Units
°C
°C
•
On the VDD pin
-0.3 to +6.5
V
•
On the VDDIO2 pin
-0.3 to +6.5
V
•
On the RESET pin
-0.3 to (VDD + 0.3)
V
•
On all other pins
-0.3 to (VDD + 0.3)
V
350
120
350
120
350
120
mA
mA
mA
mA
mA
mA
±50
mA
±20
mA
800
mW
Maximum Current
•
On the GND pin(1)
•
On the VDD pin(1)
•
On the VDDIO2 pin(1)
•
On any standard I/O pin
-40°C ≤ TA ≤ +85°C
+85°C < TA ≤ +125°C
-40°C ≤ TA ≤ +85°C
+85°C < TA ≤ +125°C
-40°C ≤ TA ≤ +85°C
+85°C < TA ≤ +125°C
Clamp current, IK (VPIN < 0 or VPIN
> VDD)
Total power dissipation(2)
Note:
1.
2.
39.3
The maximum current rating requires even load distribution across I/O pins. The maximum current rating may
be limited by the device package power dissipation characterizations, see the Thermal Characteristics section
to calculate device specifications.
Power dissipation is calculated as follows: PDIS = VDD × {IDD - Σ IOH} + Σ {(VDD - VOH) × IOH} + Σ (VOI × IOL)
Standard Operating Conditions
The device must operate within the ratings listed in this section for all other electrical characteristics and typical
characteristics of the device to be valid.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 598
AVR128DB28/32/48/64
Electrical Characteristics
Table 39-2. General Operating Conditions
Operating Voltage
VDDMIN ≤ VDD ≤ VDDMAX
Operating Temperature
TA_MIN ≤ TA ≤ TA_MAX
The standard operating conditions for any device, are defined as:
Table 39-3. Standard Operating Conditions
Parameter
VDD — Operating Supply Voltage(1)
Industrial and Extended temperature
Ratings
Units
VDDMIN
VDDMAX
+1.8
+5.5
V
V
TA_MIN
TA_MAX
TA_MIN
TA_MAX
-40
+85
-40
+125
°C
°C
°C
°C
TA — Operating Ambient Temperature Range
Industrial temperature
Extended temperature
Note:
1. Refer to section 39.4.1 Supply Voltage.
39.4
DC Characteristics
39.4.1
Supply Voltage
Table 39-4. Supply Voltage
Symbol
Min.
Typ.
Max.
Units
VDD
1.8
—
5.5
V
VDDIO2
1.62
—
5.5
V
—
—
250
V/ms
1.7
—
—
V
Device in Power-Down mode
VPOR
—
1.6
—
V
BOD disabled(2)
tPOR
—
1
—
μs
BOD disabled(2)
VPORR
—
1.25
—
V
BOD disabled(2)
tPORR
—
2.7
—
μs
BOD disabled(2)
—
V/ms
BOD disabled(2)
Supply
Conditions
Voltage(1)(2)
SVDD
1.8V < VDD < 5.5V
RAM Data Retention(3)
VDR
Power-on Reset
Release(4)
Power-on Reset Rearm(4)
VDD Slope to Ensure Internal Power-on Reset Signal(4)
SVDD
© 2020 Microchip Technology Inc.
0.05
—
Preliminary Datasheet
DS40002247A-page 599
AVR128DB28/32/48/64
Electrical Characteristics
...........continued
Symbol
Min.
Typ.
Max.
Units
Conditions
Notes:
1. During Chip Erase, the Brown-out Detector (BOD) configured with BODLEVEL0 is forced ON. If the supply
voltage VDD is below VBOD for BODLEVEL0, the erase attempt will fail.
2. Refer to section 39.5.2 Reset Controller Specifications for BOD trip point information.
3. This is the limit to which VDD can be lowered in sleep mode without losing RAM data.
4. Refer to Figure 39-1.
Figure 39-1. POR and PORR with Slow Rising VDD
POR
VDD
VPOR
VPORR
SVDD
tPOR
tPORR
Note:
• When POR is low, the device is held in Reset
39.4.2
Power Consumption
Table 39-5. Device Power Consumption
Operating Conditions:
• VDD = VDDIO2 = 3V
• TA = 25°C
• System power consumption measured with peripherals disabled and I/O ports driven low with inputs
disabled
Symbol
IDD
Description
Active power consumption
© 2020 Microchip Technology Inc.
Min. Typ. Max. Units Conditions
—
4.1
—
mA
OSCHF = 24 MHz
—
1.0
—
mA
OSCHF = 4 MHz
—
7.0
—
μA
OSC32K = 32.768 kHz
—
3.8
—
mA
EXTCLK = 24 MHz
—
—
—
μA
EXTCLK = 4 MHz
—
9.0
—
μA
XOSC32K = 32.768 kHz (High Power)
—
7.5
—
μA
XOSC32K = 32.768 kHz (Low Power)
Preliminary Datasheet
DS40002247A-page 600
AVR128DB28/32/48/64
Electrical Characteristics
...........continued
Operating Conditions:
• VDD = VDDIO2 = 3V
• TA = 25°C
• System power consumption measured with peripherals disabled and I/O ports driven low with inputs
disabled
Symbol
IDD_IDLE
Description
Idle power consumption
IDD_STDBY(1) Standby power consumption
IDD_BASE(1)
IRST
Minimum power consumption in
different sleep modes
Reset power consumption
Min. Typ. Max. Units Conditions
—
1.9
—
mA
OSCHF = 24 MHz
—
580
—
μA
OSCHF = 4 MHz
—
4.0
—
μA
OSC32K = 32.768 kHz
—
1.7
—
mA
EXTCLK = 24 MHz
—
—
—
μA
EXTCLK = 4 MHz
—
7.5
—
μA
XOSC32K = 32.768 kHz (High Power)
—
6.0
—
μA
XOSC32K = 32.768 kHz (Low Power)
—
3.2
—
μA
RTC running at 1.024 kHz from
XOSC32K (CL = 9 pF, High Power)
—
1.6
—
μA
RTC running at 1.024 kHz from
XOSC32K (CL = 9 pF, Low Power)
—
1.2
—
μA
RTC running at 1.024 kHz from
OSC32K
—
2
—
µA
Idle mode, all peripherals disabled(2)
—
700
—
nA
Standby mode, all peripherals
disabled(3)
—
700
—
nA
Power-Down mode, all peripherals
disabled(3)
—
170
—
μA
RESET line pulled low
Notes:
1. Single supply mode.
2. PMODE configured to AUTO in the Voltage Regulator Control (SLPCTRL.VREGCTRL) register and 32 kHz
clock source selected.
3. PMODE configured to AUTO in the Voltage Regulator Control (SLPCTRL.VREGCTRL) register.
39.4.3
Peripherals Power Consumption
The table below can be used to calculate the additional current consumption for the different I/O peripherals in the
various operating modes. Some peripherals will request the clock to be enabled when operating in STANDBY. Refer
to the peripheral section for further information.
© 2020 Microchip Technology Inc.
Preliminary Datasheet
DS40002247A-page 601
AVR128DB28/32/48/64
Electrical Characteristics
Table 39-6. Peripherals Power Consumption(1)
Operating Conditions:
• VDD = VDDIO2 = 3V
• TA = 25°C
• OSCHF at 4 MHz used as clock source
• Device in Standby sleep mode
Symbol
Description
IDD_WDT
IDD_MVIO
IDD_VREF
IDD_BOD
Min.
Typ.
Max.
Units
Watchdog Timer (WDT)
—
350
—
nA
Dual supply configuration,
Multi-Voltage I/O (MVIO)
—
500
—
nA
—
—
—
μA
—
—
—
μA
ACREF enabled, VREF =
2.048V
—
—
—
μA
DACREF enabled, VREF =
2.048V
—
20
—
μA
Brown-out Detect (BOD)
continuous, including bandgap
—
1.3
—
μA
BOD sampling @128 Hz,
BODLEVEL3, including
bandgap
—
600
—
nA
Voltage Reference (VREF)
Brown-out Detector (BOD)
Conditions
32 kHz Internal Oscillator
ADC0REF enabled, VREF =
2.048V
BOD sampling @32 Hz,
BODLEVEL3, including
bandgap
IDD_TCA
16-bit Timer/Counter Type A
(TCA)
—
13
—
μA
IDD_TCB
16-bit Timer/Counter Type B
(TCB)
—
10
—
μA
IDD_TCD
12-bit Timer/Counter Type D
(TCD)
—
—
—
μA
IDD_RTC
Real-Time Counter (RTC)
—
700
—
nA
IDD_OSCHF
Internal High-Frequency
Oscillator (OSCHF)
—
150
—
µA
Internal Oscillator running at 4
MHz
IDD_XOSCHF
High Frequency Crystal
Oscillator (XOSCHF)
—
350
—
μA
20 MHz XTAL, CL = 15 pF
—
350
—
nA
—
2.4
—
μA
XOSC32K High Power, CL = 9
pF
—
580
—
nA
XOSC32K Low Power, CL = 9
pF
IDD_OSC32K
32.768 kHz Internal Oscillator
(OSC32K)
IDD_XOSC32K
32.768 kHz Crystal Oscillator
(XOSC32K)
© 2020 Microchip Technology Inc.
Preliminary Datasheet
CLK_PER = HFOSC/4 = 1
MHz
CLK_RTC = 32.768 kHz
Internal Oscillator
DS40002247A-page 602
AVR128DB28/32/48/64
Electrical Characteristics
...........continued
Operating Conditions:
• VDD = VDDIO2 = 3V
• TA = 25°C
• OSCHF at 4 MHz used as clock source
• Device in Standby sleep mode
Symbol
Description
IDD_ADC
Analog-to-Digital Converter
(ADC)
IDD_AC
IDD_OPAMP
IDD_DAC
Min.
Typ.
Max.
Units
—
—
—
μA
ADC - Non-converting
—
980
—
μA
ADC @60 ksps(2)
—
1.1
—
mA
ADC @120 ksps(2)
—
70
—
μA
Power Profile 0
—
12
—
μA
Power Profile 1
—
6
—
μA
Power Profile 2
Analog Signal Conditioning
(OPAMP), IRSEL = 0(3)
—
1.2
—
mA
One OPAMP in voltage
follower mode, VCM = VDD/2
Analog Signal Conditioning
(OPAMP), IRSEL = 1(3)
—
880
—
μA
One OPAMP in voltage
follower mode, VCM = VDD/2
Digital-to-Analog Converter
(DAC)
—
—
—
μA
DAC + DACOUT, DACVREF =
VDD/2
—
—
—
μA
DAC (VDD), DACVREF = VDD/2
Analog Comparator (AC)
Conditions
IDD_USART
Universal Synchronous and
Asynchronous Receiver and
Transmitter (USART)
—
9
—
μA
USART Enabled @9600 Baud
IDD_SPI
Serial Peripheral Interface
(SPI)
—
—
—
μA
SPI Master @100 kHz
IDD_TWI
Two-Wire Interface (TWI)
—
9
—
μA
TWI Master @100 kHz
—
7
—
μA
TWI Slave @100 kHz
IDD_NVM_ERASE
Flash Programming Erase
—
—
—
mA
Flash Programming Erase
IDD_NVM_WRITE
Flash Programming Write
—
—
—
mA
Flash Programming Write
IDD_ZCD
Zero Cross Detector (ZCD)
—
12
—
μA
Excluding sink/source currents
Notes:
1. Current consumption of the module only. To calculate the total internal power consumption of the
microcontroller, add this value to the base power consumption given in the Power Consumption section in
Electrical Characteristics.
2. Average power consumption with ADC active in Free-Running mode.
3. Without resistor ladder.
39.4.4
I/O Pin Characteristics
Table 39-7. I/O Pin Characteristics(1)
Symbol Description
Input Low Voltage
© 2020 Microchip Technology Inc.
Min.
Typ.
Max.
Units Conditions
Preliminary Datasheet
DS40002247A-page 603
AVR128DB28/32/48/64
Electrical Characteristics
...........continued
Symbol Description
VIL
I/O PORT:
Typ.
Max.
Units Conditions
•
With Schmitt Trigger buffer
—
—
0.2×VDD
V
•
With I2C levels
—
—
0.3×VDD
V
•
With SMBus 3.0 levels
—
—
0.8
V
—
—
—
—
0.2×VDD
0.8
V
V
RESET Pin
TTL level
Input High Voltage
VIH
Min.
I/O PORT:
•
With Schmitt Trigger buffer 0.8×VDD
—
—
V
•
With I2C levels
0.7×VDD
—
—
V
•
With SMBus 3.0 levels
1.35
—
—
V
0.8×VDD
1.6
—
—
—
—
V
V
—
VDD - 0.1
—
V
Iload= 1.5 mA
System gain accuracy with
internal resistor ladder
—
工商网监
湘ICP备2023018690号
