PCU9669B [NXP]
I2C BUS CONTROLLER, PQFP48, 7 X 7 MM, 1.40 MM HEIGHT, PLASTIC, MS-026, SOT313-2, LQFP-48;型号: | PCU9669B |
厂家: | NXP |
描述: | I2C BUS CONTROLLER, PQFP48, 7 X 7 MM, 1.40 MM HEIGHT, PLASTIC, MS-026, SOT313-2, LQFP-48 时钟 数据传输 驱动 外围集成电路 驱动器 |
文件: | 总69页 (文件大小:1310K) |
中文: | 中文翻译 | 下载: | 下载PDF数据表文档文件 |
PCU9669
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus
controller
Rev. 2 — 1 July 2011
Product data sheet
1. General description
The PCU9669 is an advanced single master mode I2C-bus controller. It is a fourth
generation bus controller designed for data intensive I2C-bus data transfers. It has three
independent I2C-bus channels, one of them with data rates up to 1 Mbits/s using the
Fast-mode Plus (Fm+) open-drain topology and two with a much larger transmit only
transfer rate of up to 5 Mbits/s using the new Ultra Fast-mode (UFm) bus with push-pull
topology. Each channel has a generous 4352 byte data buffer which makes the PCU9669
the ideal companion to any CPU that needs to transmit and receive large amounts of
serial data with minimal interruptions.
The PCU9669 is a 8-bit parallel-bus to I2C-bus protocol converter. It can be configured to
communicate with up to 64 slaves in one serial sequence with no intervention from the
CPU. The controller also has a sequence loop control feature that allows it to
automatically retransmit a stored sequence.
Its onboard oscillator and PLL allow the controller to generate the clocks for the I2C-bus
and for the interval timer used in sequence looping. This feature greatly reduces CPU
overhead when data refresh is required in fault tolerant applications.
An external trigger input allows data synchronization with external events. The trigger
signal controls the rate at which a stored sequence is re-transmitted over the I2C-bus.
Error reporting is handled at the transaction level, channel level, and controller level.
A simple interrupt tree and interrupt masks allow further customization of interrupt
management.
The controller parallel bus interface runs at 3.3 V and the I2C-bus I/Os logic levels are
referenced to a dedicated VDD(IO) input pin with a range of 3.0 V to 5.5 V.
2. Features and benefits
Parallel-bus to I2C-bus protocol converter and interface
5 Mbit/s unidirectional data transfer on Ultra Fast-mode (UFm) channel (push-pull
driver)
1 Mbit/s and up to 30 mA SCL/SDA IOL Fast-mode Plus (Fm+) capability
Internal oscillator trimmed to 1 % accuracy reduces external components
Individual 4352-byte buffers for the Fm+ and UFm channels for a total of 13056 bytes
of buffer space
Three levels of reset: individual software channel reset, global software reset, global
hardware RESET pin
Communicates with up to 64 slaves in one serial sequence
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
Sequence looping with interval timer
Supports SCL clock stretching (Fm+ only)
JTAG port available for boundary scan testing during board manufacturing process
Trigger input synchronizes serial communication exactly with external events
Maskable interrupts
Fast-mode Plus I2C-bus capable and compatible with SMBus
Operating supply voltage: 3.0 V to 3.6 V (device and host interface)
I2C-bus I/O supply voltage: 3.0 V to 5.5 V
Latch-up testing is done to JEDEC Standard JESD78 which exceeds 100 mA
ESD protection exceeds 8000 V HBM per JESD22-A114 and 1000 V CDM per
JESD22-C101
Packages offered: LQFP48
3. Applications
Add I2C-bus port to controllers/processors that do not have one
Add additional I2C-bus ports to controllers/processors that need multiple I2C-bus ports
Converts 8 bits of parallel data to serial data stream to prevent having to run a large
number of traces across the entire printed-circuit board
Entertainment systems
LED matrix control
Data intensive I2C-bus transfers
4. Ordering information
Table 1.
Ordering information
Type number Topside
mark
Package
Name
Description
Version
PCU9669B
PCU9669 LQFP48
plastic low profile quad flat package;
SOT313-2
48 leads; body 7 7 1.4 mm
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
2 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
5. Block diagram
CE WR RD INT RESET
Channel 0
PCU9669
2
Fm+ I C-bus control
STATUS0_[n]
CONTROL
CHSTATUS
INTMSK
SDA0
SCL0
TRIG
SLATABLE
TRANCONFIG
DATA
TRANSEL
TRANOFS
BYTECOUNT
FRAMECNT
REFRATE
SCLL
SCLH
MODE
TIMEOUT
PRESET
INTERRUPT
CONTROL
A0
A1
A2
A3
A4
A5
A6
A7
BUFFER
CONTROL
4352-BYTE
BUFFER
BUS
INTERFACE
D0
D1
D2
D3
D4
D5
D6
D7
Channel 1
2
UFm I C-bus control
STATUS1_[n]
CONTROL
CHSTATUS
INTMSK
SLATABLE
TRANCONFIG
DATA
TRANSEL
TRANOFS
BYTECOUNT
FRAMECNT
REFRATE
SCLPER
USDA1
USCL1
CTRLSTATUS
CTRLINTMSK
DEVICE_ID
CTRLPRESET
CTRLRDY
4352-BYTE
BUFFER
CONTROL BLOCK
SDADLY
MODE
PRESET
POWER-ON/
POWER-DOWN
RESET
V
DD
Channel 2
2
DC/DC
UFm I C-bus control
REGULATOR
STATUS2_[n]
CONTROL
CHSTATUS
INTMSK
SLATABLE
TRANCONFIG
DATA
USDA2
USCL2
TCK
TRST
TMS
TDI
JTAG
TDO
TRANSEL
TRANOFS
BYTECOUNT
FRAMECNT
REFRATE
SCLPER
OSCILLATOR
PLL
4352-BYTE
BUFFER
V
DD(IO)
SDADLY
MODE
PRESET
002aaf479
Fig 1. Block diagram
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
3 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
6. Pinning information
6.1 Pinning
1
2
36
35
34
33
32
D6
D7
A0
A1
A2
A3
RESET
V
SS
3
TRIG
CE
4
5
RD
6
31 WR
PCU9669B
7
30
29
28
27
26
25
V
V
V
DD
DD
SS
8
V
SS
A4
9
SCL0
10
11
12
A5
A6
A7
SDA0
USCL1
USDA1
002aaf480
Fig 2. Pin configuration for LQFP48
6.2 Pin description
Table 2.
Pin description
Symbol Pin
Type
Description
A0
A1
A2
A3
A4
A5
A6
A7
D0
D1
D2
D3
D4
D5
D6
D7
3
I
Address inputs: selects the bus controller’s internal registers and
ports for read/write operations. Address is registered when CE is
LOW and whether WR or RD transitions LOW. A0 is the least
significant bit.
4
I
5
I
6
I
9
I
10
11
12
37
38
41
42
45
46
1
I
I
I
I/O
I/O
I/O
I/O
I/O
I/O
I/O
I/O
Data bus: bidirectional 3-state data bus used to transfer
commands, data and status between the bus controller and the
host. D0 is the least significant bit. Data is registered on the rising
edge of WR when CE is LOW.
2
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
4 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
Table 2.
Pin description …continued
Symbol Pin
Type
Description
TRST
TMS
13
14
I
I
JTAG test reset input. For normal operation, hold LOW (VSS).
JTAG test mode select input. For normal operation, hold HIGH
(VDD).
TCK
TDI
15
16
17
I
JTAG test clock input. For normal operation, hold HIGH (VDD).
JTAG test data in input. For normal operation, hold HIGH (VDD).
I
TDO
O
JTAG test data out output. For normal operation, do not connect
(n.c.).
INT
20
O
O
Interrupt request: Active LOW, open-drain, output. This pin
requires a pull-up device.
Channel 2 Ultra Fast-mode I2C-bus serial data output.
USDA2 21
Push-pull drive. No pull-up device is needed.
Channel 2 Ultra Fast-mode I2C-bus serial clock output.
USCL2
22
O
Push-pull drive. No pull-up device is needed.
Channel 1 Ultra Fast-mode I2C-bus serial data output.
USDA1 25
O
Push-pull drive. No pull-up device is needed.
Channel 1 Ultra Fast-mode I2C-bus serial clock output.
USCL1
SDA0
SCL0
WR
26
27
28
31
O
Push-pull drive. No pull-up device is needed.
Channel 0 I2C-bus serial data input/output (open-drain).
I/O
I/O
I
This pin requires a pull-up device.
Channel 0 I2C-bus serial clock input/output (open-drain).
This pin requires a pull-up device.
Write strobe: When LOW and CE is also LOW, the content of the
data bus is loaded into the addressed register. Data are latched on
the rising edge of WR. CE may remain LOW or transition with WR.
RD
CE
32
33
I
I
Read strobe: When LOW and CE is also LOW, causes the
contents of the addressed register to be presented on the data
bus. The read cycle begins on the falling edge of RD. Data lines
are driven when RD and CE are LOW. CE may transition with RD.
Chip Enable: Active LOW input signal. When LOW, data transfers
between the host and the bus controller are enabled on D0 to D7
as controlled by the WR, RD and A0 to A7 inputs. When HIGH,
places the D0 to D7 lines in the 3-state condition.
During the initialization period, CE must transition with RD until
controller is ready.
TRIG
34
I
I
Trigger input: provides the trigger to start a new frame.
RESET 36
Reset: Active LOW input. A LOW level resets the device to the
power-on state. Internally pulled HIGH through weak pull-up
current.
VDD(IO)
24
23
power I/O power supply: 3.0 V to 5.5 V. Power supply reference for
I2C-bus pins. Sets the voltage reference point for VIL/VIH and the
output drive rail for the UFm channel.
VSS(IO)
VDD
power I/O supply ground. Can be tied to VSS.
7, 18, 30, power Power supply: 3.0 V to 3.6 V. All VDD pins should be connected
40, 44, 48 together externally.
VSS
8, 19, 29, power Supply ground. All VSS pins should be connected together
35, 39,
43, 47
externally.
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
5 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
7. Functional description
7.1 General
The PCU9669 acts as an interface device between standard high-speed parallel buses
and the serial I2C-bus. On the I2C-bus, it acts as a master. Data transfer between the
I2C-bus and the parallel-bus host is carried out on a buffered basis, using either an
interrupt or polled handshake.
7.2 Internal oscillator and PLL
The PCU9669 contains an internal 12.0 MHz oscillator and 156 MHz PLL which are used
for all internal and I2C-bus timing. The oscillator and PLL require up to tinit(po) to start up
and lock after power-up. The oscillator is not shut down if the serial bus is disabled.
7.3 Buffer description
Remark: In the following section a ‘transaction’ is defined as a contiguous set of
commands and/or data sent/received to/from a single slave. A ‘sequence’ is a set of
transactions stored in the buffer.
The PCU9669 channels have individual 4352-byte data buffers (see Section 7.3.2 “Buffer
sizes”) that allow several transactions to be executed before an interrupt is generated.
This allows the host to request several transactions (up to maximum buffer size on each
channel) in a single sequence and lets the PCU9669 perform it without the intervention of
the host each time a requested transaction is performed. The host can then perform other
tasks while the PCU9669 executes the requested sequences.
By following a simple procedure, the I2C-bus controller can store several I2C-bus
transactions directed to different slaves addresses on any of the channels. The
transaction stored in the buffer can be of any type, thus reads and writes can be interlaced
in a sequence. When multiple slave reads are requested in a sequence, the read data is
stored in-line in the sequence and the buffer number must be specified in the TRANSEL
to provide the read location and the TRANOFS byte offset value. By default, the
TRANOFS is set to 00h. So let us consider the scenario where the host has done the
initialization (mode, masks, and other configuration) and writes data into the buffer of one
of the three channels.
The host starts by programming the buffer configuration registers TRANCONFIG (number
of slaves and bytes per slave) and then the SLATABLE (slave addresses). Then the host
programs the TRANSEL (Transaction Data Buffer Selection) and the TRANOFS (byte
offset selection) to 00h to set the memory pointers to the beginning of the buffer (the
default value is 00h after a power-on or RESET). Next, the host transfers the data into
DATA until the entire sequence is loaded. If the transaction is a read transaction, the host
must write a dummy byte (i.e., FFh) for each expected serial read byte to reserve the
memory space in the buffer for the transaction.
Care should be taken so as to not overflow the buffer with excessive read/write
commands. In the event of an overflow, represented by the BE bit in the CTRLSTATUS
register, will be set to logic 1. The INT pin will be set LOW if the BEMSK bit in the
CTRLINTMSK register is logic 0. To recover the channel, a channel reset is required. All
configuration and data needs to be checked by the host and resent to the I2C-bus
controller. (See Section 7.3.2 “Buffer sizes”.)
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
6 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
After sending all the commands and data it wanted to the I2C-bus controller, the host
could either continue to program data for other channels or write to the CONTROL
register to begin data transmission on the current channel. The transactions will be sent
on the I2C-bus in the order in which the slave addresses are listed in the SLATABLE,
separated by a RESTART condition. The last transaction in the sequence will end with a
STOP condition.
If during a READ command a NACK on the slave address is received, the buffer space
allocated for the read will remain untouched and will contain the last information written in
that location. A buffer read on the parallel bus should only be done after a valid buffer
state is reached to guarantee data valid (see Section 7.5.1.1 “STATUS0_[n],
STATUS1_[n], STATUS2_[n] — Transaction status registers”).
To program data for another channel, that channel is selected and data programmed as
described above. One or more channels can be busy with serial transmission while
additional parallel-bus data is sent to the buffer of an idle channel.
7.3.1 Buffer management assumptions
• Repeated STARTs will be sent between two consecutive transactions.
• After the last operation on a channel is completed, a STOP will be sent.
• In a READ transaction, after the last data byte has been received from a particular
slave, a NACK is sent to the slave.
7.3.2 Buffer sizes
The PCU9669 channels have individual buffers assigned to them. The contents of the
buffers should only be modified during channel idle states.
The memory allocation is 4352 bytes per channel.
The buffer sizes represent the memory allocated for the data block only. The slave
address table and configuration bytes are contained in other locations and do not need to
be included in the required buffer size calculation.
For example, to calculate the size of the memory needed to write 26 bytes to 10 slaves
and to read 2 bytes from 4 slaves (no command bytes required for the read):
10 slaves 26 bytes/slave = 260 bytes for the write transactions
4 slaves 2 bytes/slave = 8 bytes for the read transactions
A total of 268 bytes of buffer space is required to complete the sequence.
Remark: Note that the bytes required to store the 30 slave addresses are not included in
the calculation since they are stored in the SLATABLE register.
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
7 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
7.4 Error reporting and handling
In case of any transaction error conditions, the device will load the transaction error status
in the STATUSx_[n], generate an interrupt, if unmasked, by pulling down the INT pin and
update the CHSTATUS and CTRLSTATUS registers. The status for the individual SLA
addresses will be stored in the STATUSx_[n] registers.
In the event of a NACK from a slave, there are two possible courses of action. The first is
that an interrupt will be generated and the current transaction and sequence terminated.
The second is that while the WEMSK and/or REMSK is a logic 1, a NACKed byte will be
ignored, and the transmission will continue with the next transaction in the sequence until
the end of the sequence. The controller will skip the slave address and/or data where the
NACK occurred and move on to the next transaction in the sequence. Any error will be
reported in the corresponding STATUSx_[n] register (where ‘n’ is the buffer number of the
slave) or the CHSTATUS or CTRLSTATUS registers.
7.5 Registers
The PCU9669 contains several registers that are used to configure the operation of the
device, status reporting, and to send and receive data. The device also contains global
registers for chip level control and status reporting.
The STATUSx_[n] registers are channel-level direct access registers. The DATA,
SLATABLE, TRANCONFIG, and BYTECOUNT registers are auto-increment registers.
The memory access pointer to the DATA registers can be programmed using the
TRANSEL and TRANOFS registers. See Section 7.5.1.2 “CONTROL — Control register”,
for information on the pointer reset bits BPTRRST and AIPTRRST.
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
8 of 69
xxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxx x x x xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxx xx xx xxxxx
xxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxx xxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxx x x
xxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxx
xxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxx
Table 3.
PCU9669 register address map - direct register access
7
6
5
4
3
2
1
0
Register name Access Write access
Description
Default
Size
while
(bytes)
CH active
Channel status registers
0
0
1
0
1
0
channel 0 transaction number STATUS0_[n]
(hex)
R
R
R
no
no
no
individual transaction status (direct address)
00h
00h
00h
64
64
64
channel 1 transaction number STATUS1_[n]
(hex)
individual transaction status (direct address)
([7:2] = 0 in UFm)
channel 2 transaction number STATUS2_[n]
(hex)
individual transaction status (direct address)
([7:2] = 0 in UFm)
Channel 0 (Fm+) registers
1
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
1
1
0
0
1
0
1
0
CONTROL
CHSTATUS
INTMSK
R/W
R
yes[1]
no
channel 0 control
00h
00h
00h
1
channel 0 status
1
R/W
R/W
yes
channel 0 interrupt mask
1
SLATABLE
no
channel 0 slave address table (auto-increment) 00h
64
65
TRANCONFIG R/W
yes, for
channel 0 transaction configuration
00h
TRANCOUNT[2] (auto-increment)
0
0
0
1
1
1
1
0
0
1
1
0
1
0
1
0
DATA
R/W
R/W
R/W
R
yes
yes
yes
no
channel 0 data (auto-increment)
00h
00h
00h
00h
bufsize[3]
TRANSEL
TRANOFS
BYTECOUNT
channel 0 transaction data buffer select
1
channel 0 transaction data buffer byte offset
1
channel 0 transmitted byte count
(auto-increment)
64
1
1
1
1
1
1
1
0
0
0
1
1
1
1
0
1
1
0
0
1
1
1
0
1
0
1
0
1
FRAMECNT
REFRATE
SCLL
R/W
R/W
R/W
R/W
R/W
R/W
R/W
no
no
no
no
no
no
yes
channel 0 frame count
channel 0 frame refresh rate
channel 0 clock LOW state
channel 0 clock HIGH state
channel 0 mode
01h
00h
5Eh
3Fh
92h
00h
00h
1
1
1
1
1
1
1
SCLH
MODE
TIMEOUT
PRESET
channel 0 time-out
channel 0 parallel reset
xxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxx x x x xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxx xx xx xxxxx
xxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxx xxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxx x x
xxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxx
xxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxx
PCU9669 register address map - direct register access …continued
Table 3.
7
6
5
4
3
2
1
0
Register name Access Write access
Description
Default
Size
while
(bytes)
CH active
Channel 1 (UFm) registers
1
1
0
1
0
0
0
0
0
0
0
0
0
1
0
0
1
1
0
0
1
0
1
0
CONTROL
CHSTATUS
INTMSK
R/W
R
yes[1]
no
channel 1 control ([7] = 1)
00h
00h
00h
1
channel 1 status ([5:1] = 0 in UFm)
channel 1 interrupt mask ([5:1] = don’t care)
1
R/W
R/W
yes
1
SLATABLE
no
channel 1 slave address table (auto-increment) 00h
64
65
TRANCONFIG R/W
yes, for
channel 1 transaction configuration
00h
TRANCOUNT[2] (auto-increment)
0
0
0
1
1
1
1
0
0
1
1
0
1
0
1
0
DATA
R/W
R/W
R/W
R
yes
yes
yes
no
channel 1 data (auto-increment)
00h
00h
00h
00h
bufsize[3]
TRANSEL
TRANOFS
BYTECOUNT
channel 1 transaction data buffer select
1
channel 1 transaction data buffer byte offset
1
channel 1 transmitted byte count
(auto-increment)
64
1
1
1
1
1
1
1
0
0
0
1
1
1
1
0
1
1
0
0
1
1
1
0
1
0
1
0
1
FRAMECNT
REFRATE
SCLPER
SDADLY
MODE[4]
-
R/W
R/W
R/W
R/W
R/W
-
no
no
no
no
no
-
channel 1 frame count
channel 1 frame refresh rate
channel 1 clock period
channel 1 SDA delay
channel 1 mode
01h
00h
20h
08h
83h
00h
00h
1
1
1
1
1
1
1
reserved
PRESET
R/W
yes
channel 1 parallel reset
xxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxx x x x xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxx xx xx xxxxx
xxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxx xxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxx x x
xxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxx
xxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxx
PCU9669 register address map - direct register access …continued
Table 3.
7
6
5
4
3
2
1
0
Register name Access Write access
Description
Default
Size
while
(bytes)
CH active
Channel 2 (UFm) registers
1
1
1
0
0
0
0
0
0
0
0
0
0
1
0
0
1
1
0
0
1
0
1
0
CONTROL
CHSTATUS
INTMSK
R/W
R
yes[1]
no
channel 2 control ([7] = 1)
00h
00h
00h
1
channel 2 status ([5:1] = 0 in UFm)
channel 2 interrupt mask ([5:1] = don’t care)
1
R/W
R/W
yes
1
SLATABLE
no
channel 2 slave address table (auto-increment) 00h
64
65
TRANCONFIG R/W
yes, for
channel 2 transaction configuration
00h
TRANCOUNT[2] (auto-increment)
0
0
0
1
1
1
1
0
0
1
1
0
1
0
1
0
DATA
R/W
R/W
R/W
R
yes
yes
yes
no
channel 2 data (auto-increment)
00h
00h
00h
00h
bufsize[3]
TRANSEL
TRANOFS
BYTECOUNT
channel 2 transaction data buffer select
1
channel 2 transaction data buffer byte offset
1
channel 2 transmitted byte count
(auto-increment)
64
1
1
1
1
1
1
1
0
0
0
1
1
1
1
0
1
1
0
0
1
1
1
0
1
0
1
0
1
FRAMECNT
REFRATE
SCLPER
SDADLY
MODE[4]
-
R/W
R/W
R/W
R/W
R/W
-
no
no
no
no
no
no
yes
channel 2 frame count
channel 2 frame refresh rate
channel 2 clock period
channel 2 SDA delay
channel 2 mode
01h
00h
20h
08h
83h
00h
00h
1
1
1
1
1
1
1
reserved
PRESET
R/W
channel 2 parallel reset
xxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxx x x x xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxx xx xx xxxxx
xxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxx xxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxx x x
xxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxx
xxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxx
PCU9669 register address map - direct register access …continued
Table 3.
7
6
5
4
3
2
1
0
Register name Access Write access
Description
Default
Size
while
(bytes)
CH active
Global registers
1
1
1
1
0
0
0
0
0
0
0
0
1
0
0
0
0
1
1
1
1
1
0
0
1
1
0
0
1
1
1
0
1
0
1
0
1
0
1
1
CTRLSTATUS
R
yes
yes
no
controller status
master interrupt mask
reserved
00h
00h
08h
00h
00h
00h
E9h
00h
FFh
1
1
CTRLINTMSK R/W
-
R
R
R
R
R
-
no
reserved
-
no
reserved
-
no
reserved
DEVICE_ID
no
device ID
CTRLPRESET R/W
CTRLRDY[5]
yes
no
master parallel reset
controller ready register
1
1
R
[1] Except TP and TE. Changing polarity of TP while TE is active will cause a false trigger.
[2] The transaction count (TRANCONFIG[0]) can be written to during the idle period between sequences.
[3] Refer to Section 7.3.2 “Buffer sizes” for channel memory allocation.
[4] Unused bits in the UFm register set will return 0b when read and writes will be ignored.
[5] Controller ready = FFh immediately after POR or after a hardware reset or global reset. It will clear (00h) once the initialization routine is done.
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
7.5.1 Channel registers
7.5.1.1 STATUS0_[n], STATUS1_[n], STATUS2_[n] — Transaction status registers
STATUS0_[n], STATUS1_[n], and STATUS2_[n] are 8-bit 64 read-only registers that
provide status information for a given transaction. Only the 5 lower bits are used; the
top bits will always read 0. When bits [4:2] are set, a channel interrupt is requested (the
INT pin is asserted LOW). A read to STATUSx_[n] register will clear its status. To clear all
the STATUSx_[n] registers, a byte-by-byte read of all STATUSx_[n] registers is required.
The controller will auto-clear the STATUSx_[n] registers at each START of a sequence
when FRAMECNT = 1 and only at the first START when FRAMECNT 1.
Each register byte can be accessed by direct addressing so that the host can choose to
read the status on one or more individual transactions without having to read all
64 status bytes.
Table 4.
STATUSx_[n] - Transaction status code register bit description
Bit Symbol Description
7:5 ST[7:5] always reads 000
4
3
2
1
0
RSN[1] Read slave NACK. When HIGH, a NACK was received after a slave address was
transmitted on the serial bus on a read transaction. An interrupt will be requested.
WSN[1] Write slave NACK. When HIGH, a NACK was received after a slave address was
transmitted on the serial bus on a write transaction. An interrupt will be requested.
WDN[1] Write data NACK. When HIGH, a NACK was received for a data byte during a
write transaction on the serial bus. An interrupt will be requested.
TA
Transaction active. When 1, the transaction is currently active on the serial bus.
No interrupt is requested.
TR
Transaction ready. When 1, a transaction is loaded in the buffer and waiting to be
executed. No interrupt is requested.
[1] Does not apply to the UFm channel.
Remark: When STATUSx_[n] = 00h, no interrupt is requested and the transaction is in the
Done/Idle state.
During program execution, the TR and TA bits behave as follows:
Example, we are to transfer 3 transactions in a sequence. All initialization is completed
(loading of SLA, TRANCONFIG, DATA) and device is ready for serial transfer.
Before the STA bit is set, the STATUSx_[n] register will contain:
STATUSx_[0] = 0
STATUSx_[1] = 0
STATUSx_[2] = 0
STATUSx_[3] = 0
:
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
13 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
After STA is set:
STATUSx_[0] = 2
STATUSx_[1] = 1
STATUSx_[2] = 1
STATUSx_[3] = 0
:
Since there is no timing requirement in setting the STA bit after the initialization, the device
will update the first status when the STA bit is set and will always go from 0 to 2 (Idle to
Transaction active).
7.5.1.2 CONTROL — Control register
CONTROL is an 8-bit register. The STO bit is affected by the bus controller hardware: it is
cleared when a STOP condition is present on the I2C-bus.
Table 5.
CONTROL - Control register bit description
Address: Channel 0 = C0h; Channel 1 = D0h; Channel 2 = E0h.
Legend: * reset value
Bit Symbol
Access Value Description
7
STOSEQ
R/W
Stop sequence bit.
1
When the STOSEQ bit is set while the channel is active, a STOP condition will be
transmitted immediately following the end of the current sequence being transferred
on the I2C-bus. No further buffered transactions will be carried out and the channel
will return to the idle state. Normal error reporting will occur up until the last bit. When
a STOP condition is detected on the bus, the hardware clears the STOSEQ flag.
0*
1
When STOSEQ is reset, no action will be taken.
The START flag.
6
STA
R/W
When the STA bit is set to begin a sequence, the bus controller hardware checks the
status of the I2C-bus and generates a START condition if the bus is free (does not
apply to the UFm channel). If the bus is not idle, then INT will go LOW and the
CHSTATUS register will contain a bus error code (either DAE or CLE will be set).
The STA bit may be set only at a valid idle state. The controller will reset the bit under
the following conditions:
• A sequence is done and FRAMECNT = 1.
• A sequence loop is done and FRAMECNT > 1.
• The STOSEQ bit is set, FRAMECNT = 0, and the current sequence is done.
• The STOSEQ bit is set, FRAMECNT > 1, and the current sequence is done.
• The STO bit is set and the current byte transaction is done. This bit cannot be
set if the CHEN bit is 0.
0*
1
When the STA bit is reset, no START condition will be generated.
The STOP flag.
5
STO
R/W
When the STO bit is set while the channel is active, a STOP condition will be
transmitted immediately following the current data or slave address byte being
transferred on the I2C-bus. If a read is in progress, a NACK will be generated before
the STOP. No further buffered transactions will be carried out and the channel will
return to the idle state. Normal error reporting will occur up until the last bit.
When a STOP condition is detected on the bus, the hardware clears the STO flag.
When the STO bit is reset, no action will be taken.
0*
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
14 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
Table 5.
CONTROL - Control register bit description …continued
Address: Channel 0 = C0h; Channel 1 = D0h; Channel 2 = E0h.
Legend: * reset value
Bit Symbol
Access Value Description
4
TP
R/W
Trigger polarity bit. Cannot be changed while channel is active.
1
Trigger will be detected on a falling edge.
Trigger will be detected on a rising edge.
0*
3
TE
R/W
Trigger Enable (TE) bit controls the trigger input used for frame refresh. TE cannot be
changed while channel is active. When the trigger input is enabled, the trigger will
override the contents of the FRAMECNT register and will start triggering when STA
bit is set. Thereafter, when a trigger tick is detected, the controller will issue a START
command and the stored sequence will be transferred on the serial bus.
1
When TE = 1, the sequence is controlled by the Trigger input.
When TE = 0, the trigger inputs are ignored.
0*
1
2
1
BPTRRST
W
Resets auto increment pointers for BYTECOUNT. Reads back as 0.
AIPTRRST W
1
Resets auto increment pointers for SLATABLE and TRANCONFIG. The DATA
register auto-increment pointer will be set to the value that corresponds to TRANSEL
and TRANOFS registers. Reads back as 0.
Remark: To reset the data pointer, write 00h to TRANSEL.
0
-
W
0
Reserved. User must write 0 to this bit.
Remark: Due to a small latency between setting the STA bit and the ability to detect a
trigger pulse, if the STA bit is set simultaneously to an incoming trigger pulse, the pulse
will be ignored and the controller will wait for the next trigger to send the START.
If the STO or STOSEQ bit are set at anytime while the STA bit is 0, then no action will be
taken and the write to these bits is ignored.
Remark: STO has priority over STOSEQ.
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
15 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
Table 6.
CONTROL register bits STA, STO, STOSEQ operation/behavior
Channel state
(initialization steps)
Next write action by host
FRAMECNT TE STA STO STOSEQ
Results
Idle (reset, TRANCONFIG,
SLATABLE, DATA, STA = 0)
1
1
0
0
0
1
X
X
X
X
No action.
START transmitted on serial bus followed by
sequence stored in buffer.
Active (reset, load
TRANCONFIG, SLATABLE,
DATA, STA = 1
1
1
0
0
X
X
0
1
X
X
No change; cannot write STA while active.
When the STO bit is set, two actions are
possible:
1. If the transaction is a read, a STOP is
sent after the first read byte (NACK sent)
and the byte count is updated.
2. If the transaction is a write, a STOP is
sent after the end of ACK cycle of the
current byte and BYTECNT is updated.
The SD bits will be set.
No action.
REFRATE Loop idle (reset,
load TRANCONFIG,
SLATABLE, DATA STA = 1)[1]
1
1
0
0
0
X
0
X
1
X
Channel will go immediately to the inactive
state and SD and FLD bits will be set.[2]
1
0
X
1
X
Channel will go immediately to the inactive
state and SD and FLD bits will be set.[2]
REFRATE Loop active (reset, 1
0
0
X
X
0
0
0
1
No action.
load, TRANCONFIG,
1
STOP at end of current frame. The SD and
FLD bits will be set.
SLATABLE, DATA, STA = 1)
1
0
X
1
X
When the STO bit is set, two actions are
possible:
1. If the transaction is a read, a STOP is
sent after the first read byte (NACK sent)
and the byte count is updated.
2. If the transaction is a write, a STOP is
sent after the end of ACK cycle of the
current byte and BYTECNT is updated.
The SD and FLD bits will be set.
No action.
Trigger Loop Idle (reset, load
TRANCONFIG, SLATABLE,
DATA, STA = 1)
X
X
1
1
0
X
0
X
1
X
STOP at end of current frame. The SD and
FLD bits will be set.
X
1
X
1
X
When the STO bit is set, two actions are
possible:
1. If the transaction is a read, a STOP is
sent after the first read byte (NACK sent)
and the byte count is updated.
2. If the transaction is a write, a STOP is
sent after the end of ACK cycle of the
current byte and the BYTECNT is
updated.
The SD and FLD bits will be set.
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
16 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
Table 6.
CONTROL register bits STA, STO, STOSEQ operation/behavior …continued
Channel state
(initialization steps)
Next write action by host
FRAMECNT TE STA STO STOSEQ
Results
Trigger Loop active (reset, load
TRANCONFIG, SLATABLE,
DATA, STA = 1)
X
X
1
1
X
X
0
0
0
1
No action.
Channel will go immediately to the inactive
state and SD and FLD bits will be set.[2]
X
1
X
1
X
Channel will go immediately to the inactive
state and SD and FLD bits will be set.[2]
[1] Loop Idle is defined as the time elapsed from a STOP to the START of the next sequence while STA = 1.
[2] Channel Active is defined by the CTRLSTATUS[5:3] bits.
7.5.1.3 CHSTATUS — Channel status register
CHSTATUS is an 8-bit read-only register that provides status information for a given
channel. Some of these status bits are error codes that cannot be masked (NMI) by the
INTMSK register and need attention from the host. All these status drive the INT pin
active LOW. To clear the individual channel interrupt request, you must read the
CHSTATUS register. The BE interrupt is cleared by reading the CTRLSTATUS register.
After the CHSTATUS register is cleared, only new errors or status updates will cause the
CHSTATUS bits to be set.
Table 7.
CHSTATUS - Channel and buffer status codes register bit description
Address: Channel 0 = C1h; Channel 1 = D1h; Channel 2 = E1h.
Bit Symbol Description
7
6
5
4
SD
Sequence Done. The sequence loaded in the buffer was sent and STOP issued
on the serial bus.
FLD
WE[1]
RE[1]
Frame Loop Done. The FRAMECNT value has been reached. A STOP has been
issued on the bus.
Write Error detected in transaction. An SLA NACK or data NACK was detected in
a write transaction of the sequence.
Read Error detected in transaction. An SLA NACK was detected in a read
transaction of the sequence.
3
2
1
0
DAE[1]
CLE[1]
SSE[1]
FE
Bus error, SDA stuck LOW.
Bus error, SCL stuck LOW.
Bus error, illegal START or STOP detected.
Frame Error detected. The time required to send the sequence exceeds refresh
rate programmed to the REFRATE register or the time between trigger ticks.
[1] Does not apply to UFm channel. Always read as logic 0.
The DAE, CLE and SSE bits correspond to bus error states, and the FE bit corresponds to
host programming errors.
DAE - SDA error bit: This bit indicates that the SDA line is stuck LOW when the
PCU9669 is trying to send a START condition.
CLE - SCL error bit: This bit indicates that the SCL line is stuck LOW.
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
17 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
SSE - illegal START/STOP detected bit: This bit indicates that a bus error has occurred
during a serial transfer. A bus error is caused when a START or STOP condition occurs at
an illegal position in the format frame. Examples of such illegal positions are during the
serial transfer of an address byte, a data byte, or an acknowledge bit. A bus error may
also be caused when external interference disturbs the internal PCU9669 signals.
FE - Frame Error bit: This bit indicates that the time required to send the sequence
exceeds the refresh rate programmed in the REFRATE register or the time between
trigger ticks. Solving frame errors include programming longer refresh rates, speeding up
the bus frequency, shortening the amount of bytes sent/received in the sequence, or
increasing the time between trigger ticks. If the frame error is masked by the FEMSK, the
device will continue to transmit transactions until the end of the sequence without
re-starting the sequence even if new triggers are detected. The total number of sequences
transmitted will be the number stored in the FRAMECNT register. Once a complete
sequence is transmitted, a new sequence will initiate when a subsequent trigger appears.
The FE flag will be held HIGH and sequences will still be transmitted unless CHSTATUS is
read. If the frame error is unmasked, the sequence will be aborted at the next logical
stopping point (i.e., for a read transaction a NACK will be sent), a STOP transmitted and
an interrupt will be generated. Since the controller terminates the sequence in a controlled
mechanism, there may be a 2-byte delay if a frame error (FE) is detected during a read
transaction. The FE bit is set after the STOP is detected on the bus.
sequence A
10 ms
sequence A
10 ms
sequence A
10 ms
time
002aaf247
a. Sequence fully executed within the period programmed in REFRATE register
frame error detected, data not sent after FE
sequence B
time
10 ms
10 ms
10 ms
002aaf627
This condition causes a frame error and the FE bit to be set.
b. Sequence exceeds period programmed in REFRATE register, FEMSK = 0
frame error detected, FEMSK = 1, data sent after FE
sequence C
10 ms
sequence C
time
10 ms
10 ms
002aaf628
c. Sequence exceeds period programmed in REFRATE register, FEMSK = 1
Fig 3. Frame Error detection
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
18 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
Table 8.
Error detection operation/behavior
Channel state
AR (MODE Error detected
Next Action
register)
(CHSTATUS)
DAE CLE SSE
Active or idle
X
0
0
1
Interrupt set, if a transaction is active it will be
immediately aborted and no further action taken by
controller. Host to re-initialize bus (i.e., force a bus
recovery), reset slaves, or take other appropriate
recovery action. After bus is recovered, host to
re-start transaction.
Active or idle, time-out enabled,
and clock line is LOW
X
1
0
0
1
0
0
0
Interrupt set, active transaction will be immediately
aborted and no further action taken by controller.
No bus recovery possible by bus-controller. Host to
recover bus by resetting slaves or system. After
bus is recovered, host to re-start transaction.
Active and at a START or
repeated-START condition
Interrupt not set, active transaction will be
immediately aborted and a bus recovery will be
attempted by the bus-controller. If successful, a
start will be issued automatically and the serial
transfer will continue normally at the location of the
failed transaction. No host action is required.
1
0
1
1
0
0
0
0
Interrupt set, an auto-recovery was attempted and
failed. Active transaction will be immediately
aborted and the bus-controller determines bus
recovery actions, for example setting the BR bit or
resetting the slaves.
Interrupt set, active transaction will be immediately
aborted and no bus recovery will be attempted by
the bus-controller. Host may attempt a bus
recovery by setting the BR bit or determine other
bus recovery action.
7.5.1.4 INTMSK — Interrupt mask register
Through the INTMSK register, there is the option to manage which states generate an
interrupt, allowing more control from the host on the transaction. The interrupt mask
applies to all transactions in a given channel. A bit set to 1 indicates that the mask is
active. The INTMSK register default is all interrupts are un-masked (00h).
Table 9.
INTMSK - Interrupt mask register bit description
Address: Channel 0 = C2h; Channel 1 = D2h; Channel 2 = E2h.
Bit
7
Symbol
Description
SDMSK
Sequence Done Mask. The end of sequence interrupt will not be generated.
6
FLDMSK Frame loop done mask. A frame loop done interrupt will not be generated. The
controller will enter the idle state.
5
4
WEMSK[1] Write Error Mask. An SLA NACK or data NACK interrupt will not be generated
and the controller will skip the remaining write data in the transaction and
continue with the START of the next transaction in the sequence.
REMSK[1] Read Error detected in transaction. An SLA NACK interrupt will not be
generated and the controller will skip the read transaction and continue with
the START of the next transaction in the sequence.
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
19 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
Table 9.
INTMSK - Interrupt mask register bit description …continued
Address: Channel 0 = C2h; Channel 1 = D2h; Channel 2 = E2h.
Bit
3:1
0
Symbol
-
Description
reserved
FEMSK
Frame Error Mask. A frame error interrupt will not be generated.
Remark: Use caution and good judgement when using this mask.
Unexpected/erratic behavior may result in the slave devices.
[1] Does not apply to UFm channel.
7.5.1.5 SLATABLE — Slave address table register
SLATABLE is an 8-bit 64 register set that makes up a table that stores the slave address
for each transaction in the sequence. The table is loaded by using an auto-increment
pointer that is not user-accessible. To reset the pointer, the AIPTRRST bit must be set in
the CONTROL register. The slave addresses in the SLATABLE register are stored with a
zero-based (N 1) index. The first slave address occupies the 00h position.
Remark: Slave address entries greater than the transaction count are not part of the
sequence. TRANCONFIG[0] contains the transaction count that will be included in the
sequence.
Table 10. SLATABLE - Slave address table register bit description
Address: Channel 0 = C3h; Channel 1 = D3h; Channel 2 = E3h.
Bit
7:1
0
Symbol
SLATABLE[7:1] Slave address.
SLATABLE[0] When 1, a read transaction is requested.
When 0, a write transaction is requested.
Description
Table 11. Example of SLATABLE registers
Transaction
Slave address
00h
01h
02h
03h
04h
:
10h
12h
28h
40h
14h
:
3Fh
36h
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
20 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
7.5.1.6 TRANCONFIG — Transaction configuration register
The TRANCONFIG register is an 8-bit 65 register set that makes up a table that
contains the number of transactions that will be executed in a sequence and the number
of data bytes involved in the transaction.
The first byte of the register is the Transaction Count register. The remaining 64 registers
are the Transaction Length registers.
Table 12. TRANCONFIG, byte 0 - Transaction configuration register bit description
Address: Channel 0 = C4h; Channel 1 = D4h; Channel 2 = E4h.
Bit
Symbol
Description
7:0
Number of transactions in the sequence. Maximum is 40h.
Table 13. TRANCONFIG, byte 1 to 40h - Transaction configuration register bit description
Bit
Symbol
Description
7:0
Number of bytes per transaction in the sequence. Maximum is FFh.
Table 14. Example of TRANCONFIG register loaded
Register
Value
10h
0Ah
12h
28h
40h
:
Description
Transaction count
Transaction length 00h
Transaction length 01h
Transaction length 02h
Transaction length 03h
:
16 transactions = 16 slave addresses in the SLATABLE
10 byte transaction
18 byte transaction
40 byte transaction
64 byte transaction
:
Transaction length 3Fh
12h
18 byte transaction
Remark: Even if the Transaction length (TRANCONFIG[1:40h]) and the
SLATABLE([0:3Fh]) are fully initialized, only the specified number of transactions in the
Transaction count (TRANCONFIG[0]) will be part of the sequence.
If the Transaction count is 0, then there will be no activity on the serial bus if the STA bit is
set. In addition, there will be no interrupts generated or status updated. The controller will
simply reset the CONTROL.STA bit without performing any transactions.
If the Transaction length is 0, a read transaction will be skipped and a write transaction will
send the slave address plus write bit (SLA+W) on the serial bus with no data bytes.
7.5.1.7 DATA — I2C-bus Data register
DATA is an 8-bit read/write, auto-increment register. It is the interface port to the channel
buffer. When accessing the buffer, the host writes a byte of serial data to be transmitted or
reads bytes that have just been received at this location. The host can read from the
DATA at any time and can only write to this 8-bit register while the channel is idle.
Remark: Reading the DATA when the serial interface is active may return outdated or
erroneous data.
The host can read or write data up to the amount of memory space allotted to the channel.
The location at which the data is accessed is stored in the TRANSEL and TRANOFS
register (both default at 00h).
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
21 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
To return to the data location pointed by the contents of the TRANSEL and TRANOFS
register after read or write access to the DATA register, set the AIPTRRST
(auto-increment pointer reset) bit in the control register.
To return to the first DATA register location in the buffer set the TRANSEL to 00h.
Table 15. DATA - Data register bit description
Address: Channel 0 = C5h; Channel 1 = D5h; Channel 2 = E5h.
Bit
Symbol Description
D[7:0] Eight bits to be transmitted or just received. A logic 1 in DATA corresponds to a
HIGH level on the I2C-bus. A logic 0 corresponds to a LOW level on the bus.
7:0
7.5.1.8 TRANSEL — Transaction data buffer select register
The TRANSEL register is used to select the pointer to a specific transaction in the DATA
buffer. This allows the user to update the data of a specific slave without having to re-write
the entire data buffer or to read back the stored serial data from a read transaction. The
value of this register is the slave address position in the SLATABLE register. The
TRANSEL register is zero-based (N 1) register.
For example, if a change to the 22nd slave address data is required, the host would set
the TRANSEL register to 15h. This register can be used in conjunction with the
TRANSOFS register to access a specific byte in the data buffer. The host would then
proceed to write the new data to the DATA register. The auto-increment feature continues
to operate from this new position in the DATA register.
Setting TRANSEL to an uninitialized TRANCONFIG entry may cause a request to
read/write data outside the data buffer. If this occurs, the BE bit in the CTRLSTATUS
register will be set to a logic 1. Write data will be ignored and read data will be invalid.
When a new transaction is selected by programming the TRANSEL registers, the
TRANSOFS register will automatically be reset to 00h.
Remark: When updating the data buffer, if the number of bytes to be updated or read
exceeds the number of bytes that were specified in the TRANCONFIG register, the
auto-increment will go over the transaction boundary into the next transaction stored in
the buffer.
Remark: To reset the DATA pointer, write 00h to the TRANSEL register.
Table 16. TRANSEL - Transaction data buffer select register bit description
Address: Channel 0 = C6h; Channel 1 = D6h; Channel 2 = E6h.
Bit
7
Symbol
Description
Reserved.
Reserved.
-
-
6
5:0
TRANSEL[5:0] Slave address position in the SLATABLE. The maximum number is 3Fh.
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
22 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
7.5.1.9 TRANOFS — Transaction data buffer byte select register
In conjunction with the TRANSEL register, the TRANOFS register is used to select the
pointer to a specific byte in a transaction in the data buffer. This allows the user to read or
re-write a specific data byte of a specific slave without having to read/re-write the entire
data buffer. The TRANOFS register is zero-based (N 1), so the maximum bytes this
register will point to is 256.
For example, if the tenth byte in the 40th slave address data is required, the host would
set the TRANSEL register to 27h and the TRANSOFS register to 09h. The host would
then proceed with a read to the DATA register.
Setting TRANOFS to a byte offset outside of the data buffer will cause the BE bit in the
CTRLSTATUS register will be set to a logic 1. Write data will be ignored and read data will
be invalid.
Remark: The number of bytes to be updated or read should not exceed the number of
bytes that were specified in the TRANCONFIG register. Doing so will cause the
auto-increment to go over the transaction boundary into the next transaction stored in the
buffer.
Table 17. TRANOFS - Transaction data buffer byte select register bit description
Address: Channel 0 = C7h; Channel 1 = D7h; Channel 2 = E7h.
Bit
Symbol
Description
7:0
TRANOFS[7:0]
Byte index for the specified transaction buffer in TRANSEL.
7.5.1.10 BYTECOUNT — Transmitted and received byte count register
The BYTECOUNT register stores the number of bytes that have been sent or received.
The count is continuously updated, therefore the BYTECOUNT is a real time reporting of
transmitted and received bytes. This is a read-only register. The BYTECOUNT includes
only the bytes that have been ACKed in a write transaction and all bytes received in a
read transaction including in transactions where the WEMSK or REMSK are enabled and
part or complete transactions have been skipped (see Figure 9). The BYTECOUNT
register is cleared at the START of every sequence.
Table 18. BYTECOUNT, byte 0 - Transaction configuration register bit description
Address: Channel 0 = C8h; Channel 1 = D8h; Channel 2 = E8h.
Bit
Symbol
Description
7:0
BYTECOUNT[7:0]
Number of bytes sent/received per transaction in the sequence.
Maximum is FFh.
7.5.1.11 FRAMECNT — Frame count register
Table 19. FRAMECNT - Frame count register bit description
Address: Channel 0 = C9h; Channel 1 = D9h; Channel 2 = E9h.
Bit
Symbol
Description
7:0
FRAMECNT[7:0] Bit 7 to bit 0 indicate the number of times buffered commands are to be
re-transmitted. Default is 01h.
This register is a read/write register. The contents of this register holds the programmed
value by the host and is not a real-time count of frames sent on the serial bus.
If the FRAMECNT is 00h, the sequence stored in the buffer will loop continuously. A
STOP will be sent at the end of each sequence.
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
23 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
If the FRAMECNT is 01h, it is defined as the default state and the sequence stored in the
buffer will be sent once and a STOP will be sent at the end of the sequence.
If the FRAMECNT is greater than 01h, the sequence stored in the buffer will loop
FRAMECNT times and a STOP will be sent at the end of each sequence.
Remark: The FRAMECNT can only be set to loop on the sequence stored in the buffer.
7.5.1.12 REFRATE — Refresh rate register
The REFRATE register defines the time period between each sequence start when
REFRATE looping is enabled (FRAMECNT 1, and TE = 0).
The refresh period defined by REFRATE should always be programmed to be greater
than the time it takes for the sequence to be transferred on the I2C-bus. If the REFRATE
values is too small, the frame error (FE) bit will be set and an interrupt will be requested.
Table 20. REFRATE - Refresh rate register bit description
Address: Channel 0 = CAh; Channel 1 = DAh; Channel 2 = EAh.
Bit
Symbol
Description
7:0
REFRATE[7:0] Bit 7 to bit 0 indicate the sequence refresh period. The resolution is
100 s. The default value is 00h, the timer is disabled, and the sequences
will be sent back-to-back if the FRAMECNT is = 0 or FRAMECNT is > 1.
Remark: If the FRAMECNT is 1, then the refresh rate function will be disabled.
7.5.1.13 SCLL, SCLH and SCLPER, SDADLY — Clock rate registers
Table 21. SCLL - Clock Rate Low register bit description (Standard-mode, Fast-mode,
Fast-mode Plus)
Address: Channel 0 = CBh.
Bit
Symbol
Description
7:0
L[7:0]
Eight bits defining the LOW state of SCL. Default: 94 (5Eh).
Table 22. SCLH - Clock Rate High register bit description (Standard-mode, Fast-mode,
Fast-mode Plus)
Address: Channel 0 = CCh.
Bit
Symbol
Description
7:0
H[7:0]
Eight bits defining the HIGH state of SCL. Default: 63 (3Fh).
The clock rate register for the Standard-mode, Fast-mode, and Fast-mode Plus (Fm+) is
controlled by the SCLL and SCLH registers and the for the Ultra Fast-mode channel by
the SCLPER and SDADLY registers. They define the data rate for the serial bus of the
PCU9669. The actual frequency on the serial bus is determined by tHIGH (time where SCL
is HIGH), tLOW (time where SCL is LOW), tr (rise time), and tf (fall time) values. Writing
illegal values into the SCLL and SCLH registers or SCLPER registers will cause the part
to operate at the respective maximum channel frequency.
For Standard, Fast, and Fast-mode Plus, tHIGH and tLOW are calculated based on the
values that are programmed into SCLH and SCLL registers and the PLL clock frequency.
For UFm mode, the clock is a fixed 50 % duty cycle defined by the SCLPER. In both
cases tr and tf are system/application dependent.
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
24 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
Remark: The MODE register needs to be programmed before programming the SCLL
and SCLH registers in order to know which I2C-bus mode is selected. See Section
7.5.1.14 “MODE — I2C-bus mode register” for more detail.
Fast-mode Plus (Fm+) is the default selected mode at power-up or after reset.
The clock is derived from the internal PLL frequency which is set at 156 MHz (13 OSC
clock). Given a 1 % accuracy on the internal clock, the worst case TPLL is
1
1
---------------------------------------
------------------------------
=
= 6.347 ns .
12.12 MHz 13
157.56 MHz
Calculating clock settings for Standard, Fast, and Fast-mode Plus:
1
----------------------------
TOTAL_SCLLH =
scale factor
(1)
TPLL freq
The scale factor is set by the MODE register and used in the TOTAL_SCLLH calculation.
The scale factor is 8 for Standard-mode, 4 for Fast-mode, and 1 for Fast-mode Plus.
The SCLL and SCLH can be found by:
SCLL = 0.6 TOTAL_SCLLH
SCLH = 0.4 TOTAL_SCLLH
(2)
(3)
Remark: The contributions for the rise time (tr) and fall time (tf) are adjusted internally by
hardware to match the desired frequency. If an invalid number is written to SCLL or SCLH
such that it violates the specification, then the controller will adjust the bus frequency to
the allowable SCLL and SCLH minimums.
Sample resulting SCL frequencies:
Table 23. SCL calculation scale factor
I2C-bus mode
Frequency
100 kHz
Scale factor
Standard
8
4
1
Fast
400 kHz
Fast-mode Plus
1000 kHz
Table 24. Typical SCL frequencies
Data shown under following conditions:
Pull-up resistor RPU = 500 ; bus capacitance Cb = ~170 pF.
Desired frequency (kHz)
Actual frequency (kHz)
SCLL
SCLH
Standard-mode (Sm)
100
90
80
70
60
50
99.3
90.0
80.0
69.5
59.7
50.0
116
129
145
168
194
233
79
87
98
112
132
156
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
25 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
Table 24. Typical SCL frequencies …continued
Data shown under following conditions:
Pull-up resistor RPU = 500 ; bus capacitance Cb = ~170 pF.
Desired frequency (kHz)
Actual frequency (kHz)
SCLL
SCLH
Fast-mode (Fm)
400
398.4
348.7
298.2
250.2
198.0
150.1
100.0
58
39
350
66
45
300
78
52
250
93
62
200
117
155
233
79
150
104
156
100
Fast-mode Plus (Fm+)
1000
900
800
700
600
500
400
999.0
900.0
798.3
698.5
599.9
499.5
399.7
90
63
100
113
130
152
183
229
70
79
90
105
126
158
Remark: The correct MODE setting should be programmed based on desired frequency
since the bus controller will internally select the appropriate tr and tf for the selected mode.
The minimum I2C-bus frequency is 50 kHz.
Remark: The actual SCL frequency will be affected by the PLL frequency and the bus
load. The controller will adjust the SCL timing by monitoring the rise time on the SCL line
and bring the output frequency as close to the programmed value as possible without
violating the I2C-bus specification for minimum clock HIGH and LOW timing.
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
26 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
Table 25. SCLPER - Clock Period register bit description (Ultra Fast mode)
Address: Channel 1 = DBh; Channel 2 = EBh.
Bit
Symbol
Description
7:0
L[7:0]
Eight bits defining the clock period (Ultra Fast mode). Default 32 (20h).
Table 26. SDADLY - SDA delay register bit description (Ultra Fast mode)
Address: Channel 1 = DCh; Channel 2 = ECh.
Bit
7:6
5:0
Symbol
H[7:6]
Description
Reserved. Read only read back zero.
Six bits defining the SDA delay (Ultra Fast mode). Default: 8 (08h).
H[5:0]
Calculating clock settings for Ultra Fast mode (UFm):
The clock period is defined as follows (50 % duty cycle):
1
----------------------------
SCLPERmin
=
(4)
(5)
TPLL freq
The data will be delayed with respect to the falling edge of the clock as follows:
SCLPER
----------------------
=
SDADLYmax
4
Table 27. Sample clock period and allowable data delay
Frequency
5.0 MHz
4.0 MHz
3.0 MHz
2.0 MHz
1.0 MHz
SCLPER[1]
SDADLY[2]
32
39
53
2 to 8
2 to 9
2 to 13
2 to 19
2 to 39
79
158
[1] The minimum allowable value that can be stored in SCLPER is 32.
[2] The minimum allowable value that can be stored in SDADLY is 2.
The PCU9669 will force a 50 % duty cycle by shifting the contents of the SCLPER register
right by 1.
When the user writes the SCLPER register, the SDADLY will be loaded automatically with
a value 1⁄4 the value of SCLPER (SCLPER register value right shifted twice). The user can
then overwrite the SDADLY register if desired.
The order in which the registers should be written is first the SCLPER, then the SDADLY
register to adjust the delay.
The maximum value for SDADLY is the preferred value to be loaded.
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
27 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
7.5.1.14 MODE — I2C-bus mode register
MODE is a read/write register. It contains the control bits that select the bus recovery
options, and the correct timing parameters. Timing parameters involved with AC[1:0] are
tBUF, tHD;STA, tSU;STA, tSU;STO, tHIGH, tLOW. The auto recovery and bus recovery bits are
contained in this register. They control the bus recovery sequence as defined in Section
8.5.1 “I2C-bus obstructed by a LOW level on SDA (DAE)”.
Table 28. MODE - I2C-bus mode register bit description
Address: Channel 0 = CDh; Channel 1 = DDh; Channel 2 = EDh.
Bit Symbol Description
7
CHEN
Channel Enable bit. R/W.
0: Channel is disabled, SCL and SDA high-impedance, USDA and USCL driven
HIGH. All registers are accessible for setup and configuration, however a
sequence cannot be started if the CHEN bit is 0 (STA cannot be set).
1 (default): Channel is enabled.
Reserved.
6
5
-
BR
Bus Recovery. When BR is set to 1, the bus controller will attempt a bus recovery
by sending 9 clock pulses on the bus. Once the bus recovery is complete, the
controller will reset the bit to 0. This bit is not intended to generate random or
asynchronous 9 clock pulses on the bus. This function is performed automatically
when the AR bit is 1.
4
AR
Auto Recovery.
When AR = 1 (default), the bus controller will automatically attempt to recover the
bus as described in Section 8.5.1 “I2C-bus obstructed by a LOW level on SDA
(DAE)”.
When AR = 0, the bus controller will abort the current transaction and generate an
error code by setting the DAE bit in the CHSTATUS register and pulling the INT
pin LOW.
3:2
-
Reserved.
Fm+ Channel 0
1:0 AC[1:0] I2C-bus mode selection to ensure proper timing parameters (see Table 29 and
Table 40).
AC[1:0] = 00: Standard-mode AC parameters selected.
AC[1:0] = 01: Fast-mode AC parameters selected.
AC[1:0] = 10 (default): Fast-mode Plus AC parameters selected.
AC[1:0] = 11: Reserved.
UFm Channel 1 and Channel 2
1:0 AC[1:0] I2C-bus mode selection to ensure proper timing parameters (see Table 29 and
Table 40).
AC[1:0] = 00: Reserved.
AC[1:0] = 01: Reserved.
AC[1:0] = 10: Reserved.
AC[1:0] = 11 (default): Ultra Fast-mode AC parameters selected. Read-only
bits.
Remark: CHEN bit value must be changed only when the I2C-bus is idle.
Remark: Any change in the AC[1:0] bits (Fast-mode to Standard-mode, for example) may
cause the HIGH and LOW timings of SCL to be violated. It is then required to program the
SCLL and SCLH registers with values in accordance with the selected mode.
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
28 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
Remark: The AC[1:0], BR and AR bits are not applicable to the UFm channel, they are
read-only bits. The UFm channel AC parameters are controlled internally.
Table 29. I2C-bus mode selection example
I2C-bus frequency (kHz)[1]
Scale factor
AC[1:0]
00
Mode
100
400
1000
-
8
4
1
-
Standard
Fast
01
10
Fast-mode Plus
reserved
11
1
----------------------------------------------------------------------------------------
=
[1] Using the formula fSCL
TPLLSCLL + SCLH sf + tr + tf
7.5.1.15 TIMEOUT — Time-out register
TIMEOUT is an 8-bit read/write register. It is used to determine the maximum time that
SCL is allowed to be in a LOW logic state before a CLE interrupt is generated.
Remark: The TIMEOUT does not apply to the UFm channel of the controller.
When the I2C-bus interface is operating, TIMEOUT is loaded in the time-out counter at
every LOW SCL transition.
Table 30. TIMEOUT - Time-out register bit description
Address: Channel 0 = CEh.
Bit
Symbol
Description
7
TE
Time-out enable/disable
TE = 1: Time-out function enabled
TE = 0: Time-out function disabled
6:0
TO[6:0]
Time-out value. The time-out period = (TIMEOUT[6:0] + 1) 200 s.
The time-out value may vary some, and is an approximate value.
The Time-out register can be used in the following cases:
• When the bus controller wants to send a START condition and the SCL line is held
LOW by some other device. Then the bus controller waits a time period equivalent to
the time-out value for the SCL to be released. In case it is not released, the bus
controller concludes that there is a bus error, sets the CLE bit in the CHSTATUS
register, generates an interrupt signal and releases the SCL and SDA lines.
• The time-out feature starts every time the SCL goes LOW. If SCL stays LOW for a
time period equal to or greater than the time-out value, the bus controller concludes
there is a bus error and behaves in the manner described above. When the I2C-bus
interface is operating, TIMEOUT is loaded in the time-out counter at every SCL
transition. See Section 8.7 “Global reset” for more information.
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
29 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
7.5.1.16 PRESET — I2C-bus channel parallel software reset register
Table 31. PRESET - I2C-bus channel parallel software reset register bit description
Address: Channel 0 = CFh; Channel 1 = DFh; Channel 2 = EFh.
Bit Symbol
Description
7:0 PRESET[7:0] Read/Write register used during an I2C-bus channel parallel reset command.
PRESET is an 8-bit write-only register. Programming the PRESET register allows the user
to reset each individual PCU9669 channel under software control. The software reset is
achieved by writing two consecutive bytes to this register. The first byte must be A5h while
the second byte must be 5Ah. The writes must be consecutive and the values must match
A5h and 5Ah. If this sequence is not followed as described, the reset is aborted.
The PRESET resets state-machines, registers, and buffer pointers to the default values,
zeroes the TRANCONFIG, SLATABLE, BYTECOUNT, and DATA arrays of the respective
channel and will not reset the entire chip. The parallel bus remains active while a software
reset is active. The user can read the PRESET register to determine when the reset has
completed, PRESET returns all 1s when the reset is active and all 0s when complete.
7.5.2 Global registers
7.5.2.1 CTRLSTATUS — Controller status register
The CTRLSTATUS register reports the status of the controller, including the interrupts
generated by the parallel bus. There are six status bits. When CTRLSTATUS contains
00h, it indicates the idle state and therefore no serial interrupts are requested. The content
of this register is continuously updated during the operation of the controller.
The lower 3 bits represent the channels that have an interrupt request pending. To clear
the individual channel interrupt request, you must read the CHSTATUS register. Bits [5:3]
indicate if a channel is currently active or if it is in the idle state.
Table 32. CTRLSTATUS - Interrupt status register bit description
Address: F0h.
Bit
7
Symbol
BE
Description
Buffer Error. A buffer error such as overflow has been detected.
6
-
5
CH2ACT
CH1ACT
CH0ACT
CH2INTP
CH1INTP
CH0INTP
Channel 2 is active.
4
Channel 1 is active.
3
Channel 0 is active.
2
Channel 2 interrupt pending.
Channel 1 interrupt pending.
Channel 0 interrupt pending.
1
0
Remark: A global reset will reset all channels and configuration settings.
BE - Buffer Error bit: This bit indicates that a buffer error has been detected. For
example, a buffer overflow due to the host programming too many bytes will set this bit. A
software or hardware reset is necessary to recover from a buffer error.
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
30 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
The buffer error may occur when a data location is being read or written to that has not
previously been configured by the TRANCONFIG register. The buffer error can occur on a
parallel data write or read beyond the buffer capacity, or setting the TRANSEL and
TRANOFS pointers beyond the buffer boundary.
When the DATA register is loaded with data that goes beyond the capacity of the buffer,
the bytes that go over the buffer size will be ignored and a Buffer Error (BE) will be
generated.
Special case: The BE interrupt is cleared by reading the CTRLSTATUS register. All other
interrupts are cleared by reading the respective CHSTATUS register.
SD
FLD
WE
RE
DAE
CLE
SSE
FE
CH0INTP (Fm+)
CH1INTP (UFm)
SD
FLD
FE
SD
FLD
FE
CH2INTP (UFm)
002aag093
Fig 4. PCU9669 status reporting logic
See Table 7 for channel status.
7.5.2.2 CTRLINTMSK — Control Interrupt mask register
The CTRLINTMSK masks all interrupts generated by the masked channel. This allows the
host MCU to complete other operations before servicing the interrupt without being
interrupted by the same channel.
Table 33. CTRLINTMSK - Control interrupt mask register bit description
Address: F1h.
Bit
Symbol
Description
7
BEMSK
Buffer Error Mask. A buffer error interrupt will not be generated.
Remark: Use caution and good judgement when using this mask.
Unexpected/erratic behavior may result in the slave devices.
6:3
2
-
reserved
CH2MSK
When this bit is set to 1, all interrupts for the channel will be masked and
the INT pin will not be pulled LOW.
1
0
CH1MSK
CH0MSK
When this bit is set to 1, all interrupts for the channel will be masked and
the INT pin will not be pulled LOW.
When this bit is set to 1, all interrupts for the channel will be masked and
the INT pin will not be pulled LOW.
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
31 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
SD
SDMSK
WE
WEMSK
RE
REMSK
CH0 interrupt
sources and masks
FE
FEMSK
CH0MSK
FLD
FLDMSK
DAE
CLE
SSE
SD
SDMSK
FE
FEMSK
CH1 interrupt
sources and masks
to INT pin
CH1MSK
FLD
FLDMSK
BE
BEMSK
SD
SDMSK
RE
REMSK
CH2 interrupt
sources and masks
FE
FEMSK
CH2MSK
FLD
FLDMSK
002aag094
Fig 5. PCU9669 interrupt logic
See Table 9 for interrupt mask.
7.5.2.3 DEVICE_ID — Device ID
The DEVICE_ID register stores the bus controller part number so it can be identified on
the parallel bus.
Table 34. DEVICE_ID - Device ID register bit description
Address: F6h.
Bit
Symbol
Description
7
U/A
Selects PCU or PCA device.
1 = PCU96xx
0 = PCA96xx
6:0
BCD
BCD (Binary Coded Decimal) code of the ending 2 digits for ID.
Range is 00h to 79h. The code for the PCU9669 is E9h.
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
32 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
7.5.2.4 CTRLPRESET — Parallel software reset register
Table 35. CTRLPRESET - Parallel software reset register bit description
Address: F7h.
Bit
Symbol
Description
7:0
CTRLPRESET[7:0] Write-only register used during a device parallel reset command.
CTRLPRESET is an 8-bit write-only register. Programming the CTRLPRESET register
allows the user to reset the PCU9669 under software control. The software reset is
achieved by writing two consecutive bytes to this register. The first byte must be A5h while
the second byte must be 5Ah. The writes must be consecutive and the values must match
A5h and 5Ah. If this sequence is not followed as described, the reset is aborted.
7.5.2.5 CTRLRDY — Controller ready register
Table 36. CTRLRDY - Controller ready register bit description
Address: FFh.
Bit
Symbol
Description
7:0
CTRLRDY[7:0] Read-only register indicates the internal state of the controller. FFh
indicates the controller is initializing, 00h indicates controller is in normal
operating mode.
CTRLRDY (address FFh) is an 8-bit read-only register. It indicates the internal state of the
controller. When the register is FFh, the controller is in the initialization state. The
initialization state will be entered at power-up, after a hardware reset, or after a global
software reset.
The oscillator and the PLL will be initialized only after a Power-On Reset (POR), a
hardware reset, or a global software reset (CTRLPRESET).
When the register is 00h, the controller is in the normal operating mode.
Access while the controller is initializing requires CE pin follow the RD pin transitions to
update the state of the controller that is read back. After controller is ready, the CE pin can
be held LOW while RD and WR pins transition. See Figure 6, Figure 7 and Figure 8.
CE
RD
DATA
FFh
FFh
00h
ready
initializing
002aag095
Fig 6. During initialization, CE must transition with RD at each read operation
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
33 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
ADDR
read address X
read address Y
read address Z
CE
RD
DATA
address X
address Y
address Z
002aag096
Fig 7. During normal operation, CE may remain LOW while RD transitions during
multiple reads
ADDR
CE
write address X
write address Y
write address Z
WR
DATA
data X
data Y
data Z
002aag097
Fig 8. During normal operation, CE may remain LOW while WR transitions during
multiple writes
8. PCU9669 operation
The PCU9669 is designed to efficiently transmit and receive large amounts of data on a
single master bus. There are three major components that compose the architecture of
the I2C-bus controller that interact with each other to provide a high throughput and a high
level of automation when it conducts transactions:
• Slave address table: specifies the address of the slaves on the bus and the direction
(read or write).
• Transaction configuration: specifies the size of the transaction.
• Data buffer: contains the data to be transmitted or received from the slave.
These three components are integrated in the PCU9669 to build a sequence. A sequence
is a set of read or write transactions and the minimum sequence size is one read or write
transaction. Several transactions can be stored in one sequence and be executed without
the intervention of the host controller (CPU) through loop control and using the built-in
refresh rate timers.
The PCU9669 executes transactions in the order they were loaded into the buffer without
interrupting the host. Once the end of a sequence is reached, the Sequence Done (SD) bit
will be asserted in the CHSTATUS register and the controller will request an interrupt, if
SDMSK = 0. At this point, the host can reload the buffer with a new sequence or resend
the one that is currently loaded in the buffer.
When a sequence is in progress, no interrupts are generated unless there is an error
when a transaction is conducted. The host will only receive an interrupt when the
sequence is done. The PCU9669 will dynamically shift between being a Master
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
34 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
Transmitter or a Master Receiver according to the direction bits specified in the
SLATABLE. The host has the ability to retrieve stored serial data as soon as a read
transaction is done, while the controller carries on the remaining transactions in the
sequence.
8.1 Sequence execution
Sequences can have transactions of two types:
• Write transactions, where the PCU9669 will behave as a Master Transmitter
• Read transactions, where the PCU9669 will behave as a Master Receiver on the
Fm+ channel only, since UFm channels are uni-directional
Data transfers in each direction are shown in Figure 9. This figure contains the following
abbreviations:
S — START condition
SLA — 7-bit slave address
R — Read bit (HIGH level at SDA)
W — Write bit (LOW level at SDA)
A — Acknowledge bit (LOW level at SDA)
A — Not acknowledge bit (HIGH level at SDA)
Data — 8-bit data byte
P — STOP condition
In Figure 9, circles are used to indicate when a bit is set in the CHSTATUS register. A
channel interrupt is not requested when CHSTATUS = 00h and the INT pin is not asserted
when the interrupt is masked (see Section 7.5.2.2).
For a successful sequence execution, all three components mentioned above must exist
in the memory and must be correctly set up. There are not safeguards against
programming incorrect transaction sizes, data buffer lengths, or direction bits. If the
transaction length is set to 00h, then only the slave address with direction bit will be
transmitted.
Once the host has configured the serial port and programmed the TRANCONFIG (number
of slaves and bytes per slave), the SLATABLE (slave addresses), TRANSEL (transaction
data buffer selection) and the TRANOFS (byte offset selection) and loaded the serial data
into the DATA buffer, the sequence is ready to be transmitted.
To send the sequence, the host will set the STA bit in the CONTROL register and the
controller will immediately send a START on the serial bus. Then, the transactions will be
carried out in the order they appear in the SLATABLE, each being separated by a
ReSTART command.
If the interrupts are unmasked, the serial transfer will be conducted without generating
interrupts in between transactions. Once all transactions are successfully completed, the
controller will generate a STOP, the Sequence Done bit (SD) will be set in the CHSTATUS
and an interrupt will be generated.
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
35 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
When the interrupts are unmasked, a NACK on slave address or data (in a write cycle) will
terminate the serial transfer, generate a STOP, and the INT pin will be asserted. The host
can read the CTRLSTATUS (Controller status register) to determine which channel
generated the interrupt, then it can read the CHSTATUS register of the channel and the
STATUSx_[n] to determine which slave address caused the error.
If the interrupts WEMSK and REMSK are set, then a NACK on slave address or data (in a
write cycle) will not terminate the serial transfer, the error will be stored in the
STATUSx_[n] register and the serial transfer will continue with the next transaction in the
sequence. Once all transactions are completed, the controller will generate a STOP and
the Sequence Done bit (SD) and other error bits (WE or RE) will be set in the CHSTATUS
and an interrupt will be generated.
If the host wants to poll the PCU9669, it can mask all registers including the SD bit and
read the CTRLSTATUS, CHSTATUS, STATUSx_[n], and/or the CONTROL registers to
determine the state of the controller.
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
36 of 69
xxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxx x x x xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxx xx xx xxxxx
xxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxx xxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxx x x
xxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxx
xxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxx
a.) Transactions with WEMSK and REMSK = 0
S
SLA 0
W
A
DATA
A
S
SLA 0
R
A
DATA
P
A
DATA
S
SLA 1
W
A
DATAn
P
A
S
P
S
SLA n
W
A
DATAn
P
A
P
A
P
P
P
00h
80h
A
A
A
A
A
A
A
data available
to be read on
parallel bus
01h 01h
08h
04h
10h
F8h
04h
08h
04h
C0h
20h
20h
10h
A
20h
A
20h
A
20h
A
20h
A
b.) Transactions with WEMSK and REMSK = 1
S
SLA 0
W
A
DATA
A
S
SLA 0
R
DATA
A
DATA
S
SLA 1
W
DATAn
S
S
SLA n
W
DATAn
P
A
01h 01h
02h
02h
02h
00h
A
A
A
A
A
02h
80h
E0h
C0h
A0h
data available
to be read on
parallel bus
08h
04h
10h
08h
08h
20h
20h
10h
20h
20h
from master to slave
from slave to master
STATUSx_[n] register, no interrupt
CHSTATUS register, interrupt requested; interrupt goes LOW at the STOP
n
any number of data bytes and their associated Acknowledge bits
last byte is NACK
DATA
A
A
002aaf619
Example CHSTATUS codes:
80h: sequence done with no errors
C0h: frame loop and sequence done with no errors
A0h: sequence done with a write error
D0h: frame loop and sequence done with a read error
Fig 9. PCU9669 I2C status codes
xxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxx x x x xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxx xx xx xxxxx
xxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxx xxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxx x x
xxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxx
xxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxx
number of
slave addresses
to be included
in a sequence
TRANCONFIG
DATA
internal memory pointer
0000h
SLATABLE
-
40h Transaction count
10h Transaction 0, data byte 0
00h Transaction 1, data byte 0
00h Transaction 1, data byte 1
00h Transaction 1, data byte 2
00h Transaction 1, data byte 3
00h Transaction 1, data byte 4
02h Transaction 2, data byte 0
55h Transaction 2, data byte 1
00h 10h
SLAW
SLAR
SLAW
SLAW
:
00h 01h Transaction 0 length, 1 byte
01h 05h Transaction 1 length, 5 bytes
02h 10h Transaction 2 length, 16 bytes
03h 08h Transaction 3 length, 8 bytes
01h 11h
02h 40h
03h E0h
:
:
:
:
:
3Dh 20h
3Eh 33h
3Fh 20h
SLAW
SLAR
SLAW
3Dh 10h Transaction 61 length, 16 bytes
3Eh 05h Transaction 62 length, 5 bytes
3Fh 08h Transaction 63 length, 8 bytes
sequence read
and write data
memory space
:
:
AAh Transaction 2, data byte 15
:
:
slave address
plus direction bit
transaction length
corresponding to each
slave address in the
SLATABLE
44h Transaction 63, data byte 0
AAh Transaction 63, data byte 1
:
:
55h Transaction 63, data byte 7
The slave address plus transaction count,
direction bit, the transaction length and the
transaction data make up one complete
serial bus transaction or sequence.
:
:
:
:
:
:
:
:
unused
memory
space
internal memory pointer
A00h or F00h
The memory pointers are managed
internally by the buffer controller.
002aaf620
Status and configuration registers are not shown.
Shaded areas are comments/indexes that are not user-accessible.
Fig 10. PCU9669 sequence block diagram; sample sequence loaded
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
8.2 Read transactions (Fm+ channel only)
Many I2C-bus slave devices need a command or register offset to setup a read operation.
In this case, a read transaction is actually a multi-part transaction consisting of a write
transaction followed by a read transaction. This is done by setting the transactions in that
order when programming the sequence.
If no write is required prior to a read, then the read transaction can be placed in any
location of the sequence. Once the read transaction is completed (i.e., the TR bit is
cleared to 0) the data is immediately available for the host to retrieve it on the parallel bus.
8.3 Stopping a sequence
If the host needs to stop the execution of a sequence, it should set the STO bit in the
CONTROL register. For write transactions, the host will issue a STOP after the
acknowledge cycle of the current byte being transferred on the serial bus. For read
transactions, if the host sets the STO bit while an address + read bit (SLA+R) is sent, the
controller will complete the read of one byte by sending 9 clocks and a NACK on the ninth
clock before sending the STOP condition. If the host sets the STO bit while a read
transaction is in progress, the current byte will be NACKed before sending a STOP
condition. No interrupts will be generated and all the status registers will be up to date.
The Sequence Done bit (SD) will be set to indicate to the host that the STOP condition
was completed and the bus is idle. The Sequence Done and the Frame Loop Done will be
set if the channel is in Loop mode (FRAMECNT 1) and a STO or STOSEQ bit is set.
If the host issues a STOP (by setting the STO) in the middle of a sequence followed by a
START (by setting the STA), then the controller will re-send the sequence from the
beginning, not from the point where the sequence was last stopped.
8.4 Looping a sequence
A sequence can be set to automatically loop several times using the FRAMECNT and one
of the following:
• The REFRATE register. The REFRATE register contains the value of the refresh rate
which is timing required between the START of two sequences. The refresh rate is
derived from the internal clock of the bus controller. If the REFRATE is programmed to
00h, the sequences will be looped back-to-back.
• Trigger enable (TE) bit. When TE is set, the refresh rate is controlled by the external
trigger input and the contents of the REFRATE registers is ignored. There is no
maximum timing requirement for the trigger interval.
The FRAMECNT register sets the number of times the sequence will be repeated. A
frame is defined as a sequence associated with its respective refresh rate. As described
above, the frame refresh rate is determined by the REFRATE register or an external
trigger source.
During looping, there is no host intervention required and all status and error reporting
remains active. The SD (Sequence Done) bit can be masked to avoid getting interrupted
each time a frame is completed while the other error reporting bits remain unmasked. In
this manner, normal transactions can run without host intervention and errors will be
reported at the STOP of the current byte where the error occurred.
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
39 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
Once the FRAMECNT values is reached, the FLD bit in the CHSTATUS register is set and
no further transactions will be executed and the channel will go to the idle state. The FLD
interrupt can be masked with the FLDMSK bit in the CTRLINTMSK register. The host can
poll the CTRLSTATUS register to check if the channel is active (looping) or if it is idle.
For indefinite or long term looping the host can do the following:
1. A sequence can be set to loop indefinitely by setting the FRAMECNT register to 00h.
Each frame will be sent out following the REFRATE settings or the Trigger input if the
TE bit is set. To end the Loop mode, the host sets the STO or STOSEQ bits in the
CONTROL register.
2. A frame will be sent out continuously and back-to-back if FRAMECNT and REFRATE
are set to 00h. To end the Loop mode, the hose sets the STO or STOSEQ bits in the
CONTROL register.
8.4.1 Looping with REFRATE control
When using the REFRATE register (TE bit is 0) the refresh rate timing is controlled
internally. Once the STA bit is set, the START command will be immediately sent on the
serial bus followed by the sequence. Thereafter, the controller will issue a START
command followed by the stored sequence every time the REFRATE value is reached. It
is important to program enough time in the REFRATE to allow a complete sequence to
reach the Sequence Done state. If the refresh rate is not long enough, the Frame Error
(FE) bit will be set and an interrupt will be generated. The FE bit is maskable, however,
masking the FE bit may yield undesired results on the serial interface. If the FE bit is
masked, the Loop mode will continue to operate and the FE flag will remain set. To exit
the Loop mode, the STO or the STOSEQ bit should be set.
8.4.2 Looping with Trigger control
The PCU9669 has one trigger input. The trigger enable (TE) bit in the CONTROL register
is used to control the use of external triggering. Once enabled, the trigger will override the
contents of the REFRATE register, and will start triggering when the STA bit is set.
Therefore, a significant time delay can occur between setting the STA bit and the
detection of a trigger. When a trigger edge is detected, the controller will issue a START
command and the stored sequence will be transferred on the serial bus. The trigger will
control the timing of the frame, therefore, enough time should be allowed by the trigger to
allow the sequence to reach the Sequence Done state.
If a trigger edge is detected while a sequence is actively being transmitted on the bus, the
Frame Error (FE) bit will be set and an interrupt will be generated. The FE bit is maskable,
however, masking the FE bit may yield undesired results on the serial interface. If the FE
bit is masked, the Loop mode will continue to operate and the FE flag will remain set. The
polarity of the trigger edge detect is controlled by the TP bit in the CONTROL register. To
exit the Trigger mode, the STO or the STOSEQ bit should be set.
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
40 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
8.5 Bus errors (Fm+ channel only)
Bus errors are a rare occurrence in a well designed I2C-bus system. The PCU9669 has a
robust error detection mechanism that detects hang-ups such as if SDA or SCL is pulled
LOW by an external source, or if an illegal START or STOP condition appears on the bus.
8.5.1 I2C-bus obstructed by a LOW level on SDA (DAE)
An I2C-bus hang-up occurs if SDA is pulled LOW by an uncontrolled source (e.g., a slave
device out of bit synchronization). If the SDA line is obstructed by another device on the
bus, the problem can be solved by transmitting additional clock pulses on the SCL line
(see Figure 11). The SDA stuck fault detection is only active during a START or
repeated-START condition.
When the error is detected, if the auto-recovery bit is set (AR = 1), the PCU9669 sends
out nine clock pulses followed by the STOP condition (see Figure 11). If the SDA line is
released by the slave pulling it LOW, a normal START condition is transmitted by the
PCU9669, the TA bit is set in the STATUSx_[n] register and the serial transfer continues. If
the SDA line is not released by the slave pulling it LOW, then the PCU9669 concludes that
there is a bus error, sets the DAE bit in the CHSTATUS register, generates an interrupt
signal, and releases the SCL and SDA lines.
If the auto-recovery bit is reset (AR = 0) during error detection, the PCU9669 loads the
bus error (sets the DAE bit in the CHSTATUS register), generates an interrupt signal, and
releases the SCL and SDA lines. After the host reads the status register, it can force a bus
recovery sequence by setting the bus recovery bit to 1 (BR = 1). The PCU9669 will
transmit additional clock pulses on the SCL line and the host must re-start the
transmission by setting the STA bit.
If a repeated START condition is transmitted while SDA is obstructed (pulled LOW), the
PCU9669 performs the same action as described above. In each case, the TA bit is set
after a successful START condition is transmitted and normal serial transfer continues.
Note that the host is not involved in solving these bus hang-up problems when the
auto-recovery bit is set (AR = 1).
When a host is unable to recover the bus by having the AR bit set or forcing a bus
recovery sequence by setting the bus recovery by setting the BR, then it may be
necessary to reset the slaves or the system.
Remark: If the AR bit is set and an SDA stuck LOW is detected, the transaction will
continue normally after an auto-recovery from the failed location in the sequence. If the
AR bit is zero and a manual bus recovery is performed, the transaction will be re-started
from the beginning of the sequence.
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
41 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
line released by slave,
bus recovered
line driven
by master
S
P
SDA line
line held LOW by slave
1
2
3
4
5
6
7
8
9
SCL line
START
condition
9 clocks driven by master
fault detected
at START
STOP condition
002aaf621
Fig 11. Recovering from a bus obstruction caused by a LOW level on SDA (AR = 1)
8.5.2 I2C-bus obstructed by a LOW level on SCL (CLE)
An I2C-bus hang-up occurs if SDA or SCL is pulled LOW by an uncontrolled source. If the
SCL line is obstructed (pulled LOW) by a device on the bus, no further serial transfer is
possible, and the PCU9669 cannot resolve this type of problem. When this occurs, the
problem must be resolved by the device that is pulling the SCL bus line LOW. To resolve
this type of a problem, resetting the slaves or the system may be required.
When the SCL line stays LOW for a period equal to the time-out value, the PCU9669
concludes that this is a bus error and behaves in a manner described in Section 7.5.1.15
“TIMEOUT — Time-out register”.
The bus recovery function (setting the BR bit) will not have any effect on an SCL stuck
LOW error.
8.5.3 Illegal START or STOP (SSE)
The illegal START or STOP detection is active immediately after the CTRLRDY register is
set to 00h at device start-up. The SSE condition will be monitored and detected at any
time the bus controller is not the one initiating the transition.
An SSE occurs when a START or STOP condition is present at an illegal position.
Examples of illegal positions are during the serial transfer of an address byte, a data or an
acknowledge bit.
When an SSE condition is detected, the PCU9669 releases the SDA and SCL lines, sets
the interrupt flag, and sets the SSE bit in the channel status register (CHSTATUS).
8.6 Power-on reset
When power is applied to VDD, an internal Power-On Reset holds the PCU9669 in a reset
condition until VDD has reached VPOR. At this point, the reset condition is released and the
PCU9669 goes to the power-up initialization phase where the following operations are
performed:
1. The oscillator and PLL will be re-initialized.
2. Internal register initialization is performed.
3. The memory space will be zeroed out.
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
42 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
The complete power-up initialization phase takes trst to be performed. During this time,
writes to the PCU9669 through the parallel port are ignored. However, the parallel port
can be read. This allows the device connected to the parallel port of the PCU9669 to poll
the CTRLRDY register.
8.7 Global reset
Reset of the PCU9669 to its default state can be performed in 2 different ways:
• By holding the RESET pin LOW for a minimum of tw(rst)
.
• By using the Parallel Software Reset sequence as described in Figure 12. The host
must write to the CTRLPRESET register of the target channel in two successive
parallel bus writes to the bus controller. The first byte is A5h and the second byte is
5Ah.
CTRLPRESET register selected
A[7:0]
D[7:0]
WR
data byte 1
A5h
data byte 2
5Ah
If D[7:0] ≠ A5h, following byte
is ignored and reset is aborted.
If D[7:0] ≠ 5Ah, reset is aborted.
If Data 1 = A5h and Data 2 = 5Ah,
PCU9669 is reset to its default state.
CE
internal
global reset
signal
002aaf622
Fig 12. Parallel Software Reset sequence
The RESET hardware pin and the global software reset function behave the same as the
power-on reset. A complete power-up initialization phase will be performed as defined in
Section 8.6. The RESET pin has an internal pull-up resistor (through a series diode) to
guarantee proper operation of the device. This pin should not be left floating and should
always be driven.
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
43 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
8.8 Channel reset
In addition to the above chip reset options, each channel can be individually reset by
programming the PRESET register for that channel as described in Figure 13. The
channel will reset to its default power-up state. The host must write to the PRESET
register of the target channel in two successive parallel bus writes to the bus controller.
The first byte is A5h and the second byte is 5Ah.
channel PRESET register selected
A[7:0]
D[7:0]
WR
data byte 1
A5h
data byte 2
5Ah
If D[7:0] ≠ A5h, following byte
is ignored and reset is aborted.
If D[7:0] ≠ 5Ah, reset is aborted.
If Data 1 = A5h and Data 2 = 5Ah,
PCU9669 is reset to its default state.
CE
internal
channel reset
signal
002aaf623
Fig 13. I2C-bus Channel Parallel Software Reset sequence
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
44 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
8.9 I2C-bus timing diagrams
The diagrams Figure 14 and Figure 15 illustrate typical timing diagrams for the PCU9669.
SCL
SDA
INT
(1)
7-bit address
interrupt
(after STOP)
(1)
(1)
first byte
n byte
R/W = 0
STOP
condition
START
condition
ACK
ACK
ACK
002aaf301
from slave receiver
PCU9669 writes data to slave.
(1) 7-bit address + R/W = 0 byte and number of bytes sent = value programmed in Transaction length
register in TRANCONFIG register.
Fig 14. Bus timing diagram; write transactions
SCL
SDA
INT
interrupt
(after STOP)
7-bit address
R/W = 1
(1)
(1)
first byte
n byte
STOP
START
condition
ACK
ACK
no ACK
from PCU9669
condition
002aaf624
from slave
PCU9669 reads data from slave.
(1) Number of bytes received = value programmed in the Transaction length register in
TRANCONFIG.
Fig 15. Bus timing diagram; read transactions (does not apply to the UFm channel)
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
45 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
9. Characteristics of the I2C-bus
The I2C-bus is for 2-way, 2-line communication between different ICs or modules. The two
lines are a serial data line (SDA) and a serial clock line (SCL). Both lines must be
connected to a positive supply via a pull-up resistor when connected to the output stages
of a device. Data transfer may be initiated only when the bus is not busy.
9.1 Bit transfer
One data bit is transferred during each clock pulse. The data on the SDA line must remain
stable during the HIGH period of the clock pulse as changes in the data line at this time
will be interpreted as control signals (see Figure 16).
SDA
SCL
data line
stable;
data valid
change
of data
allowed
mba607
Fig 16. Bit transfer
9.1.1 START and STOP conditions
Both data and clock lines remain HIGH when the bus is not busy. A HIGH-to-LOW
transition of the data line while the clock is HIGH is defined as the START condition (S). A
LOW-to-HIGH transition of the data line while the clock is HIGH is defined as the STOP
condition (P) (see Figure 17).
SDA
SCL
S
P
STOP condition
START condition
mba608
Fig 17. Definition of START and STOP conditions
9.2 System configuration
A device generating a message is a ‘transmitter’; a device receiving is the ‘receiver’. The
device that controls the message is the ‘master’ and the devices which are controlled by
the master are the ‘slaves’ (see Figure 18).
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
46 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
SDA
SCL
SLAVE
TRANSMITTER/
RECEIVER
SLAVE
TRANSMITTER/
RECEIVER
PCU9669
MASTER
TRANSMITTER/
RECEIVER
2
SLAVE
RECEIVER
SLAVE
RECEIVER
I C-BUS
MULTIPLEXER
SLAVE
TRANSMITTER/
RECEIVER
002aaf625
Fig 18. System configuration
9.3 Acknowledge
The number of data bytes transferred between the START and the STOP conditions from
transmitter to receiver is not limited. Each byte of eight bits is followed by one
acknowledge bit. The acknowledge bit is a HIGH level put on the bus by the transmitter,
whereas the master generates an extra acknowledge related clock pulse.
A slave receiver which is addressed must generate an acknowledge after the reception of
each byte. Also a master must generate an acknowledge after the reception of each byte
that has been clocked out of the slave transmitter. The device that acknowledges has to
pull down the SDA line during the acknowledge clock pulse, so that the SDA line is stable
LOW during the HIGH period of the acknowledge related clock pulse; set-up and hold
times must be taken into account.
A master receiver must signal an end of data to the transmitter by not generating an
acknowledge on the last byte that has been clocked out of the slave. In this event, the
transmitter must leave the data line HIGH to enable the master to generate a STOP
condition.
data output
by transmitter
not acknowledge
data output
by receiver
acknowledge
SCL from master
1
2
8
9
S
clock pulse for
START
condition
acknowledgement
002aaa987
Fig 19. Acknowledgement on the I2C-bus
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
47 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
10. Characteristics of the I2C-bus — Ultra Fast-mode (UFm)
The PCU9669 UFm bus is a 2-wire push-pull serial bus that operates from 50 kHz to
5 MHz transmitting data in one direction. The UFm protocol is based on the I2C-bus
protocol that consists of a START, slave address, command bit, ninth clock, and a STOP
bit. The command bit is a ‘write’ only, and the data bit on the ninth clock is driven HIGH,
ignoring the ACK cycle due to the unidirectional nature of the bus. The 2-wire pull-pull
drivers consists of a UFm clock (USCL) and data (USDA), requiring external series
resistors to allow proper line termination. The UFm bus is designed to be used in high
performance single master multi-drop applications.
V
DD(IO)
USCL or
USDA pin
R , external
s
V
SS
002aaf143
Fig 20. Simplified schematic of USCL, USDA outputs
The external resistors are chosen based upon the characteristic impedance of the UFm
bus. For example, if the characteristic input impedance of the line is 175 , a series
resistance of 175 can be used. Since the output resistance of the driver is
approximately 50 , the value of the series resistance used would then be 125 . The
final value of the resistance also depends upon the electrical length of the bus and the
signal settling time required to meet the UFm timing characteristics. Larger values result
in longer time for the signal to settle to its final valid value. Lower values can result in
overshoot and ringing on the bus. Careful consideration must be made in designing the
I2C-bus routing and selecting the series resistance.
10.1 Bit transfer
One data bit is transferred during each clock pulse. The data on the USDA line must
remain stable during the HIGH period of the clock pulse as changes in the data line at this
time will be interpreted as control signals (see Figure 21).
USDA
SDADLY
USCL
change
data line
of data
allowed
stable
SCLPER
002aaf626
Fig 21. UFm I2C-bus bit transfer
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
48 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
10.2 START and STOP conditions
Both data and clock lines remain HIGH when the bus is not busy. A HIGH-to-LOW
transition of the data line while the clock is HIGH is defined as the START condition (S).
A LOW-to-HIGH transition of the data line while the clock is HIGH is defined as the STOP
condition (P) (see Figure 22).
USDA
USCL
S
P
STOP condition
START condition
002aaf145
Fig 22. Definition of START and STOP conditions for UFm I2C-bus
10.3 Acknowledge (ninth clock)
The UFm bus functions as a transmitter only (unidirectional). The number of data bytes
transferred between the START and the STOP conditions from transmitter to receiver is
not limited. Each byte of eight bits of real data is followed by a dummy bit which is a HIGH
level put on the bus by the transmitter, which also generates an associated clock pulse.
Since the UFm bus is unidirectional, a slave receiver shall not generate an acknowledge
pulse. The slave USCLn and USDAn pins are input only.
data output
by transmitter
Master drives the line HIGH on 9th clock cycle.
Slave never drives the USDA line.
SCL from master
1
2
8
9
clock pulse for
acknowledgement
START
condition
002aaf144
Fig 23. Acknowledgement on the UFm I2C-bus
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
49 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
11. JTAG port
The PCU9669 has a JTAG IEEE 1149.1 compliant port. All signals (TDI, TMS, TCK, TRST
and TDO) are accessible. Only EXTEST functions are enabled, for example to conduct
board-level continuity tests. Device debug/emulation functionality such as INTEST
commands are not supported. The JTAG port is used for boundary scan testing (i.e.,
opens/shorts) during PCB manufacturing.
The following EXTEST JTAG instructions are supported:
• BYPASS
• EXTEST
• IDCODE
• SAMPLE
• PRELOAD
• CLAMP
• HIGHZ
If the JTAG boundary scan is not being used, then the JTAG pins must be held in the
following states:
• TDI, TCK, TMS: VDD
• TRST: VSS
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
50 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
12. Application design-in information
V
DD
address bus
V
DD(IO)
V
V
DD(IO)
DD
A0
A1
A2
A3
A4
A5
A6
A7
SCL0
SDA0
80C51
PCU9669
SLAVE
INT
SLAVE
RESET
DECODER
ALE
CE
USCL1
USDA1
8
D0 to D7
SLAVE
SLAVE
RD
WR
TRIG
USCL2
USDA2
V
DD
V
DD
INT
RESET
V
SS
V
SS
002aaf481
Fig 24. Application diagram using the 80C51
12.1 Specific applications
The PCU9669 is a parallel bus to I2C-bus controller that is designed to allow ‘smart’
devices to interface with I2C-bus or SMBus components, where the ‘smart’ device does
not have an integrated I2C-bus port and the designer does not want to ‘bit-bang’ the
I2C-bus port. The PCU9669 can also be used to add more I2C-bus ports to ‘smart’
devices, provide a higher frequency, lower voltage migration path for the PCF8584,
PCA9564 and PCA9665 and convert 8 bits of parallel data to a serial bus to avoid running
multiple traces across the printed-circuit board.
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
51 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
12.2 Add I2C-bus port
As shown in Figure 25, the PCU9669 converts 8-bits of parallel data into a single master
capable I2C-bus port for microcontrollers, microprocessors, custom ASICs, DSPs, etc.,
that need to interface with I2C-bus or SMBus components.
SDA0
SCL0
control signals
MICROCONTROLLER,
MICROPROCESSOR,
OR ASIC
USDA1
USCL1
PCU9669
8 bits data
USDA2
USCL2
002aaf482
Fig 25. Adding I2C-bus port application
12.3 Add additional I2C-bus ports
The PCU9669 can be used to convert 8-bit parallel data into additional single master
capable I2C-bus port as shown in Figure 26. It is used if the microcontroller,
microprocessor, custom ASIC, DSP, etc., already have an I2C-bus port but need one or
more additional I2C-bus ports to interface with more I2C-bus or SMBus components or
components that cannot be located on the same bus (e.g., 100 kHz and 400 kHz slaves
on different buses so that each bus can operate at its maximum potential).
SDA0
control signals
8 bits data
SLAVE
SLAVE
SCL0
PCU9669
MICROCONTROLLER,
MICROPROCESSOR,
OR ASIC
USDA1
USCL1
MASTER
USDA2
USCL2
SLAVE
002aaf483
2
I C-bus
Fig 26. Adding additional I2C-bus ports application
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
52 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
13. Limiting values
Table 37. Limiting values
In accordance with the Absolute Maximum Rating System (IEC 60134).
Symbol Parameter
Conditions
Min
0.3
0.3
Max
+4.6
+7.0
Unit
V
VDD
supply voltage
VDD(IO)
input/output supply voltage
power supply reference
for I2C-bus I/O pins
V
VI
input voltage
parallel bus interface
I2C-bus pins
any input
0.3
0.3
10
10
-
+4.6
+7.0
+10
+10
106
110
300
50
V
[1]
V
II
input current
mA
mA
mA
mA
mW
mW
C
IO
output current
any output
IOSH
IOSL
Ptot
P/out
Tstg
Tamb
HIGH-level short-circuit output current
LOW-level short-circuit output current
total power dissipation
power dissipation per output
storage temperature
I/O D0 to D7
I/O D0 to D7
-
-
-
65
40
+150
+85
ambient temperature
operating
C
[1] 5.5 V steady state voltage tolerance on inputs and outputs is valid only when the supply voltage is present. 4.6 V steady state voltage
tolerance on inputs and outputs when no supply voltage is present.
14. Static characteristics
Table 38. Static characteristics
VDD = 3.0 V to 3.6 V; Tamb = 40 C to +85 C; unless otherwise specified.
Symbol
Supply
VDD
Parameter
Conditions
Min
Typ
Max
Unit
supply voltage
monotonic supply during power-up
and power-down with a ramp time
(tramp): 5 s < tr < 20 ms
3.0
-
3.6
V
(5 % VDD(min) to 95 % VDD(min)
)
VDD(PLL)
VDD(IO)
phase-locked loop
supply voltage
input/output supply voltage power supply reference for I2C-bus
I/O pins
power supply for PLL bias circuit
3.0
3.0
-
-
3.6
5.5
V
V
IDD
supply current
operating mode; no load
-
-
15
-
25
1
mA
mA
IDD(IO)
input/output supply current VDD(IO) = 5.5 V; VDD = 3.6 V;
I/O not switching
VPOR
power-on reset voltage
LOW to HIGH
HIGH to LOW
-
-
2.75
2.60
-
-
V
V
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
53 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
Table 38. Static characteristics …continued
VDD = 3.0 V to 3.6 V; Tamb = 40 C to +85 C; unless otherwise specified.
Symbol
Parameter
Conditions
Min
Typ
Max
Unit
Inputs WR, RD, A0 to A7, CE, TRIG
VIL
VIH
Vhys
IL
LOW-level input voltage
HIGH-level input voltage
hysteresis voltage
0
-
0.3VDD
3.6
-
V
[1]
0.7VDD
0.1VDD
1
-
V
-
V
leakage current
input; VI = 0 V or 3.6 V
VI = VSS or VDD
-
+1
A
pF
Ci
input capacitance
-
2.0
4.5
Input RESET
VIL
VIH
Vhys
IL
LOW-level input voltage
0
-
0.3VDD
3.6
V
[1]
HIGH-level input voltage
hysteresis voltage
leakage current
0.7VDD
0.1VDD
1
-
V
-
-
V
input; VI = 0 V or 3.6 V
VI = VSS or VDD
-
+75
4.5
A
pF
Ci
input capacitance
-
2.0
Inputs/outputs D0 to D7
VIL
VIH
IOH
IOL
IL
LOW-level input voltage
0
-
0.3VDD
V
HIGH-level input voltage
HIGH-level output current
LOW-level output current
leakage current
0.7VDD
3.2
2.0
1
-
3.6
-
V
VOH = VDD(IO) 0.4 V
VOL = 0.4 V
-
mA
mA
A
pF
-
-
input; VI = 0 V or 5.5 V
VI = VSS or VDD
-
+1
5
Cio
input/output capacitance
-
2.8
USDAn and USCLn
IOL
IOH
Cio
RON
IL
LOW-level output current
VOL = 0.4 V
5
-
-
mA
mA
pF
HIGH-level output current
input/output capacitance
ON resistance
VOH = VDD(IO) 0.4 V
VI = VSS or VDD(IO)
4.8
-
-
-
5.6
50
-
7
-
-
leakage current
VDD = 3.6 V
VDD = 5.5 V
1
10
+1
+10
A
A
-
SDAn and SCLn
VIL
VIH
IL
LOW-level input voltage
0
-
0.3VDD(IO)
V
[1]
HIGH-level input voltage
leakage current
0.7VDD(IO)
-
5.5
+1
+1
-
V
input/output; VI = 0 V or 3.6 V
input/output; VI = 0 V or 5.5 V
VOL = 0.4 V
75
75
30
-
-
A
A
mA
pF
-
IOL
LOW-level output current
input/output capacitance
-
Cio
VI = VSS or VDD(IO)
5.6
7
Output INT
IOL
IL
LOW-level output current
leakage current
VOL = 0.4 V
6.0
1
-
-
-
mA
A
pF
VO = 0 V or 3.6 V
VI = VSS or VDD
-
+75
5.5
Co
output capacitance
3.8
[1] 5.5 V steady state voltage tolerance on inputs and outputs is valid only when the supply voltage is present. 4.6 V steady state voltage
tolerance on inputs and outputs when no supply voltage is present.
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
54 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
15. Dynamic characteristics
Table 39. Dynamic characteristics (3.3 volt)[1][2][3]
VDD = 3.3 V 0.3 V; Tamb = 40 C to +85 C; unless otherwise specified.
Symbol
Initialization timing
tinit(po) power-on initialization time
tinit initialization time
Parameter
Conditions
Min
Typ
Max
Unit
VDD 3.0 V
-
-
-
-
650
70
s
s
channel initialization time from
Channel Software Reset
controller initialization time from
POR, RESET, or Global Software
Reset inactive
-
-
650
s
RESET timing
tw(rst)
reset pulse width
4
-
-
-
-
s
s
[4][5]
trst
reset time
1.5
INT timing
tas(int)
interrupt assert time
-
-
-
-
500
100
ns
ns
tdas(int)
TRIG timing
tw(trig)
interrupt de-assert time
trigger pulse width
HIGH or LOW
100
-
-
ns
Bus timing (see Figure 27 and Figure 29)
tsu(A)
address set-up time
address hold time
to RD, WR LOW
from RD, WR LOW
to RD, WR LOW
from RD, WR LOW
0
-
-
-
-
-
-
-
-
-
-
-
-
-
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
th(A)
14
0
-
tsu(CE_N)
th(CE_N)
tw(RDL)
tw(WRL)
td(DV)
CE set-up time
-
CE hold time
0
-
RD LOW pulse width
WR LOW pulse width
data valid delay time
data output float delay time
data output set-up time
data output hold time
RD HIGH pulse width
WR HIGH pulse width
40
40
-
-
-
after RD and CE LOW
after RD or CE HIGH
before WR HIGH
45
7
-
td(QZ)
-
tsu(Q)
5
th(Q)
after WR HIGH
2
-
tw(RDH)
tw(WRH)
40
40
-
-
[1] Parameters are valid over specified temperature and voltage range.
[2] All voltage measurements are referenced to ground (VSS). For testing, all inputs swing between 0 V and 3.0 V with a transition time of
5 ns maximum. All time measurements are referenced at input voltages of 1.5 V and output voltages shown in Figure 27 and Figure 29.
[3] Test conditions for outputs: CL = 50 pF; RL = 500 , except open-drain outputs.
Test conditions for open-drain outputs: CL = 50 pF; RL = 1 k pull-up to VDD
.
[4] Resetting the device while actively communicating on the bus may cause glitches or an errant STOP condition.
[5] Upon reset, the full delay will be the sum of trst and the RC time constant of the SDA and SCL bus.
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
55 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
A0 to A7
t
su(A)
t
h(A)
CE
RD
t
t
h(CE_N)
t
w(RDH)
su(CE_N)
t
w(RDL)
t
d(QZ)
t
d(DV)
D0 to D7
(read)
float
not valid
valid
float
002aaf458
Fig 27. Bus timing (read cycle)
A0 to A7
t
su(A)
t
h(A)
CE
t
t
h(CE_N)
su(CE_N)
t
t
w(WRH)
w(WRL)
WR
t
h(Q)
t
su(Q)
D0 to D7
(write)
valid
002aaf459
Fig 28. Parallel bus timing (write cycle)
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
56 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
V
I
RD, CE input
V
M
V
M
V
SS
t
t
d(QZL)
d(QLZ)
V
DD
Dn output
LOW-to-float
float-to-LOW
V
M
V
X
V
OL
t
d(QZH)
t
d(QHZ)
V
OH
V
Y
Dn output
HIGH-to-float
float-to-HIGH
V
M
V
SS
outputs
enabled
outputs
enabled
outputs
floating
002aaf172
VM = 1.5 V
VX = VOL + 0.2 V
Y = VOH 0.2 V
V
VOL and VOH are typical output voltage drops that occur with the output load.
Fig 29. Data timing
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
57 of 69
xxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxx x x x xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxx xx xx xxxxx
xxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxx xxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxx x x
xxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxx
xxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxx
Table 40. I2C-bus frequency and timing specifications
All the timing limits are valid within the operating supply voltage and ambient temperature range; VDD = 2.5 V 0.2 V and 3.3 V 0.3 V; Tamb = 40 C to +85 C;
and refer to VIL and VIH with an input voltage of VSS to VDD
.
Symbol Parameter Conditions
Standard-mode Fast-mode I2C-bus
I2C-bus
Fast-mode Plus Ultra Fast-mode Unit
I2C-bus I2C-bus
Min
0
Max
100
-
Min
0
Max
400
-
Min
Max
1000
-
Min
Max
[1]
fSCL
tBUF
SCL clock frequency
0
0
5000 kHz
bus free time between a STOP and
START condition
4.7
1.3
0.5
0.08
-
s
tHD;STA
tSU;STA
hold time (repeated) START condition
4.0
4.7
-
-
0.6
0.6
-
-
0.26
0.26
-
-
0.05
0.05
-
-
s
s
set-up time for a repeated START
condition
tSU;STO
tHD;DAT
tVD;ACK
tVD;DAT
tSU;DAT
tLOW
tHIGH
tf
set-up time for STOP condition
data hold time
4.0
0
-
0.6
0
-
-
0.26
0
-
0.05
10
-
s
ns
s
ns
ns
s
s
ns
ns
ns
-
-
0.45
-
-
[2]
[4]
[3]
[3]
data valid acknowledge time
data valid time
0.1
100
100
4.7
4.0
-
3.45
0.1
0.9
-
0.1
100
100
0.5
0.26
-
-
-
-
100
10
30
-
-
-
-
data set-up time
-
100
-
-
LOW period of the SCL clock
HIGH period of the SCL clock
fall time of both SDA and SCL signals
rise time of both SDA and SCL signals
-
-
1.3
-
-
0.05
0.05
0.6
-
-
[5][6]
[9]
[7]
[7]
[8]
300
20 + 0.1Cb
300
300
50
120
120
50
-
50
[8]
tr
-
1000 20 + 0.1Cb
50
-
-
50
[10]
[10]
tSP
pulse width of spikes that must be
suppressed by the input filter
-
-
-
-
-
[1] Minimum SCL clock frequency is limited by the bus time-out feature, generates a CLE error if the SCL is held LOW for the TIMEOUT period.
[2] tVD;ACK = time for Acknowledgement signal from SCL LOW to SDA (out) LOW.
[3]
t
VD;ACK is not applicable to the Ultra Fast-mode (UFm) I2C-bus.
[4] tVD;DAT = minimum time for SDA data out to be valid following SCL LOW.
[5] A master device must internally provide a hold time of at least 300 ns for the SDA signal (refer to the VIL of the SCL signal) in order to bridge the undefined region SCL’s falling
edge. Does not apply to the UFm channel.
[6] The maximum tf for the SDA and SCL bus lines is specified at 300 ns. The maximum fall time for the SDA output stage tf is specified at 250 ns. This allows series protection
resistors to be connected between the SDAn and the SCLn pins and the SDA/SCL bus lines without exceeding the maximum specified tf. Does not apply to the UFm channel.
[7] Cb = total capacitance of one bus line in pF.
[8] Typical rise/fall times for UFm signals is 25 ns measured from the 20 % level to the 80 % (rise time) or from the 80 % level to the 20 % level (fall time).
[9] Input filters on the SDAn and SCLn inputs suppress noise spikes less than 50 ns.
[10] tSP is not applicable to the Ultra Fast-mode (UFm) I2C-bus.
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
SDA
t
BUF
t
t
LOW
t
t
SP
SU;DAT
HD;STA
t
f
t
r
t
r
t
f
SCL
t
t
t
SU;STO
HD;STA
SU;STA
t
HIGH
S
Sr
P
S
t
HD;DAT
002aab271
Fig 30. Definition of timing on the I2C-bus
START
bit 7
STOP
condition
(P)
acknowledge
(A)
protocol
condition
(S)
bit 6
bit n
bit 0
MSB
t
t
t
HIGH
SU;STA
LOW
1
/f
SCL
SCL
SDA
t
t
BUF
f
t
r
t
t
t
t
t
t
HD;DAT
VD;DAT
VD;ACK
SU;STO
HD;STA
SU;DAT
002aac696
Rise and fall times refer to VIL and VIH.
Fig 31. I2C-bus timing diagram
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
59 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
16. Test information
V
open
× 2
DD
V
SS
V
R
500 Ω
DD
L
V
V
O
I
PULSE
DUT
GENERATOR
R
L
500 Ω
C
50 pF
L
R
T
002aac694
Test data are given in Table 41.
RL = load resistance.
CL = load capacitance includes jig and probe capacitance.
RT = termination resistance should be equal to the output impedance ZO of the pulse generators.
Fig 32. Test circuitry for switching times
Table 41. Test data
Test
Conditions
Load
CL
S1
RL
td(DV), td(QZ)
Dn outputs active LOW
Dn outputs active HIGH
50 pF
50 pF
500
500
VDD 2
open
V
DD
open
V
SS
V
R
L
1 kΩ
DD
V
V
O
I
PULSE
DUT
GENERATOR
C
50 pF
L
R
T
002aac695
Test data are given in Table 42.
RL = load resistance.
CL = load capacitance includes jig and probe capacitance.
RT = termination resistance should be equal to the output impedance ZO of the pulse generators.
Fig 33. Test circuitry for open-drain switching times
Table 42. Test data INT pin
Test
Load
CL
S1
RL
tas(int)
50 pF
50 pF
1 k
1 k
VDD
VDD
tdas(int)
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
60 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
17. Package outline
LQFP48: plastic low profile quad flat package; 48 leads; body 7 x 7 x 1.4 mm
SOT313-2
c
y
X
36
25
A
E
37
24
Z
E
e
H
E
A
2
A
(A )
3
A
1
w M
p
θ
pin 1 index
b
L
p
L
13
48
detail X
1
12
Z
v M
D
A
e
w M
b
p
D
B
H
v
M
B
D
0
2.5
5 mm
scale
DIMENSIONS (mm are the original dimensions)
A
(1)
(1)
(1)
(1)
UNIT
A
A
A
b
c
D
E
e
H
D
H
L
L
v
w
y
Z
Z
E
θ
1
2
3
p
E
p
D
max.
7o
0o
0.20 1.45
0.05 1.35
0.27 0.18 7.1
0.17 0.12 6.9
7.1
6.9
9.15 9.15
8.85 8.85
0.75
0.45
0.95 0.95
0.55 0.55
1.6
mm
0.25
0.5
1
0.2 0.12 0.1
Note
1. Plastic or metal protrusions of 0.25 mm maximum per side are not included.
REFERENCES
OUTLINE
EUROPEAN
PROJECTION
ISSUE DATE
VERSION
IEC
JEDEC
JEITA
00-01-19
03-02-25
SOT313-2
136E05
MS-026
Fig 34. Package outline SOT313-2 (LQFP48)
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
61 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
18. Handling information
All input and output pins are protected against ElectroStatic Discharge (ESD) under
normal handling. When handling ensure that the appropriate precautions are taken as
described in JESD625-A or equivalent standards.
19. Soldering of SMD packages
This text provides a very brief insight into a complex technology. A more in-depth account
of soldering ICs can be found in Application Note AN10365 “Surface mount reflow
soldering description”.
19.1 Introduction to soldering
Soldering is one of the most common methods through which packages are attached to
Printed Circuit Boards (PCBs), to form electrical circuits. The soldered joint provides both
the mechanical and the electrical connection. There is no single soldering method that is
ideal for all IC packages. Wave soldering is often preferred when through-hole and
Surface Mount Devices (SMDs) are mixed on one printed wiring board; however, it is not
suitable for fine pitch SMDs. Reflow soldering is ideal for the small pitches and high
densities that come with increased miniaturization.
19.2 Wave and reflow soldering
Wave soldering is a joining technology in which the joints are made by solder coming from
a standing wave of liquid solder. The wave soldering process is suitable for the following:
• Through-hole components
• Leaded or leadless SMDs, which are glued to the surface of the printed circuit board
Not all SMDs can be wave soldered. Packages with solder balls, and some leadless
packages which have solder lands underneath the body, cannot be wave soldered. Also,
leaded SMDs with leads having a pitch smaller than ~0.6 mm cannot be wave soldered,
due to an increased probability of bridging.
The reflow soldering process involves applying solder paste to a board, followed by
component placement and exposure to a temperature profile. Leaded packages,
packages with solder balls, and leadless packages are all reflow solderable.
Key characteristics in both wave and reflow soldering are:
• Board specifications, including the board finish, solder masks and vias
• Package footprints, including solder thieves and orientation
• The moisture sensitivity level of the packages
• Package placement
• Inspection and repair
• Lead-free soldering versus SnPb soldering
19.3 Wave soldering
Key characteristics in wave soldering are:
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
62 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
• Process issues, such as application of adhesive and flux, clinching of leads, board
transport, the solder wave parameters, and the time during which components are
exposed to the wave
• Solder bath specifications, including temperature and impurities
19.4 Reflow soldering
Key characteristics in reflow soldering are:
• Lead-free versus SnPb soldering; note that a lead-free reflow process usually leads to
higher minimum peak temperatures (see Figure 35) than a SnPb process, thus
reducing the process window
• Solder paste printing issues including smearing, release, and adjusting the process
window for a mix of large and small components on one board
• Reflow temperature profile; this profile includes preheat, reflow (in which the board is
heated to the peak temperature) and cooling down. It is imperative that the peak
temperature is high enough for the solder to make reliable solder joints (a solder paste
characteristic). In addition, the peak temperature must be low enough that the
packages and/or boards are not damaged. The peak temperature of the package
depends on package thickness and volume and is classified in accordance with
Table 43 and 44
Table 43. SnPb eutectic process (from J-STD-020C)
Package thickness (mm) Package reflow temperature (C)
Volume (mm3)
< 350
235
350
220
< 2.5
2.5
220
220
Table 44. Lead-free process (from J-STD-020C)
Package thickness (mm) Package reflow temperature (C)
Volume (mm3)
< 350
260
350 to 2000
> 2000
260
< 1.6
260
250
245
1.6 to 2.5
> 2.5
260
245
250
245
Moisture sensitivity precautions, as indicated on the packing, must be respected at all
times.
Studies have shown that small packages reach higher temperatures during reflow
soldering, see Figure 35.
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
63 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
maximum peak temperature
= MSL limit, damage level
temperature
minimum peak temperature
= minimum soldering temperature
peak
temperature
time
001aac844
MSL: Moisture Sensitivity Level
Fig 35. Temperature profiles for large and small components
For further information on temperature profiles, refer to Application Note AN10365
“Surface mount reflow soldering description”.
20. Abbreviations
Table 45. Abbreviations
Acronym
ASIC
CDM
CPU
Description
Application Specific Integrated Circuit
Charged-Device Model
Central Processing Unit
Digital Signal Processor
ElectroStatic Discharge
Fast-mode Plus
DSP
ESD
Fm+
HBM
I2C-bus
I/O
Human Body Model
Inter-Integrated Circuit bus
Input/Output
LED
Light Emitting Diode
Phase-Locked Loop
PLL
SMBus
UFm
System Management Bus
Ultra Fast-mode
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
64 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
21. Revision history
Table 46. Revision history
Document ID
PCU9669 v.2
Modifications:
Release date
Data sheet status
Change notice
Supersedes
20110701
Product data sheet
-
PCU9669 v.1
• Table 40 “I2C-bus frequency and timing specifications”, Table note [8]: unit of measure is
corrected from “25 ms” to “25 ns”.
PCU9669 v.1
20110606
Product data sheet
-
-
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
65 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
22. Legal information
22.1 Data sheet status
Document status[1][2]
Product status[3]
Development
Definition
Objective [short] data sheet
This document contains data from the objective specification for product development.
This document contains data from the preliminary specification.
This document contains the product specification.
Preliminary [short] data sheet Qualification
Product [short] data sheet Production
[1]
[2]
[3]
Please consult the most recently issued document before initiating or completing a design.
The term ‘short data sheet’ is explained in section “Definitions”.
The product status of device(s) described in this document may have changed since this document was published and may differ in case of multiple devices. The latest product status
information is available on the Internet at URL http://www.nxp.com.
malfunction of an NXP Semiconductors product can reasonably be expected
22.2 Definitions
to result in personal injury, death or severe property or environmental
damage. NXP Semiconductors accepts no liability for inclusion and/or use of
NXP Semiconductors products in such equipment or applications and
therefore such inclusion and/or use is at the customer’s own risk.
Draft — The document is a draft version only. The content is still under
internal review and subject to formal approval, which may result in
modifications or additions. NXP Semiconductors does not give any
representations or warranties as to the accuracy or completeness of
information included herein and shall have no liability for the consequences of
use of such information.
Applications — Applications that are described herein for any of these
products are for illustrative purposes only. NXP Semiconductors makes no
representation or warranty that such applications will be suitable for the
specified use without further testing or modification.
Short data sheet — A short data sheet is an extract from a full data sheet
with the same product type number(s) and title. A short data sheet is intended
for quick reference only and should not be relied upon to contain detailed and
full information. For detailed and full information see the relevant full data
sheet, which is available on request via the local NXP Semiconductors sales
office. In case of any inconsistency or conflict with the short data sheet, the
full data sheet shall prevail.
Customers are responsible for the design and operation of their applications
and products using NXP Semiconductors products, and NXP Semiconductors
accepts no liability for any assistance with applications or customer product
design. It is customer’s sole responsibility to determine whether the NXP
Semiconductors product is suitable and fit for the customer’s applications and
products planned, as well as for the planned application and use of
customer’s third party customer(s). Customers should provide appropriate
design and operating safeguards to minimize the risks associated with their
applications and products.
Product specification — The information and data provided in a Product
data sheet shall define the specification of the product as agreed between
NXP Semiconductors and its customer, unless NXP Semiconductors and
customer have explicitly agreed otherwise in writing. In no event however,
shall an agreement be valid in which the NXP Semiconductors product is
deemed to offer functions and qualities beyond those described in the
Product data sheet.
NXP Semiconductors does not accept any liability related to any default,
damage, costs or problem which is based on any weakness or default in the
customer’s applications or products, or the application or use by customer’s
third party customer(s). Customer is responsible for doing all necessary
testing for the customer’s applications and products using NXP
Semiconductors products in order to avoid a default of the applications and
the products or of the application or use by customer’s third party
customer(s). NXP does not accept any liability in this respect.
22.3 Disclaimers
Limiting values — Stress above one or more limiting values (as defined in
the Absolute Maximum Ratings System of IEC 60134) will cause permanent
damage to the device. Limiting values are stress ratings only and (proper)
operation of the device at these or any other conditions above those given in
the Recommended operating conditions section (if present) or the
Characteristics sections of this document is not warranted. Constant or
repeated exposure to limiting values will permanently and irreversibly affect
the quality and reliability of the device.
Limited warranty and liability — Information in this document is believed to
be accurate and reliable. However, NXP Semiconductors does not give any
representations or warranties, expressed or implied, as to the accuracy or
completeness of such information and shall have no liability for the
consequences of use of such information.
In no event shall NXP Semiconductors be liable for any indirect, incidental,
punitive, special or consequential damages (including - without limitation - lost
profits, lost savings, business interruption, costs related to the removal or
replacement of any products or rework charges) whether or not such
damages are based on tort (including negligence), warranty, breach of
contract or any other legal theory.
Terms and conditions of commercial sale — NXP Semiconductors
products are sold subject to the general terms and conditions of commercial
sale, as published at http://www.nxp.com/profile/terms, unless otherwise
agreed in a valid written individual agreement. In case an individual
agreement is concluded only the terms and conditions of the respective
agreement shall apply. NXP Semiconductors hereby expressly objects to
applying the customer’s general terms and conditions with regard to the
purchase of NXP Semiconductors products by customer.
Notwithstanding any damages that customer might incur for any reason
whatsoever, NXP Semiconductors’ aggregate and cumulative liability towards
customer for the products described herein shall be limited in accordance
with the Terms and conditions of commercial sale of NXP Semiconductors.
Right to make changes — NXP Semiconductors reserves the right to make
changes to information published in this document, including without
limitation specifications and product descriptions, at any time and without
notice. This document supersedes and replaces all information supplied prior
to the publication hereof.
No offer to sell or license — Nothing in this document may be interpreted or
construed as an offer to sell products that is open for acceptance or the grant,
conveyance or implication of any license under any copyrights, patents or
other industrial or intellectual property rights.
Export control — This document as well as the item(s) described herein
may be subject to export control regulations. Export might require a prior
authorization from national authorities.
Suitability for use — NXP Semiconductors products are not designed,
authorized or warranted to be suitable for use in life support, life-critical or
safety-critical systems or equipment, nor in applications where failure or
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
66 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
Non-automotive qualified products — Unless this data sheet expressly
states that this specific NXP Semiconductors product is automotive qualified,
the product is not suitable for automotive use. It is neither qualified nor tested
in accordance with automotive testing or application requirements. NXP
Semiconductors accepts no liability for inclusion and/or use of
own risk, and (c) customer fully indemnifies NXP Semiconductors for any
liability, damages or failed product claims resulting from customer design and
use of the product for automotive applications beyond NXP Semiconductors’
standard warranty and NXP Semiconductors’ product specifications.
non-automotive qualified products in automotive equipment or applications.
22.4 Trademarks
Notice: All referenced brands, product names, service names and trademarks
are the property of their respective owners.
In the event that customer uses the product for design-in and use in
automotive applications to automotive specifications and standards, customer
(a) shall use the product without NXP Semiconductors’ warranty of the
product for such automotive applications, use and specifications, and (b)
whenever customer uses the product for automotive applications beyond
NXP Semiconductors’ specifications such use shall be solely at customer’s
I2C-bus — logo is a trademark of NXP B.V.
23. Contact information
For more information, please visit: http://www.nxp.com
For sales office addresses, please send an email to: salesaddresses@nxp.com
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
67 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
24. Contents
1
2
3
4
5
General description. . . . . . . . . . . . . . . . . . . . . . 1
7.5.2.5
CTRLRDY — Controller ready register . . . . . 33
Features and benefits . . . . . . . . . . . . . . . . . . . . 1
Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Ordering information. . . . . . . . . . . . . . . . . . . . . 2
Block diagram . . . . . . . . . . . . . . . . . . . . . . . . . . 3
8
8.1
8.2
8.3
PCU9669 operation . . . . . . . . . . . . . . . . . . . . . 34
Sequence execution . . . . . . . . . . . . . . . . . . . 35
Read transactions (Fm+ channel only) . . . . . 39
Stopping a sequence . . . . . . . . . . . . . . . . . . . 39
Looping a sequence. . . . . . . . . . . . . . . . . . . . 39
Looping with REFRATE control . . . . . . . . . . . 40
Looping with Trigger control. . . . . . . . . . . . . . 40
Bus errors (Fm+ channel only). . . . . . . . . . . . 41
I2C-bus obstructed by a LOW level on
8.4
6
6.1
6.2
Pinning information. . . . . . . . . . . . . . . . . . . . . . 4
Pinning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Pin description . . . . . . . . . . . . . . . . . . . . . . . . . 4
8.4.1
8.4.2
8.5
7
7.1
7.2
7.3
7.3.1
7.3.2
7.4
7.5
7.5.1
7.5.1.1
Functional description . . . . . . . . . . . . . . . . . . . 6
General. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Internal oscillator and PLL . . . . . . . . . . . . . . . . 6
Buffer description . . . . . . . . . . . . . . . . . . . . . . . 6
Buffer management assumptions. . . . . . . . . . . 7
Buffer sizes. . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Error reporting and handling. . . . . . . . . . . . . . . 8
Registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Channel registers . . . . . . . . . . . . . . . . . . . . . . 13
STATUS0_[n], STATUS1_[n],
STATUS2_[n] — Transaction status registers 13
CONTROL — Control register . . . . . . . . . . . . 14
CHSTATUS — Channel status register . . . . . 17
INTMSK — Interrupt mask register. . . . . . . . . 19
SLATABLE — Slave address table register . . 20
TRANCONFIG — Transaction configuration
register . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
DATA — I2C-bus Data register . . . . . . . . . . . . 21
TRANSEL — Transaction data buffer
8.5.1
SDA (DAE). . . . . . . . . . . . . . . . . . . . . . . . . . . 41
I2C-bus obstructed by a LOW level on
8.5.2
SCL (CLE) . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
Illegal START or STOP (SSE) . . . . . . . . . . . . 42
Power-on reset. . . . . . . . . . . . . . . . . . . . . . . . 42
Global reset . . . . . . . . . . . . . . . . . . . . . . . . . . 43
Channel reset. . . . . . . . . . . . . . . . . . . . . . . . . 44
I2C-bus timing diagrams. . . . . . . . . . . . . . . . . 45
Characteristics of the I2C-bus . . . . . . . . . . . . 46
Bit transfer . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
START and STOP conditions. . . . . . . . . . . . . 46
System configuration . . . . . . . . . . . . . . . . . . . 46
Acknowledge . . . . . . . . . . . . . . . . . . . . . . . . . 47
8.5.3
8.6
8.7
8.8
8.9
9
9.1
9.1.1
9.2
9.3
7.5.1.2
7.5.1.3
7.5.1.4
7.5.1.5
7.5.1.6
10
Characteristics of the I2C-bus — Ultra
Fast-mode (UFm). . . . . . . . . . . . . . . . . . . . . . . 48
Bit transfer . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
START and STOP conditions. . . . . . . . . . . . . 49
Acknowledge (ninth clock) . . . . . . . . . . . . . . . 49
10.1
10.2
10.3
7.5.1.7
7.5.1.8
select register . . . . . . . . . . . . . . . . . . . . . . . . . 22
TRANOFS — Transaction data buffer byte
11
JTAG port. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
7.5.1.9
12
Application design-in information. . . . . . . . . 51
Specific applications. . . . . . . . . . . . . . . . . . . . 51
Add I2C-bus port . . . . . . . . . . . . . . . . . . . . . . 52
Add additional I2C-bus ports . . . . . . . . . . . . . 52
select register . . . . . . . . . . . . . . . . . . . . . . . . . 23
7.5.1.10 BYTECOUNT — Transmitted and received
byte count register . . . . . . . . . . . . . . . . . . . . . 23
12.1
12.2
12.3
7.5.1.11 FRAMECNT — Frame count register. . . . . . . 23
7.5.1.12 REFRATE — Refresh rate register. . . . . . . . . 24
7.5.1.13 SCLL, SCLH and SCLPER, SDADLY —
Clock rate registers. . . . . . . . . . . . . . . . . . . . . 24
7.5.1.14 MODE — I2C-bus mode register . . . . . . . . . . 28
7.5.1.15 TIMEOUT — Time-out register. . . . . . . . . . . . 29
7.5.1.16 PRESET — I2C-bus channel parallel software
reset register. . . . . . . . . . . . . . . . . . . . . . . . . . 30
13
14
15
16
17
18
Limiting values . . . . . . . . . . . . . . . . . . . . . . . . 53
Static characteristics . . . . . . . . . . . . . . . . . . . 53
Dynamic characteristics. . . . . . . . . . . . . . . . . 55
Test information . . . . . . . . . . . . . . . . . . . . . . . 60
Package outline. . . . . . . . . . . . . . . . . . . . . . . . 61
Handling information . . . . . . . . . . . . . . . . . . . 62
7.5.2
7.5.2.1
7.5.2.2
Global registers . . . . . . . . . . . . . . . . . . . . . . . 30
CTRLSTATUS — Controller status register . . 30
CTRLINTMSK — Control Interrupt mask
register . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
DEVICE_ID — Device ID . . . . . . . . . . . . . . . . 32
CTRLPRESET — Parallel software reset
19
Soldering of SMD packages. . . . . . . . . . . . . . 62
Introduction to soldering. . . . . . . . . . . . . . . . . 62
Wave and reflow soldering. . . . . . . . . . . . . . . 62
Wave soldering . . . . . . . . . . . . . . . . . . . . . . . 62
Reflow soldering . . . . . . . . . . . . . . . . . . . . . . 63
19.1
19.2
19.3
19.4
7.5.2.3
7.5.2.4
20
Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . 64
register . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
continued >>
PCU9669
All information provided in this document is subject to legal disclaimers.
© NXP B.V. 2011. All rights reserved.
Product data sheet
Rev. 2 — 1 July 2011
68 of 69
PCU9669
NXP Semiconductors
Parallel bus to 1 channel Fm+ and 2 channel UFm I2C-bus controller
21
Revision history. . . . . . . . . . . . . . . . . . . . . . . . 65
Legal information. . . . . . . . . . . . . . . . . . . . . . . 66
22
22.1
22.2
22.3
22.4
Data sheet status . . . . . . . . . . . . . . . . . . . . . . 66
Definitions. . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
Disclaimers. . . . . . . . . . . . . . . . . . . . . . . . . . . 66
Trademarks. . . . . . . . . . . . . . . . . . . . . . . . . . . 67
23
24
Contact information. . . . . . . . . . . . . . . . . . . . . 67
Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
Please be aware that important notices concerning this document and the product(s)
described herein, have been included in section ‘Legal information’.
© NXP B.V. 2011.
All rights reserved.
For more information, please visit: http://www.nxp.com
For sales office addresses, please send an email to: salesaddresses@nxp.com
Date of release: 1 July 2011
Document identifier: PCU9669
相关型号:
PCU9956ATWY
PCU9956A - 24-channel UFm 5 MHz I2C-bus 57 mA/20 V constant current LED driver TSSOP 38-Pin
NXP
©2020 ICPDF网 联系我们和版权申明