AT32UC3L Series - Summary

Features
• High-performance, Low-power 32-bit Atmel® AVR® Microcontroller
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
– Compact Single-cycle RISC Instruction Set Including DSP Instructions
– Read-modify-write Instructions and Atomic Bit Manipulation
– Performance
• Up to 64DMIPS Running at 50MHz from Flash (1 Flash Wait State)
• Up to 36DMIPS Running at 25MHz from Flash (0 Flash Wait State)
– Memory Protection Unit (MPU)
• Secure Access Unit (SAU) providing User-defined Peripheral Protection
picoPower® Technology for Ultra-low Power Consumption
Multi-hierarchy Bus System
– High-performance Data Transfers on Separate Buses for Increased Performance
– 12 Peripheral DMA Channels Improve Speed for Peripheral Communication
Internal High-speed Flash
– 64Kbytes, 32Kbytes, and 16Kbytes Versions
– Single-cycle Access up to 25MHz
– FlashVault Technology Allows Pre-programmed Secure Library Support for End
User Applications
– Prefetch Buffer Optimizing Instruction Execution at Maximum Speed
– 100,000 Write Cycles, 15-year Data Retention Capability
– Flash Security Locks and User-defined Configuration Area
Internal High-speed SRAM, Single-cycle Access at Full Speed
– 16Kbytes (64Kbytes and 32Kbytes Flash), or 8Kbytes (16Kbytes Flash)
Interrupt Controller (INTC)
– Autovectored Low-latency Interrupt Service with Programmable Priority
External Interrupt Controller (EIC)
Peripheral Event System for Direct Peripheral to Peripheral Communication
System Functions
– Power and Clock Manager
– SleepWalking Power Saving Control
– Internal System RC Oscillator (RCSYS)
– 32KHz Oscillator
– Multipurpose Oscillator and Digital Frequency Locked Loop (DFLL)
Windowed Watchdog Timer (WDT)
Asynchronous Timer (AST) with Real-time Clock Capability
– Counter or Calendar Mode Supported
Frequency Meter (FREQM) for Accurate Measuring of Clock Frequency
Six 16-bit Timer/Counter (TC) Channels
– External Clock Inputs, PWM, Capture, and Various Counting Capabilities
36 PWM Channels (PWMA)
– 8-bit PWM with a Source Clock up to 150MHz
Four Universal Synchronous/Asynchronous Receiver/Transmitters (USART)
– Independent Baudrate Generator, Support for SPI
– Support for Hardware Handshaking
One Master/Slave Serial Peripheral Interfaces (SPI) with Chip Select Signals
– Up to 15 SPI Slaves can be Addressed
Two Master and Two Slave Two-wire Interface (TWI), 400kbit/s I2C-compatible
One 8-channel Analog-to-digital Converter (ADC) with up to 12 Bits Resolution
– Internal Temperature Sensor
32-bit Atmel
AVR
Microcontroller
AT32UC3L064
AT32UC3L032
AT32UC3L016
Summary
32099IS–01/2012
AT32UC3L016/32/64
• Eight Analog Comparators (AC) with Optional Window Detection
• Capacitive Touch (CAT) Module
•
•
•
•
•
– Hardware-assisted Atmel® AVR® QTouch® and Atmel® AVR® QMatrix Touch Acquisition
– Supports QTouch and QMatrix Capture from Capacitive Touch Sensors
QTouch Library Support
– Capacitive Touch Buttons, Sliders, and Wheels
– QTouch and QMatrix Acquisition
On-chip Non-intrusive Debug System
– Nexus Class 2+, Runtime Control, Non-intrusive Data and Program Trace
– aWire Single-pin Programming Trace and Debug Interface Muxed with Reset Pin
– NanoTrace Provides Trace Capabilities through JTAG or aWire Interface
48-pin TQFP/QFN/TLLGA (36 GPIO Pins)
Five High-drive I/O Pins
Single 1.62-3.6 V Power Supply
2
32099IS–01/2012
AT32UC3L016/32/64
1. Description
The Atmel® AVR® AT32UC3L016/32/64 is a complete system-on-chip microcontroller based on
the AVR32 UC RISC processor running at frequencies up to 50MHz. AVR32 UC is a high-performance 32-bit RISC microprocessor core, designed for cost-sensitive embedded applications,
with particular emphasis on low power consumption, high code density, and high performance.
The processor implements a Memory Protection Unit (MPU) and a fast and flexible interrupt controller for supporting modern and real-time operating systems. The Secure Access Unit (SAU) is
used together with the MPU to provide the required security and integrity.
Higher computation capability is achieved using a rich set of DSP instructions.
The AT32UC3L016/32/64 embeds state-of-the-art picoPower technology for ultra-low power
consumption. Combined power control techniques are used to bring active current consumption
down to 165 µA/MHz, and leakage down to 9nA while still retaining a bank of backup registers.
The device allows a wide range of trade-offs between functionality and power consumption, giving the user the ability to reach the lowest possible power consumption with the feature set
required for the application.
The Peripheral Direct Memory Access (DMA) controller enables data transfers between peripherals and memories without processor involvement. The Peripheral DMA controller drastically
reduces processing overhead when transferring continuous and large data streams.
The AT32UC3L016/32/64 incorporates on-chip Flash and SRAM memories for secure and fast
access. The FlashVault technology allows secure libraries to be programmed into the device.
The secure libraries can be executed while the CPU is in Secure State, but not read by nonsecure software in the device. The device can thus be shipped to end customers, who will be
able to program their own code into the device to access the secure libraries, but without risk of
compromising the proprietary secure code.
The External Interrupt Controller (EIC) allows pins to be configured as external interrupts. Each
external interrupt has its own interrupt request and can be individually masked.
The Peripheral Event System allows peripherals to receive, react to, and send peripheral events
without CPU intervention. Asynchronous interrupts allow advanced peripheral operation in low
power sleep modes.
The Power Manager (PM) improves design flexibility and security. The Power Manager supports
SleepWalking functionality, by which a module can be selectively activated based on peripheral
events, even in sleep modes where the module clock is stopped. Power monitoring is supported
by on-chip Power-on Reset (POR), Brown-out Detector (BOD), and Supply Monitor (SM). The
device features several oscillators, such as Digital Frequency Locked Loop (DFLL), Oscillator 0
(OSC0), and system RC oscillator (RCSYS). Either of these oscillators can be used as source
for the system clock. The DFLL is a programmable internal oscillator from 40 to 150MHz. It can
be tuned to a high accuracy if an accurate refernce clock is running, e.g. the 32 KHz crystal
oscillator.
The Watchdog Timer (WDT) will reset the device unless it is periodically serviced by the software. This allows the device to recover from a condition that has caused the system to be
unstable.
The Asynchronous Timer (AST) combined with the 32KHz crystal oscillator supports powerful
real-time clock capabilities, with a maximum timeout of up to 136 years. The AST can operate in
counter mode or calendar mode.
3
32099IS–01/2012
AT32UC3L016/32/64
The Frequency Meter (FREQM) allows accurate measuring of a clock frequency by comparing it
to a known reference clock.
The device includes six identical 16-bit Timer/Counter (TC) channels. Each channel can be independently programmed to perform frequency measurement, event counting, interval
measurement, pulse generation, delay timing, and pulse width modulation.
The Pulse Width Modulation controller (PWMA) provides 8-bit PWM channels which can be synchronized and controlled from a common timer. One PWM channel is available for each I/O pin
on the device, enabling applications that require multiple PWM outputs, such as LCD backlight
control. The PWM channels can operate independently, with duty cycles set independently from
each other, or in interlinked mode, with multiple channels changed at the same time.
The AT32UC3L016/32/64 also features many communication interfaces, like USART, SPI, and
TWI, for communication intensive applications. The USART supports different communication
modes, like SPI Mode and LIN Mode.
A general purpose 8-channel ADC is provided, as well as eight analog comparators (AC). The
ADC can operate in 10-bit mode at full speed or in enhanced mode at reduced speed, offering
up to 12-bit resolution. The ADC also provides an internal temperature sensor input channel.
The analog comparators can be paired to detect when the sensing voltage is within or outside
the defined reference window.
The Capacitive Touch (CAT) module senses touch on external capacitive touch sensors, using
the QTouch technology. Capacitive touch sensors use no external mechanical components,
unlike normal push buttons, and therefore demand less maintenance in the user application.
The CAT module allows up to 17 touch sensors, or up to 16 by 8 matrix sensors to be interfaced.
One touch sensor can be configured to operate autonomously without software interaction,
allowing wakeup from sleep modes when activated.
Atmel offers the QTouch library for embedding capacitive touch buttons, sliders, and wheels
functionality into AVR microcontrollers. The patented charge-transfer signal acquisition offers
robust sensing and includes fully debounced reporting of touch keys as well as Adjacent Key
Suppression® (AKS®) technology for unambiguous detection of key events. The easy-to-use
QTouch Suite toolchain allows you to explore, develop, and debug your own touch applications.
The AT32UC3L016/32/64 integrates a class 2+ Nexus 2.0 On-chip Debug (OCD) System, with
non-intrusive real-time trace and full-speed read/write memory access, in addition to basic runtime control. The NanoTrace interface enables trace feature for aWire- or JTAG-based
debuggers. The single-pin aWire interface allows all features available through the JTAG interface to be accessed through the RESET pin, allowing the JTAG pins to be used for GPIO or
peripherals.
4
32099IS–01/2012
AT32UC3L016/32/64
2. Overview
Block Diagram
DATAOUT
aWire
NEXUS
CLASS 2+
OCD
MEMORY PROTECTION UNIT
INSTR
INTERFACE
DATA
INTERFACE
M
M
M
SAU
S
HSB-PB
BRIDGE B
GENERALPURPOSE I/Os
S
POWER MANAGER
CLOCK
CONTROLLER
SLEEP
CONTROLLER
RESET
CONTROLLER
64/32/16 KB
FLASH
M
S
CONFIGURATION
PA
PB
S
HIGH SPEED
BUS MATRIX
S/M
LOCAL BUS
16/8 KB
SRAM
REGISTERS BUS
PERIPHERAL
DMA
CONTROLLER
HSB-PB
BRIDGE A
DMA
RESET_N
JTAG
INTERFACE
LOCAL BUS
INTERFACE
CAPACITIVE TOUCH
MODULE
DMA
EVTO_N
TCK
TDO
TDI
TMS
AVR32UC CPU
FLASH
CONTROLLER
MCKO
MDO[5..0]
MSEO[1..0]
EVTI_N
MEMORY INTERFACE
Block Diagram
USART0
USART1
USART2
USART3
DMA
Figure 2-1.
SPI
DMA
2.1
TWI MASTER 0
TWI MASTER 1
DIS
VDIVEN
CSA[16:0]
CSB[16:0]
SMP
SYNC
RXD
TXD
CLK
RTS, CTS
RCSYS
RC32K
XIN32
XOUT32
OSC32K
XIN0
XOUT0
OSC0
SYSTEM CONTROL
INTERFACE
DFLL
INTERRUPT
CONTROLLER
EXTINT[5..1]
NMI
PWMA[35..0]
NPCS[3..0]
TWCK
TWD
TWALM
TWCK
DMA
RC120M
TWI SLAVE 0
TWI SLAVE 1
8-CHANNEL ADC
INTERFACE
FREQUENCY METER
TRIGGER
AD[8..0]
A[2..0]
TIMER/COUNTER 0
TIMER/COUNTER 1
B[2..0]
CLK[2..0]
ASYNCHRONOUS
TIMER
WATCHDOG
TIMER
TWALM
ADVREFP
EXTERNAL INTERRUPT
CONTROLLER
PWM CONTROLLER
TWD
PA
PB
ADP[1..0]
DMA
RC32OUT
MISO, MOSI
GENERAL PURPOSE I/Os
SCK
GCLK[4..0]
AC INTERFACE
GLUE LOGIC
CONTROLLER
ACBP[3..0]
ACBN[3..0]
ACAP[3..0]
ACAN[3..0]
ACREFN
OUT[1:0]
IN[7..0]
5
32099IS–01/2012
AT32UC3L016/32/64
2.2
Configuration Summary
Table 2-1.
Configuration Summary
Feature
AT32UC3L064
AT32UC3L032
AT32UC3L016
Flash
64KB
32KB
16KB
SRAM
16KB
16KB
8KB
GPIO
36
High-drive pins
5
External Interrupts
6
TWI
2
USART
4
Peripheral DMA Channels
12
Peripheral Event System
1
SPI
1
Asynchronous Timers
1
Timer/Counter Channels
6
PWM channels
36
Frequency Meter
1
Watchdog Timer
1
Power Manager
1
Secure Access Unit
1
Glue Logic Controller
1
Oscillators
ADC
Digital Frequency Locked Loop 40-150MHz (DFLL)
Crystal Oscillator 3-16MHz (OSC0)
Crystal Oscillator 32KHz (OSC32K)
RC Oscillator 120MHz (RC120M)
RC Oscillator 115kHz (RCSYS)
RC Oscillator 32kHz (RC32K)
8-channel 12-bit
Temperature Sensor
1
Analog Comparators
8
Capacitive Touch Module
1
JTAG
1
aWire
1
Max Frequency
Packages
50MHz
TQFP48/QFN48/TLLGA48
6
32099IS–01/2012
AT32UC3L016/32/64
3. Package and Pinout
3.1
Package
The device pins are multiplexed with peripheral functions as described in Section 3.2.
TQFP48/QFN48 Pinout
36
35
34
33
32
31
30
29
28
27
26
25
PA14
VDDANA
ADVREFP
GNDANA
PB08
PB07
PB06
PB09
PA04
PA11
PA13
PA20
Figure 3-1.
PA15
PA16
PA17
PA19
PA18
VDDIO
GND
PB11
GND
PA10
PA12
VDDIO
37
38
39
40
41
42
43
44
45
46
47
48
24
23
22
21
20
19
18
17
16
15
14
13
PA21
PB10
RESET_N
PB04
PB05
GND
VDDCORE
VDDIN
PB01
PA07
PA01
PA02
12
11
10
9
8
7
6
5
4
3
2
1
PA05
PA00
PA06
PA22
PB03
PB02
PB00
PB12
PA03
PA08
PA09
GND
7
32099IS–01/2012
AT32UC3L016/32/64
TLLGA48 Pinout
37
36
35
34
33
32
31
30
29
28
27
26
25
PA15
PA14
VDDANA
ADVREFP
GNDANA
PB08
PB07
PB06
PB09
PA04
PA11
PA13
PA20
Figure 3-2.
PA16
PA17
PA19
PA18
VDDIO
GND
PB11
GND
PA10
PA12
VDDIO
24
23
22
21
20
19
18
17
16
15
14
38
39
40
41
42
43
44
45
46
47
48
PA21
PB10
RESET_N
PB04
PB05
GND
VDDCORE
VDDIN
PB01
PA07
PA01
13
12
11
10
9
8
7
6
5
4
3
2
1
PA02
PA05
PA00
PA06
PA22
PB03
PB02
PB00
PB12
PA03
PA08
PA09
GND
8
32099IS–01/2012
AT32UC3L016/32/64
3.2
Peripheral Multiplexing on I/O lines
3.2.1
Multiplexed signals
Each GPIO line can be assigned to one of the peripheral functions.The following table describes
the peripheral signals multiplexed to the GPIO lines.
Table 3-1.
GPIO Controller Function Multiplexing
48pin
PIN
G
P
I
O
11
PA00
0
14
PA01
13
GPIO Function
Pin
Type
A
B
C
VDDIO
Normal
I/O
USART0
TXD
USART1
RTS
SPI
NPCS[2]
1
VDDIO
Normal
I/O
USART0
RXD
USART1
CTS
SPI
NPCS[3]
USART1
CLK
PWMA
PWMA[1]
PA02
2
VDDIO
Highdrive I/O
USART0
RTS
ADCIFB
TRIGGER
USART2
TXD
TC0
A0
4
PA03
3
VDDIO
Normal
I/O
USART0
CTS
SPI
NPCS[1]
USART2
TXD
28
PA04
4
VDDIO
Normal
I/O
SPI
MISO
TWIMS0
TWCK
12
PA05
5
VDDIO
Normal
I/O (TWI)
SPI
MOSI
10
PA06
6
VDDIO
Highdrive I/O,
5V
tolerant
15
PA07
7
VDDIO
3
PA08
8
2
PA09
46
Supply
D
E
F
G
H
SCIF
GCLK[0]
CAT
CSA[2]
ACIFB
ACAP[0]
TWIMS0
TWALM
CAT
CSA[1]
PWMA
PWMA[2]
ACIFB
ACBP[0]
USART0
CLK
CAT
CSA[3]
TC0
B0
PWMA
PWMA[3]
ACIFB
ACBN[3]
USART0
CLK
CAT
CSB[3]
USART1
RXD
TC0
B1
PWMA
PWMA[4]
ACIFB
ACBP[1]
TWIMS1
TWCK
USART1
TXD
TC0
A1
PWMA
PWMA[5]
ACIFB
ACBN[0]
SPI
SCK
USART2
TXD
USART1
CLK
TC0
B0
PWMA
PWMA[6]
Normal
I/O (TWI)
SPI
NPCS[0]
USART2
RXD
TWIMS1
TWALM
TWIMS0
TWCK
PWMA
PWMA[7]
VDDIO
Highdrive I/O
USART1
TXD
SPI
NPCS[2]
TC0
A2
ADCIFB
ADP[0]
PWMA
PWMA[8]
9
VDDIO
Highdrive I/O
USART1
RXD
SPI
NPCS[3]
TC0
B2
ADCIFB
ADP[1]
PWMA
PWMA[9]
SCIF
GCLK[2]
EIC
EXTINT[1]
CAT
CSB[4]
PA10
10
VDDIO
Normal
I/O
TWIMS0
TWD
PWMA
PWMA[10]
ACIFB
ACAP[1]
SCIF
GCLK[2]
CAT
CSA[5]
27
PA11
11
VDDIN
Normal
I/O
47
PA12
12
VDDIO
Normal
I/O
26
PA13
13
VDDIN
Normal
I/O
36
PA14
14
VDDIO
37
PA15
15
38
PA16
39
41
PWMA
PWMA[0]
TC0
A0
ACIFB
ACAN[0]
CAT
CSA[7]
TWIMS0
TWD
CAT
CSB[7]
SCIF
GCLK[1]
CAT
CSB[1]
EIC
EXTINT[0]
CAT
CSB[2]
CAT
CSA[4]
PWMA
PWMA[11]
USART2
CLK
TC0
CLK1
CAT
SMP
PWMA
PWMA[12]
ACIFB
ACAN[1]
SCIF
GCLK[3]
CAT
CSB[5]
GLOC
OUT[0]
GLOC
IN[7]
TC0
A0
SCIF
GCLK[2]
PWMA
PWMA[13]
CAT
SMP
EIC
EXTINT[2]
CAT
CSA[0]
Normal
I/O
ADCIFB
AD[0]
TC0
CLK2
USART2
RTS
CAT
SMP
PWMA
PWMA[14]
SCIF
GCLK[4]
CAT
CSA[6]
VDDIO
Normal
I/O
ADCIFB
AD[1]
TC0
CLK1
GLOC
IN[6]
PWMA
PWMA[15]
CAT
SYNC
EIC
EXTINT[3]
CAT
CSB[6]
16
VDDIO
Normal
I/O
ADCIFB
AD[2]
TC0
CLK0
GLOC
IN[5]
PWMA
PWMA[16]
ACIFB
ACREFN
EIC
EXTINT[4]
CAT
CSA[8]
PA17
17
VDDIO
Normal
I/O (TWI)
TWIMS1
TWD
PWMA
PWMA[17]
CAT
SMP
CAT
DIS
CAT
CSB[8]
PA18
18
VDDIO
Normal
I/O
GLOC
IN[4]
PWMA
PWMA[18]
CAT
SYNC
EIC
EXTINT[5]
CAT
CSB[0]
TC0
A1
ADCIFB
AD[4]
TC0
B1
USART2
CTS
9
32099IS–01/2012
AT32UC3L016/32/64
Table 3-1.
GPIO Controller Function Multiplexing
40
PA19
19
VDDIO
Normal
I/O
ADCIFB
AD[5]
TC0
A2
TWIMS1
TWALM
PWMA
PWMA[19]
CAT
SYNC
25
PA20
20
VDDIN
Normal
I/O
USART2
TXD
TC0
A1
GLOC
IN[3]
PWMA
PWMA[20]
SCIF
RC32OUT
USART2
RXD
TWIMS0
TWD
TC0
B1
ADCIFB
TRIGGER
PWMA
PWMA[21]
PWMA
PWMAOD[21]
24
PA21
21
VDDIN
Normal
I/O (TWI,
5V
tolerant
SMBus)
9
PA22
22
VDDIO
Normal
I/O
USART0
CTS
USART2
CLK
TC0
B2
CAT
SMP
PWMA
PWMA[22]
ACIFB
ACBN[2]
6
PB00
32
VDDIO
Normal
I/O
USART3
TXD
ADCIFB
ADP[0]
SPI
NPCS[0]
TC0
A1
PWMA
PWMA[23]
ACIFB
ACAP[2]
16
PB01
33
VDDIO
Highdrive I/O
USART3
RXD
ADCIFB
ADP[1]
SPI
SCK
TC0
B1
PWMA
PWMA[24]
7
PB02
34
VDDIO
Normal
I/O
USART3
RTS
USART3
CLK
SPI
MISO
TC0
A2
PWMA
PWMA[25]
8
PB03
35
VDDIO
Normal
I/O
USART3
CTS
USART3
CLK
SPI
MOSI
TC0
B2
VDDIN
Normal
I/O (TWI,
5V
tolerant
SMBus)
TC1
A0
USART1
RTS
USART1
CLK
TC1
B0
USART1
CTS
21
PB04
36
CAT
CSA[10]
CAT
CSA[12]
SCIF
GCLK[0]
CAT
SMP
CAT
CSB[10]
TC1
A0
CAT
CSA[9]
TC1
A1
CAT
CSB[9]
ACIFB
ACAN[2]
SCIF
GCLK[1]
CAT
CSB[11]
PWMA
PWMA[26]
ACIFB
ACBP[2]
TC1
A2
CAT
CSA[11]
TWIMS0
TWALM
PWMA
PWMA[27]
PWMA
PWMAOD[27]
TWIMS1
TWCK
CAT
CSA[14]
USART1
CLK
TWIMS0
TWCK
PWMA
PWMA[28]
PWMA
PWMAOD[28]
SCIF
GCLK[3]
CAT
CSB[14]
20
PB05
37
VDDIN
Normal
I/O (TWI,
5V
tolerant
SMBus)
30
PB06
38
VDDIO
Normal
I/O
TC1
A1
USART3
TXD
ADCIFB
AD[6]
GLOC
IN[2]
PWMA
PWMA[29]
ACIFB
ACAN[3]
EIC
EXTINT[0]
CAT
CSB[13]
31
PB07
39
VDDIO
Normal
I/O
TC1
B1
USART3
RXD
ADCIFB
AD[7]
GLOC
IN[1]
PWMA
PWMA[30]
ACIFB
ACAP[3]
EIC
EXTINT[1]
CAT
CSA[13]
32
PB08
40
VDDIO
Normal
I/O
TC1
A2
USART3
RTS
ADCIFB
AD[8]
GLOC
IN[0]
PWMA
PWMA[31]
CAT
SYNC
EIC
EXTINT[2]
CAT
CSB[12]
29
PB09
41
VDDIO
Normal
I/O
TC1
B2
USART3
CTS
USART3
CLK
PWMA
PWMA[32]
ACIFB
ACBN[1]
EIC
EXTINT[3]
CAT
CSB[15]
23
PB10
42
VDDIN
Normal
I/O
TC1
CLK0
USART1
TXD
USART3
CLK
EIC
EXTINT[4]
CAT
CSB[16]
44
PB11
43
VDDIO
Normal
I/O
TC1
CLK1
USART1
RXD
5
PB12
44
VDDIO
Normal
I/O
TC1
CLK2
TWIMS1
TWALM
GLOC
OUT[1]
PWMA
PWMA[33]
ADCIFB
TRIGGER
PWMA
PWMA[34]
CAT
VDIVEN
EIC
EXTINT[5]
CAT
CSA[16]
CAT
SYNC
PWMA
PWMA[35]
ACIFB
ACBP[3]
SCIF
GCLK[4]
CAT
CSA[15]
See Section 3.3 for a description of the various peripheral signals.
Refer to ”Electrical Characteristics” on page 41 for a description of the electrical properties of the
pin types used.
3.2.1.1
TWI, 5V Tolerant, and SMBUS Pins
Some Normal I/O pins offer TWI, 5V Tolerant, and SMBUS features. These features are only
available when either of the TWI functions or the PWMAOD function in the PWMA are selected
for these pins.
10
32099IS–01/2012
AT32UC3L016/32/64
Refer to the ”TWI Pin Characteristics(1)” on page 49 for a description of the electrical properties
of the TWI, 5V Tolerant, and SMBUS pins.
3.2.2
Peripheral Functions
Each GPIO line can be assigned to one of several peripheral functions. The following table
describes how the various peripheral functions are selected. The last listed function has priority
in case multiple functions are enabled on the same pin.
Table 3-2.
3.2.3
Function
Description
GPIO Controller Function multiplexing
GPIO and GPIO peripheral selection A to H
Nexus OCD AUX port connections
OCD trace system
aWire DATAOUT
aWire output in two-pin mode
JTAG port connections
JTAG debug port
Oscillators
OSC0, OSC32
JTAG Port Connections
If the JTAG is enabled, the JTAG will take control over a number of pins, irrespectively of the I/O
Controller configuration.
Table 3-3.
3.2.4
Peripheral Functions
JTAG Pinout
48-pin
Pin Name
JTAG Pin
11
PA00
TCK
14
PA01
TMS
13
PA02
TDO
4
PA03
TDI
Nexus OCD AUX Port Connections
If the OCD trace system is enabled, the trace system will take control over a number of pins, irrespectively of the I/O Controller configuration. Two different OCD trace pin mappings are
possible, depending on the configuration of the OCD AXS register. For details, see the AVR32
UC Technical Reference Manual.
Table 3-4.
Nexus OCD AUX Port Connections
Pin
AXS=1
AXS=0
EVTI_N
PA05
PB08
MDO[5]
PA10
PB00
MDO[4]
PA18
PB04
MDO[3]
PA17
PB05
MDO[2]
PA16
PB03
11
32099IS–01/2012
AT32UC3L016/32/64
Table 3-4.
3.2.5
Pin
AXS=1
AXS=0
MDO[1]
PA15
PB02
MDO[0]
PA14
PB09
EVTO_N
PA04
PA04
MCKO
PA06
PB01
MSEO[1]
PA07
PB11
MSEO[0]
PA11
PB12
Oscillator Pinout
The oscillators are not mapped to the normal GPIO functions and their muxings are controlled
by registers in the System Control Interface (SCIF). Please refer to the SCIF chapter for more
information about this.
Table 3-5.
3.2.6
Nexus OCD AUX Port Connections
Oscillator Pinout
48-pin
Pin
Oscillator Function
3
PA08
XIN0
46
PA10
XIN32
26
PA13
XIN32_2
2
PA09
XOUT0
47
PA12
XOUT32
25
PA20
XOUT32_2
Other Functions
The functions listed in Table 3-6 are not mapped to the normal GPIO functions. The aWire DATA
pin will only be active after the aWire is enabled. The aWire DATAOUT pin will only be active
after the aWire is enabled and the 2_PIN_MODE command has been sent. The WAKE_N pin is
always enabled. Please refer to Section 6.1.4 on page 40 for constraints on the WAKE_N pin.
Table 3-6.
Other Functions
48-pin
Pin
Function
27
PA11
WAKE_N
22
RESET_N
aWire DATA
11
PA00
aWire DATAOUT
12
32099IS–01/2012
AT32UC3L016/32/64
3.3
Signal Descriptions
The following table gives details on signal name classified by peripheral.
Table 3-7.
Signal Descriptions List
Signal Name
Function
Type
Active
Level
Comments
Analog Comparator Interface - ACIFB
ACAN3 - ACAN0
Negative inputs for comparators "A"
Analog
ACAP3 - ACAP0
Positive inputs for comparators "A"
Analog
ACBN3 - ACBN0
Negative inputs for comparators "B"
Analog
ACBP3 - ACBP0
Positive inputs for comparators "B"
Analog
ACREFN
Common negative reference
Analog
ADC Interface - ADCIFB
AD8 - AD0
Analog Signal
Analog
ADP1 - ADP0
Drive Pin for resistive touch screen
Output
TRIGGER
External trigger
Input
aWire - AW
DATA
aWire data
I/O
DATAOUT
aWire data output for 2-pin mode
I/O
Capacitive Touch Module - CAT
CSA16 - CSA0
Capacitive Sense A
I/O
CSB16 - CSB0
Capacitive Sense B
I/O
DIS
Discharge current control
Analog
SMP
SMP signal
Output
SYNC
Synchronize signal
VDIVEN
Voltage divider enable
Input
Output
External Interrupt Controller - EIC
NMI
Non-Maskable Interrupt
Input
EXTINT5 - EXTINT1
External interrupt
Input
Glue Logic Controller - GLOC
IN7 - IN0
Inputs to lookup tables
OUT1 - OUT0
Outputs from lookup tables
Input
Output
JTAG module - JTAG
TCK
Test Clock
Input
TDI
Test Data In
Input
TDO
Test Data Out
TMS
Test Mode Select
Output
Input
13
32099IS–01/2012
AT32UC3L016/32/64
Table 3-7.
Signal Descriptions List
Power Manager - PM
RESET_N
Reset
Input
Low
Pulse Width Modulation Controller - PWMA
PWMA35 - PWMA0
PWMA channel waveforms
Output
PWMAOD35 PWMAOD0
PWMA channel waveforms, open drain
mode
Output
Not all channels support open
drain mode
System Control Interface - SCIF
GCLK4 - GCLK0
Generic Clock Output
Output
RC32OUT
RC32K output at startup
Output
XIN0
Crystal 0 Input
Analog/
Digital
XIN32
Crystal 32 Input (primary location)
Analog/
Digital
XIN32_2
Crystal 32 Input (secondary location)
Analog/
Digital
XOUT0
Crystal 0 Output
Analog
XOUT32
Crystal 32 Output (primary location)
Analog
XOUT32_2
Crystal 32 Output (secondary location)
Analog
Serial Peripheral Interface - SPI
MISO
Master In Slave Out
I/O
MOSI
Master Out Slave In
I/O
NPCS3 - NPCS0
SPI Peripheral Chip Select
I/O
SCK
Clock
I/O
Low
Timer/Counter - TC0, TC1
A0
Channel 0 Line A
I/O
A1
Channel 1 Line A
I/O
A2
Channel 2 Line A
I/O
B0
Channel 0 Line B
I/O
B1
Channel 1 Line B
I/O
B2
Channel 2 Line B
I/O
CLK0
Channel 0 External Clock Input
Input
CLK1
Channel 1 External Clock Input
Input
CLK2
Channel 2 External Clock Input
Input
Two-wire Interface - TWIMS0, TWIMS1
TWALM
SMBus SMBALERT
I/O
TWCK
Two-wire Serial Clock
I/O
TWD
Two-wire Serial Data
I/O
Low
Universal Synchronous/Asynchronous Receiver/Transmitter - USART0, USART1, USART2, USART3
14
32099IS–01/2012
AT32UC3L016/32/64
Table 3-7.
Signal Descriptions List
CLK
Clock
CTS
Clear To Send
RTS
Request To Send
RXD
Receive Data
Input
TXD
Transmit Data
Output
Note:
I/O
Input
Low
Output
Low
1. ADCIFB: AD3 does not exist.
Table 3-8.
Signal Description List, Continued
Signal Name
Function
Type
Active
Level
Comments
Power
VDDCORE
Core Power Supply / Voltage Regulator Output
Power
Input/Output
1.62V to 1.98V
VDDIO
I/O Power Supply
Power Input
1.62V to 3.6V. VDDIO should
always be equal to or lower than
VDDIN.
VDDANA
Analog Power Supply
Power Input
1.62V to 1.98V
ADVREFP
Analog Reference Voltage
Power Input
1.62V to 1.98V
VDDIN
Voltage Regulator Input
Power Input
1.62V to 3.6V (1)
GNDANA
Analog Ground
Ground
GND
Ground
Ground
Auxiliary Port - AUX
MCKO
Trace Data Output Clock
Output
MDO5 - MDO0
Trace Data Output
Output
MSEO1 - MSEO0
Trace Frame Control
Output
EVTI_N
Event In
EVTO_N
Event Out
Input
Low
Output
Low
General Purpose I/O pin
PA22 - PA00
Parallel I/O Controller I/O Port 0
I/O
PB12 - PB00
Parallel I/O Controller I/O Port 1
I/O
1.
See Section 6.1 on page 36
15
32099IS–01/2012
AT32UC3L016/32/64
3.4
3.4.1
I/O Line Considerations
JTAG Pins
The JTAG is enabled if TCK is low while the RESET_N pin is released. The TCK, TMS, and TDI
pins have pull-up resistors when JTAG is enabled. The TCK pin always has pull-up enabled during reset. The TDO pin is an output, driven at VDDIO, and has no pull-up resistor. The JTAG
pins can be used as GPIO pins and multiplexed with peripherals when the JTAG is disabled.
Please refer to Section 3.2.3 on page 11 for the JTAG port connections.
3.4.2
PA00
Note that PA00 is multiplexed with TCK. PA00 GPIO function must only be used as output in the
application.
3.4.3
RESET_N Pin
The RESET_N pin is a schmitt input and integrates a permanent pull-up resistor to VDDIN. As
the product integrates a power-on reset detector, the RESET_N pin can be left unconnected in
case no reset from the system needs to be applied to the product.
The RESET_N pin is also used for the aWire debug protocol. When the pin is used for debugging, it must not be driven by external circuitry.
3.4.4
TWI Pins PA21/PB04/PB05
When these pins are used for TWI, the pins are open-drain outputs with slew-rate limitation and
inputs with spike filtering. When used as GPIO pins or used for other peripherals, the pins have
the same characteristics as other GPIO pins. Selected pins are also SMBus compliant (refer to
Section 3.2 on page 9). As required by the SMBus specification, these pins provide no leakage
path to ground when the AT32UC3L016/32/64 is powered down. This allows other devices on
the SMBus to continue communicating even though the AT32UC3L016/32/64 is not powered.
After reset a TWI function is selected on these pins instead of the GPIO. Please refer to the
GPIO Module Configuration chapter for details.
3.4.5
TWI Pins PA05/PA07/PA17
When these pins are used for TWI, the pins are open-drain outputs with slew-rate limitation and
inputs with spike filtering. When used as GPIO pins or used for other peripherals, the pins have
the same characteristics as other GPIO pins.
After reset a TWI function is selected on these pins instead of the GPIO. Please refer to the
GPIO Module Configuration chapter for details.
3.4.6
GPIO Pins
All the I/O lines integrate a pull-up resistor. Programming of this pull-up resistor is performed
independently for each I/O line through the GPIO Controller. After reset, I/O lines default as
inputs with pull-up resistors disabled, except PA00. PA20 selects SCIF-RC32OUT (GPIO Function F) as default enabled after reset.
3.4.7
High-Drive Pins
The five pins PA02, PA06, PA08, PA09, and PB01 have high-drive output capabilities. Refer to
Section 7. on page 41 for electrical characteristics.
16
32099IS–01/2012
AT32UC3L016/32/64
3.4.8
RC32OUT Pin
3.4.8.1
Clock output at startup
After power-up, the clock generated by the 32kHz RC oscillator (RC32K) will be output on PA20,
even when the device is still reset by the Power-On Reset Circuitry. This clock can be used by
the system to start other devices or to clock a switching regulator to rise the power supply voltage up to an acceptable value.
The clock will be available on PA20, but will be disabled if one of the following conditions are
true:
• PA20 is configured to use a GPIO function other than F (SCIF-RC32OUT)
• PA20 is configured as a General Purpose Input/Output (GPIO)
• The bit FRC32 in the Power Manager PPCR register is written to zero (refer to the Power
Manager chapter)
The maximum amplitude of the clock signal will be defined by VDDIN.
Once the RC32K output on PA20 is disabled it can never be enabled again.
3.4.8.2
3.4.9
XOUT32_2 function
PA20 selects RC32OUT as default enabled after reset. This function is not automatically disabled when the user enables the XOUT32_2 function on PA20. This disturbs the oscillator and
may result in the wrong frequency. To avoid this, RC32OUT must be disabled when XOUT32_2
is enabled.
ADC Input Pins
These pins are regular I/O pins powered from the VDDIO. However, when these pins are used
for ADC inputs, the voltage applied to the pin must not exceed 1.98V. Internal circuitry ensures
that the pin cannot be used as an analog input pin when the I/O drives to VDD. When the pins
are not used for ADC inputs, the pins may be driven to the full I/O voltage range.
17
32099IS–01/2012
AT32UC3L016/32/64
4. Processor and Architecture
Rev: 2.1.0.0
This chapter gives an overview of the AVR32UC CPU. AVR32UC is an implementation of the
AVR32 architecture. A summary of the programming model, instruction set, and MPU is presented. For further details, see the AVR32 Architecture Manual and the AVR32UC Technical
Reference Manual.
4.1
Features
• 32-bit load/store AVR32A RISC architecture
–
–
–
–
–
15 general-purpose 32-bit registers
32-bit Stack Pointer, Program Counter and Link Register reside in register file
Fully orthogonal instruction set
Privileged and unprivileged modes enabling efficient and secure operating systems
Innovative instruction set together with variable instruction length ensuring industry leading
code density
– DSP extension with saturating arithmetic, and a wide variety of multiply instructions
• 3-stage pipeline allowing one instruction per clock cycle for most instructions
– Byte, halfword, word, and double word memory access
– Multiple interrupt priority levels
• MPU allows for operating systems with memory protection
• Secure State for supporting FlashVault technology
4.2
AVR32 Architecture
AVR32 is a new, high-performance 32-bit RISC microprocessor architecture, designed for costsensitive embedded applications, with particular emphasis on low power consumption and high
code density. In addition, the instruction set architecture has been tuned to allow a variety of
microarchitectures, enabling the AVR32 to be implemented as low-, mid-, or high-performance
processors. AVR32 extends the AVR family into the world of 32- and 64-bit applications.
Through a quantitative approach, a large set of industry recognized benchmarks has been compiled and analyzed to achieve the best code density in its class. In addition to lowering the
memory requirements, a compact code size also contributes to the core’s low power characteristics. The processor supports byte and halfword data types without penalty in code size and
performance.
Memory load and store operations are provided for byte, halfword, word, and double word data
with automatic sign- or zero extension of halfword and byte data. The C-compiler is closely
linked to the architecture and is able to exploit code optimization features, both for size and
speed.
In order to reduce code size to a minimum, some instructions have multiple addressing modes.
As an example, instructions with immediates often have a compact format with a smaller immediate, and an extended format with a larger immediate. In this way, the compiler is able to use
the format giving the smallest code size.
Another feature of the instruction set is that frequently used instructions, like add, have a compact format with two operands as well as an extended format with three operands. The larger
format increases performance, allowing an addition and a data move in the same instruction in a
18
32099IS–01/2012
AT32UC3L016/32/64
single cycle. Load and store instructions have several different formats in order to reduce code
size and speed up execution.
The register file is organized as sixteen 32-bit registers and includes the Program Counter, the
Link Register, and the Stack Pointer. In addition, register R12 is designed to hold return values
from function calls and is used implicitly by some instructions.
4.3
The AVR32UC CPU
The AVR32UC CPU targets low- and medium-performance applications, and provides an
advanced On-Chip Debug (OCD) system, no caches, and a Memory Protection Unit (MPU).
Java acceleration hardware is not implemented.
AVR32UC provides three memory interfaces, one High Speed Bus master for instruction fetch,
one High Speed Bus master for data access, and one High Speed Bus slave interface allowing
other bus masters to access data RAMs internal to the CPU. Keeping data RAMs internal to the
CPU allows fast access to the RAMs, reduces latency, and guarantees deterministic timing.
Also, power consumption is reduced by not needing a full High Speed Bus access for memory
accesses. A dedicated data RAM interface is provided for communicating with the internal data
RAMs.
A local bus interface is provided for connecting the CPU to device-specific high-speed systems,
such as floating-point units and I/O controller ports. This local bus has to be enabled by writing a
one to the LOCEN bit in the CPUCR system register. The local bus is able to transfer data
between the CPU and the local bus slave in a single clock cycle. The local bus has a dedicated
memory range allocated to it, and data transfers are performed using regular load and store
instructions. Details on which devices that are mapped into the local bus space is given in the
CPU Local Bus section in the Memories chapter.
Figure 4-1 on page 20 displays the contents of AVR32UC.
19
32099IS–01/2012
AT32UC3L016/32/64
OCD interface
Reset interface
Overview of the AVR32UC CPU
Interrupt controller interface
Figure 4-1.
OCD
system
Power/
Reset
control
AVR32UC CPU pipeline
MPU
4.3.1
High
Speed
Bus slave
CPU Local
Bus
master
CPU Local Bus
High Speed
Bus master
High Speed Bus
High Speed Bus
High Speed Bus master
High Speed Bus
Data memory controller
Instruction memory controller
CPU RAM
Pipeline Overview
AVR32UC has three pipeline stages, Instruction Fetch (IF), Instruction Decode (ID), and Instruction Execute (EX). The EX stage is split into three parallel subsections, one arithmetic/logic
(ALU) section, one multiply (MUL) section, and one load/store (LS) section.
Instructions are issued and complete in order. Certain operations require several clock cycles to
complete, and in this case, the instruction resides in the ID and EX stages for the required number of clock cycles. Since there is only three pipeline stages, no internal data forwarding is
required, and no data dependencies can arise in the pipeline.
Figure 4-2 on page 21 shows an overview of the AVR32UC pipeline stages.
20
32099IS–01/2012
AT32UC3L016/32/64
Figure 4-2.
The AVR32UC Pipeline
MUL
IF
ID
Prefetch unit
Decode unit
Regfile
Read
ALU
LS
4.3.2
Multiply unit
Regfile
write
ALU unit
Load-store
unit
AVR32A Microarchitecture Compliance
AVR32UC implements an AVR32A microarchitecture. The AVR32A microarchitecture is targeted at cost-sensitive, lower-end applications like smaller microcontrollers. This
microarchitecture does not provide dedicated hardware registers for shadowing of register file
registers in interrupt contexts. Additionally, it does not provide hardware registers for the return
address registers and return status registers. Instead, all this information is stored on the system
stack. This saves chip area at the expense of slower interrupt handling.
4.3.2.1
Interrupt Handling
Upon interrupt initiation, registers R8-R12 are automatically pushed to the system stack. These
registers are pushed regardless of the priority level of the pending interrupt. The return address
and status register are also automatically pushed to stack. The interrupt handler can therefore
use R8-R12 freely. Upon interrupt completion, the old R8-R12 registers and status register are
restored, and execution continues at the return address stored popped from stack.
The stack is also used to store the status register and return address for exceptions and scall.
Executing the rete or rets instruction at the completion of an exception or system call will pop
this status register and continue execution at the popped return address.
4.3.2.2
Java Support
AVR32UC does not provide Java hardware acceleration.
4.3.2.3
Memory Protection
The MPU allows the user to check all memory accesses for privilege violations. If an access is
attempted to an illegal memory address, the access is aborted and an exception is taken. The
MPU in AVR32UC is specified in the AVR32UC Technical Reference manual.
4.3.2.4
Unaligned Reference Handling
AVR32UC does not support unaligned accesses, except for doubleword accesses. AVR32UC is
able to perform word-aligned st.d and ld.d. Any other unaligned memory access will cause an
21
32099IS–01/2012
AT32UC3L016/32/64
address exception. Doubleword-sized accesses with word-aligned pointers will automatically be
performed as two word-sized accesses.
The following table shows the instructions with support for unaligned addresses. All other
instructions require aligned addresses.
Table 4-1.
4.3.2.5
Instructions with Unaligned Reference Support
Instruction
Supported Alignment
ld.d
Word
st.d
Word
Unimplemented Instructions
The following instructions are unimplemented in AVR32UC, and will cause an Unimplemented
Instruction Exception if executed:
• All SIMD instructions
• All coprocessor instructions if no coprocessors are present
• retj, incjosp, popjc, pushjc
• tlbr, tlbs, tlbw
• cache
4.3.2.6
CPU and Architecture Revision
Three major revisions of the AVR32UC CPU currently exist. The device described in this
datasheet uses CPU revision 3.
The Architecture Revision field in the CONFIG0 system register identifies which architecture
revision is implemented in a specific device.
AVR32UC CPU revision 3 is fully backward-compatible with revisions 1 and 2, ie. code compiled
for revision 1 or 2 is binary-compatible with revision 3 CPUs.
22
32099IS–01/2012
AT32UC3L016/32/64
4.4
4.4.1
Programming Model
Register File Configuration
The AVR32UC register file is shown below.
Figure 4-3.
The AVR32UC Register File
Application
Supervisor
INT0
Bit 31
Bit 31
Bit 31
Bit 0
Bit 0
INT1
Bit 0
INT2
Bit 31
Bit 0
INT3
Bit 31
Bit 0
Bit 31
Bit 0
Exception
NMI
Bit 31
Bit 31
Bit 0
Secure
Bit 0
Bit 31
Bit 0
PC
LR
SP_APP
R12
R11
R10
R9
R8
INT0PC
R7
INT1PC
R6
FINTPC
R5
SMPC
R4
R3
R2
R1
R0
PC
LR
SP_SYS
R12
R11
R10
R9
R8
INT0PC
R7
INT1PC
R6
FINTPC
R5
SMPC
R4
R3
R2
R1
R0
PC
LR
SP_SYS
R12
R11
R10
R9
R8
INT0PC
R7
INT1PC
R6
FINTPC
R5
SMPC
R4
R3
R2
R1
R0
PC
LR
SP_SYS
R12
R11
R10
R9
R8
INT0PC
R7
INT1PC
R6
FINTPC
R5
SMPC
R4
R3
R2
R1
R0
PC
LR
SP_SYS
R12
R11
R10
R9
R8
INT0PC
R7
INT1PC
R6
FINTPC
R5
SMPC
R4
R3
R2
R1
R0
PC
LR
SP_SYS
R12
R11
R10
R9
R8
INT0PC
R7
INT1PC
R6
FINTPC
R5
SMPC
R4
R3
R2
R1
R0
PC
LR
SP_SYS
R12
R11
R10
R9
R8
INT0PC
R7
INT1PC
R6
FINTPC
R5
SMPC
R4
R3
R2
R1
R0
PC
LR
SP_SYS
R12
R11
R10
R9
R8
INT0PC
R7
INT1PC
R6
FINTPC
R5
SMPC
R4
R3
R2
R1
R0
PC
LR
SP_SEC
R12
R11
R10
R9
R8
INT0PC
R7
INT1PC
R6
FINTPC
R5
SMPC
R4
R3
R2
R1
R0
SR
SR
SR
SR
SR
SR
SR
SR
SR
SS_STATUS
SS_ADRF
SS_ADRR
SS_ADR0
SS_ADR1
SS_SP_SYS
SS_SP_APP
SS_RAR
SS_RSR
4.4.2
Status Register Configuration
The Status Register (SR) is split into two halfwords, one upper and one lower, see Figure 4-4
and Figure 4-5. The lower word contains the C, Z, N, V, and Q condition code flags and the R, T,
and L bits, while the upper halfword contains information about the mode and state the processor executes in. Refer to the AVR32 Architecture Manual for details.
Figure 4-4.
The Status Register High Halfword
Bit 31
Bit 16
SS
LC
1
-
-
DM
D
-
M2
M1
M0
EM
I3M
I2M
FE
I1M
I0M
GM
0
0
0
0
0
0
0
0
0
1
1
0
0
0
0
1
Bit nam e
Initial value
G lobal Interrupt M ask
Interrupt Level 0 M ask
Interrupt Level 1 M ask
Interrupt Level 2 M ask
Interrupt Level 3 M ask
Exception M ask
M ode Bit 0
M ode Bit 1
M ode Bit 2
Reserved
Debug State
Debug State M ask
Reserved
Secure State
23
32099IS–01/2012
AT32UC3L016/32/64
Figure 4-5.
The Status Register Low Halfword
Bit 15
Bit 0
-
T
-
-
-
-
-
-
-
-
L
Q
V
N
Z
C
Bit name
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Initial value
Carry
Zero
Sign
Overflow
Saturation
Lock
Reserved
Scratch
Reserved
4.4.3
Processor States
4.4.3.1
Normal RISC State
The AVR32 processor supports several different execution contexts as shown in Table 4-2.
Table 4-2.
Overview of Execution Modes, their Priorities and Privilege Levels.
Priority
Mode
Security
Description
1
Non Maskable Interrupt
Privileged
Non Maskable high priority interrupt mode
2
Exception
Privileged
Execute exceptions
3
Interrupt 3
Privileged
General purpose interrupt mode
4
Interrupt 2
Privileged
General purpose interrupt mode
5
Interrupt 1
Privileged
General purpose interrupt mode
6
Interrupt 0
Privileged
General purpose interrupt mode
N/A
Supervisor
Privileged
Runs supervisor calls
N/A
Application
Unprivileged
Normal program execution mode
Mode changes can be made under software control, or can be caused by external interrupts or
exception processing. A mode can be interrupted by a higher priority mode, but never by one
with lower priority. Nested exceptions can be supported with a minimal software overhead.
When running an operating system on the AVR32, user processes will typically execute in the
application mode. The programs executed in this mode are restricted from executing certain
instructions. Furthermore, most system registers together with the upper halfword of the status
register cannot be accessed. Protected memory areas are also not available. All other operating
modes are privileged and are collectively called System Modes. They have full access to all privileged and unprivileged resources. After a reset, the processor will be in supervisor mode.
4.4.3.2
Debug State
The AVR32 can be set in a debug state, which allows implementation of software monitor routines that can read out and alter system information for use during application development. This
implies that all system and application registers, including the status registers and program
counters, are accessible in debug state. The privileged instructions are also available.
All interrupt levels are by default disabled when debug state is entered, but they can individually
be switched on by the monitor routine by clearing the respective mask bit in the status register.
24
32099IS–01/2012
AT32UC3L016/32/64
Debug state can be entered as described in the AVR32UC Technical Reference Manual.
Debug state is exited by the retd instruction.
4.4.3.3
4.4.4
Secure State
The AVR32 can be set in a secure state, that allows a part of the code to execute in a state with
higher security levels. The rest of the code can not access resources reserved for this secure
code. Secure State is used to implement FlashVault technology. Refer to the AVR32UC Technical Reference Manual for details.
System Registers
The system registers are placed outside of the virtual memory space, and are only accessible
using the privileged mfsr and mtsr instructions. The table below lists the system registers specified in the AVR32 architecture, some of which are unused in AVR32UC. The programmer is
responsible for maintaining correct sequencing of any instructions following a mtsr instruction.
For detail on the system registers, refer to the AVR32UC Technical Reference Manual.
Table 4-3.
System Registers
Reg #
Address
Name
Function
0
0
SR
Status Register
1
4
EVBA
Exception Vector Base Address
2
8
ACBA
Application Call Base Address
3
12
CPUCR
CPU Control Register
4
16
ECR
Exception Cause Register
5
20
RSR_SUP
Unused in AVR32UC
6
24
RSR_INT0
Unused in AVR32UC
7
28
RSR_INT1
Unused in AVR32UC
8
32
RSR_INT2
Unused in AVR32UC
9
36
RSR_INT3
Unused in AVR32UC
10
40
RSR_EX
Unused in AVR32UC
11
44
RSR_NMI
Unused in AVR32UC
12
48
RSR_DBG
Return Status Register for Debug mode
13
52
RAR_SUP
Unused in AVR32UC
14
56
RAR_INT0
Unused in AVR32UC
15
60
RAR_INT1
Unused in AVR32UC
16
64
RAR_INT2
Unused in AVR32UC
17
68
RAR_INT3
Unused in AVR32UC
18
72
RAR_EX
Unused in AVR32UC
19
76
RAR_NMI
Unused in AVR32UC
20
80
RAR_DBG
Return Address Register for Debug mode
21
84
JECR
Unused in AVR32UC
22
88
JOSP
Unused in AVR32UC
23
92
JAVA_LV0
Unused in AVR32UC
25
32099IS–01/2012
AT32UC3L016/32/64
Table 4-3.
System Registers (Continued)
Reg #
Address
Name
Function
24
96
JAVA_LV1
Unused in AVR32UC
25
100
JAVA_LV2
Unused in AVR32UC
26
104
JAVA_LV3
Unused in AVR32UC
27
108
JAVA_LV4
Unused in AVR32UC
28
112
JAVA_LV5
Unused in AVR32UC
29
116
JAVA_LV6
Unused in AVR32UC
30
120
JAVA_LV7
Unused in AVR32UC
31
124
JTBA
Unused in AVR32UC
32
128
JBCR
Unused in AVR32UC
33-63
132-252
Reserved
Reserved for future use
64
256
CONFIG0
Configuration register 0
65
260
CONFIG1
Configuration register 1
66
264
COUNT
Cycle Counter register
67
268
COMPARE
Compare register
68
272
TLBEHI
Unused in AVR32UC
69
276
TLBELO
Unused in AVR32UC
70
280
PTBR
Unused in AVR32UC
71
284
TLBEAR
Unused in AVR32UC
72
288
MMUCR
Unused in AVR32UC
73
292
TLBARLO
Unused in AVR32UC
74
296
TLBARHI
Unused in AVR32UC
75
300
PCCNT
Unused in AVR32UC
76
304
PCNT0
Unused in AVR32UC
77
308
PCNT1
Unused in AVR32UC
78
312
PCCR
Unused in AVR32UC
79
316
BEAR
Bus Error Address Register
80
320
MPUAR0
MPU Address Register region 0
81
324
MPUAR1
MPU Address Register region 1
82
328
MPUAR2
MPU Address Register region 2
83
332
MPUAR3
MPU Address Register region 3
84
336
MPUAR4
MPU Address Register region 4
85
340
MPUAR5
MPU Address Register region 5
86
344
MPUAR6
MPU Address Register region 6
87
348
MPUAR7
MPU Address Register region 7
88
352
MPUPSR0
MPU Privilege Select Register region 0
89
356
MPUPSR1
MPU Privilege Select Register region 1
26
32099IS–01/2012
AT32UC3L016/32/64
Table 4-3.
4.5
System Registers (Continued)
Reg #
Address
Name
Function
90
360
MPUPSR2
MPU Privilege Select Register region 2
91
364
MPUPSR3
MPU Privilege Select Register region 3
92
368
MPUPSR4
MPU Privilege Select Register region 4
93
372
MPUPSR5
MPU Privilege Select Register region 5
94
376
MPUPSR6
MPU Privilege Select Register region 6
95
380
MPUPSR7
MPU Privilege Select Register region 7
96
384
MPUCRA
Unused in this version of AVR32UC
97
388
MPUCRB
Unused in this version of AVR32UC
98
392
MPUBRA
Unused in this version of AVR32UC
99
396
MPUBRB
Unused in this version of AVR32UC
100
400
MPUAPRA
MPU Access Permission Register A
101
404
MPUAPRB
MPU Access Permission Register B
102
408
MPUCR
MPU Control Register
103
412
SS_STATUS
Secure State Status Register
104
416
SS_ADRF
Secure State Address Flash Register
105
420
SS_ADRR
Secure State Address RAM Register
106
424
SS_ADR0
Secure State Address 0 Register
107
428
SS_ADR1
Secure State Address 1 Register
108
432
SS_SP_SYS
Secure State Stack Pointer System Register
109
436
SS_SP_APP
Secure State Stack Pointer Application Register
110
440
SS_RAR
Secure State Return Address Register
111
444
SS_RSR
Secure State Return Status Register
112-191
448-764
Reserved
Reserved for future use
192-255
768-1020
IMPL
IMPLEMENTATION DEFINED
Exceptions and Interrupts
In the AVR32 architecture, events are used as a common term for exceptions and interrupts.
AVR32UC incorporates a powerful event handling scheme. The different event sources, like Illegal Op-code and interrupt requests, have different priority levels, ensuring a well-defined
behavior when multiple events are received simultaneously. Additionally, pending events of a
higher priority class may preempt handling of ongoing events of a lower priority class.
When an event occurs, the execution of the instruction stream is halted, and execution is passed
to an event handler at an address specified in Table 4-4 on page 31. Most of the handlers are
placed sequentially in the code space starting at the address specified by EVBA, with four bytes
between each handler. This gives ample space for a jump instruction to be placed there, jumping to the event routine itself. A few critical handlers have larger spacing between them, allowing
the entire event routine to be placed directly at the address specified by the EVBA-relative offset
generated by hardware. All interrupt sources have autovectored interrupt service routine (ISR)
addresses. This allows the interrupt controller to directly specify the ISR address as an address
27
32099IS–01/2012
AT32UC3L016/32/64
relative to EVBA. The autovector offset has 14 address bits, giving an offset of maximum 16384
bytes. The target address of the event handler is calculated as (EVBA | event_handler_offset),
not (EVBA + event_handler_offset), so EVBA and exception code segments must be set up
appropriately. The same mechanisms are used to service all different types of events, including
interrupt requests, yielding a uniform event handling scheme.
An interrupt controller does the priority handling of the interrupts and provides the autovector offset to the CPU.
4.5.1
System Stack Issues
Event handling in AVR32UC uses the system stack pointed to by the system stack pointer,
SP_SYS, for pushing and popping R8-R12, LR, status register, and return address. Since event
code may be timing-critical, SP_SYS should point to memory addresses in the IRAM section,
since the timing of accesses to this memory section is both fast and deterministic.
The user must also make sure that the system stack is large enough so that any event is able to
push the required registers to stack. If the system stack is full, and an event occurs, the system
will enter an UNDEFINED state.
4.5.2
Exceptions and Interrupt Requests
When an event other than scall or debug request is received by the core, the following actions
are performed atomically:
1. The pending event will not be accepted if it is masked. The I3M, I2M, I1M, I0M, EM, and
GM bits in the Status Register are used to mask different events. Not all events can be
masked. A few critical events (NMI, Unrecoverable Exception, TLB Multiple Hit, and
Bus Error) can not be masked. When an event is accepted, hardware automatically
sets the mask bits corresponding to all sources with equal or lower priority. This inhibits
acceptance of other events of the same or lower priority, except for the critical events
listed above. Software may choose to clear some or all of these bits after saving the
necessary state if other priority schemes are desired. It is the event source’s responsability to ensure that their events are left pending until accepted by the CPU.
2. When a request is accepted, the Status Register and Program Counter of the current
context is stored to the system stack. If the event is an INT0, INT1, INT2, or INT3, registers R8-R12 and LR are also automatically stored to stack. Storing the Status
Register ensures that the core is returned to the previous execution mode when the
current event handling is completed. When exceptions occur, both the EM and GM bits
are set, and the application may manually enable nested exceptions if desired by clearing the appropriate bit. Each exception handler has a dedicated handler address, and
this address uniquely identifies the exception source.
3. The Mode bits are set to reflect the priority of the accepted event, and the correct register file bank is selected. The address of the event handler, as shown in Table 4-4 on
page 31, is loaded into the Program Counter.
The execution of the event handler routine then continues from the effective address calculated.
The rete instruction signals the end of the event. When encountered, the Return Status Register
and Return Address Register are popped from the system stack and restored to the Status Register and Program Counter. If the rete instruction returns from INT0, INT1, INT2, or INT3,
registers R8-R12 and LR are also popped from the system stack. The restored Status Register
contains information allowing the core to resume operation in the previous execution mode. This
concludes the event handling.
28
32099IS–01/2012
AT32UC3L016/32/64
4.5.3
Supervisor Calls
The AVR32 instruction set provides a supervisor mode call instruction. The scall instruction is
designed so that privileged routines can be called from any context. This facilitates sharing of
code between different execution modes. The scall mechanism is designed so that a minimal
execution cycle overhead is experienced when performing supervisor routine calls from timecritical event handlers.
The scall instruction behaves differently depending on which mode it is called from. The behaviour is detailed in the instruction set reference. In order to allow the scall routine to return to the
correct context, a return from supervisor call instruction, rets, is implemented. In the AVR32UC
CPU, scall and rets uses the system stack to store the return address and the status register.
4.5.4
Debug Requests
The AVR32 architecture defines a dedicated Debug mode. When a debug request is received by
the core, Debug mode is entered. Entry into Debug mode can be masked by the DM bit in the
status register. Upon entry into Debug mode, hardware sets the SR.D bit and jumps to the
Debug Exception handler. By default, Debug mode executes in the exception context, but with
dedicated Return Address Register and Return Status Register. These dedicated registers
remove the need for storing this data to the system stack, thereby improving debuggability. The
Mode bits in the Status Register can freely be manipulated in Debug mode, to observe registers
in all contexts, while retaining full privileges.
Debug mode is exited by executing the retd instruction. This returns to the previous context.
4.5.5
Entry Points for Events
Several different event handler entry points exist. In AVR32UC, the reset address is
0x80000000. This places the reset address in the boot flash memory area.
TLB miss exceptions and scall have a dedicated space relative to EVBA where their event handler can be placed. This speeds up execution by removing the need for a jump instruction placed
at the program address jumped to by the event hardware. All other exceptions have a dedicated
event routine entry point located relative to EVBA. The handler routine address identifies the
exception source directly.
AVR32UC uses the ITLB and DTLB protection exceptions to signal a MPU protection violation.
ITLB and DTLB miss exceptions are used to signal that an access address did not map to any of
the entries in the MPU. TLB multiple hit exception indicates that an access address did map to
multiple TLB entries, signalling an error.
All interrupt requests have entry points located at an offset relative to EVBA. This autovector offset is specified by an interrupt controller. The programmer must make sure that none of the
autovector offsets interfere with the placement of other code. The autovector offset has 14
address bits, giving an offset of maximum 16384 bytes.
Special considerations should be made when loading EVBA with a pointer. Due to security considerations, the event handlers should be located in non-writeable flash memory, or optionally in
a privileged memory protection region if an MPU is present.
If several events occur on the same instruction, they are handled in a prioritized way. The priority
ordering is presented in Table 4-4 on page 31. If events occur on several instructions at different
locations in the pipeline, the events on the oldest instruction are always handled before any
events on any younger instruction, even if the younger instruction has events of higher priority
29
32099IS–01/2012
AT32UC3L016/32/64
than the oldest instruction. An instruction B is younger than an instruction A if it was sent down
the pipeline later than A.
The addresses and priority of simultaneous events are shown in Table 4-4 on page 31. Some of
the exceptions are unused in AVR32UC since it has no MMU, coprocessor interface, or floatingpoint unit.
30
32099IS–01/2012
AT32UC3L016/32/64
Table 4-4.
Priority and Handler Addresses for Events
Priority
Handler Address
Name
Event source
Stored Return Address
1
0x80000000
Reset
External input
Undefined
2
Provided by OCD system
OCD Stop CPU
OCD system
First non-completed instruction
3
EVBA+0x00
Unrecoverable exception
Internal
PC of offending instruction
4
EVBA+0x04
TLB multiple hit
MPU
PC of offending instruction
5
EVBA+0x08
Bus error data fetch
Data bus
First non-completed instruction
6
EVBA+0x0C
Bus error instruction fetch
Data bus
First non-completed instruction
7
EVBA+0x10
NMI
External input
First non-completed instruction
8
Autovectored
Interrupt 3 request
External input
First non-completed instruction
9
Autovectored
Interrupt 2 request
External input
First non-completed instruction
10
Autovectored
Interrupt 1 request
External input
First non-completed instruction
11
Autovectored
Interrupt 0 request
External input
First non-completed instruction
12
EVBA+0x14
Instruction Address
CPU
PC of offending instruction
13
EVBA+0x50
ITLB Miss
MPU
PC of offending instruction
14
EVBA+0x18
ITLB Protection
MPU
PC of offending instruction
15
EVBA+0x1C
Breakpoint
OCD system
First non-completed instruction
16
EVBA+0x20
Illegal Opcode
Instruction
PC of offending instruction
17
EVBA+0x24
Unimplemented instruction
Instruction
PC of offending instruction
18
EVBA+0x28
Privilege violation
Instruction
PC of offending instruction
19
EVBA+0x2C
Floating-point
UNUSED
20
EVBA+0x30
Coprocessor absent
Instruction
PC of offending instruction
21
EVBA+0x100
Supervisor call
Instruction
PC(Supervisor Call) +2
22
EVBA+0x34
Data Address (Read)
CPU
PC of offending instruction
23
EVBA+0x38
Data Address (Write)
CPU
PC of offending instruction
24
EVBA+0x60
DTLB Miss (Read)
MPU
PC of offending instruction
25
EVBA+0x70
DTLB Miss (Write)
MPU
PC of offending instruction
26
EVBA+0x3C
DTLB Protection (Read)
MPU
PC of offending instruction
27
EVBA+0x40
DTLB Protection (Write)
MPU
PC of offending instruction
28
EVBA+0x44
DTLB Modified
UNUSED
31
32099IS–01/2012
AT32UC3L016/32/64
5. Memories
5.1
Embedded Memories
• Internal high-speed flash
– 64Kbytes (AT32UC3L064)
– 32Kbytes (AT32UC3L032)
– 16Kbytes (AT32UC3L016)
• 0 wait state access at up to 25mhz in worst case conditions
• 1 wait state access at up to 50mhz in worst case conditions
• Pipelined flash architecture, allowing burst reads from sequential Flash locations, hiding
penalty of 1 wait state access
• Pipelined flash architecture typically reduces the cycle penalty of 1 wait state operation
to only 8% compared to 0 wait state operation
• 100 000 write cycles, 15-year data retention capability
• Sector lock capabilities, bootloader protection, security bit
• 32 fuses, erased during chip erase
• User page for data to be preserved during chip erase
• Internal high-speed SRAM, single-cycle access at full speed
– 16Kbytes (AT32UC3L064, AT32UC3L032)
– 8Kbytes (AT32UC3L016)
5.2
Physical Memory Map
The system bus is implemented as a bus matrix. All system bus addresses are fixed, and they
are never remapped in any way, not even in boot. Note that AVR32 UC CPU uses unsegmented
translation, as described in the AVR32 Architecture Manual. The 32-bit physical address space
is mapped as follows:
Table 5-1.
AT32UC3L016/32/64 Physical Memory Map
Device
Table 5-2.
Start Address
Size
AT32UC3L064
AT32UC3L032
AT32UC3L016
Embedded SRAM
0x00000000
16Kbytes
16Kbytes
8Kbytes
Embedded Flash
0x80000000
64Kbytes
32Kbytes
16Kbytes
SAU Channels
0x90000000
256 bytes
256 bytes
256 bytes
HSB-PB Bridge B
0xFFFE0000
64Kbytes
64Kbytes
64Kbytes
HSB-PB Bridge A
0xFFFF0000
64Kbytes
64Kbytes
64Kbytes
Flash Memory Parameters
Part Number
Flash Size (FLASH_PW)
Number of pages
(FLASH_P)
Page size
(FLASH_W)
AT32UC3L064
64Kbytes
256
256 bytes
AT32UC3L032
32Kbytes
128
256 bytes
AT32UC3L016
16Kbytes
64
256 bytes
32
32099IS–01/2012
AT32UC3L016/32/64
5.3
Peripheral Address Map
Table 5-3.
Peripheral Address Mapping
Address
Peripheral Name
0xFFFE0000
FLASHCDW
Flash Controller - FLASHCDW
0xFFFE0400
HMATRIX
HSB Matrix - HMATRIX
0xFFFE0800
SAU
Secure Access Unit - SAU
0xFFFF0000
PDCA
Peripheral DMA Controller - PDCA
INTC
Interrupt controller - INTC
0xFFFF1000
0xFFFF1400
PM
Power Manager - PM
0xFFFF1800
SCIF
System Control Interface - SCIF
AST
Asynchronous Timer - AST
WDT
Watchdog Timer - WDT
EIC
External Interrupt Controller - EIC
0xFFFF1C00
0xFFFF2000
0xFFFF2400
0xFFFF2800
FREQM
Frequency Meter - FREQM
0xFFFF2C00
GPIO
0xFFFF3000
General Purpose Input/Output Controller - GPIO
USART0
Universal Synchronous/Asynchronous
Receiver/Transmitter - USART0
USART1
Universal Synchronous/Asynchronous
Receiver/Transmitter - USART1
USART2
Universal Synchronous/Asynchronous
Receiver/Transmitter - USART2
USART3
Universal Synchronous/Asynchronous
Receiver/Transmitter - USART3
0xFFFF3400
0xFFFF3800
0xFFFF3C00
0xFFFF4000
SPI
Serial Peripheral Interface - SPI
0xFFFF4400
TWIM0
Two-wire Master Interface - TWIM0
33
32099IS–01/2012
AT32UC3L016/32/64
Table 5-3.
Peripheral Address Mapping
0xFFFF4800
TWIM1
Two-wire Master Interface - TWIM1
TWIS0
Two-wire Slave Interface - TWIS0
TWIS1
Two-wire Slave Interface - TWIS1
PWMA
Pulse Width Modulation Controller - PWMA
0xFFFF4C00
0xFFFF5000
0xFFFF5400
0xFFFF5800
TC0
Timer/Counter - TC0
TC1
Timer/Counter - TC1
0xFFFF5C00
0xFFFF6000
ADCIFB
ADC Interface - ADCIFB
0xFFFF6400
ACIFB
Analog Comparator Interface - ACIFB
0xFFFF6800
CAT
Capacitive Touch Module - CAT
0xFFFF6C00
GLOC
Glue Logic Controller - GLOC
0xFFFF7000
AW
5.4
aWire - AW
CPU Local Bus Mapping
Some of the registers in the GPIO module are mapped onto the CPU local bus, in addition to
being mapped on the Peripheral Bus. These registers can therefore be reached both by
accesses on the Peripheral Bus, and by accesses on the local bus.
Mapping these registers on the local bus allows cycle-deterministic toggling of GPIO pins since
the CPU and GPIO are the only modules connected to this bus. Also, since the local bus runs at
CPU speed, one write or read operation can be performed per clock cycle to the local busmapped GPIO registers.
34
32099IS–01/2012
AT32UC3L016/32/64
The following GPIO registers are mapped on the local bus:
Table 5-4.
Local Bus Mapped GPIO Registers
Port
Register
Mode
Local Bus
Address
Access
0
Output Driver Enable Register (ODER)
WRITE
0x40000040
Write-only
SET
0x40000044
Write-only
CLEAR
0x40000048
Write-only
TOGGLE
0x4000004C
Write-only
WRITE
0x40000050
Write-only
SET
0x40000054
Write-only
CLEAR
0x40000058
Write-only
TOGGLE
0x4000005C
Write-only
Pin Value Register (PVR)
-
0x40000060
Read-only
Output Driver Enable Register (ODER)
WRITE
0x40000140
Write-only
SET
0x40000144
Write-only
CLEAR
0x40000148
Write-only
TOGGLE
0x4000014C
Write-only
WRITE
0x40000150
Write-only
SET
0x40000154
Write-only
CLEAR
0x40000158
Write-only
TOGGLE
0x4000015C
Write-only
-
0x40000160
Read-only
Output Value Register (OVR)
1
Output Value Register (OVR)
Pin Value Register (PVR)
35
32099IS–01/2012
AT32UC3L016/32/64
6. Supply and Startup Considerations
6.1
6.1.1
Supply Considerations
Power Supplies
The AT32UC3L016/32/64 has several types of power supply pins:
•VDDIO: Powers I/O lines. Voltage is 1.8 to 3.3V nominal.
•VDDIN: Powers I/O lines and the internal regulator. Voltage is 1.8 to 3.3V nominal.
•VDDANA: Powers the ADC. Voltage is 1.8V nominal.
•VDDCORE: Powers the core, memories, and peripherals. Voltage is 1.8V nominal.
The ground pins GND are common to VDDCORE, VDDIO, and VDDIN. The ground pin for
VDDANA is GNDANA.
When VDDCORE is not connected to VDDIN, the VDDIN voltage must be higher than 1.98V.
Refer to Section 7. on page 41 for power consumption on the various supply pins.
For decoupling recommendations for the different power supplies, please refer to the schematic
checklist.
6.1.2
Voltage Regulator
The AT32UC3L016/32/64 embeds a voltage regulator that converts from 3.3V nominal to 1.8V
with a load of up to 60mA. The regulator supplies the output voltage on VDDCORE. The regulator may only be used to drive internal circuitry in the device. VDDCORE should be externally
connected to the 1.8V domains. See Section 6.1.3 for regulator connection figures.
Adequate output supply decoupling is mandatory for VDDCORE to reduce ripple and avoid
oscillations. The best way to achieve this is to use two capacitors in parallell between
VDDCORE and GND as close to the device as possible. Please refer to Section 7.8.1 on page
55 for decoupling capacitors values and regulator characteristics.
Figure 6-1.
Supply Decoupling
3.3V
VDDIN
C IN3
CIN2
CIN1
1.8V
VDDCORE
COUT2
6.1.3
1.8V
Regulator
COUT1
Regulator Connection
The AT32UC3L016/32/64 supports three power supply configurations:
• 3.3V single supply mode
• 1.8V single supply mode
• 3.3V supply mode, with 1.8V regulated I/O lines
36
32099IS–01/2012
AT32UC3L016/32/64
6.1.3.1
3.3V Single Supply Mode
In 3.3V single supply mode the internal regulator is connected to the 3.3V source (VDDIN pin)
and its output feeds VDDCORE. Figure 6-2 shows the power schematics to be used for 3.3V
single supply mode. All I/O lines will be powered by the same power (VDDIN=VDDIO).
Figure 6-2.
3.3V Single Supply Mode
+
1.98-3.6V
-
VDDIN
VDDIO
I/O Pins
I/O Pins
VDDCORE
OSC32K
RC32K
AST
Wake
POR33
SM33
Linear
regulator
VDDANA
GND
ADC
CPU,
Peripherals,
Memories,
SCIF, BOD,
RCSYS,
DFLL
GNDANA
37
32099IS–01/2012
AT32UC3L016/32/64
6.1.3.2
1.8V Single Supply Mode
In 1.8V single supply mode the internal regulator is not used, and VDDIO and VDDCORE are
powered by a single 1.8V supply as shown in Figure 6-3. All I/O lines will be powered by the
same power (VDDIN = VDDIO = VDDCORE).
Figure 6-3.
1.8V Single Supply Mode.
+
1.62-1.98V
-
VDDIN
VDDIO
I/O Pins
I/O Pins
VDDCORE
Linear
Regulator
VDDANA
OSC32K
RC32K
AST
Wake
POR33
SM33
ADC
GNDANA
GND
CPU,
Peripherals,
Memories,
SCIF, BOD,
RCSYS,
DFLL
38
32099IS–01/2012
AT32UC3L016/32/64
6.1.3.3
3.3V Supply Mode with 1.8V Regulated I/O Lines
In this mode, the internal regulator is connected to the 3.3V source and its output is connected
to both VDDCORE and VDDIO as shown in Figure 6-4. This configuration is required in order to
use Shutdown mode.
Figure 6-4.
3.3V Supply Mode with 1.8V Regulated I/O Lines
1.98-3.6V
+
-
VDDIN
VDDIO
I/O Pins
VDDCORE
I/O Pins
Linear
Regulator
VDDANA
OSC32K
RC32K
AST
Wake
POR33
SM33
ADC
GNDANA
GND
CPU,
Peripherals,
Memories,
SCIF, BOD,
RCSYS,
DFLL
In this mode, some I/O lines are powered by VDDIN while other I/O lines are powered by VDDIO.
Refer to Section 3.2 on page 9 for description of power supply for each I/O line.
Refer to the Power Manager chapter for a description of what parts of the system are powered in
Shutdown mode.
Important note: As the regulator has a maximum output current of 60mA, this mode can only be
used in applications where the maximum I/O current is known and compatible with the core and
peripheral power consumption. Typically, great care must be used to ensure that only a few I/O
lines are toggling at the same time and drive very small loads.
39
32099IS–01/2012
AT32UC3L016/32/64
6.1.4
Power-up Sequence
6.1.4.1
Maximum Rise Rate
To avoid risk of latch-up, the rise rate of the power supplies must not exceed the values
described in Table 7-3 on page 42.
Recommended order for power supplies is also described in this chapter.
6.1.4.2
Minimum Rise Rate
The integrated Power-on Reset (POR33) circuitry monitoring the VDDIN powering supply
requires a minimum rise rate for the VDDIN power supply.
See Table 7-3 on page 42 for the minimum rise rate value.
If the application can not ensure that the minimum rise rate condition for the VDDIN power supply is met, one of the following configurations can be used:
• A logic “0” value is applied during power-up on pin PA11 until VDDIN rises above 1.2V.
• A logic “0” value is applied during power-up on pin RESET_N until VDDIN rises above 1.2V.
6.2
Startup Considerations
This chapter summarizes the boot sequence of the AT32UC3L016/32/64. The behavior after
power-up is controlled by the Power Manager. For specific details, refer to the Power Manager
chapter.
6.2.1
Starting of Clocks
After power-up, the device will be held in a reset state by the Power-on Reset (POR18 and
POR33) circuitry for a short time to allow the power to stabilize throughout the device. After
reset, the device will use the System RC Oscillator (RCSYS) as clock source. Please refer to
Table 7-17 on page 54 for the frequency for this oscillator.
On system start-up, the DFLL is disabled. All clocks to all modules are running. No clocks have
a divided frequency; all parts of the system receive a clock with the same frequency as the System RC Oscillator.
When powering up the device, there may be a delay before the voltage has stabilized, depending on the rise time of the supply used. The CPU can start executing code as soon as the supply
is above the POR18 and POR33 thresholds, and before the supply is stable. Before switching to
a high-speed clock source, the user should use the BOD to make sure the VDDCORE is above
the minimum level (1.62V).
6.2.2
Fetching of Initial Instructions
After reset has been released, the AVR32 UC CPU starts fetching instructions from the reset
address, which is 0x80000000. This address points to the first address in the internal flash.
The code read from the internal flash is free to configure the clock system and clock sources .
Please refer to the Power Manager and SCIF chapters for details.
40
32099IS–01/2012
AT32UC3L016/32/64
7. Electrical Characteristics
7.1
Absolute Maximum Ratings*
Table 7-1.
Absolute Maximum Ratings
Operating temperature..................................... -40°C to +85°C
*NOTICE:
Storage temperature...................................... -60°C to +150°C
Voltage on input pins (except for 5V pins) with respect to ground
.................................................................-0.3V to VVDD(2)+0.3V
Voltage on 5V tolerant(1) pins with respect to ground ...............
.............................................................................-0.3V to 5.5V
Total DC output current on all I/O pins - VDDIO ........... 120mA
Total DC output current on all I/O pins - VDDIN ............. 36mA
Stresses beyond those listed under
“Absolute Maximum Ratings” may cause
permanent damage to the device. This is
a stress rating only and functional operation of the device at these or other conditions beyond those indicated in the
operational sections of this specification is
not implied. Exposure to absolute maximum rating conditions for extended periods may affect device reliability.
Maximum operating voltage VDDCORE......................... 1.98V
Maximum operating voltage VDDIO, VDDIN .................... 3.6V
Notes:
1. 5V tolerant pins, see Section 3.2 ”Peripheral Multiplexing on I/O lines” on page 9
2. VVDD corresponds to either VVDDIN or VVDDIO, depending on the supply for the pin. Refer to Section 3.2 on page 9 for details.
7.2
Supply Characteristics
The following characteristics are applicable to the operating temperature range: TA = -40°C to 85°C, unless otherwise specified and are valid for a junction temperature up to T J = 100°C. Please refer to Section 6. ”Supply and Startup
Considerations” on page 36.
Table 7-2.
Supply Characteristics
Voltage
Symbol
Parameter
Min
Max
Unit
VVDDIO
DC supply peripheral I/Os
1.62
3.6
V
DC supply peripheral I/Os, 1.8V single
supply mode
1.62
1.98
V
DC supply peripheral I/Os and internal
regulator, 3.3V supply mode
1.98
3.6
V
VVDDCORE
DC supply core
1.62
1.98
V
VVDDANA
Analog supply voltage
1.62
1.98
V
VVDDIN
41
32099IS–01/2012
AT32UC3L016/32/64
Table 7-3.
Supply Rise Rates and Order(1)
Rise Rate
Symbol
Parameter
Min
Max
Unit
VVDDIO
DC supply peripheral I/Os
0
2.5
V/µs
VVDDIN
DC supply peripheral I/Os
and internal regulator
0.002
2.5
V/µs
Slower rise time requires
external power-on reset
circuit.
VVDDCORE
DC supply core
0
2.5
V/µs
Rise before or at the same
time as VDDIO
VVDDANA
Analog supply voltage
0
2.5
V/µs
Rise together with
VDDCORE
Note:
7.3
Comment
1. These values are based on simulation and characterization of other AVR microcontrollers
manufactured in the same process technology. These values are not covered by test limits in
production.
Maximum Clock Frequencies
These parameters are given in the following conditions:
• VVDDCORE = 1.62V to 1.98V
• Temperature = -40°C to 85°C
Table 7-4.
7.4
Clock Frequencies
Symbol
Parameter
fCPU
Conditions
Min
Max
Units
CPU clock frequency
50
MHz
fPBA
PBA clock frequency
50
MHz
fPBB
PBB clock frequency
50
MHz
fGCLK0
GCLK0 clock frequency
DFLLIF main reference, GCLK0
pin
150
MHz
fGCLK1
GCLK1 clock frequency
DFLLIF dithering and ssg
reference, GCLK1 pin
150
MHz
fGCLK2
GCLK2 clock frequency
AST, GCLK2 pin
80
MHz
fGCLK3
GCLK3 clock frequency
PWMA, GCLK3 pin
110
MHz
fGCLK4
GCLK4 clock frequency
CAT, ACIFB, GCLK4 pin
110
MHz
fGCLK5
GCLK5 clock frequency
GLOC
80
MHz
Power Consumption
The values in Table 7-5 are measured values of power consumption under the following conditions, except where noted:
• Operating conditions internal core supply (Figure 7-1) - this is the default configuration
– VVDDIN = 3.0V
– VVDDCORE = 1.62V, supplied by the internal regulator
42
32099IS–01/2012
AT32UC3L016/32/64
– Corresponds to the 3.3V supply mode with 1.8V regulated I/O lines, please refer to
the Supply and Startup Considerations section for more details
• Equivalent to the 3.3V single supply mode
• Consumption in 1.8V single supply mode can be estimated by subtracting the regulator static current
• Operating conditions external core supply (Figure 7-2) - used only when noted
– VVDDIN = VVDDCORE = 1.8V
– Corresponds to the 1.8V single supply mode, please refer to the Supply and Startup
Considerations section for more details
• TA = 25°C
• Oscillators
– OSC0 (crystal oscillator) stopped
– OSC32K (32KHz crystal oscillator) running with external 32KHz crystal
– DFLL running at 50MHz with OSC32K as reference
• Clocks
– DFLL used as main clock source
– CPU, HSB, and PBB clocks undivided
– PBA clock divided by 4
– The following peripheral clocks running
• PM, SCIF, AST, FLASHCDW, PBA bridge
– All other peripheral clocks stopped
• I/Os are inactive with internal pull-up
• Flash enabled in high speed mode
• POR33 disabled
43
32099IS–01/2012
AT32UC3L016/32/64
Table 7-5.
Mode
Power Consumption for Different Operating Modes
Conditions
Active(1)
Measured on
Consumption Typ
-CPU running a recursive Fibonacci algorithm
260
-CPU running a division algorithm
165
Idle(1)
92
(1)
(1)
47
Stop
37
DeepStop
23
-OSC32K and AST stopped
-Internal core supply
Static
Shutdown
Amp0
10
µA
-OSC32K running
-AST running at 1KHz
-External core supply (Figure 7-2)
5.3
-OSC32K and AST stopped
-External core supply (Figure 7-2)
4.7
-OSC32K running
-AST running at 1KHz
600
nA
-AST and OSC32K stopped
Note:
µA/MHz
58
Frozen
Standby
Unit
9
1. These numbers are valid for the measured condition only and must not be extrapolated to other frequencies.
Figure 7-1.
Measurement Schematic, Internal Core Supply
Amp0
VDDIN
VDDIO
VDDCORE
VDDANA
44
32099IS–01/2012
AT32UC3L016/32/64
Figure 7-2.
Measurement Schematic, External Core Supply
Amp0
VDDIN
VDDIO
VDDCORE
VDDANA
7.4.1
Peripheral Power Consumption
The values in Table 7-6 are measured values of power consumption under the following
conditions.
• Operating conditions internal core supply (Figure 7-1)
– VVDDIN = 3.0V
– VVDDCORE = 1.62V, supplied by the internal regulator
– Corresponds to the 3.3V supply mode with 1.8V regulated I/O lines, please refer to
the Supply and Startup Considerations section for more details
• TA = 25°C
• Oscillators
– OSC0 (crystal oscillator) stopped
– OSC32K (32KHz crystal oscillator) running with external 32KHz crystal
– DFLL running at 50MHz with OSC32K as reference
• Clocks
– DFLL used as main clock source
– CPU, HSB, and PB clocks undivided
• I/Os are inactive with internal pull-up
• Flash enabled in high speed mode
• POR33 disabled
Consumption active is the added current consumption when the module clock is turned on and
the module is doing a typical set of operations.
45
32099IS–01/2012
AT32UC3L016/32/64
Table 7-6.
Peripheral
Typical Current Consumption by Peripheral(2)
Typ Consumption Active
ACIFB
14.0
ADCIFB(1)
14.9
AST
5.6
AW USART
6.8
CAT
12.4
EIC
1.3
FREQM
3.2
GLOC
0.4
GPIO
15.9
PWMA
2.5
SPI
7.6
TC
7.2
TWIM
5.1
TWIS
3.2
USART
12.3
WDT
2.3
Unit
µA/MHz
Notes:
1. Includes the current consumption on VDDANA and ADVREFP.
2. These numbers are valid for the measured condition only and must not be extrapolated to
other frequencies.
46
32099IS–01/2012
AT32UC3L016/32/64
7.5
I/O Pin Characteristics
Table 7-7.
Normal I/O Pin Characteristics(1)
Symbol
Parameter
RPULLUP
Pull-up resistance
VIL
Input low-level voltage
VIH
Input high-level voltage
VOL
Output low-level voltage
VOH
Output high-level voltage
fMAX
Output frequency(2)
tRISE
Rise time(2)
Condition
Min
Typ
Max
Units
75
100
145
kOhm
VVDD = 3.0V
-0.3
0.3*VVDD
VVDD = 1.62V
-0.3
0.3*VVDD
VVDD = 3.6V
0.7*VVDD
VVDD + 0.3
VVDD = 1.98V
0.7*VVDD
VVDD + 0.3
VVDD = 3.0V, IOL = 3mA
0.4
VVDD = 1.62V, IOL = 2mA
0.4
VVDD = 3.0V, IOH = 3mA
VVDD - 0.4
VVDD = 1.62V, IOH = 2mA
VVDD - 0.4
V
VVDD = 3.0V, load = 10pF
45
VVDD = 3.0V, load = 30pF
23
VVDD = 3.0V, load = 10pF
4.7
VVDD = 3.0V, load = 30pF
11.5
VVDD = 3.0V, load = 10pF
4.8
VVDD = 3.0V, load = 30pF
12
1
MHz
ns
Fall time(2)
ILEAK
Input leakage current
Pull-up resistors disabled
TQFP48 package
1.4
CIN
Input capacitance, all
normal I/O pins except
PA05, PA07, PA17, PA20,
PA21, PB04, PB05
QFN48 package
1.1
TLLGA 48 package
1.1
TQFP48 package
2.7
QFN48 package
2.4
TLLGA 48 package
2.4
TQFP48 package
3.8
QFN48 package
3.5
TLLGA 48 package
3.5
Input capacitance, PA20
Input capacitance, PA05,
PA07, PA17, PA21, PB04,
PB05
CIN
Notes:
V
V
tFALL
CIN
V
µA
pF
1. VVDD corresponds to either VVDDIN or VVDDIO, depending on the supply for the pin. Refer to Section 3.2 on page 9 for details.
2. These values are based on simulation and characterization of other AVR microcontrollers manufactured in the same process technology. These values are not covered by test limits in production.
Table 7-8.
Symbol
RPULLUP
High-drive I/O Pin Characteristics(1)
Parameter
Pull-up resistance
Condition
Min
Typ
Max
PA06
30
50
110
PA02, PB01, RESET
75
100
145
PA08, PA09
10
20
45
Units
kOhm
47
32099IS–01/2012
AT32UC3L016/32/64
Table 7-8.
High-drive I/O Pin Characteristics(1)
Symbol
Parameter
Condition
Min
Typ
Max
VIL
Input low-level voltage
VVDD = 3.0V
-0.3
0.3*VVDD
VVDD = 1.62V
-0.3
0.3*VVDD
VIH
Input high-level voltage
VVDD = 3.6V
0.7*VVDD
VVDD + 0.3
VVDD = 1.98V
0.7*VVDD
VVDD + 0.3
VOL
Output low-level voltage
VOH
Output high-level voltage
Output frequency, all Highdrive I/O pins, except
PA08 and PA09(2)
VVDD = 3.0V, load = 10pF
45
fMAX
VVDD = 3.0V, load = 30pF
23
Rise time, all High-drive
I/O pins, except PA08 and
PA09(2)
VVDD = 3.0V, load = 10pF
4.7
tRISE
VVDD = 3.0V, load = 30pF
11.5
VVDD = 3.0V, load = 10pF
4.8
tFALL
Fall time, all High-drive I/O
pins, except PA08 and
PA09(2)
VVDD = 3.0V, load = 30pF
12
Output frequency, PA08
and PA09(2)
VVDD = 3.0V, load = 10pF
52
fMAX
VVDD = 3.0V, load = 30pF
39
Rise time, PA08 and
PA09(2)
VVDD = 3.0V, load = 10pF
2.9
tRISE
VVDD = 3.0V, load = 30pF
4.9
VVDD = 3.0V, load = 10pF
2.5
tFALL
Fall time, PA08 and
PA09(2)
VVDD = 3.0V, load = 30pF
4.6
ILEAK
Input leakage current
Pull-up resistors disabled
1
CIN
Input capacitance, all
High-drive I/O pins, except
PA08 and PA09
VVDD = 3.0V, IOL = 6mA
0.4
VVDD = 1.62V, IOL = 4mA
0.4
Units
V
V
V
VVDD = 3.0V, IOH = 6mA
VVDD-0.4
VVDD = 1.62V, IOH = 4mA
VVDD-0.4
V
MHz
ns
MHz
ns
TQFP48 package
2,2
QFN48 package
2.0
TLLGA 48 package
2.0
TQFP48 package
7.0
QFN48 package
6.7
TLLGA 48 package
6.7
µA
pF
Input capacitance, PA08
and PA09
CIN
Notes:
1. VVDD corresponds to either VVDDIN or VVDDIO, depending on the supply for the pin. Refer to Section 3.2 on page 9 for details.
2. These values are based on simulation and characterization of other AVR microcontrollers manufactured in the same process technology. These values are not covered by test limits in production.
Table 7-9.
High-drive I/O, 5V Tolerant, Pin Characteristics(1)
Symbol
Parameter
RPULLUP
Pull-up resistance
VIL
Input low-level voltage
Condition
Min
Typ
Max
Units
30
50
110
kOhm
VVDD = 3.0V
-0.3
0.3*VVDD
VVDD = 1.62V
-0.3
0.3*VVDD
V
48
32099IS–01/2012
AT32UC3L016/32/64
Table 7-9.
High-drive I/O, 5V Tolerant, Pin Characteristics(1)
Symbol
Parameter
VIH
Input high-level voltage
VOL
Output low-level voltage
VOH
Output high-level voltage
fMAX
Output frequency(2)
tRISE
Rise time(2)
(2)
tFALL
Fall time
ILEAK
Input leakage current
CIN
Notes:
Input capacitance
Condition
Min
Typ
Max
VVDD = 3.6V
0.7*VVDD
5.5
VVDD = 1.98V
0.7*VVDD
5.5
Units
V
VVDD = 3.0V, IOL = 6mA
0.4
VVDD = 1.62V, IOL = 4mA
0.4
V
VVDD = 3.0V, IOH = 6mA
VVDD-0.4
VVDD = 1.62V, IOH = 4mA
VVDD-0.4
V
VVDD = 3.0V, load = 10pF
87
VVDD = 3.0V, load = 30pF
58
VVDD = 3.0V, load = 10pF
2.3
VVDD = 3.0V, load = 30pF
4.3
VVDD = 3.0V, load = 10pF
1.9
VVDD = 3.0V, load = 30pF
3.7
5.5V, pull-up resistors disabled
10
MHz
ns
TQFP48 package
4.5
QFN48 package
4.2
TLLGA48 package
4.2
µA
pF
1. VVDD corresponds to either VVDDIN or VVDDIO, depending on the supply for the pin. Refer to Section 3.2 on page 9 for details.
2. These values are based on simulation and characterization of other AVR microcontrollers manufactured in the same process technology. These values are not covered by test limits in production.
Table 7-10.
TWI Pin Characteristics(1)
Symbol
Parameter
RPULLUP
Pull-up resistance
VIL
Input low-level voltage
Input high-level voltage
VIH
Condition
Min
Typ
Max
Units
25
35
60
kOhm
VVDD = 3.0V
-0.3
0.3*VVDD
VVDD = 1.62V
-0.3
0.3*VVDD
VVDD = 3.6V
0.7*VVDD
VVDD + 0.3
VVDD = 1.98V
0.7*VVDD
VVDD + 0.3
Input high-level voltage, 5V
tolerant SMBUS compliant
pins
VVDD = 3.6V
0.7*VVDD
5.5
VVDD = 1.98V
0.7*VVDD
5.5
VOL
Output low-level voltage
IOL = 3mA
ILEAK
Input leakage current
Pull-up resistors disabled
IIL
Input low leakage
1
IIH
Input high leakage
1
CIN
Input capacitance
V
V
V
0.4
V
1
TQFP48 package
3.8
QFN48 package
3.5
TLLGA48 package
3.5
µA
pF
49
32099IS–01/2012
AT32UC3L016/32/64
Table 7-10.
TWI Pin Characteristics(1)
Symbol
Parameter
tFALL
Fall time
fMAX
Max frequency
Condition
Min
Typ
Cbus = 400pF, VVDD > 2.0V
250
Cbus = 400pF, VVDD > 1.62V
470
Cbus = 400pF, VVDD > 2.0V
Max
Units
ns
400
kHz
Note:
1. VVDD corresponds to either VVDDIN or VVDDIO, depending on the supply for the pin. Refer to Section 3.2 on page 9 for details.
7.6
Oscillator Characteristics
7.6.1
Oscillator 0 (OSC0) Characteristics
7.6.1.1
Digital Clock Characteristics
The following table describes the characteristics for the oscillator when a digital clock is applied
on XIN.
Table 7-11.
Digital Clock Characteristics
Symbol
Parameter
fCPXIN
XIN clock frequency
tCPXIN
XIN clock duty cycle
tSTARTUP
Startup time
CIN
7.6.1.2
XIN input capacitance
Conditions
Min
Typ
Max
40
Units
50
MHz
60
%
0
TQFP48 package
7.0
QFN48 package
6.7
TLLGA48 package
6.7
cycles
pF
Crystal Oscillator Characteristics
The following table describes the characteristics for the oscillator when a crystal is connected
between XIN and XOUT as shown in Figure 7-3. The user must choose a crystal oscillator
where the crystal load capacitance CL is within the range given in the table. The exact value of CL
can be found in the crystal datasheet. The capacitance of the external capacitors (CLEXT) can
then be computed as follows:
C LEXT = 2 ( C L – C i ) – C PCB
where CPCB is the capacitance of the PCB and Ci is the internal equivalent load capacitance.
Table 7-12.
Crystal Oscillator Characteristics
Symbol
Parameter
fOUT
Crystal oscillator frequency
CL
Crystal load capacitance
Ci
Internal equivalent load capacitance
Conditions
Min
Typ
Max
Unit
0.45
10
16
MHz
6
18
pF
2
50
32099IS–01/2012
AT32UC3L016/32/64
Table 7-12.
Symbol
Crystal Oscillator Characteristics
Parameter
tSTARTUP
Conditions
Min
(1)
Startup time
SCIF.OSCCTRL.GAIN = 2
Current consumption
Notes:
Max
(2)
30 000
Active mode, f = 0.45MHz,
SCIF.OSCCTRL.GAIN = 0
IOSC
Typ
Unit
cycles
30
µA
Active mode, f = 10MHz,
SCIF.OSCCTRL.GAIN = 2
170
1. Please refer to the SCIF chapter for details.
2. Nominal crystal cycles.
3. These values are based on simulation and characterization of other AVR microcontrollers manufactured in the same process technology. These values are not covered by test limits in production.
Figure 7-3.
Oscillator Connection
CLEXT
XOUT
UC3L
Ci
CL
XIN
CLEXT
7.6.2
32KHz Crystal Oscillator (OSC32K) Characteristics
Figure 7-3 and the equation above also applies to the 32 KHz oscillator connection. The user
must choose a crystal oscillator where the crystal load capacitance CL is within the range given
in the table. The exact value of CL can then be found in the crystal datasheet.
Table 7-13.
32 KHz Crystal Oscillator Characteristics
Symbol
Parameter
fOUT
Crystal oscillator frequency
tSTARTUP
Startup time
CL
Crystal load capacitance(2)
Ci
Internal equivalent load
capacitance
IOSC32
Current consumption
RS
Equivalent series resistance
Conditions
Min
Typ
Max
32 768
Hz
(1)
RS = 60kOhm, CL = 9pF
30 000
6
Unit
cycles
12.5
pF
2
0.9
(2)
32 768Hz
35
µA
85
kOhm
51
32099IS–01/2012
AT32UC3L016/32/64
Notes:
1. Nominal crystal cycles.
2. These values are based on simulation and characterization of other AVR microcontrollers manufactured in the same process technology. These values are not covered by test limits in production.
7.6.3
Digital Frequency Locked Loop (DFLL) Characteristics
Table 7-14.
Symbol
Digital Frequency Locked Loop Characteristics
Parameter
Conditions
Output frequency
fOUT
(2)
Reference frequency
fREF
(2)
FINE resolution
FINE > 100, all COARSE values
Frequency drift over voltage
and temperature
Accuracy(2)
tSTARTUP
Startup time
tLOCK
Lock time
Notes:
(2)
Typ
Max
Unit
40
150
MHz
8
150
kHz
0.25
%
See
Figure 7-4
Fine lock, fREF = 32kHz, SSG disabled
0.1
0.5
Accurate lock, fREF = 32kHz, dither clk
RCSYS/2, SSG disabled
0.06
0.5
Fine lock, fREF = 8-150kHz, SSG
disabled
0.2
1
Accurate lock, fREF = 8-150kHz, dither
clk RCSYS/2, SSG disabled
0.1
1
Power consumption
IDFLL
Min
%
22
Within 90% of final values
µA/MHz
100
fREF = 32kHz, fine lock, SSG disabled
600
fREF = 32kHz, accurate lock, dithering
clock = RCSYS/2, SSG disabled
1100
µs
1. Spread Spectrum Generator (SSG) is disabled by writing a zero to the EN bit in the SCIF.DFLL0SSG register.
2. These values are based on simulation and characterization of other AVR microcontrollers manufactured in the same process technology. These values are not covered by test limits in production.
52
32099IS–01/2012
AT32UC3L016/32/64
Figure 7-4.
DFLL Open Loop Frequency Variation(1)
DFLL Open Loop Frequency variation
160
150
Frequencies (MHz)
140
130
1,98V
120
1,8V
1.62V
110
100
90
80
-40
-20
0
20
40
60
80
Tem pera ture
Note:
1. The plot shows a typical behaviour for coarse = 99 and fine = 255 in open loop mode.
7.6.4
120MHz RC Oscillator (RC120M) Characteristics
Table 7-15.
Symbol
Internal 120MHz RC Oscillator Characteristics
Parameter
Conditions
(1)
fOUT
Output frequency
IRC120M
Current consumption
tSTARTUP
Startup time
Note:
VVDDCORE = 1.8V
Min
Typ
Max
Unit
88
120
152
MHz
1.85
mA
3
µs
1. These values are based on simulation and characterization of other AVR microcontrollers manufactured in the same process technology. These values are not covered by test limits in production.
53
32099IS–01/2012
AT32UC3L016/32/64
7.6.5
32kHz RC Oscillator (RC32K) Characteristics
Table 7-16.
Symbol
32kHz RC Oscillator Characteristics
Parameter
Conditions
(1)
Min
Typ
Max
Unit
20
32
44
kHz
fOUT
Output frequency
IRC32K
Current consumption
0.6
µA
tSTARTUP
Startup time
100
µs
Note:
1. These values are based on simulation and characterization of other AVR microcontrollers manufactured in the same process technology. These values are not covered by test limits in production.
7.6.6
System RC Oscillator (RCSYS) Characteristics
Table 7-17.
System RC Oscillator Characteristics
Symbol
Parameter
Conditions
fOUT
Output frequency
Calibrated at 85°C
7.7
Min
Typ
Max
Unit
111.6
115
118.4
kHz
Flash Characteristics
Table 7-18 gives the device maximum operating frequency depending on the number of flash
wait states and the flash read mode. The FSW bit in the FLASHCDW FSR register controls the
number of wait states used when accessing the flash memory.
Table 7-18.
Maximum Operating Frequency
Flash Wait States
Read Mode
Maximum Operating Frequency
1
50MHz
High speed read mode
0
25MHz
1
30MHz
Normal read mode
0
Table 7-19.
15MHz
Flash Characteristics
Symbol
Parameter
tFPP
Page programming time
tFPE
Page erase time
tFFP
Fuse programming time
tFEA
Full chip erase time (EA)
tFCE
JTAG chip erase time (CHIP_ERASE)
Conditions
Min
Typ
Max
Unit
5
5
fCLK_HSB = 50MHz
1
ms
5
fCLK_HSB = 115kHz
170
54
32099IS–01/2012
AT32UC3L016/32/64
Table 7-20.
Flash Endurance and Data Retention
Symbol
Parameter
NFARRAY
Array endurance (write/page)
100k
NFFUSE
General Purpose fuses endurance (write/bit)
10k
tRET
Data retention
15
7.8
Min
Typ
Max
Unit
cycles
years
Analog Characteristics
7.8.1
Voltage Regulator Characteristics
Table 7-21.
VREG Electrical Characteristics
Symbol
Parameter
VVDDIN
Input voltage range
VVDDCORE
Output voltage, calibrated value
Condition
Min
Typ
Max
1.98
3.3
3.6
Units
V
Output voltage accuracy
IOUT
DC output current(1)
IVREG
Static current of internal regulator
Note:
Conditions
VVDDIN >= 1.98V
1.8
IOUT = 0.1mA to 60mA,
VVDDIN > 2.2V
2
IOUT = 0.1mA to 60mA,
VVDDIN = 1.98V to 2.2V
4
%
Normal mode
60
Low power mode
1
mA
Normal mode
20
Low power mode
6
µA
1. These values are based on simulation and characterization of other AVR microcontrollers manufactured in the same process technology. These values are not covered by test limits in production.
Table 7-22.
Decoupling Requirements
Symbol
Parameter
CIN1
Input regulator capacitor 1
33
CIN2
Input regulator capacitor 2
100
CIN3
Input regulator capacitor 3
10
µF
COUT1
Output regulator capacitor 1
100
nF
COUT2
Output regulator capacitor 2
2.2
Note:
Condition
Typ
Techno.
Units
nF
Tantalum
0.5<ESR<10Ohm
µF
1. Refer to Section 6.1.2 on page 36.
55
32099IS–01/2012
AT32UC3L016/32/64
7.8.2
Power-on Reset 18 Characteristics
Table 7-23.
POR18 Characteristics
Symbol
Parameter
Condition
VPOT+
Voltage threshold on VVDDCORE rising
VPOT-
Voltage threshold on VVDDCORE falling
tDET
Detection time
Typ
Max
1.45
1.58
Units
V
1.2
Time with VDDCORE < VPOTnecessary to generate a reset
signal
1.32
460
µs
POR18 Operating Principles
VVDDCORE
Figure 7-5.
Min
VPOT+
Reset
VPOT-
56
32099IS–01/2012
AT32UC3L016/32/64
7.8.3
Power-on Reset 33 Characteristics
Table 7-24.
POR33 Characteristics
Symbol
Parameter
Condition
VPOT+
Voltage threshold on VVDDIN rising
VPOT-
Voltage threshold on VVDDIN falling
tDET
Detection time
Time with VDDIN < VPOTnecessary to generate a reset
signal
460
µs
IPOR33
Current consumption
After tRESET
15
µA
tSTARTUP
Startup time
400
µs
Typ
Max
1.49
1.58
Units
V
1.3
1.45
POR33 Operating Principles
VVDDIN
Figure 7-6.
Min
Reset
VPOT+
VPOT-
7.8.4
Brown Out Detector Characteristics
The values in Table 7-25 describe the values of the BODLEVEL in the flash General Purpose
Fuse register.
Table 7-25.
BODLEVEL Values
BODLEVEL Value
Min
Typ
011111 binary (31) 0x1F
1.56
100111 binary (39) 0x27
1.65
Max
Units
V
57
32099IS–01/2012
AT32UC3L016/32/64
Table 7-26.
BOD Characteristics
Symbol
Parameter
Condition
VHYST
BOD hysteresis
T = 25°C
10
mV
tDET
Detection time
Time with VDDCORE <
BODLEVEL necessary to
generate a reset signal
1
µs
IBOD
Current consumption
16
µA
tSTARTUP
Startup time
5
µs
7.8.5
Min
Typ
Max
Units
Supply Monitor 33 Characteristics
Table 7-27.
Symbol
VTH
SM33 Characteristics
Parameter
Voltage threshold
Condition
(1)
Calibrated , T = 25°C
Min
Typ
Max
Units
1.675
1.75
1.825
V
Step size, between adjacent values
in SCIF.SM33.CALIB
11
VHYST
Hysteresis
30
tDET
Detection time
Time with VDDIN < VTH
necessary to generate a reset
signal
280
µs
ISM33
Current consumption
Normal mode
15
µA
tSTARTUP
Startup time
Normal mode
140
µs
Note:
mV
1. Calibration value can be read from the SCIF.SM33.CALIB field. This field is updated by the flash fuses after a reset. Refer to
SCIF chapter for details.
58
32099IS–01/2012
AT32UC3L016/32/64
7.8.6
Analog to Digital Converter Characteristics
Table 7-28.
ADC Characteristics
Symbol
Parameter
Conditions
fADC
ADC clock frequency
tSTARTUP
Startup time
Return from Idle Mode
tCONV
Conversion time (latency)
fADC = 6MHz
Throughput rate
Min
Typ
10-bit resolution mode
6
8-bit resolution mode
6
15
11
VVDD > 3.0V, fADC = 6MHz,
10-bit resolution mode,
low impedance source
460
VVDD > 3.0V, fADC = 6MHz,
8-bit resolution mode,
low impedance source
MHz
460
cycles
kSPS
Reference voltage range
VADVREFP = VVDDANA
IADC
Current consumption on VVDDANA
ADC Clock = 6MHz
300
IADVREFP
Current consumption on ADVREFP
pin
fADC = 6MHz
250
1.62
1.98
V
µA
These values are based on simulation and characterization of other AVR microcontrollers manufactured in the same process
technology. These values are not covered by test limits in production.
7.8.6.1
Inputs and Sample and Hold Aquisition Time
Table 7-29.
Analog Inputs
Symbol
Parameter
VADn
Input Voltage Range
CONCHIP
Internal Capacitance(1)
RONCHIP
Note:
Units
µs
26
VADVREFP
Note:
Max
Conditions
Internal Resistance
(1)
10-bit mode
8-bit mode
Min
0
Typ
Max
Units
VADVREFP
V
21.5
pF
VVDDIO = 3.0V to 3.6V,
VVDDCORE = 1.8V
2.55
VVDDIO = VVDDCORE = 1.62V to 1.98V
55.3
kOhm
1. These values are based on simulation and characterization of other AVR microcontrollers manufactured in the same process technology. These values are not covered by test limits in production.
An analog voltage input must be able to charge the sample and hold (S/H) capacitor in the ADC
in order to achieve maximum accuracy. Seen externally the ADC input consists of a resistor
( R ONCHIP ) and a capacitor ( CONCHIP ). In addition the resistance ( R SOURCE ) and capacitance
( C SOURCE ) of the PCB and source must be taken into account when calculating the sample and
hold time. Figure 7-7 shows the ADC input channel equivalent circuit.
59
32099IS–01/2012
AT32UC3L016/32/64
Figure 7-7.
ADC Input
RSOURCE
Positive Input
RONCHIP
CSOURCE
VIN
CONCHIP
ADCVREFP/2
The minimum sample and hold time (in ns) can be found using this formula:
t SAMPLEHOLD ≥ ( R ONCHIP + R OFFCHIP ) × ( C ONCHIP + C OFFCHIP ) × ln ( 2
n+1
)
Where n is the number of bits in the conversion. t SAMPLEHOLD is defined by the SHTIM field in the
ADCIFB ACR register. Please refer to the ADCIFB chapter for more information.
7.8.6.2
Table 7-30.
Applicable Conditions and Derating Data
Transfer Characteristics 10-bit Resolution Mode
Parameter
Conditions
Min
Resolution
Differential non-linearity
Offset error
Bit
ADC clock frequency = 6MHz
1
+/-4
LSB
+/-4
Transfer Characteristics 8-bit Resolution Mode
Parameter
Conditions
Min
Resolution
Differential non-linearity
Typ
Max
8
Integral non-linearity
Gain error
Units
+/-2
-0.9
Gain error
Offset error
Max
10
Integral non-linearity
Table 7-31.
Typ
Units
Bit
+/-0.5
ADC clock frequency = 6MHz
-0.23
0.25
+/-1
LSB
+/-1
60
32099IS–01/2012
AT32UC3L016/32/64
7.8.7
Temperature Sensor Characteristics
Temperature Sensor Characteristics(1)
Table 7-32.
Symbol
Parameter
Condition
Min
Typ
Max
Units
Gradient
1
mV/°C
ITS
Current consumption
0.5
µA
tSTARTUP
Startup time
0
µs
Note:
1. The Temperature Sensor is not calibrated. The accuracy of the Temperature Sensor is governed by the ADC accuracy.
7.8.8
Analog Comparator Characteristics
Table 7-33.
Symbol
Analog Comparator Characteristics
Parameter
Condition
Min
Typ
Max
Positive input
voltage range
-0.2
VVDDIO + 0.3
Negative input
voltage range
-0.2
VVDDIO - 0.6
Units
V
Statistical offset
VACREFN = 1.0V,
fAC = 12MHz,
filter length = 2,
hysteresis = 0(1)
Clock frequency for
GCLK4
fAC
Throughput rate(3)
fAC = 12MHz
Propagation delay
Delay from input
change to
Interrupt Status
Register Changes
IAC
Current
consumption
All channels,
VDDIO = 3.3V,
fA = 3MHz
tSTARTUP
Startup time
Input current per
pin
Notes:
20
1
⎛
+ 3⎞⎠ × t CLKACIFB
⎝ t---------------------------------------CLKACIFB × f AC
mV
12
MHz
12 000 000
Comparisons
per second
ns
420
µA
3
cycles
0.2
µA/MHz(2)
1. AC.CONFn.FLEN and AC.CONFn.HYS fields, refer to the Analog Comparator Interface chapter.
2. Referring to fAC.
3. These values are based on simulation and characterization of other AVR microcontrollers manufactured in the same process technology. These values are not covered by test limits in production.
61
32099IS–01/2012
AT32UC3L016/32/64
7.8.9
Capacitive Touch Characteristics
7.8.9.1
Table 7-34.
Discharge Current Source
DICS Characteristics
Symbol
Parameter
RREF
Internal resistor
120
kOhm
k
Trim step size
0.7
%
7.8.9.2
Table 7-35.
Min
Typ
Max
Unit
Strong Pull-up Pull-down
Strong Pull-up Pull-down
Parameter
Min
Typ
Pull-down resistor
1
Pull-up resistor
1
Max
Unit
kOhm
62
32099IS–01/2012
AT32UC3L016/32/64
7.9
Timing Characteristics
7.9.1
Startup, Reset, and Wake-up Timing
The startup, reset, and wake-up timings are calculated using the following formula:
t = t CONST + N CPU × t CPU
Where t CONST and N CPU are found in Table 7-36. t CPU is the period of the CPU clock. If
another clock source than RCSYS is selected as CPU clock the startup time of the oscillator,
t OSCSTART , must added to the wake-up time in the stop, deepstop, and static sleep modes.
Please refer to the source for the CPU clock in the ”Oscillator Characteristics” on page 50 for
more details about oscillator startup times.
Table 7-36.
Maximum Reset and Wake-up Timing(1)
Max t CONST (in µs)
Max N CPU
Parameter
Measuring
Startup time from power-up, using
regulator
Time from VDDIN crossing the VPOT+ threshold of
POR33 to the first instruction entering the decode
stage of CPU. VDDCORE is supplied by the internal
regulator.
2210
0
Startup time from power-up, no
regulator
Time from VDDIN crossing the VPOT+ threshold of
POR33 to the first instruction entering the decode
stage of CPU. VDDCORE is connected to VDDIN.
1810
0
Startup time from reset release
Time from releasing a reset source (except POR18,
POR33, and SM33) to the first instruction entering
the decode stage of CPU.
170
0
Idle
0
19
Frozen
0
110
0
110
27 + t OSCSTART
116
Deepstop
27 + t OSCSTART
116
Static
97 + t OSCSTART
116
1180
0
Standby
Wake-up
Stop
Wake-up from shutdown
Note:
From wake-up event to the first instruction of an
interrupt routine entering the decode stage of the
CPU.
From wake-up event to the first instruction entering
the decode stage of the CPU.
1. These values are based on simulation and characterization of other AVR microcontrollers manufactured in the same process technology. These values are not covered by test limits in production.
7.9.2
RESET_N Timing
Table 7-37.
RESET_N Waveform Parameters(1)
Symbol
Parameter
tRESET
RESET_N minimum pulse length
Note:
Conditions
Min
10
Max
Units
ns
1. These values are based on simulation and characterization of other AVR microcontrollers manufactured in the same process technology. These values are not covered by test limits in production.
63
32099IS–01/2012
AT32UC3L016/32/64
7.9.3
USART in SPI Mode Timing
7.9.3.1
Master mode
Figure 7-8.
USART in SPI Master Mode With (CPOL= CPHA= 0) or (CPOL= CPHA= 1)
SPCK
MISO
USPI0
USPI1
MOSI
USPI2
Figure 7-9.
USART in SPI Master Mode With (CPOL= 0 and CPHA= 1) or (CPOL= 1 and
CPHA= 0)
SPCK
MISO
USPI3
USPI4
MOSI
USPI5
Table 7-38.
Symbol
USART in SPI Mode Timing, Master Mode(1)
Parameter
Conditions
USPI0
MISO setup time before SPCK rises
USPI1
MISO hold time after SPCK rises
USPI2
SPCK rising to MOSI delay
USPI3
MISO setup time before SPCK falls
USPI4
MISO hold time after SPCK falls
USPI5
SPCK falling to MOSI delay
Notes:
Min
30.0+
VVDDIO from
3.0V to 3.6V,
maximum
external
capacitor =
40pF
Max
Units
tSAMPLE(2)
0
8.5
ns
25.5 + tSAMPLE(2)
0
13.6
1. These values are based on simulation and characterization of other AVR microcontrollers manufactured in the same process technology. These values are not covered by test limits in production.
t SPCK
1⎞
2. Where: t SAMPLE = t SPCK – ⎛ -------------------------------------- × t CLKUSART
⎝ 2×t
2⎠
CLKUSART
64
32099IS–01/2012
AT32UC3L016/32/64
Maximum SPI Frequency, Master Output
The maximum SPI master output frequency is given by the following formula:
1 f CLKSPI × 2
f SPCKMAX = MIN (f PINMAX,------------, -----------------------------)
SPIn
9
Where SPIn is the MOSI delay, USPI2 or USPI5 depending on CPOL and NCPHA. f PINMAX is
the maximum frequency of the SPI pins. Please refer to the I/O Pin Characteristics section for
the maximum frequency of the pins. f CLKSPI is the maximum frequency of the CLK_SPI. Refer
to the SPI chapter for a description of this clock.
Maximum SPI Frequency, Master Input
The maximum SPI master input frequency is given by the following formula:
f CLKSPI × 2
1
f SPCKMAX = MIN (------------------------------------,-----------------------------)
SPIn + t VALID
9
Where SPIn is the MISO setup and hold time, USPI0 + USPI1 or USPI3 + USPI4 depending
on CPOL and NCPHA. T VALID is the SPI slave response time. Please refer to the SPI slave
datasheet for T VALID . f CLKSPI is the maximum frequency of the CLK_SPI. Refer to the SPI
chapter for a description of this clock.
7.9.3.2
Slave mode
Figure 7-10. USART in SPI Slave Mode With (CPOL= 0 and CPHA= 1) or (CPOL= 1 and
CPHA= 0)
SPCK
MISO
USPI6
MOSI
USPI7
USPI8
Figure 7-11. USART in SPI Slave Mode With (CPOL= CPHA= 0) or (CPOL= CPHA= 1)
SPCK
MISO
USPI9
MOSI
USPI10
USPI11
65
32099IS–01/2012
AT32UC3L016/32/64
Figure 7-12. USART in SPI Slave Mode NPCS Timing
USPI12
USPI13
USPI14
USPI15
SPCK, CPOL=0
SPCK, CPOL=1
NSS
Table 7-39.
USART in SPI mode Timing, Slave Mode(1)
Symbol
Parameter
Conditions
USPI6
SPCK falling to MISO delay
Max
Units
27.6
USPI7
MOSI setup time before SPCK rises
USPI8
MOSI hold time after SPCK rises
USPI9
SPCK rising to MISO delay
USPI10
MOSI setup time before SPCK falls
USPI11
MOSI hold time after SPCK falls
USPI12
NSS setup time before SPCK rises
USPI13
NSS hold time after SPCK falls
USPI14
NSS setup time before SPCK falls
USPI15
NSS hold time after SPCK rises
Notes:
Min
tSAMPLE(2)
+ tCLK_USART
0
VVDDIO from
3.0V to 3.6V,
maximum
external
capacitor =
40pF
27.2
tSAMPLE(2)
+ tCLK_USART
ns
0
25.0
0
25.0
0
1. These values are based on simulation and characterization of other AVR microcontrollers manufactured in the same process technology. These values are not covered by test limits in production.
t SPCK
1
2. Where: t SAMPLE = t SPCK – ⎛ -----------------------------------+ ---⎞ × t CLKUSART
⎝ 2×t
2⎠
CLKUSART
Maximum SPI Frequency, Slave Input Mode
The maximum SPI slave input frequency is given by the following formula:
f CLKSPI × 2 1
f SPCKMAX = MIN (-----------------------------,------------)
9
SPIn
Where SPIn is the MOSI setup and hold time, USPI7 + USPI8 or USPI10 + USPI11 depending
on CPOL and NCPHA. f CLKSPI is the maximum frequency of the CLK_SPI. Refer to the SPI
chapter for a description of this clock.
Maximum SPI Frequency, Slave Output Mode
66
32099IS–01/2012
AT32UC3L016/32/64
The maximum SPI slave output frequency is given by the following formula:
f CLKSPI × 2
1
f SPCKMAX = MIN (-----------------------------, f PINMAX,------------------------------------)
9
SPIn + t SETUP
Where SPIn is the MISO delay, USPI6 or USPI9 depending on CPOL and NCPHA. T SETUP is
the SPI master setup time. Please refer to the SPI master datasheet for T SETUP . f CLKSPI is the
maximum frequency of the CLK_SPI. Refer to the SPI chapter for a description of this
clock. f PINMAX is the maximum frequency of the SPI pins. Please refer to the I/O Pin Characteristics section for the maximum frequency of the pins.
7.9.4
SPI Timing
7.9.4.1
Master mode
Figure 7-13. SPI Master Mode With (CPOL= NCPHA= 0) or (CPOL= NCPHA= 1)
SPCK
MISO
SPI0
SPI1
MOSI
SPI2
Figure 7-14. SPI Master Mode With (CPOL= 0 and NCPHA= 1) or (CPOL= 1 and NCPHA= 0)
SPCK
MISO
SPI3
SPI4
MOSI
SPI5
67
32099IS–01/2012
AT32UC3L016/32/64
Table 7-40.
SPI Timing, Master Mode(1)
Symbol
Parameter
SPI0
MISO setup time before SPCK rises
SPI1
MISO hold time after SPCK rises
SPI2
SPCK rising to MOSI delay
SPI3
MISO setup time before SPCK falls
SPI4
MISO hold time after SPCK falls
SPI5
SPCK falling to MOSI delay
Note:
Conditions
Min
Max
Units
28.4 + (tCLK_SPI)/2
VVDDIO from
3.0V to 3.6V,
maximum
external
capacitor =
40pF
0
7.1
ns
22.8 + (tCLK_SPI)/2
0
11.0
1. These values are based on simulation and characterization of other AVR microcontrollers manufactured in the same process technology. These values are not covered by test limits in production.
Maximum SPI Frequency, Master Output
The maximum SPI master output frequency is given by the following formula:
1
f SPCKMAX = MIN (f PINMAX,------------)
SPIn
Where SPIn is the MOSI delay, SPI2 or SPI5 depending on CPOL and NCPHA. f PINMAX is the
maximum frequency of the SPI pins. Please refer to the I/O Pin Characteristics section for the
maximum frequency of the pins.
Maximum SPI Frequency, Master Input
The maximum SPI master input frequency is given by the following formula:
1
f SPCKMAX = -----------------------------------SPIn + t VALID
Where SPIn is the MISO setup and hold time, SPI0 + SPI1 or SPI3 + SPI4 depending on
CPOL and NCPHA. t VALID is the SPI slave response time. Please refer to the SPI slave
datasheet for t VALID .
7.9.4.2
Slave mode
Figure 7-15. SPI Slave Mode With (CPOL= 0 and NCPHA= 1) or (CPOL= 1 and NCPHA= 0)
SPCK
MISO
SPI6
MOSI
SPI7
SPI8
68
32099IS–01/2012
AT32UC3L016/32/64
Figure 7-16. SPI Slave Mode With (CPOL= NCPHA= 0) or (CPOL= NCPHA= 1)
SPCK
MISO
SPI9
MOSI
SPI10
Figure 7-17.
SPI11
SPI Slave Mode NPCS Timing
SPI12
SPI13
SPI14
SPI15
SPCK, CPOL=0
SPCK, CPOL=1
NPCS
Table 7-41.
SPI Timing, Slave Mode(1)
Symbol
Parameter
SPI6
SPCK falling to MISO delay
SPI7
MOSI setup time before SPCK rises
SPI8
MOSI hold time after SPCK rises
SPI9
SPCK rising to MISO delay
SPI10
MOSI setup time before SPCK falls
SPI11
MOSI hold time after SPCK falls
SPI12
NPCS setup time before SPCK rises
SPI13
NPCS hold time after SPCK falls
0.2
SPI14
NPCS setup time before SPCK falls
2.2
SPI15
NPCS hold time after SPCK rises
Note:
Conditions
Min
Max
Units
30.8
0
4.1
VVDDIO from
3.0V to 3.6V,
maximum
external
capacitor =
40pF
29.9
0
ns
3.5
1.9
0
1. These values are based on simulation and characterization of other AVR microcontrollers manufactured in the same process technology. These values are not covered by test limits in production.
Maximum SPI Frequency, Slave Input Mode
69
32099IS–01/2012
AT32UC3L016/32/64
The maximum SPI slave input frequency is given by the following formula:
1
f SPCKMAX = MIN (f CLKSPI,------------)
SPIn
Where SPIn is the MOSI setup and hold time, SPI7 + SPI8 or SPI10 + SPI11 depending on
CPOL and NCPHA. f CLKSPI is the maximum frequency of the CLK_SPI. Refer to the SPI chapter for a description of this clock.
Maximum SPI Frequency, Slave Output Mode
The maximum SPI slave output frequency is given by the following formula:
1
f SPCKMAX = MIN (f PINMAX,------------------------------------)
SPIn + t SETUP
Where SPIn is the MISO delay, SPI6 or SPI9 depending on CPOL and NCPHA. t SETUP is the
SPI master setup time. Please refer to the SPI master datasheet for t SETUP . f PINMAX is the maximum frequency of the SPI pins. Please refer to the I/O Pin Characteristics section for the
maximum frequency of the pins.
7.9.5
TWIM/TWIS Timing
Figure 7-42 shows the TWI-bus timing requirements and the compliance of the device with
them. Some of these requirements (tr and tf) are met by the device without requiring user intervention. Compliance with the other requirements (tHD-STA, tSU-STA, tSU-STO, tHD-DAT, tSU-DAT-TWI, tLOWTWI, tHIGH, and fTWCK) requires user intervention through appropriate programming of the relevant
TWIM and TWIS user interface registers. Please refer to the TWIM and TWIS sections for more
information.
Table 7-42.
TWI-Bus Timing Requirements
Minimum
Symbol
Parameter
Mode
Requirement
Standard(1)
tr
TWCK and TWD rise time
tf
TWCK and TWD fall time
tHD-STA
(Repeated) START hold time
tSU-STA
(Repeated) START set-up time
tSU-STO
STOP set-up time
tHD-DAT
Data hold time
Maximum
Device
Requirement
Device
-
1000
20 + 0.1Cb
300
-
300
20 + 0.1Cb
300
Unit
ns
Fast(1)
Standard
ns
Fast
Standard
4
Fast
0.6
Standard
4.7
Fast
0.6
Standard
4.0
Fast
0.6
Standard
Fast
0.3(2)
tclkpb
-
μs
tclkpb
-
μs
4tclkpb
-
μs
3.45()
2tclkpb
0.9()
15tprescaled + tclkpb
μs
70
32099IS–01/2012
AT32UC3L016/32/64
Table 7-42.
TWI-Bus Timing Requirements
Minimum
Symbol
Parameter
tSU-DAT-TWI Data set-up time
tSU-DAT
tLOW-TWI
Mode
Requirement
Standard
250
Fast
100
-
Device
-
Standard
4.7
Fast
1.3
TWCK LOW period
tLOW
-
tHIGH
TWCK HIGH period
fTWCK
TWCK frequency
-
Standard
4.0
Fast
0.6
Standard
Notes:
Maximum
Requirement
Unit
2tclkpb
-
ns
tclkpb
-
-
4tclkpb
-
μs
tclkpb
-
-
8tclkpb
-
μs
100
-
Fast
Device
400
1
-----------------------12t clkpb
kHz
1. Standard mode: f TWCK ≤ 100 kHz ; fast mode: f TWCK > 100 kHz .
2. A device must internally provide a hold time of at least 300 ns for TWD with reference to the falling edge of TWCK.
Notations:
Cb = total capacitance of one bus line in pF
tclkpb = period of TWI peripheral bus clock
tprescaled = period of TWI internal prescaled clock (see chapters on TWIM and TWIS)
The maximum tHD;DAT has only to be met if the device does not stretch the LOW period (tLOW-TWI)
of TWCK.
71
32099IS–01/2012
AT32UC3L016/32/64
7.9.6
JTAG Timing
Figure 7-18. JTAG Interface Signals
JTAG2
TCK
JTAG0
JTAG1
TMS/TDI
JTAG3
JTAG4
JTAG7
JTAG8
TDO
JTAG5
JTAG6
Boundary
Scan Inputs
Boundary
Scan Outputs
JTAG9
JTAG10
Table 7-43.
JTAG Timings(1)
Symbol
Parameter
JTAG0
TCK Low Half-period
23.2
JTAG1
TCK High Half-period
8.8
JTAG2
TCK Period
32.0
JTAG3
TDI, TMS Setup before TCK High
JTAG4
TDI, TMS Hold after TCK High
JTAG5
TDO Hold Time
JTAG6
TCK Low to TDO Valid
JTAG7
Boundary Scan Inputs Setup Time
JTAG8
Boundary Scan Inputs Hold Time
5.0
JTAG9
Boundary Scan Outputs Hold Time
8.7
JTAG10
TCK to Boundary Scan Outputs Valid
Note:
Conditions
VVDDIO from
3.0V to 3.6V,
maximum
external
capacitor =
40pF
Min
Max
Units
3.9
0.6
ns
4.5
23.2
0
17.7
1. These values are based on simulation and characterization of other AVR microcontrollers manufactured in the same process technology. These values are not covered by test limits in production.
72
32099IS–01/2012
AT32UC3L016/32/64
8. Mechanical Characteristics
8.1
8.1.1
Thermal Considerations
Thermal Data
Table 8-1 summarizes the thermal resistance data depending on the package.
Table 8-1.
8.1.2
Thermal Resistance Data
Symbol
Parameter
Condition
Package
Typ
θJA
Junction-to-ambient thermal resistance
Still Air
TQFP48
63.2
θJC
Junction-to-case thermal resistance
TQFP48
21.8
θJA
Junction-to-ambient thermal resistance
QFN48
28.3
θJC
Junction-to-case thermal resistance
QFN48
2.5
θJA
Junction-to-ambient thermal resistance
TLLGA48
25.4
θJC
Junction-to-case thermal resistance
TLLGA48
12.7
Still Air
Still Air
Unit
°C/W
°C/W
°C/W
Junction Temperature
The average chip-junction temperature, TJ, in °C can be obtained from the following:
1.
T J = T A + ( P D × θ JA )
2.
T J = T A + ( P D × ( θ HEATSINK + θ JC ) )
where:
• θJA = package thermal resistance, Junction-to-ambient (°C/W), provided in Table 8-1.
• θJC = package thermal resistance, Junction-to-case thermal resistance (°C/W), provided in
Table 8-1.
• θHEAT SINK = cooling device thermal resistance (°C/W), provided in the device datasheet.
• PD = device power consumption (W) estimated from data provided in the Section 7.4 on page
42.
• TA = ambient temperature (°C).
From the first equation, the user can derive the estimated lifetime of the chip and decide if a
cooling device is necessary or not. If a cooling device is to be fitted on the chip, the second
equation should be used to compute the resulting average chip-junction temperature TJ in °C.
73
32099IS–01/2012
AT32UC3L016/32/64
8.2
Package Drawings
Figure 8-1.
TQFP-48 Package Drawing
Table 8-2.
Device and Package Maximum Weight
140
Table 8-3.
mg
Package Characteristics
Moisture Sensitivity Level
Table 8-4.
MSL3
Package Reference
JEDEC Drawing Reference
MS-026
JESD97 Classification
E3
74
32099IS–01/2012
AT32UC3L016/32/64
Figure 8-2.
Note:
QFN-48 Package Drawing
The exposed pad is not connected to anything internally, but should be soldered to ground to increase board level reliability.
Table 8-5.
Device and Package Maximum Weight
140
Table 8-6.
mg
Package Characteristics
Moisture Sensitivity Level
Table 8-7.
MSL3
Package Reference
JEDEC Drawing Reference
M0-220
JESD97 Classification
E3
75
32099IS–01/2012
AT32UC3L016/32/64
Figure 8-3.
TLLGA-48 Package Drawing
Table 8-8.
Device and Package Maximum Weight
39.3
Table 8-9.
mg
Package Characteristics
Moisture Sensitivity Level
Table 8-10.
MSL3
Package Reference
JEDEC Drawing Reference
N/A
JESD97 Classification
E4
76
32099IS–01/2012
AT32UC3L016/32/64
8.3
Soldering Profile
Table 8-11 gives the recommended soldering profile from J-STD-20.
Table 8-11.
Soldering Profile
Profile Feature
Green Package
Average Ramp-up Rate (217°C to Peak)
3°C/s max
Preheat Temperature 175°C ±25°C
150-200°C
Time Maintained Above 217°C
60-120s
Time within 5°C of Actual Peak Temperature
30s
Peak Temperature Range
260°C
Ramp-down Rate
6°C/s max
Time 25°C to Peak Temperature
8 minutes max
A maximum of three reflow passes is allowed per component.
77
32099IS–01/2012
AT32UC3L016/32/64
9. Ordering Information
Table 9-1.
Ordering Information
Device
Ordering Code
Carrier Type
AT32UC3L064-AUTES
ES
AT32UC3L064-AUT
Tray
AT32UC3L064-AUR
Tape & Reel
Package
Package Type
Temperature Operating
Range
TQFP 48
JESD97 Classification E3
AT32UC3L064
AT32UC3L064-ZAUES
ES
AT32UC3L064-ZAUT
Tray
AT32UC3L064-ZAUR
Tape & Reel
AT32UC3L064-D3HES
ES
AT32UC3L064-D3HT
Tray
AT32UC3L064-D3HR
Tape & Reel
AT32UC3L032-AUT
Tray
AT32UC3L032-AUR
Tape & Reel
QFN 48
TLLGA 48
JESD97 Classification E4
TQFP 48
Industrial (-40°C to 85°C)
JESD97 Classification E3
AT32UC3L032-ZAUT
Tray
AT32UC3L032-ZAUR
Tape & Reel
AT32UC3L032-D3HT
Tray
AT32UC3L032-D3HR
Tape & Reel
AT32UC3L032
QFN 48
TLLGA 48
AT32UC3L016-AUT
Tray
AT32UC3L016-AUR
Tape & Reel
AT32UC3L016-ZAUT
Tray
AT32UC3L016-ZAUR
Tape & Reel
AT32UC3L016-D3HT
Tray
AT32UC3L016-D3HR
Tape & Reel
JESD97 Classification E4
TQFP 48
JESD97 Classification E3
AT32UC3L016
QFN 48
TLLGA 48
JESD97 Classification E4
78
32099IS–01/2012
AT32UC3L016/32/64
10. Errata
10.1
10.1.1
Rev. E
Processor and Architecture
1. Hardware breakpoints may corrupt MAC results
Hardware breakpoints on MAC instructions may corrupt the destination register of the MAC
instruction.
Fix/Workaround
Place breakpoints on earlier or later instructions.
2. Privilege violation when using interrupts in application mode with protected system
stack
If the system stack is protected by the MPU and an interrupt occurs in application mode, an
MPU DTLB exception will occur.
Fix/Workaround
Make a DTLB Protection (Write) exception handler which permits the interrupt request to be
handled in privileged mode.
10.1.2
FLASHCDW
1. Flash self programming may fail in one wait state mode
Writes in flash and user pages may fail if executing code is located in address space
mapped to flash, and the flash controller is configured in one wait state mode (the Flash
Wait State bit in the Flash Control Register (FCR.FWS) is one).
Fix/Workaround
Solution 1: Configure the flash controller in zero wait state mode (FCR.FWS=0).
Solution 2: Configure the HMATRIX master 1 (CPU Instruction) to use the unlimited burst
length transfer mode (MCFG1.ULBT=0), and the HMATRIX slave 0 (FLASHCDW) to use
the maximum slot cycle limit (SCFG0.SLOT_CYCLE=255).
10.1.3
Power Manager
1. Clock sources will not be stopped in Static mode if the difference between CPU and
PBx division factor is larger than 4
If the division factor between the CPU/HSB and PBx frequencies is more than 4 when entering a sleep mode where the system RC oscillator (RCSYS) is turned off, the high speed
clock sources will not be turned off. This will result in a significantly higher power consumption during the sleep mode.
Fix/Workaround
Before going to sleep modes where RCSYS is stopped, make sure the division factor
between CPU/HSB and PBx frequencies is less than or equal to 4.
2. Clock Failure Detector (CFD) can be issued while turning off the CFD
While turning off the CFD, the CFD bit in the Status Register (SR) can be set. This will
change the main clock source to RCSYS.
Fix/Workaround
Solution 1: Enable CFD interrupt. If CFD interrupt is issues after turning off the CFD, switch
back to original main clock source.
Solution 2: Only turn off the CFD while running the main clock on RCSYS.
3. Sleepwalking in idle and frozen sleep mode will mask all other PB clocks
79
32099IS–01/2012
AT32UC3L016/32/64
If the CPU is in idle or frozen sleep mode and a module is in a state that triggers sleep walking, all PB clocks will be masked except the PB clock to the sleepwalking module.
Fix/Workaround
Mask all clock requests in the PM.PPCR register before going into idle or frozen mode.
10.1.4
SCIF
1. PCLKSR.OSC32RDY bit might not be cleared after disabling OSC32K
In some cases the OSC32RDY bit in the PCLKSR register will not be cleared when OSC32K
is disabled.
Fix/Workaround
When re-enabling the OSC32K, read the PCLKSR.OSC32RDY bit. If this bit is:
0: Follow normal procedures.
1: Ignore the PCLKSR.OSC32RDY and ISR.OSC32RDY bit. Use the Frequency Meter
(FREQM) to determine if the OSC32K clock is ready. The OSC32K clock is ready when the
FREQM measures a non-zero frequency.
2. The RC32K output on PA20 is not always permanently disabled
The RC32K output on PA20 may sometimes re-appear.
Fix/Workaround
Before using RC32K for other purposes, the following procedure has to be followed in order
to properly disable it:
- Run the CPU on RCSYS
- Disable the output to PA20 by writing a zero to PM.PPCR.RC32OUT
- Enable RC32K by writing a one to SCIF.RC32KCR.EN, and wait for this bit to be read as
one
- Disable RC32K by writing a zero to SCIF.RC32KCR.EN, and wait for this bit to be read as
zero.
10.1.5
AST
1. Reset may set status bits in the AST
If a reset occurs and the AST is enabled, the SR.ALARM0, SR.PER0, and SR.OVF bits may
be set.
Fix/Workaround
If the part is reset and the AST is used, clear all bits in the Status Register before entering
sleep mode.
2. AST wake signal is released one AST clock cycle after the BUSY bit is cleared
After writing to the Status Clear Register (SCR) the wake signal is released one AST clock
cycle after the BUSY bit in the Status Register (SR.BUSY) is cleared. If entering sleep mode
directly after the BUSY bit is cleared the part will wake up immediately.
Fix/Workaround
Read the Wake Enable Register (WER) and write this value back to the same register. Wait
for BUSY to clear before entering sleep mode.
10.1.6
WDT
1. Clearing the Watchdog Timer (WDT) counter in second half of timeout period will
issue a Watchdog reset
If the WDT counter is cleared in the second half of the timeout period, the WDT will immediately issue a Watchdog reset.
Fix/Workaround
80
32099IS–01/2012
AT32UC3L016/32/64
Use twice as long timeout period as needed and clear the WDT counter within the first half
of the timeout period. If the WDT counter is cleared after the first half of the timeout period,
you will get a Watchdog reset immediately. If the WDT counter is not cleared at all, the time
before the reset will be twice as long as needed.
2. WDT Control Register does not have synchronization feedback
When writing to the Timeout Prescale Select (PSEL), Time Ban Prescale Select (TBAN),
Enable (EN), or WDT Mode (MODE) fieldss of the WDT Control Register (CTRL), a synchronizer is started to propagate the values to the WDT clcok domain. This synchronization
takes a finite amount of time, but only the status of the synchronization of the EN bit is
reflected back to the user. Writing to the synchronized fields during synchronization can lead
to undefined behavior.
Fix/Workaround
-When writing to the affected fields, the user must ensure a wait corresponding to 2 clock
cycles of both the WDT peripheral bus clock and the selected WDT clock source.
-When doing writes that changes the EN bit, the EN bit can be read back until it reflects the
written value.
10.1.7
GPIO
1. Clearing GPIO interrupt may fail
Writing a one to the GPIO.IFRC register to clear an interrupt will be ignored if interrupt is
enabled for the corresponding port.
Fix/Workaround
Disable the interrupt, clear it by writing a one to GPIO.IFRC, then enable the interrupt.
10.1.8
SPI
1. SPI data transfer hangs with CSR0.CSAAT==1 and MR.MODFDIS==0
When CSR0.CSAAT==1 and mode fault detection is enabled (MR.MODFDIS==0), the SPI
module will not start a data transfer.
Fix/Workaround
Disable mode fault detection by writing a one to MR.MODFDIS.
2. Disabling SPI has no effect on the SR.TDRE bit
Disabling SPI has no effect on the SR.TDRE bit whereas the write data command is filtered
when SPI is disabled. Writing to TDR when SPI is disabled will not clear SR.TDRE. If SPI is
disabled during a PDCA transfer, the PDCA will continue to write data to TDR until its buffer
is empty, and this data will be lost.
Fix/Workaround
Disable the PDCA, add two NOPs, and disable the SPI. To continue the transfer, enable the
SPI and PDCA.
3. SPI disable does not work in SLAVE mode
SPI disable does not work in SLAVE mode.
Fix/Workaround
Read the last received data, then perform a software reset by writing a one to the Software
Reset bit in the Control Register (CR.SWRST).
4. SPI bad serial clock generation on 2nd chip_select when SCBR=1, CPOL=1, and
NCPHA=0
When multiple chip selects (CS) are in use, if one of the baudrates equal 1 while one
(CSRn.SCBR=1) of the others do not equal 1, and CSRn.CPOL=1 and CSRn.NCPHA=0,
then an additional pulse will be generated on SCK.
81
32099IS–01/2012
AT32UC3L016/32/64
Fix/Workaround
When multiple CS are in use, if one of the baudrates equals 1, the others must also equal 1
if CSRn.CPOL=1 and CSRn.NCPHA=0.
5. SPI mode fault detection enable causes incorrect behavior
When mode fault detection is enabled (MR.MODFDIS==0), the SPI module may not operate
properly.
Fix/Workaround
Always disable mode fault detection before using the SPI by writing a one to MR.MODFDIS.
10.1.9
TWI
1. TWIM SR.IDLE goes high immediately when NAK is received
When a NAK is received and there is a non-zero number of bytes to be transmitted,
SR.IDLE goes high immediately and does not wait for the STOP condition to be sent. This
does not cause any problem just by itself, but can cause a problem if software waits for
SR.IDLE to go high and then immediately disables the TWIM by writing a one to CR.MDIS.
Disabling the TWIM causes the TWCK and TWD pins to go high immediately, so the STOP
condition will not be transmitted correctly.
Fix/Workaround
If possible, do not disable the TWIM. If it is absolutely necessary to disable the TWIM, there
must be a software delay of at least two TWCK periods between the detection of
SR.IDLE==1 and the disabling of the TWIM.
2. TWIM TWALM polarity is wrong
The TWALM signal in the TWIM is active high instead of active low.
Fix/Workaround
Use an external inverter to invert the signal going into the TWIM. When using both TWIM
and TWIS on the same pins, the TWALM cannot be used.
3. TWIS may not wake the device from sleep mode
If the CPU is put to a sleep mode (except Idle and Frozen) directly after a TWI Start condition, the CPU may not wake upon a TWIS address match. The request is NACKed.
Fix/Workaround
When using the TWI address match to wake the device from sleep, do not switch to sleep
modes deeper than Frozen. Another solution is to enable asynchronous EIC wake on the
TWIS clock (TWCK) or TWIS data (TWD) pins, in order to wake the system up on bus
events.
4. SMBALERT bit may be set after reset
The SMBus Alert (SMBALERT) bit in the Status Register (SR) might be erroneously set after
system reset.
Fix/Workaround
After system reset, clear the SR.SMBALERT bit before commencing any TWI transfer.
5. Clearing the NAK bit before the BTF bit is set locks up the TWI bus
When the TWIS is in transmit mode, clearing the NAK Received (NAK) bit of the Status Register (SR) before the end of the Acknowledge/Not Acknowledge cycle will cause the TWIS to
attempt to continue transmitting data, thus locking up the bus.
Fix/Workaround
Clear SR.NAK only after the Byte Transfer Finished (BTF) bit of the same register has been
set.
82
32099IS–01/2012
AT32UC3L016/32/64
6. TWIS stretch on Address match error
When the TWIS stretches TWCK due to a slave address match, it also holds TWD low for
the same duration if it is to be receiving data. When TWIS releases TWCK, it releases TWD
at the same time. This can cause a TWI timing violation.
Fix/Workaround
None.
10.1.10
PWMA
1. BUSY bit is never cleared after writes to the Control Register (CR)
When writing a non-zero value to CR.TOP, CR.SPREAD, or CR.TCLR when the PWMA is
disabled (CR.EN == 0), the BUSY bit in the Status Register (SR.BUSY) will be set, but never
cleared.
Fix/Workaround
When writing a non-zero value to CR.TOP, CR.SPREAD, or CR.TCLR, make sure the
PWMA is enabled, or simultaneously enable the PWMA by writing a one to CR.EN.
2. Incoming peripheral events are discarded during duty cycle register update
Incoming peripheral events to all applied channels will be discarded if a duty cycle update is
received from the user interface in the same PWMA clock period.
Fix/Workaround
Ensure that duty cycle writes from the user interface are not performed in a PWMA period
when an incoming peripheral event is expected.
10.1.11
ADCIFB
1. Using STARTUPTIME larger than 0x1F will freeze the ADC
Writing a value larger than 0x1F to the Startup Time field in the ADC Configuration Register
(ACR.STARTUP) will freeze the ADC, and the Busy Status bit in the Status Register
(SR.BUSY) will never be cleared.
Fix/Workaround
Do not write values larger than 0x1F to ACR.STARTUP.
10.1.12
CAT
1. CAT asynchronous wake will be delayed by one AST event period
If the CAT detects a condition the should asynchronously wake the device in static mode,
the asynchronous wake will not occur until the next AST event. For example, if the AST is
generating events to the CAT every 50ms, and the CAT detects a touch at t=9200ms, the
asynchronous wake will occur at t=9250ms.
Fix/Workaround
None.
2. CAT QMatrix sense capacitors discharged prematurely
At the end of a QMatrix burst charging sequence that uses different burst count values for
different Y lines, the Y lines may be incorrectly grounded for up to n-1 periods of the peripheral bus clock, where n is the ratio of the PB clock frequency to the GCLK_CAT frequency.
This results in premature loss of charge from the sense capacitors and thus increased variability of the acquired count values.
Fix/Workaround
Enable the 1kOhm drive resistors on all implemented QMatrix Y lines (CSA 1, 3, 5, 7, 9, 11,
13, and/or 15) by writing ones to the corresponding odd bits of the CSARES register.
3. CAT module does not terminate QTouch burst on detect
83
32099IS–01/2012
AT32UC3L016/32/64
The CAT module does not terminate a QTouch burst when the detection voltage is
reached on the sense capacitor. This can cause the sense capacitor to be charged more
than necessary. Depending on the dielectric absorption characteristics of the capacitor, this
can lead to unstable measurements.
Fix/Workaround
Use the minimum possible value for the MAX field in the ATCFG1, TG0CFG1, and
TG1CFG1 registers.
4. Autonomous CAT acquisition must be longer than AST source clock period
When using the AST to trigger CAT autonomous touch acquisition in sleep modes where the
CAT bus clock is turned off, the CAT will start several acquisitions if the period of the AST
source clock is larger than one CAT acquisition. One AST clock period after the AST trigger,
the CAT clock will automatically stop and the CAT acquisition can be stopped prematurely,
ruining the result.
Fix/Workaround
Always ensure that the ATCFG1.max field is set so that the duration of the autonomous
touch acquisition is greater than one clock period of the AST source clock.
10.1.13
aWire
1. aWire CPU clock speed robustness
The aWire memory speed request command counter warps at clock speeds below approximately 5kHz.
Fix/Workaround
None.
2. The aWire debug interface is reset after leaving Shutdown mode
If the aWire debug mode is used as debug interface and the program enters Shutdown
mode, the aWire interface will be reset when the part receives a wakeup either from the
WAKE_N pin or the AST.
Fix/Workaround
None.
10.1.14
CHIP
1. Increased Power Consumption in VDDIO in sleep modes
If OSC0 is enabled in crystal mode when entering a sleep mode where the OSC0 is disabled, this will lead to an increased power consumption in VDDIO.
Fix/Workaround
Solution 1: Disable OSC0 by writing a zero to the Oscillator Enable bit in the System Control
Interface (SCIF) Oscillator Control Register (SCIF.OSC0CTRL.OSCEN) before going to a
sleep mode where OSC0 is disabled.
Solution 2: Pull down or up XIN0 or XOUT0 with 1MOhm resistor.
10.1.15
I/O Pins
1. PA17 has low ESD tolerance
PA17 only tolerates 500V ESD pulses (Human Body Model).
Fix/Workaround
Care must be taken during manufacturing and PCB design.
84
32099IS–01/2012
AT32UC3L016/32/64
10.2
10.2.1
Rev. D
Processor and Architecture
1. Hardware breakpoints may corrupt MAC results
Hardware breakpoints on MAC instructions may corrupt the destination register of the MAC
instruction.
Fix/Workaround
Place breakpoints on earlier or later instructions.
2. Privilege violation when using interrupts in application mode with protected system
stack
If the system stack is protected by the MPU and an interrupt occurs in application mode, an
MPU DTLB exception will occur.
Fix/Workaround
Make a DTLB Protection (Write) exception handler which permits the interrupt request to be
handled in privileged mode.
10.2.2
FLASHCDW
1. Flash self programming may fail in one wait state mode
Writes in flash and user pages may fail if executing code is located in address space
mapped to flash, and the flash controller is configured in one wait state mode (the Flash
Wait State bit in the Flash Control Register (FCR.FWS) is one).
Fix/Workaround
Solution 1: Configure the flash controller in zero wait state mode (FCR.FWS=0).
Solution 2: Configure the HMATRIX master 1 (CPU Instruction) to use the unlimited burst
length transfer mode (MCFG1.ULBT=0), and the HMATRIX slave 0 (FLASHCDW) to use
the maximum slot cycle limit (SCFG0.SLOT_CYCLE=255).
10.2.3
Power Manager
1. Clock sources will not be stopped in Static mode if the difference between CPU and
PBx division factor is larger than 4
If the division factor between the CPU/HSB and PBx frequencies is more than 4 when entering a sleep mode where the system RC oscillator (RCSYS) is turned off, the high speed
clock sources will not be turned off. This will result in a significantly higher power consumption during the sleep mode.
Fix/Workaround
Before going to sleep modes where RCSYS is stopped, make sure the division factor
between CPU/HSB and PBx frequencies is less than or equal to 4.
2. External reset in Shutdown mode
If an external reset is asserted while the device is in Shutdown mode, the Power Manager
will register this as a Power-on reset (POR), and not as a SLEEP reset, in the Reset Cause
register (RCAUSE)
Fix/Workaround
None.
3. Disabling POR33 may generate spurious resets
Depending on operating conditions, POR33 may generate a spurious reset in one of the following cases:
- When POR33 is disabled from the user interface
- When SM33 supply monitor is enabled
85
32099IS–01/2012
AT32UC3L016/32/64
- When entering Shutdown mode while debugging the chip using JTAG or aWire interface
In the listed cases, writing a one to the bit VREGCR.POR33MASK in the System Control
Interface (SCIF) to mask the POR33 reset will be ineffective
Fix/Workaround
- Do not disable POR33 using the user interface
- Do not use the SM33 supply monitor
- Do not enter Shutdown mode if a debugger is connected to the chip
4. Instability when exiting sleep walking
If all the following operating conditions are true, exiting sleep walking might lead to
instability:
- The OSC0 is enabled in external clock mode (OSCCTRL0.OSCEN == 1 and
OSCCTRL0.MODE == 0)
- A sleep mode where the OSC0 is automatically disabled is entered
- The device enters sleep walking
Fix/Workaround
Do not run OSC0 in external clock mode if sleepwalking is expected to be used.
5. Clock Failure Detector (CFD) can be issued while turning off the CFD
While turning off the CFD, the CFD bit in the Status Register (SR) can be set. This will
change the main clock source to RCSYS.
Fix/Workaround
Solution 1: Enable CFD interrupt. If CFD interrupt is issues after turning off the CFD, switch
back to original main clock source.
Solution 2: Only turn off the CFD while running the main clock on RCSYS.
6. Sleepwalking in idle and frozen sleep mode will mask all other PB clocks
If the CPU is in idle or frozen sleep mode and a module is in a state that triggers sleep walking, all PB clocks will be masked except the PB clock to the sleepwalking module.
Fix/Workaround
Mask all clock requests in the PM.PPCR register before going into idle or frozen mode.
10.2.4
SCIF
1. PCLKSR.OSC32RDY bit might not be cleared after disabling OSC32K
In some cases the OSC32RDY bit in the PCLKSR register will not be cleared when OSC32K
is disabled.
Fix/Workaround
When re-enabling the OSC32K, read the PCLKSR.OSC32RDY bit. If this bit is:
0: Follow normal procedures.
1: Ignore the PCLKSR.OSC32RDY and ISR.OSC32RDY bit. Use the Frequency Meter
(FREQM) to determine if the OSC32K clock is ready. The OSC32K clock is ready when the
FREQM measures a non-zero frequency.
2. The RC32K output on PA20 is not always permanently disabled
The RC32K output on PA20 may sometimes re-appear.
Fix/Workaround
Before using RC32K for other purposes, the following procedure has to be followed in order
to properly disable it:
- Run the CPU on RCSYS
- Disable the output to PA20 by writing a zero to PM.PPCR.RC32OUT
- Enable RC32K by writing a one to SCIF.RC32KCR.EN, and wait for this bit to be read as
one
86
32099IS–01/2012
AT32UC3L016/32/64
- Disable RC32K by writing a zero to SCIF.RC32KCR.EN, and wait for this bit to be read as
zero.
10.2.5
AST
1. Reset may set status bits in the AST
If a reset occurs and the AST is enabled, the SR.ALARM0, SR.PER0, and SR.OVF bits may
be set.
Fix/Workaround
If the part is reset and the AST is used, clear all bits in the Status Register before entering
sleep mode.
2. AST wake signal is released one AST clock cycle after the BUSY bit is cleared
After writing to the Status Clear Register (SCR) the wake signal is released one AST clock
cycle after the BUSY bit in the Status Register (SR.BUSY) is cleared. If entering sleep mode
directly after the BUSY bit is cleared the part will wake up immediately.
Fix/Workaround
Read the Wake Enable Register (WER) and write this value back to the same register. Wait
for BUSY to clear before entering sleep mode.
10.2.6
WDT
1. Clearing the Watchdog Timer (WDT) counter in second half of timeout period will
issue a Watchdog reset
If the WDT counter is cleared in the second half of the timeout period, the WDT will immediately issue a Watchdog reset.
Fix/Workaround
Use twice as long timeout period as needed and clear the WDT counter within the first half
of the timeout period. If the WDT counter is cleared after the first half of the timeout period,
you will get a Watchdog reset immediately. If the WDT counter is not cleared at all, the time
before the reset will be twice as long as needed.
2. WDT Control Register does not have synchronization feedback
When writing to the Timeout Prescale Select (PSEL), Time Ban Prescale Select (TBAN),
Enable (EN), or WDT Mode (MODE) fieldss of the WDT Control Register (CTRL), a synchronizer is started to propagate the values to the WDT clcok domain. This synchronization
takes a finite amount of time, but only the status of the synchronization of the EN bit is
reflected back to the user. Writing to the synchronized fields during synchronization can lead
to undefined behavior.
Fix/Workaround
-When writing to the affected fields, the user must ensure a wait corresponding to 2 clock
cycles of both the WDT peripheral bus clock and the selected WDT clock source.
-When doing writes that changes the EN bit, the EN bit can be read back until it reflects the
written value.
10.2.7
GPIO
1. Clearing GPIO interrupt may fail
Writing a one to the GPIO.IFRC register to clear an interrupt will be ignored if interrupt is
enabled for the corresponding port.
Fix/Workaround
Disable the interrupt, clear it by writing a one to GPIO.IFRC, then enable the interrupt.
87
32099IS–01/2012
AT32UC3L016/32/64
10.2.8
SPI
1. SPI data transfer hangs with CSR0.CSAAT==1 and MR.MODFDIS==0
When CSR0.CSAAT==1 and mode fault detection is enabled (MR.MODFDIS==0), the SPI
module will not start a data transfer.
Fix/Workaround
Disable mode fault detection by writing a one to MR.MODFDIS.
2. Disabling SPI has no effect on the SR.TDRE bit
Disabling SPI has no effect on the SR.TDRE bit whereas the write data command is filtered
when SPI is disabled. Writing to TDR when SPI is disabled will not clear SR.TDRE. If SPI is
disabled during a PDCA transfer, the PDCA will continue to write data to TDR until its buffer
is empty, and this data will be lost.
Fix/Workaround
Disable the PDCA, add two NOPs, and disable the SPI. To continue the transfer, enable the
SPI and PDCA.
3. SPI disable does not work in SLAVE mode
SPI disable does not work in SLAVE mode.
Fix/Workaround
Read the last received data, then perform a software reset by writing a one to the Software
Reset bit in the Control Register (CR.SWRST).
4. SPI bad serial clock generation on 2nd chip_select when SCBR=1, CPOL=1, and
NCPHA=0
When multiple chip selects (CS) are in use, if one of the baudrates equal 1 while one
(CSRn.SCBR=1) of the others do not equal 1, and CSRn.CPOL=1 and CSRn.NCPHA=0,
then an additional pulse will be generated on SCK.
Fix/Workaround
When multiple CS are in use, if one of the baudrates equals 1, the others must also equal 1
if CSRn.CPOL=1 and CSRn.NCPHA=0.
5. SPI mode fault detection enable causes incorrect behavior
When mode fault detection is enabled (MR.MODFDIS==0), the SPI module may not operate
properly.
Fix/Workaround
Always disable mode fault detection before using the SPI by writing a one to MR.MODFDIS.
10.2.9
TWI
1. TWIM SR.IDLE goes high immediately when NAK is received
When a NAK is received and there is a non-zero number of bytes to be transmitted,
SR.IDLE goes high immediately and does not wait for the STOP condition to be sent. This
does not cause any problem just by itself, but can cause a problem if software waits for
SR.IDLE to go high and then immediately disables the TWIM by writing a one to CR.MDIS.
Disabling the TWIM causes the TWCK and TWD pins to go high immediately, so the STOP
condition will not be transmitted correctly.
Fix/Workaround
If possible, do not disable the TWIM. If it is absolutely necessary to disable the TWIM, there
must be a software delay of at least two TWCK periods between the detection of
SR.IDLE==1 and the disabling of the TWIM.
2. TWIM TWALM polarity is wrong
88
32099IS–01/2012
AT32UC3L016/32/64
The TWALM signal in the TWIM is active high instead of active low.
Fix/Workaround
Use an external inverter to invert the signal going into the TWIM. When using both TWIM
and TWIS on the same pins, the TWALM cannot be used.
3. TWIS may not wake the device from sleep mode
If the CPU is put to a sleep mode (except Idle and Frozen) directly after a TWI Start condition, the CPU may not wake upon a TWIS address match. The request is NACKed.
Fix/Workaround
When using the TWI address match to wake the device from sleep, do not switch to sleep
modes deeper than Frozen. Another solution is to enable asynchronous EIC wake on the
TWIS clock (TWCK) or TWIS data (TWD) pins, in order to wake the system up on bus
events.
4. SMBALERT bit may be set after reset
The SMBus Alert (SMBALERT) bit in the Status Register (SR) might be erroneously set after
system reset.
Fix/Workaround
After system reset, clear the SR.SMBALERT bit before commencing any TWI transfer.
5. Clearing the NAK bit before the BTF bit is set locks up the TWI bus
When the TWIS is in transmit mode, clearing the NAK Received (NAK) bit of the Status Register (SR) before the end of the Acknowledge/Not Acknowledge cycle will cause the TWIS to
attempt to continue transmitting data, thus locking up the bus.
Fix/Workaround
Clear SR.NAK only after the Byte Transfer Finished (BTF) bit of the same register has been
set.
6. TWIS stretch on Address match error
When the TWIS stretches TWCK due to a slave address match, it also holds TWD low for
the same duration if it is to be receiving data. When TWIS releases TWCK, it releases TWD
at the same time. This can cause a TWI timing violation.
Fix/Workaround
None.
10.2.10
PWMA
1. BUSY bit is never cleared after writes to the Control Register (CR)
When writing a non-zero value to CR.TOP, CR.SPREAD, or CR.TCLR when the PWMA is
disabled (CR.EN == 0), the BUSY bit in the Status Register (SR.BUSY) will be set, but never
cleared.
Fix/Workaround
When writing a non-zero value to CR.TOP, CR.SPREAD, or CR.TCLR, make sure the
PWMA is enabled, or simultaneously enable the PWMA by writing a one to CR.EN.
2. Incoming peripheral events are discarded during duty cycle register update
Incoming peripheral events to all applied channels will be discarded if a duty cycle update is
received from the user interface in the same PWMA clock period.
Fix/Workaround
Ensure that duty cycle writes from the user interface are not performed in a PWMA period
when an incoming peripheral event is expected.
89
32099IS–01/2012
AT32UC3L016/32/64
10.2.11
ADCIFB
1. Using STARTUPTIME larger than 0x1F will freeze the ADC
Writing a value larger than 0x1F to the Startup Time field in the ADC Configuration Register
(ACR.STARTUP) will freeze the ADC, and the Busy Status bit in the Status Register
(SR.BUSY) will never be cleared.
Fix/Workaround
Do not write values larger than 0x1F to ACR.STARTUP.
10.2.12
CAT
1. CAT asynchronous wake will be delayed by one AST event period
If the CAT detects a condition the should asynchronously wake the device in static mode,
the asynchronous wake will not occur until the next AST event. For example, if the AST is
generating events to the CAT every 50ms, and the CAT detects a touch at t=9200ms, the
asynchronous wake will occur at t=9250ms.
Fix/Workaround
None.
2. CAT QMatrix sense capacitors discharged prematurely
At the end of a QMatrix burst charging sequence that uses different burst count values for
different Y lines, the Y lines may be incorrectly grounded for up to n-1 periods of the peripheral bus clock, where n is the ratio of the PB clock frequency to the GCLK_CAT frequency.
This results in premature loss of charge from the sense capacitors and thus increased variability of the acquired count values.
Fix/Workaround
Enable the 1kOhm drive resistors on all implemented QMatrix Y lines (CSA 1, 3, 5, 7, 9, 11,
13, and/or 15) by writing ones to the corresponding odd bits of the CSARES register.
3. CAT module does not terminate QTouch burst on detect
The CAT module does not terminate a QTouch burst when the detection voltage is
reached on the sense capacitor. This can cause the sense capacitor to be charged more
than necessary. Depending on the dielectric absorption characteristics of the capacitor, this
can lead to unstable measurements.
Fix/Workaround
Use the minimum possible value for the MAX field in the ATCFG1, TG0CFG1, and
TG1CFG1 registers.
4. Autonomous CAT acquisition must be longer than AST source clock period
When using the AST to trigger CAT autonomous touch acquisition in sleep modes where the
CAT bus clock is turned off, the CAT will start several acquisitions if the period of the AST
source clock is larger than one CAT acquisition. One AST clock period after the AST trigger,
the CAT clock will automatically stop and the CAT acquisition can be stopped prematurely,
ruining the result.
Fix/Workaround
Always ensure that the ATCFG1.max field is set so that the duration of the autonomous
touch acquisition is greater than one clock period of the AST source clock.
10.2.13
aWire
1. aWire CPU clock speed robustness
The aWire memory speed request command counter warps at clock speeds below approximately 5kHz.
Fix/Workaround
90
32099IS–01/2012
AT32UC3L016/32/64
None.
2. The aWire debug interface is reset after leaving Shutdown mode
If the aWire debug mode is used as debug interface and the program enters Shutdown
mode, the aWire interface will be reset when the part receives a wakeup either from the
WAKE_N pin or the AST.
Fix/Workaround
None.
10.2.14
CHIP
1. In 3.3V Single Supply Mode, the Analog Comparator inputs affects the device’s ability
to start
When using the 3.3V Single Supply Mode the state of the Analog Comparator input pins can
affect the device’s ability to release POR reset. This is due to an interaction between the
Analog Comparator input pins and the POR circuitry. The issue is not present in the 1.8V
Supply Mode or the 3.3V Supply Mode with 1.8V Regulated I/O Lines.
Fix/Workaround
ACREFN (pin PA16) must be connected to GND until the POR reset is released and the
Analog Comparator inputs should not be driven higher than 1.0 V until the POR reset is
released.
2. Increased Power Consumption in VDDIO in sleep modes
If OSC0 is enabled in crystal mode when entering a sleep mode where the OSC0 is disabled, this will lead to an increased power consumption in VDDIO.
Fix/Workaround
Solution 1: Disable OSC0 by writing a zero to the Oscillator Enable bit in the System Control
Interface (SCIF) Oscillator Control Register (SCIF.OSC0CTRL.OSCEN) before going to a
sleep mode where OSC0 is disabled.
Solution 2: Pull down or up XIN0 or XOUT0 with 1MOhm resistor.
10.2.15
I/O Pins
1. PA17 has low ESD tolerance
PA17 only tolerates 500V ESD pulses (Human Body Model).
Fix/Workaround
Care must be taken during manufacturing and PCB design.
10.3
Rev. C
Not sampled.
10.4
10.4.1
Rev. B
Processor and Architecture
1. RETS behaves incorrectly when MPU is enabled
RETS behaves incorrectly when MPU is enabled and MPU is configured so that system
stack is not readable in unprivileged mode.
Fix/Workaround
Make system stack readable in unprivileged mode, or return from supervisor mode using
rete instead of rets. This requires:
1. Changing the mode bits from 001 to 110 before issuing the instruction. Updating the
mode bits to the desired value must be done using a single mtsr instruction so it is done
91
32099IS–01/2012
AT32UC3L016/32/64
atomically. Even if this step is generally described as not safe in the UC technical reference
manual, it is safe in this very specific case.
2. Execute the RETE instruction.
2. Hardware breakpoints may corrupt MAC results
Hardware breakpoints on MAC instructions may corrupt the destination register of the MAC
instruction.
Fix/Workaround
Place breakpoints on earlier or later instructions.
3. Privilege violation when using interrupts in application mode with protected system
stack
If the system stack is protected by the MPU and an interrupt occurs in application mode, an
MPU DTLB exception will occur.
Fix/Workaround
Make a DTLB Protection (Write) exception handler which permits the interrupt request to be
handled in privileged mode.
10.4.2
PDCA
1. PCONTROL.CHxRES is non-functional
PCONTROL.CHxRES is non-functional. Counters are reset at power-on, and cannot be
reset by software.
Fix/Workaround
Software needs to keep history of performance counters.
2. Transfer error will stall a transmit peripheral handshake interface
If a transfer error is encountered on a channel transmitting to a peripheral, the peripheral
handshake of the active channel will stall and the PDCA will not do any more transfers on
the affected peripheral handshake interface.
Fix/Workaround
Disable and then enable the peripheral after the transfer error.
3. VERSION register reads 0x120
The VERSION register reads 0x120 instead of 0x122.
Fix/Workaround
None.
10.4.3
FLASHCDW
1. Fuse Programming
Programming fuses does not work.
Fix/Workaround
Do not program fuses. All fuses will be erased during chip erase command.
2. Chip Erase
When performing a chip erase, the device may report that it is protected (IR=0x11) and that
the erase failed, even if it was successful.
Fix/Workaround
Perform a reset before any further read and programming.
3. Wait 500 ns before reading from the flash after switching read mode
After switching between normal read mode and high-speed read mode, the application must
wait at least 500ns before attempting any access to the flash.
92
32099IS–01/2012
AT32UC3L016/32/64
Fix/Workaround
Solution 1: Make sure that the appropriate instructions are executed from RAM, and that a
waiting-loop is executed from RAM waiting 500ns or more before executing from flash.
Solution 2. Execute from flash with a clock with period longer than 500ns. This guarantees
that no new read access is attempted before the flash has had time to settle in the new read
mode.
4. Flash self programming may fail in one wait state mode
Writes in flash and user pages may fail if executing code is located in address space
mapped to flash, and the flash controller is configured in one wait state mode (the Flash
Wait State bit in the Flash Control Register (FCR.FWS) is one).
Fix/Workaround
Solution 1: Configure the flash controller in zero wait state mode (FCR.FWS=0).
Solution 2: Configure the HMATRIX master 1 (CPU Instruction) to use the unlimited burst
length transfer mode (MCFG1.ULBT=0), and the HMATRIX slave 0 (FLASHCDW) to use
the maximum slot cycle limit (SCFG0.SLOT_CYCLE=255).
5. VERSION register reads 0x100
The VERSION register reads 0x100 instead of 0x102.
Fix/Workaround
None.
10.4.4
SAU
1. The SR.IDLE bit reads as zero
The IDLE bit in the Status Register (SR.IDLE) reads as zero.
Fix/Workaround
None.
2. Open Mode is not functional
The Open Mode is not functional.
Fix/Workaround
None.
3. VERSION register reads 0x100
The VERSION register reads 0x100 instead of 0x110.
Fix/Workaround
None.
10.4.5
HMATRIX
1. In the PRAS and PRBS registers, the MxPR fields are only two bits
In the PRAS and PRBS registers, the MxPR fields are only two bits wide, instead of four bits.
The unused bits are undefined when reading the registers.
Fix/Workaround
Mask undefined bits when reading PRAS and PRBS.
10.4.6
Power Manager
1. CONFIG register reads 0x4F
The CONFIG register reads 0x4F instead of 0x43.
Fix/Workaround
None.
93
32099IS–01/2012
AT32UC3L016/32/64
2. It is not possible to mask the request clock requests
It is not possible to mask the request clock requests using PPCR.
Fix/Workaround
None.
3. Static mode cannot be entered if the WDT is using OSC32
If the WDT is using OSC32 as clock source and the user tries to enter Static mode, the
Deepstop mode will be entered instead.
Fix/Workaround
None.
4. Clock Failure Detector (CFD) does not work
Clock Failure Detector (CFD) does not work.
Fix/Workaround
None.
5. WCAUSE register should not be used
The WCAUSE register should not be used.
Fix/Workaround
None.
6. PB writes via debugger in sleep modes are blocked during sleepwalking
During sleepwalking, PB writes performed by a debugger will be discarded by all PB modules except the module that is requesting the clock.
Fix/Workaround
None.
7. Clock sources will not be stopped in Static mode if the difference between CPU and
PBx division factor is larger than 4
If the division factor between the CPU/HSB and PBx frequencies is more than 4 when entering a sleep mode where the system RC oscillator (RCSYS) is turned off, the high speed
clock sources will not be turned off. This will result in a significantly higher power consumption during the sleep mode.
Fix/Workaround
Before going to sleep modes where RCSYS is stopped, make sure the division factor
between CPU/HSB and PBx frequencies is less than or equal to 4.
8. Disabling POR33 may generate spurious resets
Depending on operating conditions, POR33 may generate a spurious reset in one of the following cases:
- When POR33 is disabled from the user interface
- When SM33 supply monitor is enabled
- When entering Shutdown mode while debugging the chip using JTAG or aWire interface
In the listed cases, writing a one to the bit VREGCR.POR33MASK in the System Control
Interface (SCIF) to mask the POR33 reset will be ineffective
Fix/Workaround
- Do not disable POR33 using the user interface
- Do not use the SM33 supply monitor
- Do not enter Shutdown mode if a debugger is connected to the chip
9. Instability when exiting sleep walking
If all the following operating conditions are true, exiting sleep walking might lead to
instability:
94
32099IS–01/2012
AT32UC3L016/32/64
- The OSC0 is enabled in external clock mode (OSCCTRL0.OSCEN == 1 and
OSCCTRL0.MODE == 0)
- A sleep mode where the OSC0 is automatically disabled is entered
- The device enters sleep walking
Fix/Workaround
Do not run OSC0 in external clock mode if sleepwalking is expected to be used.
10. Sleepwalking in idle and frozen sleep mode will mask all other PB clocks
If the CPU is in idle or frozen sleep mode and a module is in a state that triggers sleep walking, all PB clocks will be masked except the PB clock to the sleepwalking module.
Fix/Workaround
Mask all clock requests in the PM.PPCR register before going into idle or frozen mode.
11. VERSION register reads 0x400
The VERSION register reads 0x400 instead of 0x411.
Fix/Workaround
None.
10.4.7
SCIF
1. The DFLL should be slowed down before disabling it
The frequency of the DFLL should be set to minimum before disabling it.
Fix/Workaround
Before disabling the DFLL the value of the COARSE register should be zero.
2. Writing to ICR masks new interrupts received in the same clock cycle
Writing to ICR masks any new SCIF interrupt received in the same clock cycle, regardless of
write value.
Fix/Workaround
For every interrupt except BODDET, SM33DET, and VREGOK the PCLKSR register can be
read to detect new interrupts. BODDET, SM33DET and VREGOK interrupts will not be generated if they occur whilst writing to the ICR register.
3. FINE value for DFLL is not correct when dithering is disabled
In open loop mode, the FINE value used by the DFLL DAC is offset by two compared to the
value written to the DFLL0CONF.FINE field. The value used by the DFLL DAC is
DFLL0CONF.FINE-0x002. If DFLL0CONF.FINE is written to 0x000, 0x001 or 0x002 the
value used by the DFLL DAC will be 0x1FE, 0x1FF, or 0x000 respectively.
Fix/Workaround
Write the desired value added by two to the DFLL0CONF.FINE field.
4. BODVERSION register reads 0x100
The BODVERSION register reads 0x100 instead of 0x101
Fix/Workaround
None.
5. VREGCR.DEEPMODEDISABLE bit is not readable
VREGCR.DEEPMODEDISABLE bit is not readable.
Fix/Workaround
None.
6. DFLL step size should be seven or lower when below 30MHz
If max step size is above seven, the DFLL might not lock at the correct frequency if the target frequency is below 30MHz.
95
32099IS–01/2012
AT32UC3L016/32/64
Fix/Workaround
If the target frequency is below 30MHz, use a max step size (DFLL0MAXSTEP.MAXSTEP)
of seven or lower.
7. Generic clock sources are kept running in sleep modes
If a clock is used as a source for a generic clock when going to a sleep mode where clock
sources are stopped, the source of the generic clock will be kept running. Please refer to
Power Manager chapter for details about sleep modes.
Fix/Workaround
Disable generic clocks before going to sleep modes where clock sources are stopped to
save power.
8. DFLL clock is unstable with a fast reference clock
The DFLL clock can be unstable when a fast clock is used as a reference clock in closed
loop mode.
Fix/Workaround
Use the 32KHz crystal oscillator clock, or a clock with a similar frequency, as DFLLIF reference clock.
9. DFLLIF indicates coarse lock too early
The DFLLIF might indicate coarse lock too early, the DFLL will lose coarse lock and regain it
later.
Fix/Workaround
Use max step size (DFLL0MAXSTEP.MAXSTEP) of 4 or higher.
10. DFLLIF dithering does not work
The DFLLIF dithering does not work.
Fix/Workaround
None.
11. DFLLIF might lose fine lock when dithering is disabled
When dithering is disabled and fine lock has been acquired, the DFLL might lose the fine
lock resulting in up to 20% over-/undershoot.
Fix/Workaround
Solution 1: When the DFLL is used as main clock source, the target frequency of the DFLL
should be 20% below the maximum operating frequency of the CPU. Don’t use the DFLL as
clock source for frequency sensitive applications.
Solution 2: Do not use the DFLL in closed loop mode.
12. GCLK5 is non-functional
GCLK5 is non-functional.
Fix/Workaround
None.
13. BRIFA is non-functional
BRIFA is non-functional.
Fix/Workaround
None.
14. SCIF VERSION register reads 0x100
SCIFVERSION register reads 0x100 instead of 0x102.
Fix/Workaround
None.
96
32099IS–01/2012
AT32UC3L016/32/64
15. BODVERSION register reads 0x100
BODVERSION register reads 0x100 instead of 0x101.
Fix/Workaround
None.
16. DFLLVERSION register reads 0x200
DFLLVERSION register reads 0x200 instead of 0x201.
Fix/Workaround
None.
17. RCCRVERSION register reads 0x100
RCCRVERSION register reads 0x100 instead of 0x101.
Fix/Workaround
None.
18. OSC32VERSION register reads 0x100
OSC32VERSION register reads 0x100 instead of 0x101.
Fix/Workaround
None.
19. VREGVERSION register reads 0x100
VREGVERSION register reads 0x100 instead of 0x101.
Fix/Workaround
None.
20. RC120MVERSION register reads 0x100
RC120MVERSION register reads 0x100 instead of 0x101.
Fix/Workaround
None.
10.4.8
AST
1. AST wake signal is released one AST clock cycle after the BUSY bit is cleared
After writing to the Status Clear Register (SCR) the wake signal is released one AST clock
cycle after the BUSY bit in the Status Register (SR.BUSY) is cleared. If entering sleep mode
directly after the BUSY bit is cleared the part will wake up immediately.
Fix/Workaround
Read the Wake Enable Register (WER) and write this value back to the same register. Wait
for BUSY to clear before entering sleep mode.
10.4.9
WDT
1. Clearing the WDT in window mode
In window mode, if the WDT is cleared 2TBAN CLK_WDT cycles after entering the window,
the counter will be cleared, but will not exit the window. If this occurs, the SR.WINDOW bit
will not be cleared after clearing the WDT.
Fix/Workaround
Check SR.WINDOW immediately after clearing the WDT. If set then clear the WDT once
more.
2. Clearing the Watchdog Timer (WDT) counter in second half of timeout period will
issue a Watchdog reset
If the WDT counter is cleared in the second half of the timeout period, the WDT will immediately issue a Watchdog reset.
97
32099IS–01/2012
AT32UC3L016/32/64
Fix/Workaround
Use twice as long timeout period as needed and clear the WDT counter within the first half
of the timeout period. If the WDT counter is cleared after the first half of the timeout period,
you will get a Watchdog reset immediately. If the WDT counter is not cleared at all, the time
before the reset will be twice as long as needed.
3. VERSION register reads 0x400
The VERSION register reads 0x400 instead of 0x402.
Fix/Workaround
None.
4. WDT Control Register does not have synchronization feedback
When writing to the Timeout Prescale Select (PSEL), Time Ban Prescale Select (TBAN),
Enable (EN), or WDT Mode (MODE) fieldss of the WDT Control Register (CTRL), a synchronizer is started to propagate the values to the WDT clcok domain. This synchronization
takes a finite amount of time, but only the status of the synchronization of the EN bit is
reflected back to the user. Writing to the synchronized fields during synchronization can lead
to undefined behavior.
Fix/Workaround
-When writing to the affected fields, the user must ensure a wait corresponding to 2 clock
cycles of both the WDT peripheral bus clock and the selected WDT clock source.
-When doing writes that changes the EN bit, the EN bit can be read back until it reflects the
written value.
10.4.10
FREQM
1. Measured clock (CLK_MSR) sources 15-17 are shifted
CLKSEL = 14 selects the RC120M AW clock, CLKSEL = 15 selects the RC120M clock, and
CLKSEL = 16 selects the RC32K clock as source for the measured clock (CLK_MSR).
Fix/Workaround
None.
2. GCLK5 can not be used as source for the CLK_MSR
The frequency for GCLK5 can not be measured by the FREQM.
Fix/Workaround
None.
10.4.11
GPIO
1. GPIO interrupt can not be cleared when interrupts are disabled
The GPIO interrupt can not be cleared unless the interrupt is enabled for the pin.
Fix/Workaround
Enable interrupt for the corresponding pin, then clear the interrupt.
2. VERSION register reads 0x210
The VERSION register reads 0x210 instead of 0x211.
Fix/Workaround
None.
10.4.12
USART
1. The RTS output does not function correctly in hardware handshaking mode
98
32099IS–01/2012
AT32UC3L016/32/64
The RTS signal is not generated properly when the USART receives data in hardware handshaking mode. When the Peripheral DMA receive buffer becomes full, the RTS output
should go high, but it will stay low.
Fix/Workaround
Do not use the hardware handshaking mode of the USART. If it is necessary to drive the
RTS output high when the Peripheral DMA receive buffer becomes full, use the normal
mode of the USART. Configure the Peripheral DMA Controller to signal an interrupt when
the receive buffer is full. In the interrupt handler code, write a one to the RTSDIS bit in the
USART Control Register (CR). This will drive the RTS output high. After the next DMA transfer is started and a receive buffer is available, write a one to the RTSEN bit in the USART
CR so that RTS will be driven low.
10.4.13
SPI
1. SPI data transfer hangs with CSR0.CSAAT==1 and MR.MODFDIS==0
When CSR0.CSAAT==1 and mode fault detection is enabled (MR.MODFDIS==0), the SPI
module will not start a data transfer.
Fix/Workaround
Disable mode fault detection by writing a one to MR.MODFDIS.
2. Disabling SPI has no effect on the SR.TDRE bit
Disabling SPI has no effect on the SR.TDRE bit whereas the write data command is filtered
when SPI is disabled. Writing to TDR when SPI is disabled will not clear SR.TDRE. If SPI is
disabled during a PDCA transfer, the PDCA will continue to write data to TDR until its buffer
is empty, and this data will be lost.
Fix/Workaround
Disable the PDCA, add two NOPs, and disable the SPI. To continue the transfer, enable the
SPI and PDCA.
3. SPI disable does not work in SLAVE mode
SPI disable does not work in SLAVE mode.
Fix/Workaround
Read the last received data, then perform a software reset by writing a one to the Software
Reset bit in the Control Register (CR.SWRST).
4. SPI bad serial clock generation on 2nd chip_select when SCBR=1, CPOL=1, and
NCPHA=0
When multiple chip selects (CS) are in use, if one of the baudrates equal 1 while one
(CSRn.SCBR=1) of the others do not equal 1, and CSRn.CPOL=1 and CSRn.NCPHA=0,
then an additional pulse will be generated on SCK.
Fix/Workaround
When multiple CS are in use, if one of the baudrates equals 1, the others must also equal 1
if CSRn.CPOL=1 and CSRn.NCPHA=0.
5. SPI mode fault detection enable causes incorrect behavior
When mode fault detection is enabled (MR.MODFDIS==0), the SPI module may not operate
properly.
Fix/Workaround
Always disable mode fault detection before using the SPI by writing a one to MR.MODFDIS.
10.4.14
TWI
1. TWI pins are not SMBus compliant
99
32099IS–01/2012
AT32UC3L016/32/64
The TWI pins draw current when they are supplied with 3.3V and the part is left unpowered.
Fix/Workaround
None.
2. PA21, PB04, and PB05 are not 5V tolerant
Pins PA21, PB04, and PB05 are only 3.3V tolerant.
Fix/Workaround
None.
3. PB04 SMBALERT function should not be used
The SMBALERT function from TWIMS0 should not be selected on pin PB04.
Fix/Workaround
None.
4. TWIM STOP bit in IMR always reads as zero
The STOP bit in IMR always reads as zero.
Fix/Workaround
None.
5. Disabled TWIM drives TWD and TWCK low
When the TWIM is disabled, it drives the TWD and TWCK signals with logic level zero. This
can lead to communication problems with other devices on the TWI bus.
Fix/Workaround
Enable the TWIM first and then enable the TWD and TWCK peripheral pins in the GPIO
controller. If it is necessary to disable the TWIM, first disable the TWD and TWCK peripheral
pins in the GPIO controller and then disable the TWIM.
6. TWIM SR.IDLE goes high immediately when NAK is received
When a NAK is received and there is a non-zero number of bytes to be transmitted,
SR.IDLE goes high immediately and does not wait for the STOP condition to be sent. This
does not cause any problem just by itself, but can cause a problem if software waits for
SR.IDLE to go high and then immediately disables the TWIM by writing a one to CR.MDIS.
Disabling the TWIM causes the TWCK and TWD pins to go high immediately, so the STOP
condition will not be transmitted correctly.
Fix/Workaround
If possible, do not disable the TWIM. If it is absolutely necessary to disable the TWIM, there
must be a software delay of at least two TWCK periods between the detection of
SR.IDLE==1 and the disabling of the TWIM.
7. TWIM TWALM polarity is wrong
The TWALM signal in the TWIM is active high instead of active low.
Fix/Workaround
Use an external inverter to invert the signal going into the TWIM. When using both TWIM
and TWIS on the same pins, the TWALM cannot be used.
8. TWIS CR.STREN does not work in deep sleep modes
When the device is in Stop, DeepStop, or Static mode, address reception will not wake
device if both CR.SOAM and CR.STREN are one.
Fix/Workaround
Do not write both CR.STREN and CR.SOAM to one if the device needs to wake from deep
sleep modes.
9. TWI0.TWCK on PB05 is non-functional
TWI0.TWCK on PB05 is non-functional.
100
32099IS–01/2012
AT32UC3L016/32/64
Fix/Workaround
Use TWI0.TWCK on other pins.
10. TWIM Version Register reads zero
TWIM Version Register (VR) reads zero instead of 0x101
Fix/Workaround
None.
11. TWIS Version Register reads zero
TWIS Version Register (VR) reads zero instead of 0x112
Fix/Workaround
None.
12. SMBALERT bit may be set after reset
The SMBus Alert (SMBALERT) bit in the Status Register (SR) might be erroneously set after
system reset.
Fix/Workaround
After system reset, clear the SR.SMBALERT bit before commencing any TWI transfer.
13. Clearing the NAK bit before the BTF bit is set locks up the TWI bus
When the TWIS is in transmit mode, clearing the NAK Received (NAK) bit of the Status Register (SR) before the end of the Acknowledge/Not Acknowledge cycle will cause the TWIS to
attempt to continue transmitting data, thus locking up the bus.
Fix/Workaround
Clear SR.NAK only after the Byte Transfer Finished (BTF) bit of the same register has been
set.
14. TWIS stretch on Address match error
When the TWIS stretches TWCK due to a slave address match, it also holds TWD low for
the same duration if it is to be receiving data. When TWIS releases TWCK, it releases TWD
at the same time. This can cause a TWI timing violation.
Fix/Workaround
None.
10.4.15
PWMA
1. PARAMETER register reads 0x2424
The PARAMETER register reads 0x2424 instead of 0x24.
Fix/Workaround
None.
2. Writing to the duty cycle registers when the timebase counter overflows can give an
undefined result
The duty cycle registers will be corrupted if written when the timebase counter overflows. If
the duty cycle registers are written exactly when the timebase counter overflows at TOP, the
duty cycle registers may become corrupted.
Fix/Workaround
Write to the duty cycle registers only directly after the Timebase Overflow bit in the status
register is set.
3. Open Drain mode does not work
The open drain mode does not work.
Fix/Workaround
None.
101
32099IS–01/2012
AT32UC3L016/32/64
4. BUSY bit is never cleared after writes to the Control Register (CR)
When writing a non-zero value to CR.TOP, CR.SPREAD, or CR.TCLR when the PWMA is
disabled (CR.EN == 0), the BUSY bit in the Status Register (SR.BUSY) will be set, but never
cleared.
Fix/Workaround
When writing a non-zero value to CR.TOP, CR.SPREAD, or CR.TCLR, make sure the
PWMA is enabled, or simultaneously enable the PWMA by writing a one to CR.EN.
5. Incoming peripheral events are discarded during duty cycle register update
Incoming peripheral events to all applied channels will be discarded if a duty cycle update is
received from the user interface in the same PWMA clock period.
Fix/Workaround
Ensure that duty cycle writes from the user interface are not performed in a PWMA period
when an incoming peripheral event is expected.
6. VERSION register reads 0x100
The VERSION register reads 0x100 instead of 0x101.
Fix/Workaround
None.
10.4.16
TC
1. When the main clock is RCSYS, TIMER_CLOCK5 is equal to CLK_PBA
When the main clock is generated from RCSYS, TIMER_CLOCK5 is equal to CLK_PBA and
not CLK_PBA/128.
Fix/Workaround
None.
10.4.17
ADCIFB
1. Pendetect in sleep modes without CLK_ADCIFB will not wake the system
The pendetect will not wake the system from a sleep mode if the clock for the ADCIFB
(CLK_ADCIFB) is turned off.
Fix/Workaround
Use a sleep mode where CLK_ADCIFB is not turned off to wake the part using pendetect.
2. 8-bit mode is not working
Do not use the ADCIFB 8-bit mode.
Fix/Workaround
Use the 10-bit mode and shift right by two bits.
3. Using STARTUPTIME larger than 0x1F will freeze the ADC
Writing a value larger than 0x1F to the Startup Time field in the ADC Configuration Register
(ACR.STARTUP) will freeze the ADC, and the Busy Status bit in the Status Register
(SR.BUSY) will never be cleared.
Fix/Workaround
Do not write values larger than 0x1F to ACR.STARTUP.
4. ADC channels six to eight are non-functional
ADC channels six to eight are non-functional.
Fix/Workaround
None.
5. VERSION register reads 0x100
102
32099IS–01/2012
AT32UC3L016/32/64
The VERSION register reads 0x100 instead of 0x101.
Fix/Workaround
None.
10.4.18
ACIFB
1. Generic clock sources in sleep modes.
The ACIFB should not use RC32K or CLK_1K as generic clock source if the chip uses sleep
modes.
Fix/Workaround
None.
2. Negative offset
The static offset of the analog comparator is approximately -50mV
Fix/Workaround
None.
3. CONFW.WEVSRC and CONFW.WEVEN are not correctly described in the user
interface
CONFW.WEVSRC is only two bits instead of three bits wide. Only values 0, 1, and 2 can be
written to this register. CONFW.WEVEN is in bit position 10 instead of 11.
Fix/Workaround
Only write values 0, 1, and 2 to CONFW.WEVSRC. When reading CONFW.WEVSRC, disregard the third bit. Read/write bit 10 to access CONFW.WEVEN.
4. VERSION register reads 0x200
The VERSION register reads 0x200 instead of 0x212.
Fix/Workaround
None.
10.4.19
CAT
1. Switch off discharge current when reaching 0V
The discharge current will switch off when reaching MGCFG1.MAX, not when reaching 0V.
Fix/Workaround
None.
2. CAT external capacitors are not clamped to ground when CAT is idle
The CAT module does not clamp the external capacitors to ground when it is idle. The
capacitors are left floating, so they could accumulate small amounts of charge.
Fix/Workaround
None.
3. DISHIFT field is stuck at zero
The DISHIFT field in the MGCFG1, TGACFG1, TGBCFG1, and ATCFG1 registers is stuck
at zero and cannot be written to a different value. Capacitor discharge time will only be
determined by the DILEN field.
Fix/Workaround
None.
4. MGCFG2.CONSEN field is stuck at zero
The CONSEN field in the MGCFG2 register is stuck at zero and cannot be written to a different value. The CAT consensus filter does not function properly, so termination of QMatrix
data acquisition is controlled only by the MAX field in MGCFG1.
103
32099IS–01/2012
AT32UC3L016/32/64
Fix/Workaround
None.
5. MGCFG2.ACCTRL bit is stuck at zero
The ACCTRL bit in the MGCFG2 register is stuck at zero and cannot be written to one. The
analog comparators will be constantly enabled.
Fix/Workaround
None.
6. CAT asynchronous wake will be delayed by one AST event period
If the CAT detects a condition the should asynchronously wake the device in static mode,
the asynchronous wake will not occur until the next AST event. For example, if the AST is
generating events to the CAT every 50ms, and the CAT detects a touch at t=9200ms, the
asynchronous wake will occur at t=9250ms.
Fix/Workaround
None.
7. VERSION register reads 0x100
The VERSION register reads 0x100 instead of 0x200.
Fix/Workaround
None.
8. CAT module does not terminate QTouch burst on detect
The CAT module does not terminate a QTouch burst when the detection voltage is
reached on the sense capacitor. This can cause the sense capacitor to be charged more
than necessary. Depending on the dielectric absorption characteristics of the capacitor, this
can lead to unstable measurements.
Fix/Workaround
Use the minimum possible value for the MAX field in the ATCFG1, TG0CFG1, and
TG1CFG1 registers.
9. Autonomous CAT acquisition must be longer than AST source clock period
When using the AST to trigger CAT autonomous touch acquisition in sleep modes where the
CAT bus clock is turned off, the CAT will start several acquisitions if the period of the AST
source clock is larger than one CAT acquisition. One AST clock period after the AST trigger,
the CAT clock will automatically stop and the CAT acquisition can be stopped prematurely,
ruining the result.
Fix/Workaround
Always ensure that the ATCFG1.max field is set so that the duration of the autonomous
touch acquisition is greater than one clock period of the AST source clock.
10.4.20
GLOC
1. GLOC is non-functional
Glue Logic Controller (GLOC) is non-functional.
Fix/Workaround
None.
10.4.21
aWire
1. SAB multiaccess reads are not working
Reading more than one word, halfword, or byte in one command is not working correctly.
Fix/Workaround
Split the access into several single word, halfword, or byte accesses.
104
32099IS–01/2012
AT32UC3L016/32/64
2. If a reset happens during the last SAB write, the aWire will stall
If a reset happens during the last word, halfword or byte write the aWire will wait forever for
an acknowledge from the SAB.
Fix/Workaround
Reset the aWire by keeping the RESET_N line low for 100ms.
3. aWire enable does not work in Static mode
aWire enable does not work in Static mode.
Fix/Workaround
None.
4. aWire CPU clock speed robustness
The aWire memory speed request command counter warps at clock speeds below approximately 5kHz.
Fix/Workaround
None.
5. The aWire debug interface is reset after leaving Shutdown mode
If the aWire debug mode is used as debug interface and the program enters Shutdown
mode, the aWire interface will be reset when the part receives a wakeup either from the
WAKE_N pin or the AST.
Fix/Workaround
None.
6. aWire PB mapping and PB clock mask number
The aWire PB has a different PB address and PB clock mask number.
Fix/Workaround
Use aWire PB address 0xFFFF6C00 and PB clock (PBAMASK) 24.
7. VERSION register reads 0x200
The VERSION register reads 0x200 instead of 0x210.
Fix/Workaround
None.
10.4.22
Chip
1. WAKE_N pin can only wake up the chip from Shutdown mode
It is not possible to wake up the chip from any other sleep mode than Shutdown using the
WAKE_N pin. If the WAKE_N pin is asserted during a sleep mode other than Shutdown,
nothing will happen.
Fix/Workaround
Use an EIC pin to wake up from sleep modes higher than Shutdown.
2. Power consumption in static mode is too high
Power consumption in static mode is too high when PA21 is high.
Fix/Workaround
Ensure PA21 is low.
3. Shutdown mode is not functional
Do not enter Shutdown mode.
Fix/Workaround
None.
4. VDDIN current consumption increase above 1.8V
105
32099IS–01/2012
AT32UC3L016/32/64
When VDDIN increases above 1.8V, current on VDDIN increases with up to 40uA.
Fix/Workaround
None.
5. Increased Power Consumption in VDDIO in sleep modes
If OSC0 is enabled in crystal mode when entering a sleep mode where the OSC0 is disabled, this will lead to an increased power consumption in VDDIO.
Fix/Workaround
Solution 1: Disable OSC0 by writing a zero to the Oscillator Enable bit in the System Control
Interface (SCIF) Oscillator Control Register (SCIF.OSC0CTRL.OSCEN) before going to a
sleep mode where OSC0 is disabled.
Solution 2: Pull down or up XIN0 and XOUT0 with 1MOhm resistor.
10.4.23
I/O Pins
1. PB10 is not 3.3V tolerant.
PB10 should be grounded on the PCB and left unused.
Fix/Workaround
None.
2. Analog multiplexing consumes extra power
Current consumption on VDDIO increases when the voltage on analog inputs is close to
VDDIO/2.
Fix/Workaround
None.
3. PA02, PB01, PB04, PB05, and RESET_N have half of the pull-up strength
Pins PA02, PB01, PB04, PB05, and RESET_N have half of the specified pull-up strength.
Fix/Workaround
None.
4. JTAG is enabled at power up
The JTAG function on pins PA00, PA01, PA02, and PA03, are enabled after startup. Normal
I/O module functionality is not possible on these pins.
Fix/Workaround
Add a 10kOhm pullup on the reset line.
5. MCKO and MDO[3] are swapped in the AUX1 mapping
When using the OCD AUX1 mapping of trace signals MDO[3] is located on pin PB05 and
MCKO is located on PB01.
Fix/Workaround
Swap pins PB01 and PB05 if using OCD AUX1.
106
32099IS–01/2012
AT32UC3L016/32/64
11. Datasheet Revision History
Please note that the referring page numbers in this section are referred to this document. The
referring revision in this section are referring to the document revision.
11.1
11.2
Rev. I - 01/2012
1.
Overview - Block diagram: CAT SMP corrected from I/O to output. SPI NPCS corrected from
output to I/O.
2.
Package and Pinout: PRND signal removed from Signal Descriptions List table and GPIO
Controller Function Multiplexing table
3.
ADCIFB: PRND signal removed from block diagram.
4.
Electrical Characteristics: Added pin input capacitance CIN for TLLGA package for “all normal
I/O pins except PA05 , PA07, PA17, PA20, PA21, PB04, PB05”.
5.
Errata: Added more errata for TWI and CAT modules.
Rev. H - 12/2011
1.
Mechanical Characteristics: Updated the note related to the QFN48 Package Drawing.Updated
thermal reistance data for TLLGA48 package. Updated package drawings for TQFP48 and
QFN48 packages. Soldering profile updated (Time maintained above 217°C.)
2.
Memories: Local bus address map corrected: The address offset for port 1 registers is 0x100,
not 0x200.
3.
SCIF DFLL: Removed “not” from “DFLLnSTEP.CSTEP and DFLLnSTEP.FSTEP should not be
lower than 50% of the maximum value of DFLLnCONF.COARSE and DFLLnCONF.FINE”.
4.
SCIF VREG POR descriptions updated. POR33 bits added in VREGCR.
5.
SCIF VREG: Removed reference to flash fuses for CALIB field, user is recommended not
writing to SELVDD field. Removed references to Electrical Characteristics.
6.
SCIF: Fuses text removed from some submodules (SM33, VREG).
7.
SCIF VREG: Flash recalibration is always done at POR.
8.
SCIF SM33: Enabling SM33 will disable the POR33 detector.
9,
Erratum regarding OSC32 disabling is not valid for RevB.
10.
Flash Controller: Serial number address updated.
11.
Block diagram and ADCIFB: Removed PRND signal from block diagram and ADCIFB block
diagram.
12.
USART: Added CTSIC bit description.
13.
Power Manager: Updated Clock division and Clock ready flag sections.
14.
ADCIFB: Added DMA section in Product Dependencies.
107
32099IS–01/2012
AT32UC3L016/32/64
11.3
11.4
11.5
15.
Electrcal Characteristics: Updated SPI timing data.
16.
Electrical Characteristics, I/O Pin Characteristics: Added Input capacitance for TLLGA48
package.
17.
Errata: Removed erratum regarding SPI RDR.PCS field, as the PCS field has been removed
(refer to Section 11.8 on page 109).
Rev. G - 06/2011
1.
FLASHCDW: FSR register is a Read-only register. Added info about QPRUP.
2.
PM: Clarified POR33 masking requirements before shutdown. Added more info about wakeup
sources. Added AWEN description. PPCR register reset value corrected.
3.
SAU: SR.IDLE definition and reset value corrected.
4.
DFLL: Open- and closed-loop operation clarified.
5.
OSC32: Added note about OSC32RDY bit not always cleared when disabling OSC32.
6.
USART: Major cleanup.
1.
Features: Removed superfluous R mark.
2.
Package and Pinout, GPIO function multiplexing:TWIMS0-TWCK on PA20 removed. ADCIFBAD[3] on PA17 removed, number of ADC channels are 8, not 9. These were removed from rev.
C, but reappeared in rev. E.
1.
Overview: Added missing signals in block diagram.
2.
Package and pinout: Added note about TWI, SMBUS and 5V tolerant pads in peripheral
multiplexing. Added CAT DIS signal to signal descriptions. Removed TBD on ADVREFP
minimum voltage.
3.
Memories: Added SAU slave address to physical memory map.
4.
Supply and startup considerations: VDDIN is using GND as ground pin. Clarified references to
PORs in startup considerations.
5.
FLASHCDW: Added serial number location to module configuration section.
6.
PM: Added more info about the WAKE_N pin. Added info about CLK_PM, Updated the
selection main clock source section.
7.
SCIF: Major chapter update.
8.
AST: Updated digital tuner formula and conditions.
9.
GPIO: Updated GPER reset value and added more registers with non-zero reset value.
10.
CAT: Added info about VDIVEN and discharge current formula.
Rev. F- 11/2010
Rev. E- 10/2010
108
32099IS–01/2012
AT32UC3L016/32/64
11.6
11.
ADCIFB: Fixed Sample and Hold time formula.
12.
GLOC: Added info about pullup control and renamed LUTCR register to CR.
13.
TC: Added features and version register.
14.
SAU: Added OPEN bit to config register. Added description of unlock fields.
15.
TWIS: SCR is Write-only. Improved explanation of slave transmitter mode. Updated data
transfer diagrams.
16.
Electrical Characteristics: Added more values. Added notes on simulated and characterized
values. Added pin capacitance, rise, and fall times. Added timing characteristics. Removed all
TBDs. Added ADC analog input characteristics. Symbol cleanup.
17.
Errata: Updated errata list.
Rev. D - 06/2010
1.
11.7
11.8
Ordering Information: Ordering code for TQFP ES changed from AT32UC3L064-AUES to
AT32UC3L064-AUTES. TLLGA48 Tray option added.
Rev. C - 06/2010
1.
Features and Description: Added QTouch library support.
2.
USART: Description of unimplemented features removed.
3.
Electrical Characteristics: Power Consumption numbers updated. Flash timing numbers
added.
Rev. B - 05/2010
1.
Package and Pinout: Added pinout figure for TLLGA48 package.
2.
Package and Pinout, GPIO function multiplexing:TWIMS0-TWCK on PA20 removed. ADCIFBAD[3] on PA17 removed, number of ADC channels are 8, not 9.
3.
I/O Lines Considerations: Added: Following pins have high-drive capability: PA02, PA06,
PA08, PA09, and PB01.
Some TWI0 pins are SMBUS compliant (PA21, PB04, PB05).
4.
HMATRIX Masters: PDCA is master 4, not master 3. SAU is master 3, not master 4.
5.
SAU: IDLE bit added in the Status Register.
6.
PDCA: Number of PDCA performance monitors is device dependent.
7.
Peripheral Event System: Chapter updated.
8.
PM: Bits in RCAUSE registers removed and renamed (JTAGHARD and AWIREHARD renamed
to JTAG and AWIRE respectively, JTAG and AWIRE removed. BOD33 bit removed).
9.
PM: RCAUSE.BOD33 bit removed. SM33 reset will be detected as a POR reset.
10.
PM: WDT can be used as wake-up source if WDT is clocked from 32KHz oscillator.
109
32099IS–01/2012
AT32UC3L016/32/64
11.9
11.
PM: Entering Shutdown mode description updated.
12.
SCIF: DFLL output frequency is 40-150MHz, not 20-150MHz or 30-150MHz.
13.
SCIF: Temperature sensor is connected to ADC channel 9, not 7.
14.
SCIF: Updated the oscillator connection figure for OSC0
15.
GPIO: Removed unimplemented features (pull-down, buskeeper, drive strength, slew rate,
Schmidt trigger, open drain).
16.
SPI: RDR.PCS field removed (RDR[19:16]).
17.
TWIS: Figures updated.
18.
ADCIFB: The sample and hold time and the startup time formulas have been corrected (ADC
Configuration Register).
19.
ADCIFB: Updated ADC signal names.
20.
ACIFB: CONFW.WEVSRC is bit 8-10, CONFW.EWEVEN is bit 11. CONF.EVENP and
CONF.EVENN bits are swapped.
21.
CAT: Matrix size is 16 by 8, not 18 by 8.
22.
Electrical Characteristics: General update.
23.
Mechanical Characteristics: Added numbers for package drawings.
24.
Mechanical Characteristics: In the TQFP-48 package drawing the Lead Coplanarity is
0.102mm, not 0.080mm.
25.
Ordering Information: Ordering code for TLLGA-48 package updated.
Rev. A – 06/2009
1.
Initial revision.
110
32099IS–01/2012
AT32UC3L016/32/64
Table of Contents
Features ..................................................................................................... 1
1
Description ............................................................................................... 3
2
Overview ................................................................................................... 5
3
4
5
6
7
2.1
Block Diagram ...................................................................................................5
2.2
Configuration Summary .....................................................................................6
Package and Pinout ................................................................................. 7
3.1
Package .............................................................................................................7
3.2
Peripheral Multiplexing on I/O lines ...................................................................9
3.3
Signal Descriptions ..........................................................................................13
3.4
I/O Line Considerations ...................................................................................16
Processor and Architecture .................................................................. 18
4.1
Features ..........................................................................................................18
4.2
AVR32 Architecture .........................................................................................18
4.3
The AVR32UC CPU ........................................................................................19
4.4
Programming Model ........................................................................................23
4.5
Exceptions and Interrupts ................................................................................27
Memories ................................................................................................ 32
5.1
Embedded Memories ......................................................................................32
5.2
Physical Memory Map .....................................................................................32
5.3
Peripheral Address Map ..................................................................................33
5.4
CPU Local Bus Mapping .................................................................................34
Supply and Startup Considerations ..................................................... 36
6.1
Supply Considerations .....................................................................................36
6.2
Startup Considerations ....................................................................................40
Electrical Characteristics ...................................................................... 41
7.1
Absolute Maximum Ratings* ...........................................................................41
7.2
Supply Characteristics .....................................................................................41
7.3
Maximum Clock Frequencies ..........................................................................42
7.4
Power Consumption ........................................................................................42
7.5
I/O Pin Characteristics .....................................................................................47
7.6
Oscillator Characteristics .................................................................................50
7.7
Flash Characteristics .......................................................................................54
i
32099IS–01/2012
AT32UC3L016/32/64
8
9
7.8
Analog Characteristics .....................................................................................55
7.9
Timing Characteristics .....................................................................................63
Mechanical Characteristics ................................................................... 73
8.1
Thermal Considerations ..................................................................................73
8.2
Package Drawings ...........................................................................................74
8.3
Soldering Profile ..............................................................................................77
Ordering Information ............................................................................. 78
10 Errata ....................................................................................................... 79
10.1
Rev. E ..............................................................................................................79
10.2
Rev. D ..............................................................................................................85
10.3
Rev. C ..............................................................................................................91
10.4
Rev. B ..............................................................................................................91
11 Datasheet Revision History ................................................................ 107
11.1
Rev. I - 01/2012 .............................................................................................107
11.2
Rev. H - 12/2011 ...........................................................................................107
11.3
Rev. G - 06/2011 ...........................................................................................108
11.4
Rev. F- 11/2010 .............................................................................................108
11.5
Rev. E- 10/2010 .............................................................................................108
11.6
Rev. D - 06/2010 ...........................................................................................109
11.7
Rev. C - 06/2010 ...........................................................................................109
11.8
Rev. B - 05/2010 ............................................................................................109
11.9
Rev. A – 06/2009 ...........................................................................................110
Table of Contents....................................................................................... i
ii
32099IS–01/2012
Atmel Corporation
2325 Orchard Parkway
San Jose, CA 95131
USA
Tel: (+1)(408) 441-0311
Fax: (+1)(408) 487-2600
www.atmel.com
Atmel Asia Limited
Unit 1-5 & 16, 19/F
BEA Tower, Millennium City 5
418 Kwun Tong Road
Kwun Tong, Kowloon
HONG KONG
Tel: (+852) 2245-6100
Fax: (+852) 2722-1369
Atmel Munich GmbH
Business Campus
Parkring 4
D-85748 Garching b. Munich
GERMANY
Tel: (+49) 89-31970-0
Fax: (+49) 89-3194621
Atmel Japan
16F, Shin Osaki Kangyo Bldg.
1-6-4 Osaka Shinagawa-ku
Tokyo 104-0032
JAPAN
Tel: (+81) 3-6417-0300
Fax: (+81) 3-6417-0370
© 2012 Atmel Corporation. All rights reserved.
Atmel ®, logo and combinations thereof, AVR ®, picoPower ®, QTouch ®, AKS ® and others are registered trademarks or trademarks of
Atmel Corporation or its subsidiaries. Other terms and product names may be trademarks of others.
Disclaimer: The information in this document is provided in connection with Atmel products. No license, express or implied, by estoppel or otherwise, to
any intellectual property right is granted by this document or in connection with the sale of Atmel products. EXCEPT AS SET FORTH IN THE ATMEL
TERMS AND CONDITIONS OF SALES LOCATED ON THE ATMEL WEBSITE, ATMEL ASSUMES NO LIABILITY WHATSOEVER AND DISCLAIMS ANY
EXPRESS, IMPLIED OR STATUTORY WARRANTY RELATING TO ITS PRODUCTS INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTY OF
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, OR NON-INFRINGEMENT. IN NO EVENT SHALL ATMEL BE LIABLE FOR ANY DIRECT,
INDIRECT, CONSEQUENTIAL, PUNITIVE, SPECIAL OR INCIDENTAL DAMAGES (INCLUDING, WITHOUT LIMITATION, DAMAGES FOR LOSS AND PROFITS, BUSINESS INTERRUPTION, OR LOSS OF INFORMATION) ARISING OUT OF THE USE OR INABILITY TO USE THIS DOCUMENT, EVEN IF ATMEL
HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. Atmel makes no representations or warranties with respect to the accuracy or completeness of the contents of this document and reserves the right to make changes to specifications and product descriptions at any time without notice.
Atmel does not make any commitment to update the information contained herein. Unless specifically provided otherwise, Atmel products are not suitable for, and shall not be used in, automotive applications. Atmel products are not intended, authorized, or warranted for use as components in applications intended to support or sustain life.
32099IS–01/2012
Similar pages