z Digital Gain and Offset Corrections
z Test DAC Bit-stream Generator
Digital Sine Wave Output
z Time Break Controller, General Purpose I/O
z Secondary SPI™ Port, Boundary Scan JTAG
z Microcontroller or EEPROM Configuration
z Small-footprint, 64-pin TQFP Package
z Low Power Consumption
9 mW per Channel at 500 SPS
z Flexible Power Supplies
I/O Interface: 3.3 V or 5.0 V
Digital Logic Core: 3.0 V, 3.3 V or 5.0 V
I
Description
The CS5376A is a multi-function digital filter utilizing a
low-power signal processing architecture to achieve efficient filtering for up to four ∆Σ modulators. By
combining the CS5376A with CS3301A/02A differential
amplifiers, CS5371A/72A ∆Σ modulators, and the
CS4373A ∆Σ test DAC a synchronous, high-resolution,
self-testing, multi-channel me as ur em e nt s yst em can be
designed quickly and easily.
Digital filter coefficients for the CS5376A FIR and IIR filters are included on-chip for a simple setup, or they can
be programmed for custom applications. Selectable digital filter decimation ratios produce output word rates
from 4000 SPS to 1 SPS, resulting in measurement
bandwidths ranging from 1600 Hz down to 400 mHz
when using the on-chip coefficient sets.
The CS5376A includes integrated peripherals to simplify
system design: offset and gain corrections, a test DAC
bit stream generator, a time-break controller, 12 general-purpose I/O pins, a secondary SPI port, and a
boundary scan JTAG port.
SPI1CTRL00 - 02R/W8, 8, 8SPI 1 Control
SPI1CMD03 - 05R/W8, 8, 8SPI 1 Command
SPI1DAT106 - 08R/W8, 8, 8SPI 1 Data 1
SPI1DAT209 - 0BR/W8, 8, 8SPI 1 Data 2
Digital Filter Registers
NameAddr.Type# BitsDescription
CONFIG00R/W24Hardware Configuration
RESERVED01-0DR/W24Reserved
GPCFG00ER/W24GPIO[7:0] Direction, Pull-up Enable, and Data
GPCFG10FR/W24GPIO[11:8] Direction, Pull-up Enable, and Data
SPI2CTRL10R/W24SPI 2 Control
SPI2CMD11R/W16SPI 2 Command
SPI2DAT12R/W24SPI 2 Data
RESERVED13-1FR/W24Reserved
FILTCFG20R/W24Digital Filter Configuration
GAIN121R/W24Gain Correction Channel 1
GAIN222R/W24Gain Correction Channel 2
GAIN323R/W24Gain Correction Channel 3
GAIN424R/W24Gain Correction Channel 4
OFFSET125R/W24Offset Correction Channel 1
OFFSET226R/W24Offset Correction Channel 2
OFFSET327R/W24Offset Correction Channel 3
OFFSET428R/W24Offset Correction Channel 4
TIMEBRK29R/W24Time Break Delay
TBSCFG2AR/W24Test Bit Stream Configuration
TBSGAIN2BR/W24Test Bit Stream Gain
SYSTEM12CR/W24User Defined System Register 1
SYSTEM22DR/W24User Defined System Register 2
VERSION2ER/W24Hardware Version ID
SELFTEST2FR/W24Self-Test Result Code
CS5376A
T able 3. SPI 1 and Digital Filter Registers
DS612F412
CS5376A
2. CHARACTERISTICS AND SPECIFICATIONS
•Min / Max characteristics and specifications are guaranteed over the Specified Operating Conditions.
•Typical performance characteristics and specifications are derived from measurements taken at nominal supply voltages and TA = 25°C.
•GND, GND1, GND2 = 0 V, all voltages with respect to 0 V.
SPECIFIED OPERATING CONDITIONS
ParameterSymbol Min NomMaxUnit
Logic Core Power SupplyVD2.853.05.25V
Microcontroller Interface Power SupplyVDD13.1353.35.25V
Modulator Interface Power SupplyVDD23.1353.35.25V
Ambient Operating TemperatureIndustrial (-IQ)T
A
-40-85°C
ABSOLUTE MAXIMUM RATINGS
ParameterSymbolMinMaxUnits
DC Power SuppliesLogic Core
Microcontroller Interface
Modulator Interface
Input Current, Any Pin Except Supplies(Note 1)I
Input Current, Power Supplies(Note 1)I
Output Current(Note 1)I
Power DissipationP
Digital Input VoltagesV
Ambient Operating Temperature (Power Applied)T
Storage Temperature RangeT
1. Transient currents up to 100 mA will not cause SCR latch-up.
VDD1
VDD2
VD
IN
IN
OUT
DN
IND
A
STG
-0.3
-0.3
-0.3
-±10mA
-±50mA
-±25mA
-500mW
-0.5VDD+0.5V
-4085°C
-65150°C
6.0
6.0
6.0
V
V
V
DS612F413
THERMAL CHARACTERISTICS
V
ParameterSymbol Min TypMaxUnit
Allowable Junction TemperatureT
Junction to Ambient Thermal ImpedanceΘ
Ambient Operating Temperature (Power Applied)T
Low-Level Output Drive VoltageI
Rise Times, Digital Inputst
Fall Times, Digital Inputst
Rise Times, Digital Outputst
Fall Times, Digital Outputst
Input Leakage Current(Note 2)I
3-State Leakage CurrentI
Digital Input CapacitanceC
Digital Output Pin CapacitanceC
= -40 µAV
out
= +40 µAV
out
J
JA
A
IH
IL
OH
OL
RISE
FALL
RISE
FALL
IN
OZ
IN
OUT
CS5376A
--135°C
-65°C / W
-40-+85°C
0.6 * VDD-VDDV
0.0-0.8V
VDD - 0.3-VDDV
0.0-0.3V
--100ns
--100ns
--100ns
--100ns
-± 1± 10µA
--± 10µA
-9-pF
-9-pF
Notes: 2. Max leakage for pins with pull-up resistors (TRST, TMS, TDI, SSI, GPIO, MOSI, SCK1) is ±250 µA.
t
risein
t
fa llin
2.6 V
0.9 * VDD
0.1 * VDD
0.7 V
t
rise out
t
fallo ut
0.9 * VDD
4.6
0.1 * VDD
0.4 V
POWER CONSUMPTION
ParameterSymbol Min TypMaxUnit
Operational Power Consumption
1.024 MHz Digital Filter ClockPWR
2.048 MHz Digital Filter ClockPWR
4.096 MHz Digital Filter ClockPWR
8.192 MHz Digital Filter ClockPWR
16.384 MHz Digital Filter ClockPWR
Standby Power Consumption
32 kHz Digital Filter Clock, Filter StoppedPWR
1
2
4
8
16
S
-21-mW
-26-mW
-37-mW
-57-mW
-85-mW
-40-µW
DS612F414
SWITCHING CHARACTERISTICS
SPI 1 Interface Timing (External Master)
SSI
CS5376A
MOSI
SCK1
SCLK
MSBMSB - 1
t
1
t
2
t
t
3
t
5
4
Figure 4. MOSI Write Timing in SPI Slave Mode
SSI
MISO
SCK1
SCLK
MSBMSB - 1LSB
t
t
9
t
7
8
Figure 5. MISO Read Timing in SPI Slave Mode
ParameterSymbol Min TypMaxUnit
MOSI Write Timing
SSI
Enable to Valid Latch Clockt
Data Set-up Time Prior to SCK1 Risingt
Data Hold Time After SCK1 Risingt
SCK1 High Timet
SCK1 Low Timet
SCK1 Falling Prior to SSI
Disablet
MISO Read Timing
SCK1 Falling to New Data Bitt
SCK1 High Timet
SCK1 Low Timet
SSI
Rising to MISO Hi-Zt
10
LSB
t
6
t
10
1
2
3
4
5
6
7
8
9
60--ns
60--ns
120--ns
120--ns
120--ns
60--ns
--60ns
120--ns
120--ns
--150ns
DS612F415
SWITCHING CHARACTERISTICS
Serial Data Port (SD Port)
SDRDY
SDCLK
t
3
SDDAT
t4t
SDTKI
SDTKO
t
1
t
2
5
CS5376A
t
t
6
7
t
t
9
8
Figure 6. SD Port Read Timing
ParameterSymbol Min TypMaxUnit
SDTKI to SDRDY Falling Edget
SDTKI High Time Widtht
SDRDY
Falling Edge to SDCLK Falling Edget
Data Setup Time Prior to SDCLK Risingt
Data Hold Time After SDCLK Risingt
SDCLK High Timet
SDCLK Low Timet
SDCLK Rising to SDRDY
Data Hold Time After SDRDY
SDRDY
High to SDTKO Rising Edget
Risingt
Risingt
SDTKO High Timet
10
t10t
11
1
2
3
4
5
6
7
8
9
60--ns
60-1000ns
50--ns
60--ns
60--ns
120--ns
120--ns
60--ns
--150ns
--60ns
11
90--ns
DS612F416
SWITCHING CHARACTERISTICS
CLK, SYNC, MCLK, MSYNC, and MDATAx
SYNC
MCLK
CS5376A
MSYNC
t
msd
MDATAx
Note: SYNC input latched on MCLK rising edge. MSYNC output triggered by MCLK falling edge.
TBS Data Bit Rate-256-kbps
TBS Data Rising to TBS Clock Rising Setup Timet
TBS Clock Rising to TBS Data Falling Hold Time(Note 6)t
4
5
60--ns
60--ns
5. TBSCLK phase can be delayed in 1/8 increments. The timing diagram shows no TBSCLK delay.
6. TBSDATA can be delayed from 0 to 63 full bit period s. The tim ing diag ram sho ws no T BSDATA delay.
DS612F418
CS5376A
Geophone
or
Hydrophone
Sensor
Geophone
or
Hydrophone
Sensor
Geophone
or
Hydrophone
Sensor
Geophone
or
Hydrophone
Sensor
CS3301A
AMP
AMP
AMP
AMP
CS3302A
CS3301A
CS3302A
CS3301A
CS3302A
CS3301A
CS3302A
Switch
Switch
MUX
MUX
CS5371A
CS5372A
∆Σ
Modulator
CS5371A
CS5372A
∆Σ
Modulator
CS5376A
Digital F ilter
CS4373A
DAC
M
U
X
M
U
X
M
U
X
M
U
X
Figure 9. Multi-Channel System Block Diagram
Test
System T e lemetry
µController
or
Configuration
EEPROM
Communication
Interface
3. SYSTEM DESIGN WITH CS5376A
Figure 9 illustrates a simplified block diagram of
the CS5376A in a multi-channel measurement system.
Up to four differential sensors are connected
through CS3301A/02A differential amplifiers to
the CS5371A/72A ∆Σ modulators, where analog to
digital conversion occurs. Each modulators 1-bit
output connects to a CS5376A MDATA input,
where the oversampled ∆Σ data is decimated and
filtered to 24-bit output samples at a programmed
output rate. These output samples are buffered in
an 8-deep data FIFO and passed to the system telemetry on command.
System self tests are performed by connecting the
CS5376A test bit stream (TBS) generator to the
CS4373A test DAC. Analog tests drive differential
signals from the CS4373A test DAC into the multiplexed inputs of the CS3301A/02A amplifiers or
directly to the sensors through external analog
switches. Digital loopback tests internally connect
the TBS digital output directly to the CS5376A
modulator inputs.
3.1 Power Supplies
The multi-channel system shown in Figure 9 typically operates from a ±2.5 V analog power supply
and a 3.3 V digital power supply. The CS5376A
logic core can be powered from 3 V to minimize
power consumption, if required.
3.2 Reset Control
System reset is required only for the CS5376A device, and is a standard active low signal that can be
generated by a power supply monitor or microcontroller. Other system devices default to a powerdown state when the CS5376A is reset.
DS612F419
CS5376A
3.3 Clock Generation
A single 32.768 MHz low-jitter clock input, which
can be generated from a VCXO based PLL, is required to drive the CS5376A device. Clock inputs
for other system devices are driven by clock outputs from the CS5376A.
3.4 Synchronization
Digital filter phase and analog sample timing of the
four ∆Σ modulators connected to the CS5376A are
synchronized by a rising edge on the SYNC pin. If
a synchronization signal is received identically by
all CS5376A devices in a measurement network,
synchronous sampling across the network is guaranteed.
3.5 System Configuration
Through the SPI 1 serial port, filter coefficients and
digital filter register settings can either be programmed by a microcontroller or automatically
loaded from an external EEPROM after reset. System configuration is only required for the
CS5376A device, as other devices are configured
via the CS5376A General Purpose I/O pins.
Two registers in the digital filter, SYSTEM1 and
SYSTEM2 (0x2C, 0x2D), are provided for user defined system information. These are general purpose registers that will hold any 24-bit data values
written to them.
3.6 Digital Filter Operation
After analog to digital conversion occurs in the
modulators, the oversampled 1-bit ∆Σ data is read
into the CS5376A through the MDATA pins. The
digital filter then processes data through the enabled filter stages, decimating it to 24-bit words at
a programmed output word rate. The final 24-bit
samples are concatenated with 8-bit status words
and placed into an output FIFO.
3.7 Data Collection
Data is collected from the CS5376A through the
Serial Data port (SD port). Automatically or upon
request, depending how the SDTKI pin is connected, the SD port initiates serial transactions to transfer 32-bit data from the output FIFO to the system
telemetry. The output FIFO has eight data locations
to permit latency in data collection.
3.8 Integrated peripherals
Test Bit Stream (TBS)
A digital signal generator built into the CS5376A
produces a 1-bit ∆Σ sine wave. This digital test bit
stream can be connected to the CS4373A test DAC
to create high quality analog test signals or it can be
internally looped back to the CS5376A MDATA
inputs to test the digital filter and data collection
circuitry.
Time Break
Timing information is recorded during data collection by strobing the TIMEB pin. A dedicated flag
in the sample status bits, TB, is set high to indicate
over which measurement the timing event occurred.
General Purpose I/O (GPIO)
Twelve general purpose pins are available on the
CS5376A for system control. Each pin can be set as
input or output, high or low, with an internal pullup enabled or disabled. The CS3301A/02A,
CS5371A/72A and CS4373A devices in Figure 9
are configured by simple pin settings controlled
through the CS5376A GPIO pins.
Serial Peripheral Interface 2 (SPI 2)
A secondary master mode serial port to communicate with external serial peripherals.
The CS5376A has three sets of power supply inputs. Two sets supply power to the I/O pins of the
device (VDD1, VDD2), and the third supplies
power to the logic core (VD). The I/O pin power
supplies determine the maximum input and output
voltages when interfacing to peripherals, and the
logic core power supply largely determines the
power consumption of the CS5376A.
4.1 Pin Descriptions
VDD1, GND1 - Pins 54,53
Sets the interface voltage to a microcontroller and
system telemetry. Can be driven with voltages from
3.3 V to 5 V.
VDD1 powers pins 1-5 and 41-64:
TRST, TMS, TCK, TDI, TDO
GND
GND2
MFLAG1
SI4
VDD2
SO
SI3
SI2
SI1
SCK2
GPIO0:CS0
GPIO6 - GPIO11:EECS
SSO
, SCK1, SSI, MISO, MOSI, SINT,
RESET
SDDAT, SDRDY
, BOOT, TIMEB, CLK, SYNC
, SDCLK, SDTKO, SDTKI
VDD2, GND2 - Pins 11, 25, 24, 38
Sets the interface voltage to the modulators, test
DAC, and serial peripherals. Can be driven with
voltages from 3.3 V to 5 V.
Sets the operational voltage of the CS5376A logic
core. Can be driven with voltages from 3 V to 5 V.
A 3 V supply minimizes total power consumption.
4.2 Bypass Capacitors
Each power supply pin should be bypassed with
parallel 1 µF and 0.01 µF caps, or by a single
0.1 µF cap, placed as close as possible to the
CS5376A. Bypass capacitors should be ceramic
(X7R, C0G), tantalum, or other good quality dielectric type.
4.3 Power Consumption
Power consumption of the CS5376A depends primarily on the power supply voltage of the logic
core (VD) and the programmed digital filter clock
rate. Digital filter clock rates are selected based on
the required output word rate as explained in “Digital Filter Initialization” on page 41.
22DS612F4
CS5376A
RESET
Figure 11. Reset Control Block Diagram
Self-Tests
SELFTEST
Register
5. RESET CONTROL
The CS5376A reset signal is active low. When released, a series of self-tests are performed and the
device either actively boots from an external EEPROM or enters an idle state waiting for microcontroller configuration.
5.1 Pin Descriptions
RESET - Pin 55
Reset input, active low.
BOOT - Pin 56
Boot mode select, latched following a RESET rising edge.
After RESET is released but before booting, a series of digital filter self-tests are run. Results are
Self-Test
Type
Program ROM0x00000A0x00000F
Data ROM0x0000A00x0000F0
Program RAM0x000A000x000F00
Data RAM0x00A0000x00F000
Execution Unit0x0A00000x0F0000
Pass
Code
Fail
Code
BOOT
Pin
1
EEPROM
Boot
0
µController
Boot
combined into the SELFTEST register (0x2F),
with 0x0AAAAA indicating all passed. Self-tests
require 60 ms to complete, after which configuration commands are serviced.
5.3 Boot Configurations
The logic state of the BOOT pin after reset determines if the CS5376A actively reads configuration
information from EEPROM or enters an idle state
waiting for a microcontroller to write configuration
commands.
EEPROM Boot
When the BOOT pin is high after reset, the
CS5376A actively reads data from an external serial EEPROM and then begins operation in the specified configuration. Configuration commands and
data are encoded in the EEPROM as specified in
the ‘Configuration By EEPROM’ section of this
data sheet, starting on page 26.
Microcontroller Boot
When the BOOT pin is low after reset, the
CS5376A enters an idle state waiting for a microcontroller to write configuration commands and
initialize filter operation. Configuration commands
and data are written as specified in the ‘Configuration By Microcontroller’ section of this data sheet,
starting on page 32.
DS612F423
CS5376A
Clock DividerCLK
MCLK
Generator
DSPCFG Register
Figure 12. Clock Generation Block Diagram
6. CLOCK GENERATION
The CS5376A requires a 32.768 MHz master clock
input, which is used to generate internal digital filter clocks and external modulator clocks.
6.1 Pin Description
CLK - Pin 58
Clock input, nominal frequency 32.768 MHz.
Internal
and
Clocks
MCLK
Output
ensure recovered clocks have identical phase, system PLL designs should use a phase/frequency detector architecture.
6.3 Master Clock Jitter and Skew
Care must be taken to minimize jitter and skew in
the received master clock as both parameters affect
measurement performance.
6.2 Synchronous Clocking
To guarantee synchronous measurements throughout a sensor network, the CS5376A master clock
should be distributed to arrive at all nodes in phase.
The 32.768 MHz master clock can either be directly distributed through the system telemetry, or reconstructed locally using a VCXO based PLL. To
Jitter in the master clock causes jitter in the generated modulator clocks, resulting in sample timing
errors and increased noise.
Skew in the master clock from node to node creates
a sample timing offset, resulting in systematic measurement errors in the reconstructed signal.
24DS612F4
CS5376A
0
SYNC
1
MSEN
MSYNC
Generator
Figure 13. Synchronization Block Diagram
7. SYNCHRONIZATION
The CS5376A has a dedicated SYNC input that
aligns the internal digital filter phase and generates
an external signal for synchronizing modulator analog sampling. By providing simultaneous rising
edges to the SYNC pins of multiple CS5376A devices, synchronous sampling across a network can
be guaranteed.
7.1 Pin Description
SYNC - Pin 59
Synchronization input, rising edge triggered.
7.2 MSYNC Generation
Digital
Filter
MSYNC
Output
0
1
TSYNC
Tes t Bit
Stream
phase. Filter convolutions restart, and the next output word is available one full sample period later.
Repetitive synchronization is supported when
SYNC events occur at exactly the selected output
word rate. In this case, re-synchronization occurs at
the start of a convolution cycle when the digital filter state machine is already reset.
7.4 Modulator Synchronization
The external MSYNC signal phase aligns modulator analog sampling when connected to the
CS5371A/72A MSYNC input. This ensures synchronous analog sampling relative to MCLK.
The SYNC signal rising edge is used to generate a
retimed synchronization signal, MSYNC. The
MSYNC signal reinitializes internal digital filter
phase and is driven onto the MSYNC output pin to
Repetitive synchronization of the modulators is
supported when SYNC events occur at exactly the
selected output word rate. In this case, synchronization will occur at the start of analog sampling.
phase align modulator analog sampling.
The MSEN bit in the digital filter CONFIG register
(0x00) enables MSYNC generation. See “Modulator Interface” on page 39 for more information
about MSYNC.
7.5 Test Bit Stream Synchronization
When the test bit stream generator is enabled, an
MSYNC signal can reset the internal data pointer.
This restarts the test bit stream from the first data
point to establish a known output signal phase.
7.3 Digital Filter Synchronization
The internal MSYNC signal resets the digital filter
state machine to establish a known digital filter
The TSYNC bit in the digital filter TBSCFG register (0x2A) enables synchronization of the test bit
stream by MSYNC. When TSYNC is disabled, the
test bit stream phase is not affected by MSYNC.
DS612F425
GPIO11:EECS
CS5376AAT25640
Figure 14. EEPROM Configuration Block Diagram
8. CONFIGURATION BY EEPROM
SCK1
MISO
MOSI
CS5376A
VD
387
46
48
50
51
WP VCC HOLD
1
CS
6
SCK
2
SO
5
SI
4
GND
After reset, the CS5376A reads the state of the
BOOT pin to determine a source for configuration
commands. If BOOT is high, the CS5376A initiates serial transactions through the SPI 1 port to
read configuration information from an external
EEPROM.
8.1 Pin Descriptions
Pins required for EEPROM boot are listed here,
other SPI 1 pins are inactive.
GPIO11:EECS - Pin 46
EEPROM chip select output, active low.
SCK1 - Pin 48
Serial clock output, nominally 1.024 MHz.
MOSI - Pin 51
Serial data output pin. Valid on rising edge of
SCK1, transition on falling edge.
MISO - Pin 50
Serial data input pin. Valid on rising edge of SCK1,
transition on falling edge.
8.2 EEPROM Hardware Interface
When booting from EEPROM the CS5376A SPI 1
port actively performs serial transactions, as shown
in Figure 15, to read configuration commands and
data. 8-bit SPI opcodes and 16-bit addresses are
combined to read back 8-bit configuration commands and 24-bit configuration data.
System design should include a connection to the
configuration EEPROM for in-circuit reprogramming. The CS5376A SPI 1 pins go high impedance
when inactive to support external connections to
the serial bus.
8.3 EEPROM Organization
The boot EEPROM holds the 8-bit commands and
24-bit data required to initialize the CS5376A into
an operational state. Configuration information
starts at memory location 0x10, with addresses
0x00 to 0x0F free for use as manufacturing header
information.
The first serial transaction reads a 1-byte command
from memory location 0x10 and then, depending
on the command type, reads multiple 3-byte data
words to complete the command. Command and
data reads continue until the ‘Filter Start’ command
is recognized.
The maximum number of bytes that can be written
for a single configuration is approximately
26DS612F4
CS5376A
InstructionOpcodeAddressDefinition
Read0x03ADDR[15:0]Read data beginning at the address given in ADDR.
SPI 1 Read from EEPROM
SSI
MOSI
MISO
EECS
Cycle
SCK1
MOSI
READ
CMD
0x03ADDR
2 BYTE
ADDR
ADDR
DATA1DATA3DATA2
1 BYTE / 3 BYTE
DATA
18276543
MSBLSB
612345
MISO
EECS
MSBLSB612345
Figure 15. SPI 1 EEPROM Read Transactions
X
DS612F427
Write DF Register - 0x01
CS5376A
0000h
Mfg Header
0010h
EEPROM
Manufacturing
Information
8-bit Command
N x 24-bit Data
EEPROM
Command and
8-bit Command
Data Values
N x 24-bit Data
. . .
1FFFh
Figure 16. 8 Kbyte EEPROM Memory Organization
5 KByte (40 Kbit), which includes command overhead:
Memory RequirementBytes
Digital Filter Registers (22)154
FIR Coefficients (255+255)1537
IIR Coefficients (3+5)25
Test Bit Stream Data (1024)3076
This EEPROM command writes a data value to the
specified digital filter register. Digital filter registers control hardware peripherals and filtering
functions. See “Digital Filter Registers” on page 86
for the bit definitions of the digital filter registers.
Sample Command:
Write digital filter register 0x00 with data value
0x070431. Then write 0x20 with data 0x000240.
01 00 00 00 07 04 31
01 00 00 20 00 02 40
Write FIR Coefficients - 0x02
This EEPROM command writes custom coefficients for the FIR1 and FIR2 filters. The first two
data words set the number of FIR1 and FIR2 coefficients to be written. The remaining data words are
the concatenated FIR1 and FIR2 coefficients.
A maximum of 255 coefficients can be written for
each FIR filter, though the available digital filter
computation cycles will limit their practical size.
See “FIR Filter” on page 47 for more information
about FIR filter coefficients.
SPI mode 0 (0,0) compatible, 16-bit addresses, 8bit data, larger than 5 KByte (40 KBit). ATMEL
AT25640, AT25128, or similar serial EEPROMs
are recommended.
This EEPROM command writes custom coefficients for the two stage IIR filter. The IIR architecture and number of coefficients is fixed, so eight
data words containing coefficient values always
8.4 EEPROM Configuration
Commands
immediately follow the command byte. The IIR coefficient write order is: a11, b10, b11, a21, a22,
b20, b21, and b22. See “IIR Filter” on page 55 for
03
84 BC 9D 7D A1 B1 82 5E 4F 83 69 4F
3C AD 5F 3E 51 04 83 5D F8 3E 51 04
Write ROM Coefficients - 0x04
This EEPROM command selects the on-chip coefficients for the FIR1, FIR2, IIR 1st order, and IIR
2nd order filters for use by the digital filter. One
data word is required to select which internal coefficient sets to use. See “Filter Coefficient Selection” on page 41 for information about selecting
on-chip FIR and IIR coefficient sets.
Sample Command:
Select IIR1 and IIR2 3 Hz @ 500 SPS low-cut coefficients, with FIR1 and FIR2 linear phase highcut coefficients. Data word 0x002200.
04 00 22 00
Write TBS Data - 0x05
This EEPROM command writes a custom data set
for the test bit stream (TBS) generator. This command, along with the ability to program the test bit
stream generator interpolation and clock rate, can
create custom frequency test signals.
The first data word sets the number of TBS data to
be written and the remaining data words are the
TBS data values. See “Test Bit Stream Generator”
on page 64 for information about using custom test
bit stream data sets.
NameCMD
8-bit
NOP00-No Operation
WRITE DF REGISTER01REG
WRITE FIR COEFFICIENTS02NUM FIR1
WRITE IIR COEFFICIENTS03a11
WRITE ROM COEFFICIENTS04COEF SELUse On-Chip Coefficients
WRITE TBS DATA05NUM TBS
WRITE ROM TBS06-Use On-Chip TBS Data
FILTER START07-Start Digital Filter Operation
DATA
24-bit
DATA
NUM FIR2
(FIR COEF)
b10
b11
a21
a22
b20
b21
b22
(TBS DATA)
Description
Write Digital Filter Register
Write Custom FIR Coefficients
Write Custom IIR Coefficients
Write Custom Test Bit Stream Data
(DATA) indicates multiple words of this type are to be written.
Table 5. EEPROM Boot Configuration Commands
DS612F429
CS5376A
Sample Command:
Write test bit stream data 0x000000, 0x0007DA,
0x000FB5, 0x00178F.
05 00 00 04
00 00 00 00 07 DA 00 0F B5 00 17 8F
Write TBS ROM Data - 0x06
This EEPROM command selects the on-chip test
bit stream (TBS) data for use by the TBS generator.
No data words are required for this EEPROM command. See “Test Bit Stream Generator” on page 64
for more information about the on-chip test bit
stream data set.
Sample Command:
06
Filter Start - 0x07
This EEPROM command initializes and starts the
digital filter. Measurement data becomes available
one full sample period after this command is received. No data words are required for this EEPROM command.
Sample Command:
07
8.5 Example EEPROM Configuration
Table 6 shows an example EEPROM file for a minimal CS5376A configuration.
Figure 17. Serial Peripheral Interface 1 (SPI 1) Blo ck Diagram
SPI 1
Registers
9. CONFIGURATION BY MICROCONTROLLER
After reset, the CS5376A reads the state of the
BOOT pin to determine a source for configuration
commands. If BOOT is low, the CS5376A receives
configuration commands from a microcontroller.
9.1 Pin Descriptions
Pins required for microcontroller boot are listed
here, other SPI 1 pins are inactive.
SSI - Pin 49
Slave select input pin, active low. Serial chip select
input from a microcontroller.
SCK1 - Pin 48
9.2 Microcontroller Hardware Interface
When booting from a microcontroller the
CS5376A SPI 1 port receives configuration commands and configuration data through serial transactions, as shown in Figure 18. 8-bit SPI opcodes
and 8-bit addresses are combined to read and write
24-bit configuration commands and data.
Microcontroller serial transactions require toggling
the SSI pin as the CS5376A chip select and writing
a serial clock to the SCK1 input. Serial data is input
to the CS5376A on the MOSI pin, and output from
the CS5376A on the MISO pin.
SPI 1
Pin Logic
SSI
SCK1
MOSI
MISO
SINT
Serial clock input pin. Serial clock input from microcontroller, maximum 4.096 MHz.
MOSI - Pin 51
Serial data input pin. Valid on rising edge of SCK1,
transition on falling edge.
9.3 Microcontroller Serial Transactions
Microcontroller configuration commands are written to the digital filter through the SPI 1 registers.
A 24-bit command and two 24-bit data words can
be written to the SPI 1 registers in any single serial
transaction. Some commands require additional
MISO - Pin 50
Serial data output pin. Valid on rising edge of
data words through additional serial transactions to
complete.
SCK1, transition on falling edge. Open drain output requiring a 10 kΩ pull-up resistor.
SINT - Pin 52
Serial interrupt output pin, active low. 1 uS active
low pulse output when ready for next serial transaction.
32DS612F4
9.3.1 SPI opcodes
A microcontroller communicates with the
CS5376A SPI 1 port using standard 8-bit SPI op-
codes and an 8-bit SPI address. The standard SPI
‘Read’ and ‘Write’ opcodes are listed in Figure 18.
CS5376A
InstructionOpcodeAddressDefinition
Write0x02ADDR[7:0]Write SPI1 registers beginning at the address in ADDR.
Read0x03ADDR[7:0]Read SPI 1 registers beginning at the address in ADDR.
Microcontroller Write to SPI 1
SSI
MOSI0x02ADDR Data1
MISO
Microcontroller Read from SP I 1
SSI
MOSI
MISO
Cycle
SCK1
18276543
0x03ADDR
DataNData2
Data1DataNData2
MOSI
MISO
SSI
MSBLSB
MSBLSB612345
612345
X
Figure 18. Microcontroller Serial Transac tions
DS612F433
CS5376A
9.3.2 SPI 1 registers
The SPI 1 registers are shown in Figure 19 and are
24-bit registers mapped into an 8-bit register space
as high, mid, and low bytes. See “SPI 1 Registers”
on page 81 for the bit definitions of the SPI 1 registers.
9.3.3 SPI 1 transactions
A serial transaction to the SPI 1 registers starts with
an SPI opcode, followed by an address, and then
some number of data bytes written or read starting
at that address.
Typical serial write transactions require sending
groups of 5, 8, or 11 total bytes to the SPI1CMD or
SPI1DAT1 registers.
Example 5-byte write transaction to SPI1CMD
02 03 12 34 56
Example 5-byte write transaction to SPI1DAT1
02 06 12 34 56
Example 8-byte write transaction to SPI1CMD
02 03 12 34 56 AB CD EF
MOSI: 03 01 00
MISO: xx xx 12
5-byte read transaction of SPI1DAT1
MOSI: 03 06 00 00 00
MISO: xx xx 12 34 56
9.3.4 Multiple serial transactions
Some configuration commands require multiple se-
rial transactions to complete. There must be a small
delay between transactions for the CS5376A to
process the incoming data. Three methods can be
used to ensure the CS5376A is ready to receive the
next configuration command.
1) Delay a fixed 1 ms period to guarantee enough
time for the command to be completed.
2) Monitor the SINT pin for a 1 us active low pulse.
This pulse output occurs once the CS5376A com-
pletes processing the current command.
3) Verify the status of the E2DREQ bit by reading
the SPI1CTRL register. When low, the CS5376A is
ready for the next command.
Example 8-byte write transaction to SPI1DAT1
02 06 12 34 56 AB CD EF
Example 11-byte write transaction to SPI1CMD
02 03 12 34 56 AB CD EF 65 43 21
Typical serial read transactions require groups of 3
or 5 bytes, split between writing into MOSI and
reading from MISO.
3-byte read transaction of mid-byte of SPI1CTRL
NameAddr.Type# BitsDescription
SPI1CTRL00 - 02R/W8, 8, 8SPI 1 Control
SPI1CMD03 - 05R/W8, 8, 8SPI 1 Command
SPI1DAT106 - 08R/W8, 8, 8SPI 1 Data 1
SPI1DAT209 - 0BR/W8, 8, 8SPI 1 Data 2
Figure 19. SPI 1 Registers
9.3.5 Polling E2DREQ
One transaction type that can always be performed
no matter the delay from the previous configuration
command is reading E2DREQ in the mid-byte of
the SPI1CTRL register. A 3-byte read transaction.
MOSI: 03 01 00
MISO: xx xx 01 <- E2DREQ bit high
MISO: xx xx 00 <- E2DREQ bit low
34DS612F4
CS5376A
The E2DREQ bit reads high while a configuration
command is being processed. When low, the digital
filter is ready to receive a new configuration command.
9.4 Microcontroller Configuration
Commands
A summary of available microcontroller configuration commands is listed in Table 7.
Write DF Register - 0x01
This configuration command writes a specified
digital filter register. Digital filter registers control
hardware peripherals and filtering functions. See
“Digital Filter Registers” on page 86 for the bit definitions of the digital filter registers.
Sample Command:
Write digital filter register 0x00 with data value
0x070431. Then write 0x20 with data 0x000240.
This command reads a specified digital filter register. The register value is requested in the first SPI
transaction, with the register value copied to
SPI1DAT1 and read in a subsequent SPI transaction.
Sample Command:
Read digital filter registers 0x00 and 0x20.
02 03 00 00 02 00 00 00
Delay 1 ms, monitor SINT, or poll E2DREQ
MOSI: 03 06 00 00 00
MISO: xx xx 07 04 31
02 03 00 00 02 00 00 20
Delay 1 ms, monitor SINT, or poll E2DREQ
MOSI: 03 06 00 00 00
MISO: xx xx 00 02 40
Write FIR Coefficients - 0x03
This command writes custom coefficients for the
FIR1 and FIR2 filters. The first two data words set
the number of FIR1 and FIR2 coefficients to be
written. The remaining data words are the concatenated FIR1 and FIR2 coefficients.
A maximum of 255 coefficients can be written for
each FIR filter, though the available digital filter
computation cycles will limit their practical size.
See “FIR Filter” on page 47 for more information
about FIR filter coefficients.
Sample Command:
Write FIR1 coefficients 0x00022E, 0x000771 then
FIR2 coefficients 0xFFFFB9, 0xFFFE8D.
02 03 00 00 03 00 00 02 00 00 02
Delay 1 ms, monitor SINT
02 06 00 02 2E 00 07 71
Delay 1 ms, monitor SINT
02 06 FF FF B9 FF FE 8D
02 03 00 00 04 84 BC 9D 7D A1 B1
Delay 1 ms, monitor SINT, or poll E2DREQ
02 06 82 5E 4F 83 69 4F
Delay 1 ms, monitor SINT, or poll E2DREQ
02 06 3C AD 5F 3E 51 04
Delay 1 ms, monitor SINT
02 06 83 5D F8 3E 51 04
Delay 1 ms, monitor SINT
Write ROM Coefficients - 0x05
This configuration command selects the on-chip
coefficients for FIR1, FIR2, IIR 1st order, and IIR
2nd order filters for use by the digital filter. One
data word is required to select which internal coefficient sets to use. See “Filter Coefficient Selection” on page 41 for information about selecting
on-chip FIR and IIR coefficient sets.
, or poll E2DREQ
, or poll E2DREQ
Sample Command:
Select IIR1 and IIR2 3 Hz @ 500 SPS low-cut coefficients, with FIR1 and FIR2 linear phase highcut coefficients. Data word 0x002200.
This command writes custom coefficients for the
two stage IIR filter. The IIR architecture and number of coefficients is fixed, so eight coefficient values immediately follow this command. The IIR
coefficient write order is: a11, b10, b11, a21, a22,
b20, b21, and b22. See “IIR Filter” on page 55 for
more information about IIR filter coefficients.
36DS612F4
Write TBS Data - 0x06
This command writes a custom data set for the test
bit stream (TBS) generator. This command, along
with the ability to program the test bit stream generator interpolation and clock rate, can create custom frequency test signals.
The first data word sets the number of TBS data to
be written and the remaining data words are the
TBS data values. See “Test Bit Stream Generator”
CS5376A
on page 64 for information about using custom test
bit stream data sets.
Sample Command:
Write test bit stream data 0x000000, 0x0007DA,
0x000FB5, 0x00178F.
This command selects the on-chip test bit stream
(TBS) data for use by the TBS generator. No data
words are required for this configuration command. See “Test Bit Stream Generator” on page 64
for information about the on-chip test bit stream
data set.
This command initializes and starts the digital filter. Measurement data becomes available one full
sample period after this command is issued. No
data words are required for this configuration command.
This command disables the digital filter. Measurement data output stops immediately after this command is issued. No data words are required for this
configuration command.
The CS5376A performs digital filtering for up to
four ∆Σ modulators. Signals from the modulators
are connected through the modulator data interface
(MDI).
FIR
Filters
Output Rate 4000 SPS ~ 1 SPS
IIR
Filter
10.2Modulator Clock Generation
The MCLK and MCLK/2 outputs are low-jitter,
low-skew modulator clocks generated from the
32.768 MHz master clock.
10.1Pin Descriptions
MCLK, MCLK/2 - Pins 13, 12
Modulator clock outputs. Nominally 2.048 MHz
and 1.024 MHz.
MSYNC - Pin 14
Modulator synchronization signal output. Generated from the SYNC input.
MDATA1 - MDATA4 - Pins 15, 17, 19, 21
Modulator data inputs, nominally 512 kbit/s.
MFLAG1 - MFLAG4 - Pins 16, 18, 20, 22
Modulator flag inputs. Driven high when modulator is unstable due to an analog over-range signal.
MCLK typically operates at 2.048 MHz unless analog low-power modes require a 1.024 MHz modulator clock. MCLK/2 always produces a clock at
half the selected MCLK rate.
The MCLK rate is selected and the MCLK and
MCLK/2 outputs are enabled by bits in the digital
filter CONFIG register (0x00). By default MCLK
and MCLK/2 are disabled and driven low.
10.3Modulator Synchronization
The MSYNC output signal follows an input on the
SYNC pin. MSYNC phase aligns the modulator
sampling instant to guarantee synchronous analog
sampling across a measurement network.
MSYNC is enabled by a bit in the CONFIG register
(0x00). By default SYNC inputs do not cause an
MSYNC output.
DS612F439
CS5376A
10.4Modulator Data Inputs
The MDATA input expects 1-bit ∆Σ data at a
512 kHz or 256 kHz rate. The input rate is selected
by a bit in the CONFIG register (0x00). By default,
MDATA is expected at 512 kHz.
The MDATA input one’s density is designed for
full scale positive at 86% and full scale negative at
14%, with absolute maximum over-range capability to 93% and 7%. These raw ∆Σ inputs are decimated and filtered by the digital filter to create 24bit samples at the output rate.
10.5Modulator Flag Inputs
A high MFLAG input signal indicates the corresponding ∆Σ modulator has become unstable due
to an analog over-range input signal. Once the
over-range signal is reduced, the modulator recovers stability and the MFLAG signal is cleared.
The MFLAG inputs are mapped to status bits in the
SD port, and are associated with each sample when
written. See “Serial Data Port” on page 61 for more
information on the MFLAG error bits in the SD
port status byte.
40DS612F4
CS5376A
Modulator
Input
512 kHz
SINC Filter
2 - 64000
DC Offset
& Gain
Correction
FIR1
4
Output to High Speed Serial Data Port (SD Port)
Figure 21. Digital Filter Stages
11.DIGITAL FILTER INITIALIZATION
The CS5376A digital filter consists of three multistage sections: a three stage SINC filter, a two stage
FIR filter, and a two stage IIR filter.
To initialize the digital filter, FIR and IIR coefficient sets are selected using configuration commands and the FILTCFG register (0x20) is written
to select the output filter stage, the output word
rate, and the number of enabled channels. The digital filter clock rate is selected by writing the CONFIG register (0x00).
11.1Filter Coefficient Selection
FIR2
2
Output Rate 4000 SPS ~ 1 SPS
IIR1IIR2
1st Order
2nd Order
word, and the available coefficient sets for each selection.
Characteristics of the on-chip digital filter coefficients are discussed in the ‘SINC Filter’, ‘FIR Filter’, and ‘IIR Filter’ sections of this data sheet.
11.2Filter Configuration Options
Digital filter parameters are selected by bits in the
FILTCFG register (0x20), and the digital filter
clock rate is selected by bits in the CONFIG register (0x00).
Selection of SINC filter coefficients is not required
as they are selected automatically based on the programmed output word rate.
Digital filter FIR and IIR coefficients are selected
using the ‘Write FIR Coefficients’ and ‘Write IIR
Coefficients’, or the ‘Write ROM Coefficients’
configuration commands. When writing the FIR
and IIR coefficients from ROM, a data word selects
an on-chip coefficient set for each filter stage. Fig-
11.2.1 Output Filter Stage
The digital filter can output data following any
stage in the filter chain. The output filter stage is selected by the FSEL bits in the FILTCFG register.
Taking data from the SINC or FIR1 filter stages reduces the overall decimation of the filter chain and
increases the output rate, as discussed in the following section. Taking data from FIR2, IIR1, IIR2,
or IIR3 results in data at the selected rate.
ure 22 shows the format of the coefficient selection
The CS5376A digital filter supports output word
rates (OWRs) between 4000 SPS and 1 SPS. The
output word rate is selected by the DEC bits in the
FILTCFG register.
When taking data directly from the SINC filter, the
decimation of the FIR1 and FIR2 stages is bypassed and the actual output word rate is multiplied
by a factor of eight compared with the register selection. When taking data directly from FIR1, the
decimation of the FIR2 stage is bypassed and the
actual output word rate is multiplied by a factor of
two. Data taken from the FIR2, IIR1, IIR2, or IIR3
filtering stages is output at the selected rate.
11.2.3 Channel Enable
Digital filtering can be performed simultaneously
for up to four ∆Σ modulators. The number of enabled channels is selected by the CH bits in the
FILTCFG register.
Channels are enabled sequentially. Selecting one
channel operation enables channel 1 only, selecting
two channel operation enables channels 1 and 2, se-
Bits 3:0FIR1 Coefficients
0000Linear Phase
0001Minimum Phase
Bits 7:4FIR2 Coefficients
0000Linear Phase
0001Minimum Phase
lecting three channel operation enables channels 1,
2, and 3, and selecting four channel operation enables all four channels.
11.2.4 Digital Filter Clock
The digital filter clock rate is programmable between 16.384 MHz and 32 kHz by bits in the CONFIG register.
Computation Cycles
The minimum digital filter clock rate for a configuration depends on the computation cycles required
to complete digital filter convolutions at the selected output word rate. All configurations work for a
maximum digital filter clock, but lower clock rates
consume less power.
Standby Mode
The CS5376A can be placed in a low-power standby mode by sending the ‘Filter Stop’ configuration
command and programming the digital filter clock
to 32 kHz. In this mode the digital filter idles, consuming minimal power until re-enabled by later
configuration commands.
42DS612F4
CS5376A
4th order
sinc3
stage3
5
1-bit
∆−Σ
Input
5th order
sinc1
4th order
sinc3
stage1
5
8
4th order
4th order
sinc3
stage2
5
sinc2
stage1
2
4th order
Figure 23. SINC Filter Block Diagram
12.SINC FILTER
The SINC filters primary purpose is to attenuate
out-of-band noise components from the ∆Σ modulators. While doing so, they decimate 1-bit ∆Σ data
into lower frequency 24-bit data suitable for the
FIR and IIR filters.
The SINC filter has three cascaded sections,
SINC1, SINC2, and SINC3, which are each made
up of the smaller stages shown in Figure 23.
sinc2
stage2
2
5th order
sinc3
stage4
2
4th order
sinc2
stage3
2
6th order
sinc3
stage5
4th order
2
sinc2
stage4
2
6th order
sinc3
stage6
3
12.2SINC2 Filter
The second section is SINC2, a multi-stage, variable order, variable decimation SINC filter. Depending on the selected output word rate in the
FILTCFG register, different cascaded SINC2 stages are enabled, as shown in Table 9.
12.3SINC3 Filter
24-bit
Output
The selected output word rate in the FILTCFG register automatically determines the coefficients and
decimation ratios selected for the SINC filters.
Once the SINC filter configuration is set, all enabled channels are filtered and decimated using an
identical hardware algorithm.
The last section is SINC3, a flexible multi-stage
variable order, variable decimation SINC filter.
Depending on the selected output word rate in the
FILTCFG register, different SINC3 stages are enabled, as shown in Table 9.
12.4SINC Filter Synchronization
12.1SINC1 Filter
The first section is SINC1, a single stage 5th order
fixed decimate by 8 SINC filter. This SINC filter
decimates the incoming 1-bit ∆Σ bit stream from
the modulators down to a 64 kHz rate.
DS612F443
The SINC filter is synchronized to the external system by the MSYNC signal, which is generated
from the SYNC input. The MSYNC signal sets a
reference time (time 0) for all filter operations, and
the SINC filter is restarted to phase align with this
reference time.
FIR1 Filter - decimate by 4FIR2 Filter - decimate by 2
13.FIR FILTER
CS5376A
Figure 25. FIR Filter Block Diagram
The finite impulse response (FIR) filter block consists of two cascaded stages, FIR1 and FIR2. It
compensates for SINC filter droop and creates a
low-pass corner to block aliased components of the
input signal.
On-chip linear phase or minimum phase coefficients can be selected using a configuration command, or the coefficients can be programmed for a
custom filter response.
13.1FIR1 Filter
The FIR1 filter stage has a decimate by four architecture. It compensates for SINC filter droop and
flattens the magnitude response of the pass band.
The on-chip linear and minimum phase coefficient
sets are 48-tap, with a maximum 255 programmable coefficients. All coefficients are normalized to
24-bit two’s complement full scale, 0x7FFFFF.
The characteristic equation for FIR1 is a convolution of the input values, X(n), and the filter coefficients, h(k), to produce an output value, Y.
Y = [h(k)*X(n-k)] + [h(k+1)*X(n-(k+1))] + ...
13.2FIR2 Filter
The FIR2 filter stage has a decimate by two architecture. It creates a low-pass brick wall filter to
block aliased components of the input signal.
The on-chip linear and minimum phase coefficient
sets are 126-tap, with a maximum 255 programmable coefficients. All coefficients are normalized to
24-bit two’s complement full scale, 0x7FFFFF.
The characteristic equation for FIR2 is a convolution of the input values, X(n), and the filter coefficients, h(k), to produce an output value, Y.
Y = [h(k)*X(n-k)] + [h(k+1)*X(n-(k+1))] + ...
13.3On-Chip FIR Coefficients
Two sets of on-chip linear phase and minimum
phase coefficients are available for FIR1 and FIR2.
Performance of the on-chip coefficient sets is very
good, with excellent ripple and stop band characteristics as described in Figure 26 and Table 12.
Which on-chip coefficient set to use is selected by
a data word following the ‘Write ROM Coefficients’ configuration command. See “Filter Coefficient Selection” on page 41 for information about
selecting on-chip coefficient sets.
DS612F447
CS5376A
13.4Programmable FIR Coefficients
A maximum of 255 + 255 coefficients can be programmed into FIR1 and FIR2 to create a custom
filter response. The total number of coefficients for
the FIR filter is fundamentally limited by the available computation cycles in the digital filter, which
itself is determined by the digital filter clock rate.
Custom filter sets should normalize the maximum
coefficient value to 24-bit two’s complement full
scale, 0x7FFFFF, and scale all other coefficients
accordingly. To maintain maximum internal dynamic range, the CS5376A FIR filter performs
double precision calculations with an automatic
gain correction to scale the final output.
Custom FIR coefficients are uploaded using the
‘Write FIR Coefficients’ configuration command.
See “EEPROM Configuration Commands” on
page 28 or “Microcontroller Configuration Commands” on page 35 for information about writing
custom FIR coefficients.
13.5FIR Filter Synchronization
The FIR1 and FIR2 filters are synchronized to the
external system by the MSYNC signal, which is
generated from the SYNC input. The MSYNC signal sets a reference time (time 0) for all filter operations, and the FIR filters are restarted to phase
align with this reference time.
48DS612F4
FIR1 – Single stage, fixed decimate by 4
Coefficient set 0: linear phase decimate by 4, 48 coefficients
Coefficient set 1: minimum phase decimate by 4, 48 coefficients
SINC droop compensation filter
FIR2 – Single stage, fixed decimate by 2
Coefficient set 0: linear phase decimate by 2, 126 coefficients
Coefficient set 1: minimum phase decimate by 2, 126 coefficients
Brick wall low-pass filter, flat to 40% f
Combined SINC + FIR digital filter specifications
Passband ripple less than +/- 0.01 dB below 40% f
Transition band -3 dB frequency at 42.89% f
Stopband attenuation greater than 130 dB above 50% f
FIR1 (Coefficient set 0)
Low pass, SINC compensation
Linear phase decimate by 4
48 coefficients
= 558 h24 = 8388607
h
0
= 1905 h25 = 7042723
h
1
= 3834 h26 = 4768946
h
2
= 5118 h27 = 2266428
h
3
= 365 h28 = 189436
h
4
= -14518 h29 = -1053303
h
5
= -39787 h30 = -1392827
h
6
= -67365 h31 = -1084130
h
7
= -69909 h32 = -496361
h
8
= -19450 h33 = 39864
h
9
= 97434 h34 = 332367
h
10
= 258881 h35 = 375562
h
11
= 375562 h36 = 258881
h
12
= 332367 h37 = 97434
h
13
= 39864 h38 = -19450
h
14
= -496361 h39 = -69909
h
15
= -1084130 h40 = -67365
h
16
= -1392827 h41 = -39787
h
17
= -1053303 h42 = -14518
h
18
= 189436 h43 = 365
h
19
= 2266428 h44 = 5118
h
20
= 4768946 h45 = 3834
h
21
= 7042723 h46 = 1905
h
22
= 8388607 h47 = 558
h
23
FIR1 (Coefficient set 1)
Low pass, SINC compensation
Minimum phase decimate by 4
48 coefficients
= 3337 h24 = 555919
h
0
= 22258 h25 = -165441
h
1
= 88284 h26 = -581479
h
2
= 266742 h27 = -617500
h
3
= 655747 h28 = -388985
h
4
= 1371455 h29 = -99112
h
5
= 2502684 h30 = 114761
h
6
= 4031988 h31 = 186557
h
7
= 5783129 h32 = 141374
h
8
= 7396359 h33 = 58582
h
9
= 8388607 h34 = -12664
h
10
= 8325707 h35 = -42821
h
11
= 6988887 h36 = -35055
h
12
= 4531706 h37 = -16792
h
13
h
= 1507479 h38 = 367
14
= -1319126 h39 = 7929
h
15
= -3207750 h40 = 5926
h
16
= -3736028 h41 = 2892
h
17
= -2980701 h42 = 23
h
18
= -1421498 h43 = -1164
h
19
= 237307 h44 = -538
h
20
= 1373654 h45 = -238
h
21
= 1711919 h46 = 18
h
22
= 1322371 h47 = 113
h
23
Table 14. FIR1 Coefficients
CS5376A
DS612F452
Filter Type Filter Coefficients
(normalized 24-bit)
FIR2 (Coefficient set 0)
Low pass, passband to 40% f
Linear phase decimate by 2
126 coefficients
s
h0 = -71 h63 = 8388607
h
= -371 h64 = 3875315
1
h
= -870 h65 = -766230
2
= -986 h66 = -1854336
h
3
h
= 34 h67 = -137179
4
= 1786 h68 = 1113788
h
5
h
= 2291 h69 = 454990
6
= 291 h70 = -642475
h
7
h
= -2036 h71 = -553873
8
= -943 h72 = 298975
h
9
h
= 2985 h73 = 533334
10
= 3784 h74 = -49958
h
11
h
= -1458 h75 = -443272
12
h
= -5808 h76 = -116005
13
h
= -1007 h77 = 318763
14
= 7756 h78 = 208018
h
15
h
= 5935 h79 = -187141
16
= -7135 h80 = -238025
h
17
h
= -11691 h81 = 68863
18
= 3531 h82 = 221211
h
19
h
= 17500 h83 = 22850
20
= 4388 h84 = -174452
h
21
h
= -20661 h85 = -81993
22
= -15960 h86 = 114154
h
23
h
= 18930 h87 = 109009
24
= 29808 h88 = -54172
h
25
= -9795 h89 = -109189
h
26
h
= -42573 h90 = 4436
27
= -7745 h91 = 90744
h
28
h
= 49994 h92 = 29702
29
= 33021 h93 = -62651
h
30
h
= -47092 h94 = -47092
31
= -62651 h95 = 33021
h
32
h
= 29702 h96 = 49994
33
= 90744 h97 = -7745
h
34
h
= 4436 h98 = -42573
35
= -109189 h99 = -9795
h
36
h
= -54172 h
37
= 109009 h
h
38
h
= 114154 h
39
= -81993 h
h
40
h
= -174452 h
41
= 22850 h
h
42
h
= 221211 h
43
= 68863 h
h
44
h
= -238025 h
45
= -187141 h
h
46
h
= 208018 h
47
= 318763 h
h
48
h
= -116005 h
49
= -443272 h
h
50
= -49958 h
h
51
h
= 533334 h
52
= 298975 h
h
53
h
= -553873 h
54
= -642475 h
h
55
h
= 454990 h
56
= 1113788 h
h
57
h
= -137179 h
58
= -1854336 h
h
59
h
= -766230 h
60
= 3875315 h
h
61
h
= 8388607 h
62
= 29808
100
= 18930
101
= -15960
102
= -20661
103
= 4388
104
= 17500
105
= 3531
106
= -11691
107
= -7135
108
= 5935
109
= 7756
110
= -1007
111
= -5808
112
= -1458
113
= 3784
114
= 2985
115
= -943
116
= -2036
117
= 291
118
= 2291
119
= 1786
120
= 34
121
= -986
122
= -870
123
= -371
124
= -71
125
Table 15. FIR2 Linear Phase Coefficients
CS5376A
DS612F453
Filter Type Filter Coefficients
(normalized 24-bit)
FIR2 (Coefficient set 1)
Low pass, passband to 40% f
Minimum phase decimate by 2
126 coefficients
s
h0 = 4019 h63 = 67863
h
= 43275 h64 = -190800
1
h
= 235427 h65 = -128546
2
= 848528 h66 = 114197
h
3
h
= 2240207 h67 = 147750
4
h
= 4525758 h68 = -46352
5
h
= 7077833 h69 = -143269
6
h
= 8388607 h70 = -13290
7
h
= 6885673 h71 = 114721
8
h
= 2483461 h72 = 51933
9
h
= -2538963 h73 = -75952
10
h
= -4800543 h74 = -68746
11
h
= -2761696 h75 = 38171
12
h
= 1426109 h76 = 68492
13
h
= 3624338 h77 = -7856
14
h
= 1820814 h78 = -57526
15
h
= -1695825 h79 = -12540
16
= -2885148 h80 = 41717
h
17
h
= -605252 h81 = 23334
18
h
= 2135021 h82 = -25516
19
h
= 1974197 h83 = -26409
20
h
= -630111 h84 = 11717
21
h
= -2168177 h85 = 24246
22
h
= -750147 h86 = -1620
23
h
= 1516192 h87 = -19248
24
h
= 1550127 h88 = -4610
25
h
= -508445 h89 = 13356
26
h
= -1686937 h90 = 7526
27
h
= -437822 h91 = -7887
28
h
= 1308705 h92 = -8016
29
h
= 1069556 h93 = 3559
30
h
= -657282 h94 = 7023
31
h
= -1301014 h95 = -598
32
h
= -30654 h96 = -5350
33
h
= 1173754 h97 = -1097
34
h
= 579643 h98 = 3579
35
h
= -803111 h99 = 1806
36
h
= -895851 h
37
h
= 328399 h
38
h
= 962522 h
39
h
= 124678 h
40
h
= -820948 h
41
h
= -466657 h
42
h
= 545674 h
43
= 652827 h
h
44
h
= -220448 h
45
h
= -680495 h
46
h
= -80886 h
47
h
= 578844 h
48
= 306445 h
h
49
h
= -395302 h
50
= -431004 h
h
51
h
= 181900 h
52
= 454403 h
h
53
h
= 15856 h
54
= -395525 h
h
55
h
= -166123 h
56
h
= 284099 h
57
h
= 253485 h
58
h
= -152407 h
59
h
= -277888 h
60
h
= 28526 h
61
h
= 250843 h
62
= -2058
100
= -1859
101
= 936
102
= 1558
103
= -224
104
= -1129
105
= -152
106
= 718
107
= 290
108
= -395
109
= -290
110
= 178
111
= 227
112
= -53
113
= -151
114
= -5
115
= 86
116
= 23
117
= -42
118
= -22
119
= 17
120
= 14
121
= -5
122
= -7
123
= 1
124
= 3
125
Table 16. FIR2 Minimum Phase Coefficients
CS5376A
DS612F454
CS5376A
1st Order IIR1
b
10
-1
Z
-a
11
b
11
3rd Order IIR3 implemented by
running both IIR1 and IIR2 stages
Figure 28. IIR Filter Block Diagram
14.IIR FILTER
The infinite impulse response (IIR) filter block
consists of two cascaded stages, IIR1 and IIR2. It
creates a high-pass corner to block very low-frequency and DC components of the input signal.
2nd Order IIR2
b
-a
-a
21
22
20
-1
Z
b
21
-1
Z
b
22
The characteristic equations for the 1st order IIR
include an input value, X, an output value, Y, and
two intermediate values, W1 and W2, separated by
a delay element (z
-1
).
On-chip IIR1 and IIR2 coefficients can be selected
using a configuration command, or the coefficients
can be programmed for a custom filter response.
14.1IIR Architecture
The architecture of the IIR filter is automatically
determined when the output filter stage is selected
in the FILTCFG register. Selecting the 1st order
IIR1 filter bypasses the 2nd order stage, while selecting the 2nd order IIR2 filter bypasses the 1st order stage. Selection of the 3rd order IIR3 filter
enables both the 1st and 2nd order stages.
14.2IIR1 Filter
The 1st order IIR filter stage is a direct form filter
with three coefficients: a11, b10, and b11. Coefficients of a 1st order IIR are inherently normalized
to one, and should be scaled to 24-bit two’s complement full scale, 0x7FFFFF.
W2 = W1
W1 = X + (-a11 * W2)
Y = (W1 * b10) + (W2 * b11)
14.3IIR2 Filter
The 2nd order IIR filter stage is a direct form filter
with five coefficients: a21, a22, b20, b21, and b22.
Coefficients of a 2nd order IIR are inherently normalized to two, and should be scaled to 24-bit
two’s complement full scale, 0x7FFFFF. Normalization effectively divides the 2nd order coefficients in half relative to the input, and requires
modification of the characteristic equations.
The characteristic equations for the 2nd order IIR
include an input value, X, an output value, Y, and
three intermediate values, W3, W4, and W5, each
separated by a delay element (z-1). The following
DS612F455
CS5376A
characteristic equations model the operation of the
2nd order IIR filter with unnormalized coefficients.
cients to perform the 2nd order IIR filter calculation, which changes the algorithm slightly. The
following characteristic equations model the operation of the 2nd order IIR filter when using normalized coefficients.
The 3rd order IIR filter is implemented by running
both the 1st order and 2nd order IIR filter stages. It
can be modeled by cascading the characteristic
equations of the 1st order and 2nd order IIR stages.
14.5On-Chip IIR Coefficients
Five sets of on-chip coefficients are available for
IIR1 and IIR2, each providing a 3 Hz high-pass
Butterworth response at different output word
rates. Characteristics of the on-chip coefficient sets
are described in Figure 29 and Table 17.
Which on-chip coefficient set to use is selected by
a data word following the ‘Write ROM Coefficients’ configuration command. See “Filter Coefficient Selection” on page 41 for information about
selecting on-chip coefficient sets.
14.6Programmable IIR Coefficients
A maximum of 3 + 5 coefficients can be programmed into IIR1 and IIR2 to create a custom filter response. Custom filter sets should normalize
the coefficients to 24-bit two’s complement full
scale, 0x7FFFFF. To maintain maximum internal
dynamic range, the CS5376A IIR filter performs
double precision calculations with an automatic
gain correction to scale the final output.
Custom IIR coefficients are uploaded using the
‘Write IIR Coefficients’ configuration command.
See “EEPROM Configuration Commands” on
page 28 or “Microcontroller Configuration Commands” on page 35 for information about writing
custom IIR coefficients.
14.7IIR Filter Synchronization
The IIR filter is not synchronized to the external
system directly, only indirectly through the synchronization of the SINC and FIR filters. Because
IIR filters have ‘infinite’ memory, a discontinuity
in the input data stream from a synchronization
event can require significant time to settle out. The
exact settling time depends on the size of the discontinuity and the filter coefficient characteristics.
56DS612F4
IIR1 – Single stage, no decimation
st
order no decimation, 3 coefficients
1
Coefficient set 0: high-pass, corner 0.15% f
Coefficient set 1: high-pass, corner 0.30% f
Coefficient set 2: high-pass, corner 0.60% f
Coefficient set 3: high-pass, corner 0.90% f
Coefficient set 4: high-pass, corner 1.20% f
IIR2 – Single stage, no decimation
nd
order no decimation, 5 coefficients
2
Coefficient set 0: high-pass, corner 0.15% f
Coefficient set 1: high-pass, corner 0.30% f
Coefficient set 2: high-pass, corner 0.60% f
Coefficient set 3: high-pass, corner 0.90% f
Coefficient set 4: high-pass, corner 1.20% f
IIR3 – Two stage, no decimation
rd
order no decimation, 8 coefficients
3
(Combined IIR1 and IIR2 filter responses)
Coefficient set 0,0: high-pass, corner 0.20% f
Coefficient set 1,1: high-pass, corner 0.41% f
Coefficient set 2,2: high-pass, corner 0.82% f
Coefficient set 3,3: high-pass, corner 1.22% f
Coefficient set 4,4: high-pass, corner 1.63% f
The CS5376A digital filter can apply independent
gain and offset corrections to the data of each measurement channel. Also, an offset calibration algorithm can automatically calculate offset correction
values for each channel.
Gain correction values are written to the GAINx
registers (0x21-0x24), while offset correction values are written to the OFFSETx registers (0x250x28). Gain and offset corrections are enabled by
the USEGR and USEOR bits in the FILTCFG register (0x20).
FIR
Filters
Output to High Speed Serial Data Port (SD Port)
Output Rate 4000 SPS ~ 1 SPS
IIR
Filter
nally calculated correction values to be written into
the GAINx registers (0x21-0x24).
Gain correction values are 24-bit two’s complement with unity gain defined as full scale,
0x7FFFFF. Gain correction always scales to a fractional value, and can never gain the digital filter
data greater than one.
Output Value = Data * (GAIN / 0x7FFFFF)
Unity Gain: GAIN = 0x7FFFFF
50% Gain: GAIN = 0x3FFFFF
When enabled, the offset calibration algorithm will
automatically calculate offset correction values for
each channel and write them into the OFFSETx
registers. Offset calibration is enabled by writing
Zero Gain: GAIN = 0x000000
Once the GAIN registers are written, the USEGR
bit in the FILTCFG register enables gain correction.
the EXP and ORCAL bits in FILTCFG.
15.2Offset Correction
15.1Gain Correction
Offset correction in the CS5376A cancels the DC
Gain correction in the CS5376A normalizes sensor
gains in multi-sensor networks. It requires exter-
DS612F459
bias of a measurement channel by subtracting the
CS5376A
value in the OFFSETx registers (0x25-0x28) from
the digital filter output data word.
Offset correction values are 24-bit two’s complement with a maximum positive value of 0x7FFFFF,
and a maximum negative value of 0x800000. If applying an offset correction causes the final result to
exceed a 24-bit two’s complement maximum, the
output data will saturate to that maximum value.
Output Data = Input Data - Offset Correction
Max Positive Output Value = 0x7FFFFF
Max Negative Output Value = 0x800000
Once the OFFSET registers are written, the USEOR bit in the FILTCFG register enables offset correction.
15.3Offset Calibration
An offset calibration algorithm in the CS5376A
can automatically calculate offset correction values. When using the offset calibration algorithm,
background noise data should be used as the basis
for calculating the offset value of each measurement channel.
The offset calibration algorithm is an exponential
averaging function that places increased weight on
more recent digital filter data. The exponential
weighting factor is set by the EXP bits in the
FILTCFG register, with larger exponent values
producing a smoother averaging function that requires a longer settling time, and smaller values
producing a noisier averaging function that requires a shorter settling time. Typical exponential
values range from 0x05 to 0x0F, depending on the
available settling time.
The characteristic equations of the offset calibration algorithm include an input value, X, an output
value, Y, a summation value, YSUM, a sample index, n, and an exponential value, EXP.
Once the EXP bits are written, the ORCAL bit in
the FILTCFG register is set to enable offset calibration. When enabled, updated offset correction values are automatically written to the OFFSETx
registers. When the offset calibration algorithm is
fully settled, the ORCAL bit is cleared to maintain
the final values in the OFFSETx registers.
60DS612F4
CS5376A
System Telemetry
Token Out
Data Ready
Clock Out
Data In
Token In
Figure 31. Serial Data Port Block Diagram
16.SERIAL DATA PORT
Once digital filtering is complete, each 24-bit output sample is combined with an 8-bit status byte.
These 32-bit data words are written to an 8-deep
FIFO buffer and then transmitted to the communications channel through a high speed serial data
port (SD port).
16.1Pin Descriptions
SDTKI - Pin 64
Token input, requests an SD port transaction.
SDRDY - Pin 61
Data ready output signal, active low. Open drain
output requiring a 10 kΩ pull-up resistor.
CS5376A
SDTKI
SDRDY
SDCLK
SDDAT
SDTKO
16.2SD Port Data Format
Serial data transactions transfer 32-bit words. Each
word consists of an 8-bit status byte followed by a
24-bit output sample. The status byte, shown in
Figure 32, has an MFLAG bit, channel bits, a time
break bit, and a FIFO overflow bit.
MFLAG Bit - MFLAG
The MFLAG bit is set when an MFLAG signal is
received on the MFLAG1-MFLAG4 pins. When
received, that channel MFLAG bit is set in the next
output word. See “Modulator Interface” on page 39
for more information about MFLAG.
Channel Bits - CH[1:0]
SDCLK - Pin 62
Serial clock input.
Channel bits indicate from which conversion channel the data word is from. The channel number,
CH[1:0], is zero based.
SDDAT - Pin 60
CH[1:0] = 00 = Channel 1
Serial data output. Data valid on rising edge of
SDCLK, transition on falling edge.
SDTKO - Pin 63
Token output, ends an SD port transaction. Passes
through the SDTKI signal when no data is available
in the SD port output FIFO.
The time break bit marks a timing reference based
on a rising edge into the TIMEB pin. After a programmed delay, the TB bit in the status byte is set
for one output sample in all channels. The TIME-
BRK digital filter register (0x29) programs the
sample delay for the TB bit output. See “Time
Break Controller” on page 67 for more information
about time break.
FIFO Overflow Bit - W
The FIFO overflow bit indicates an error condition
in the SD port data FIFO, and is set if new digital
filter data overwrites a FIFO location containing
data which has not yet been sent.
The W bit is sticky, meaning it persists indefinitely
once set. Clearing the W bit requires sending the
‘Filter Stop’ and ‘Filter Start’ configuration commands to reinitialize the data FIFO.
Conversion Data Word
The lower 24-bits of the SD port output data word
is the conversion sample for the specified channel.
Conversion data is 24-bit two’s complement format.
16.3SD Port Transactions
The SD port can operate in two modes depending
how the SDTKI pin is connected: request mode
where data is output when requested by the communications channel, or continuous mode where
data is output immediately when ready.
16.3.1Request Mode
To initiate SD port transactions on request, SDTKI
is connected to an active high polling signal from
the communications channel. A rising edge into
SDTKI when new data is available in the SD port
FIFO causes the CS5376A to initiate an SD port
transaction by driving SDRDY
yet available in the SD port FIFO, the SDTKI signal is passed through to the SDTKO output.
Once an SD port transaction is initiated, serial
clocks into SDCLK cause data to be output to
SDDAT, as shown in Figure 33. When all available
low. If data is not
62DS612F4
SDTKI
SDTKO
SDRDY
SDCLK
CS5376A
SDDAT
MSB
Figure 33. SD Port Transaction
data is read from the SD port data FIFO, SDRDY is
released and SDTKO is pulsed high for 100 nS.
16.3.2Continuous Mode
To have the CS5376A automatically initiate SD
port transactions whenever data becomes available,
connect SDTKI to a 4 MHz or slower clock source
such as MCLK/2. The first rising edge into SDTKI
after data becomes available in the SD port FIFO
LSB
causes the CS5376A to initiate an SD port transaction by driving SDRDY
low. If data is not available
in the SD port FIFO, the SDTKI signal is passed
through to the SDTKO output.
Once an SD port transaction is initiated, serial
clocks into SDCLK cause data to be output to
SDDAT, as shown in Figure 33. When all available
data is read from the SD port data FIFO, SDRDY
is
released and SDTKO is pulsed high for 100 nS.
DS612F463
Digital Filter
Data Bus
CS5376A
24-bit
TBSGAIN Register
24-bit
Digital ∆Σ Modulator
1-bit
TBSDATA
Figure 34. Test Bit Stream Generator Block Diagram
17.TEST BIT STREAM GENERATOR
The CS5376A test bit stream (TBS) generator creates sine wave ∆Σ bit stream data to drive an external test DAC. The TBS digital output can also be
internally connected to the MDATA inputs for
loopback testing of the digital filter.
17.1Pin Descriptions
TBSDATA - Pin 9
Test bit stream 1-bit ∆Σ data output.
TBSCLK - Pin 8
Test bit stream clock output. Not used by the
CS4373A test DAC.
17.2TBS Architecture
The test bit stream generator consists of a data interpolator and a digital ∆Σ modulator. It receives
periodic 24-bit data from the digital filter to create
a 1-bit ∆Σ data output on the TBSDATA pin. It also
creates a clock signal at the data rate, output to the
TBSCLK pin.
The TBS input data from the digital filter is scaled
by the TBSGAIN register (0x2B). Maximum stable
amplitude is 0x04FFFF, with 0x04B8F2 approximately full scale for the CS4373A test DAC. The
TBSCFG Register
Clock Generation
TBSCLK
full scale 1-bit ∆Σ output from the TBS generator is
defined as 25% minimum and 75% maximum
one’s density.
17.3TBS Configuration
Configuration options for the TBS generator are set
through the TBSCFG register (0x2A). Gain scaling
of the TBS generator output is set by the TBSGAIN
register (0x2B).
Interpolation Factor - INTP[7:0]
Selects how many times the interpolator uses a data
point when generating the output bit stream. Interpolation is zero based and represents one greater
than the programmed register value.
Clock Rate - RATE[2:0]
Selects the TBSDATA and TBSCLK output rate.
Synchronization - TSYNC
Enables synchronization of the TBS output phase
to the MSYNC signal.
Clock Delay - CDLY[2:0]
Programs a fractional delay for TBSCLK with a 1/8
clock period resolution.
Enables digital loopback from the TBS output to
the MDATA inputs.
Run - RUN
Enables the test bit stream generator.
Data Delay - DDLY[5:0]
Programs full period delays for TBSDATA, up to a
maximum of 63 bits.
Gain - TBSGAIN[23:0]
Scales the amplitude of the sine wave output. Maximum 0x04FFFF, nominal 0x04B8F2.
17.4TBS Data Source
Data to create test signals is loaded into digital filter memory by configuration commands. The onchip sine wave data is suitable for most tests,
Output Rate
Selection
(RATE)
Interpolation
Selection
(INTP)
though custom data is required to support custom
signal frequencies. See “EEPROM Configuration
Commands” on page 28 or “Microcontroller Configuration Commands” on page 35 for information
about programming TBS data.
TBS ROM Data
An on-chip 24-bit 1024 point digital sine wave is
stored on the CS5376A. When selected by the
‘Write TBS ROM Data’ configuration command,
the TBS generator can produce the test signal frequencies listed in Table 19. Additional discrete test
frequencies and output rates can be programmed
with the on-chip data by varying the interpolation
factor and output rate.
Custom TBS Data
If a required test frequency cannot be generated using the on-chip test bit stream data, a custom data
DS612F465
CS5376A
set can be written into the CS5376A. The number
of data points to write, up to a maximum of 1024,
depends on the required test signal frequency, output rate, and available interpolation factors. Custom data sets must be continuous on the ends; i.e.
when copied end-to-end the data set must produce
a smooth curve.
17.5TBS Sine Wave Output
The TBS generator uses data from digital filter
memory to create a sine wave test signal that can
drive a test DAC. Sine wave frequency and output
data rate are calculated as shown by the characteristic equation of Table 19.
The sine wave maximum ∆Σ one’s density output
from the TBS generator is set by the TBSGAIN
register. TBSGAIN can be programmed up to a
maximum of 0x04FFFF, with the TBS generator
unstable for higher amplitudes. For the CS4373A
test DAC, a gain value of 0x04B8F2 produces an
approximately full scale sine wave output (5 V
differential).
pp
MDATA inputs. This loopback mode provides a
fully digital signal path to test the TBS generator,
digital filter, and data collection interface. Digital
loopback testing expects 512 kHz ∆Σ data for the
MDATA inputs.
A mismatch of the TBS generator full scale output
and the MDATA full scale input results in an amplitude mismatch when testing in loopback mode.
The TBS generator outputs a 75% maximum one’s
density, while the MDATA inputs expect an 86%
maximum one’s density from a ∆Σ modulator, resulting in a measured full scale error of -3.6 dB.
17.7TBS Synchronization
When the TSYNC bit is set in the TBSCFG register, the MSYNC signal resets the sine wave data
pointer and phase aligns the TBS signal output.
Once the digital filter is settled, all CS5376A devices receiving the SYNC signal will have identical
TBS signal phase. See “Synchronization” on
page 25 for more information about the SYNC and
MSYNC signals.
17.6TBS Loopback Testing
Included as part of the CS5376A test bit stream
generator is a feedback path to the digital filter
If TSYNC is clear, MSYNC has no effect on the
TBS data pointer and no change in the TBS output
phase will occur during synchronization.
66DS612F4
CS5376A
TIMEB
Figure 35. Time Break Block Diagram
TIMEBRK
Delay Counter
18.TIME BREAK CONTROLLER
A time break signal is used to mark timing events
that occur during measurement. An external signal
sets a flag in the status byte of an output sample to
mark when the external event occurred.
A rising edge input to the TIMEB pin causes the
TB timing reference flag to be set in the SD port
status byte. When set, the TB flag appears for only
one output sample in the status byte of all enabled
channels. The TB flag output can be delayed by
programming a sample delay value into the TIMEBRK digital filter register.
TB Flag
in SD Port
Status Byte
18.3Time Break Delay
The TIMEBRK register (0x29) sets a sample delay
between a received rising edge on the TIMEB pin
and writing the TB flag into the SD port status byte.
The programmable sample counter can compensate
for group delay through the digital filters. When the
proper group delay value is programmed into the
TIMEBRK register, the TB flag will be set in the
status byte of the measurement sample taken when
the timing reference signal was received.
18.1Pin Description
TIMEB - Pin 57
Time break input pin, rising edge triggered.
18.2Time Break Operation
An externally generated timing reference signal applied to the TIMEB pin initiates an internal sample
counter. After a number of output samples have
passed, programmed in the TIMEBRK digital filter
register (0x29), the TB flag is set in the status byte
of the SD port output word for all enabled channels.
The TB flag is automatically cleared for subsequent data words, and appears for only one output
sample in each channel.
18.3.1Step Input and Group Delay
A simple method to empirically measure the step
response and group delay of a CS5376A measurement channel is to use the time break signal as both
a timing reference input and an analog step input.
When a rising edge is received on the TIMEB pin
with no delay programmed into the TIMEBRK register, the TB flag is set in the next SD port status
byte. The same rising edge can act as a step input to
the analog channel, propagating through the digital
filter to appear as a rising edge in the measurement
data. By comparing the timing of the TB status flag
output and the rising edge in the measurement data,
the measurement channel group delay can be determined.
DS612F467
CS5376A
GP_PULL
CS output from SPI
Data bit
GP_DATA
GP_DIR
Figure 36. GPIO Bi-directional Structure
19.GENERAL PURPOSE I/O
The General Purpose I/O (GPIO) block provides 12
general purpose pins to interface with external
hardware.
19.1Pin Descriptions
GPIO[4:0]:CS[4:0] - Pins 32 - 36
Standard GPIO pins also used as SPI 2 chip selects.
GPIO[5:10] - Pins 37, 41 - 45
Standard GPIO pins.
GPIO11:EECS - Pin 46
Standard GPIO pin also used as an SPI 1 chip select
when booting from an external EEPROM.
Pull Up
Logic
R
GPIO/CS
sponding GPIO pin should be initialized as output
mode and logical 1 to produce the chip select falling edge.
19.3GPIO Registers
When used as standard GPIO pins, settings are programmed in the GPCFG0 and GPCFG1 registers.
GP_DIR bits set the input/output mode, GP_PULL
bits enable/disable the internal pull-up resistor, and
GP_DATA bits set the output data value. After reset, GPIO pins default as inputs with pull-up resistors enabled.
19.4GPIO Input Mode
19.2GPIO Architecture
Each GPIO pin can be configured as input or output, high or low, with a weak (~200 kΩ) internal
pull-up resistor enabled or disabled. Several GPIO
pins also double as chip selects for the SPI 1 and
When reading a value from the GP_DATA bits, the
returned data reports the current state of the pins. If
a pin is externally driven high it reads a logical 1, if
externally driven low it reads a logical 0. When a
GPIO pin is used as an input, the pull-up resistor
should be disabled to save power if it isn’t required.
SPI 2 serial ports. Figure 36 shows the structure of
a bi-directional GPIO pin with SPI chip select func-
19.5GPIO Output Mode
tionality.
When a GPIO pin is programmed as an output with
When the CS5376A is used as an SPI master, either
when booting from EEPROM using SPI 1 or performing master mode transactions using SPI 2, the
chip select signals from SPI 1 and SPI 2 are logically AND-ed with the GPIO data bit. The corre-
68DS612F4
a data value of 0, the pin is driven low and the internal pull-up resistor is automatically disabled.
When programmed as an output with a data value
of 1, the pin is driven high and the pull-up resistor
is inconsequential.
CS5376A
Any GPIO pin can be used as an open-drain output
by setting the data value to 0, enabling the pull-up,
and using the GP_DIR direction bits to control the
pin value. This open-drain output configuration
uses the internal pull-up resistor to hold the pin
high when GP_DIR is set as an input, and drives the
pin low when GP_DIR is set as an output.
19.5.1GPIO Reads in Output Mode
When reading GPIO pins the GP_DATA register
value always reports the current state of the pins, so
a value written in output mode does not necessarily
read back the same value. If a pin in output mode is
written as a logical 1, the CS5376A attempts to
drive the pin high. If an external device forces the
pin low, the read value reflects the pin state and returns a logical 0. Similarly, if an output pin is written as a logical 0 but forced high externally, the
read value reflects the pin state and returns a logical
1. In both cases the CS5376A is in contention with
the external device resulting in increased power
consumption.
DS612F469
Digital
Filter
SCKFS[2:0] / SCKPO / SCKPH
SPI2EN[4:1] / RCH[1:0]
Pin logic
4:1
CS5376A
SCK2
SO
SI1
SI2
SI3
SI4
CS[4:0]
Figure 37. Serial Peripheral Interface 2 (SPI 2) Block Diagram
20.SERIAL PERIPHERAL INTERFACE 2
The Serial Peripheral Interface 2 (SPI 2) port is a
master mode SPI port designed to interface with serial peripherals. By writing the SPI2 digital filter
registers, multiple serial slave devices can be controlled through the CS5376A.
20.1Pin Descriptions
CS[4:0] - Pins 32 - 36
Serial chip selects. Multiplexed with GPIO pins.
SCK2 - Pin 31
Serial clock output, common to all channels.
SO - Pin 30
Serial data output, common to all channels.
SI[4:1] - Pins 26 - 29
CS0
Select
logic
CS1
CS2
CS3
CS4
To GPIO Block
is selected by bits in the SPI2CTRL digital filter
register.
SPI 2 chip select outputs are multiplexed with
GPIO pins, which cannot perform both functions
simultaneously. When used as a chip select, the
GPIO output must be programmed high to permit
the chip select to operate as an active low signal.
See “General Purpose I/O” on page 68 for information about programming the GPIO pins.
The SPI 2 interface transfers data from the SPI 2
registers to a slave serial device and back through a
bi-directional 8-bit shift register. Serial transactions are automatic once control, command, and
data values are written into the SPI 2 digital filter
registers.
Serial data inputs.
20.3SPI 2 Registers
SPI 2 transactions are initiated by first writing
20.2SPI 2 Architecture
The SPI 2 pin interface has multiple chip selects
and serial data inputs, but a common serial clock
and serial data output. Which chip select and serial
input to use for a particular slave serial transaction
command, address, and data values to the
SPI2CMD and SPI2DAT digital filter registers,
and then writing the SPI2CTRL register to set the
D2SREQ bit. The D2SREQ bit initiates a serial
transaction using the programmed SPI2CTRL configuration.
70DS612F4
CS5376A
20.3.1SPI 2 Control Register
The SPI 2 hardware is configured by the
SPI2CTRL digital filter register (0x10).
Bits in this register select the serial input pin and
chip select pin used for a transaction, set the total
number of bytes in a transaction, initiate a serial
transaction, and report status information about a
transaction. Other bits in SPI2CTRL set hardware
configuration options such as the serial clock rate,
the SPI mode, and the state of internal pull-up resistors.
Chip Select Enable - CS[4:0]
The chip select pin to use during a transaction is selected by the CS0, CS1, CS2, CS3, and CS4 bits.
Multiple chip selects can be enabled to send a
transaction to more than one serial peripheral.
Serial Input Select - SPI2EN[4:1], RCH[1:0]
Which serial input pin will receive data is selected
using the SPI2EN bits and the RCH bits. The
SPI2EN bits enable the serial input, while the RCH
bits select it for the SPI 2 transaction.
A channel’s SPI2EN bit should always be enabled,
even when transactions do not expect to receive
data from the slave device.
Transaction Bytes - DNUM[2:0]
ports all four SPI modes, with mode 0 and mode 3
the most commonly used. Supported modes are:
The SPI 2 pins can operate in two modes depending on the WOM bit. A default push-pull configuration drives output signals both high and low.
Wired-Or mode only drives low, relying on a weak
internal pull-up resistor to pull the output high.
Wired-Or mode permits multiple serial controllers
to access the same bus without contention.
Initiating Serial Transactions - D2SREQ
Writing the D2SREQ bit starts an SPI 2 serial
transaction. When complete, the D2SREQ bit is automatically cleared by the SPI 2 hardware.
Status and Error Bits - D2SOP, SWEF, TM
Three bits in the SPI2CTRL register report status
and error information.
D2SOP is set when the SPI 2 port is busy performing a transaction. It is automatically cleared when
the transaction is completed.
DNUM bits specify the total number of bytes to
transfer during a serial transaction, including command and address bytes. DNUM is zero based and
represents one greater than the number programmed.
Serial Clock Rate - SCKFS[2:0]
The serial clock rate output from the SCK2 pin is
selected by the SCKFS bits. Serial clock rates
range from 32 kHz to 4.096 MHz.
SPI Mode - SCKPO, SCKPH
The serial mode used for a transaction depends on
the SCKPO and SCKPH bits. The SPI 2 port sup-
DS612F471
SWEF is set if a request to initiate a new transaction occurs during the current transaction. This flag
is latched and must be cleared manually.
TM is set to indicate the SPI 2 port timed out on the
requested transaction. This flag is latched and must
be cleared manually.
20.3.2SPI 2 Command Register
The SPI2CMD register (0x11) is a 16-bit digital filter register with the high byte designated as an SPI
command and the low byte designated as an address. The high byte holds an 8-bit SPI ‘write’ or
‘read’ opcode, as shown in Figure 38, and the low
byte holds an 8-bit serial address.
CS5376A
During a transaction, bits in SPI2CMD are output
MSB first, with data in SPI2DAT written or read
following.
20.3.3SPI 2 Data Register
The SPI2DAT register (0x12) is a 24-bit digital filter register containing three SPI data bytes. Data in
SPI2DAT is always LSB aligned, with 1-byte data
written or received using the low byte, 2-byte data
written or received using the middle and low bytes,
and 3-byte data written or received using all three
bytes.
Data in SPI2DAT is written or read after writing
the command and address bytes from the
SPI2CMD register.
20.4SPI 2 Transactions
The SPI 2 port operates as an SPI master to perform
write and read transactions with serial slave peripherals. The exact format of the SPI transactions depends on the SPI mode, selected using the SCKPO
and SCKPH bits in the SPI2CTRL register.
Write Transactions
Write transactions start by writing an SPI ‘write’
(0x02) opcode and an 8-bit destination address into
the SPI2CMD register and the output data value to
the SPI2DAT register. Writing the D2SREQ bit in
the SPI2CTRL register initiates the SPI 2 transaction based on the SPI2CTRL configuration.
A write transaction outputs 1 or 2 bytes from the
SPI2CMD register followed by 1, 2, or 3 bytes
from the SPI2DAT register. Write transactions are
therefore a minimum of 1 byte (DNUM = 0) and a
maximum of 5 bytes (DNUM = 4). The SPI 2 port
uses the DNUM bits in the SPI2CTRL register to
determine the total number of bytes to send during
a write transaction.
Write transactions are not required to use standard
SPI commands. If serial peripherals use non-stan-
dard write commands they can be written into
SPI2CMD and SPI2DAT as required.
Read Transactions
Read transactions start by writing an SPI ‘read’
(0x03) opcode and an 8-bit source address to the
SPI2CMD register. Writing the D2SREQ bit in the
SPI2CTRL register initiates the SPI 2 transaction
based on the SPI2CTRL configuration, with the
data value automatically received into the
SPI2DAT register.
A read transaction outputs 2 bytes from the
SPI2CMD register and can receive 1, 2, or 3 bytes
into the SPI2DAT register. Read transactions are a
minimum of 3 bytes (DNUM = 2) and a maximum
of 5 bytes (DNUM = 4). The SPI 2 port uses the
DNUM bits in the SPI2CTRL register to determine
the total number of bytes to send and receive during
a read transaction.
Read transactions are not required to use standard
SPI commands. If serial peripherals use non-standard read commands they can be written to the
SPI2CMD register, as long as they conform to the
format of 2 bytes out with 1, 2, or 3 bytes in.
SPI Modes
The SPI mode for the SPI 2 port is selected in the
SPI2CTRL register using the SCKPO and SCKPH
bits. The most commonly used SPI modes are
mode 0 and mode 3, both of which define the serial
clock with data valid on rising edges and transitioning on falling edges.
In SPI mode 0, the SCK2 serial clock is defined initially in a low state. Output data on the SO pin is
valid immediately after the chip select pin goes
low, and the first rising edge of SCK2 latches valid
data.
In SPI mode 3, the SCK2 serial clock is defined initially in a high state. Output data on the SO pin is
invalid until the initial falling edge of SCK2, and
the first rising edge of SCK2 latches valid data.
72DS612F4
CS5376A
InstructionOpcodeAddressDefinition
Write0x02SPI2CMD[7:0]Write serial peripheral beginning at the address
given in SPI2CMD[7:0].
Read0x03SPI2CMD[7:0]Read serial peripheral beginning at the address
given in SPI2CMD[7:0].
SPI 2 Write to External Slave
SPI2CMD[15:8]
SPI2CMD[7:0]
SPI2DAT
SO
0x02ADDRData1
SI
CS
SPI 2 Read from External Slave
SO
SI
CS
SPI2CMD[15:8]
0x03ADDR
Figure 38. SPI 2 Master Mode Transactions
SPI2CMD[7:0]
Data2
Data1Data3Data2
SPI2DAT
Data3
SPI modes 1 and 4 work similarly to modes 0 and
3, with the serial clock defined to have data valid on
falling edges and transitioning on rising edges.
DS612F473
SPI 2 Transaction with SCKPH=0
CS5376A
Cycle
SCK2
SCK2
SCKPO = 0
SCKPO = 1
SO
SI
Slave devices only drive SI after being selected and responding to a read command.
18276543
MSBLSB
MSBLSB612345
612345
CS
SPI 2 Transaction with SCKPH=1
Cycle
18276543
X
SCK2
SCK2
SO
SI
Slave devices only drive SI after being selected and responding to a read command.
SCKPO = 0
SCKPO = 1
MSBLSB
X
612345
CS
Figure 39. SPI 2 Transaction Details
LSBMSB612345
DS612F474
TRST
TMS
TCK
TDI
TAP
Controller
Boundary Scan Cells
Figure 40. JTAG Block Diagram
CS5376A
TDO
21.BOUNDARY SCAN JTAG
The CS5376A includes an IEEE 1149.1 boundary scan JTAG port to test PCB interconnections. Refer to
the IEEE 1149.1 specification for more information about boundary scan testing.
21.1Pin Descriptions
TRST - Pin 1
Reset input for the test access port (TAP) controller and all boundary scan cells, active low. Connect to
GND to disable the JTAG port.
TMS - Pin 2
Serial input to select the JTAG test mode.
TCK - Pin 3
Clock input to the TAP controller.
TDI - Pin 4
Serial input to the scan chain or TAP controller.
TDO - Pin 5
Serial output from the scan chain or TAP controller.
21.2JTAG Architecture
The JTAG test circuitry consists of a test access port (TAP) controller and boundary scan cells connected
to each pin. The boundary scan cells are linked together to create a scan chain around the CS5376A.
DS612F475
21.2.1JTAG Reset
As required by the IEEE 1149.1 specification, the
JTAG TRST
RESET
TRST
should be connected to ground. In systems
using the JTAG port, TRST
signal is independent of the CS5376A
signal. In systems not using the JTAG port,
and RESET should be
independently driven to provide reset capability
during boundry scan.
21.2.2TAP Controller
The test access port (TAP) controller manages
commands and data through the boundary scan
chain. It supports the four JTAG instructions and
contains the IDCODE listed in Table 20.
The TAP controller also implements the 16 JTAG
state assignments from the IEEE 1149.1 specification, which are sequenced using TMS and TCK.
21.2.3Boundary Scan Cells
The CS5376A JTAG test port provides access to all
device pins via internal boundary scan cells. When
the JTAG port is disabled, boundary scan cells are
transparent and do not affect CS5376A operation.
When the JTAG port is enabled, boundary scan
cells can write and read each pin independent of
CS5376A operation.
Boundary scan cells are serially linked to create a
scan chain around the CS5376A controlled by the
TAP controller. Table 21 lists the scan cell mapping of the CS5376A.
Encoding
76DS612F4
CS5376A
BRCPinFunctionBRCPinFunctionBRCPinFunction
1TBSCLKdata out36GPIO3data in68GPIO11data in
2TBSDATAdata out37data out69data out
3DNCdata out38output enable70output enable
4MCLK/2data out39pullup71pullup
5MCLKdata out40GPIO4data in72SSO
6MSYNCdata out41data out73output enable
7MDATA4data in42output enable74WOM
8MFLAG4data in43pullup75SCK1data in
9MDATA3data in44GPIO5data in76data out
10MFLAG3data in45data out77output enable
11MDATA2data in46output enable78WOM
12MFLAG2data in47pullup79pullup
13MDATA1data in48GPIO6data in80SSI
14MFLAG1data in49data out81MISOdata in
15GNDdata in50output enable82data out
16SI4data in51pullup83output enable
17SI3data in52GPIO7data in84WOM
18SI2data in53data out85pullup
19SI1data in54output enable86MOSIdata in
20SOdata out55pullup87data out
21WOM56GPIO8data in88output enable
22SCK2data out57data out89WOM
23WOM58output enable90pullup
24GPIO0data in59pullup91SINT
25data out60GPIO9data in92RESET
26output enable61data out93BOOTdata in
27pullup62output enable94TIMEBdata in
28GPIO1data in63pullup95CLKdata in
29data out64GPIO10data in96SYNCdata in
30output enable65data out97SDDATdata out
31pullup66output enable98output enable
32GPIO2data in67pullup99SDRDY
33data out
34output enable101SDTKOdata out
35pullup102SDTKIdata in
100SDCLKdata in
data out
data in
data out
data in
data out
Table 21. JTAG Scan Cell Mapping
DS612F477
22.DEVICE REVISION HISTORY
CS5376A
The CS5376A is a pin compatible upgrade to the
CS5376. The part family has had three revisions:
CS5376 rev A
CS5376 rev B
CS5376A rev A
The part number change for CS5376A reflects additional functionality built into the device.
22.1Changes from CS5376 rev A to
CS5376 rev B
New Sinc Filter, SINC3
Added a new sinc filter, SINC3, between the previous sinc filters and FIR1. Will permit higher decimation rates for seismology applications. Not used
for 0.25 ms, 0.5 ms, 1 ms, or 2 ms output rates to
maintain backward compatibility.
Added FIR1 Coefficients
Modified ROM Coefficient Selection Method
Changed the ROM coefficient selection routines
(SPI and EEPROM) to require a 24 bit data word.
Previously no data word was required, only the
command byte. The data word is parsed to select
the FIR1, FIR2, IIR1, and IIR2 coefficient sets.
Modified ROM TBS Data Selection Method
Changed the ROM test bit stream selection routine
(SPI and EEPROM) to require a 24 bit data word.
Previously no data word was required, only the
command byte. The data word scales the ROM test
bit stream data to a user selected amplitude.
Modified SPI port to strobe SINT pin
The SPI port now pulses the SINT pin whenever
data is received. Can be used by a microcontroller
to trigger additional data writes. Eliminates the
need to poll the e2dreq bit.
Included an improved FIR1 filter to compensate for
sinc filter droop. Previous filter had stop band frequency components up to -100 dB not removed by
the FIR2 brick wall filter. Required stop band attenuation is 130 dB minimum. Previous FIR1 filter
coefficients still included to maintain backwards
compatibility.
Added IIR Coefficients
Included 3 Hz IIR1 and IIR2 filter coefficients for
the 0.5 ms, 1 ms, 2 ms, 3 ms, and 4 ms configurations (5 sets IIR1, 5 sets IIR2). Previous
2 Hz @ 1 ms coefficient set was removed.
Modified Output Word Rate Selection
Changed the DEC bit settings in the FILTCFG register used to select an output word rate. Re-numbered to include the new 120 Hz, 60 Hz, 30 Hz,
15 Hz, and 7.5 Hz output rates. Other settings the
same for backward compatibility.
Fixed continuous synchronization operation
The synchronization operation was modified to
permit continuous re-sync. The SD port FIFO is no
longer reset by the SYNC interrupt.
Corrected EEPROM loader bug
The EEPROM loader bug is fixed. A preamble to
write required constants into memory is no longer
required.
22.2Changes from CS5376 rev B to
CS5376A rev A
Fixed synchronization repeatability bug
Identical synchronization signals previously
caused different impulse responses from multiple devices. Synchronization is now repeatable.
DS612F478
CS5376A
Modified SINC2 filter to correct gain and
timing errors
Corrected SINC2 decimate by 2 gain error
which affected 4000 SPS operation. Also modified SINC2 decimate by 16 output timing to
match output of other SINC2 rates. Previous
SINC2 decimate by 16 output was one sample
later than expected.
Corrected gain error of 333 SPS output rate
SINC architecture was modified to correct gain
error in SINC2 decimate by 12 by moving decimate by 3 stage into SINC3.
Modified SINC3 filter for new low bandwidth rates.
Newly supported output word rates are 200,
125, 100, 50, 40, 25, 20, 10, 5, 1 SPS. Older low
bandwidth rates of 120, 60, 30, 15, 7.5 SPS
were removed. No changes to 4000, 2000,
1000, 500, 333, 250 SPS rates for backwards
compatibility to CS5376 revision A/B.
Added minimum phase FIR coefficients
Minimum phase FIR1 coefficient set 1 and
FIR2 coefficient set 1 are newly available as selections for the SPI and EEPROM 'Write ROM
Coefficients' command.
Corrected IIR2/IIR3 channels 2, 3, 4 bug
When selecting IIR2 or IIR3 output, data from
channels 2, 3, and 4 were corrupted. IIR2 and
IIR3 now operate correctly for these channels.
Corrected IIR2 coefficient DC offset
IIR2 coefficient sets 0, 1, and 3 did not perfectly cancel DC due to coefficient b20, b21, b22
mismatch. New b21 IIR2 coefficients correct
this offset error.
Removed gain scale factor from 'Write TBS
ROM' command
TBS data was previously scaled during configuration by a data word following the 'Write
TBS ROM' command. Added a new TBSGAIN
register (0x2B, replacing WD_CFG) that scales
the TBS amplitude and can be modified during
normal operation.
Removed watchdog timer
The watchdog timer was removed. Replaced
WD_CFG register (0x2B) with TBSGAIN register.
Set GPIO11 as tri-state when EEPROM boot
completed
After stand-alone boot from EEPROM,
GPIO11 (acting as EEPROM chip select) was
previously driven high. This pin now tri-states
with an internal pull-up to hold it high.
Modified Test Bit Stream (TBS) to disable
loopback when TBS disabled.
If TBS loopback mode was enabled, the external MDATA inputs were disconnected from the
SINC filter even if the TBS was disabled. Now
when the TBS is disabled, loopback mode is
automatically disabled also.
DS612F479
CS5376A
Added Test Bit Stream (TBS) synchronization in sine wave mode.
The TBS sine wave phase will reset if bit 11 of
the TBSCFG register is set (TBSCFG bit 11 =
1) and a rising edge is received on the SYNC
pin. When TBSCFG bit 11 is set low (TBSCFG
bit 11 = 0), TBS phase is unaffected by the
SYNC input similar to CS5376 revision A/B.
Modified Time Break delay function.
The timing delay between receiving a rising
edge on the TIMEB pin and asserting the
TIMEB flag in the output word status bits is
corrected. In CS5376 revision A/B a '0' value in
the TIMEBREAK register (0x29) disabled the
TIMEB status bit write, and a '1' value set the
status bit in the current output word. Now, a '0'
value sets the TIMEB status bit in the current
output word, and a '1' value delays until the following word.
80DS612F4
23.REGISTER SUMMARY
23.1SPI 1 Registers
The CS5376A SPI 1 registers interface the serial port to the digital filter.
NameAddr.Type# BitsDescription
SPI1CTRLH00R/W8SPI 1 Control Register, High Byte
SPI1CTRLM01R/W8SPI 1 Control Register, Middle Byte
SPI1CTRLL02R/W8SPI 1 Control Register, Low Byte
SPI1CMDH03R/W8SPI 1 Command, High Byte
SPI1CMDM04R/W8SPI 1 Command, Middle Byte
SPI1CMDL05R/W8SPI 1 Command, Low Byte
SPI1DAT1H06R/W8SPI 1 Data 1, High Byte
SPI1DAT1M07R/W8SPI 1 Data 1, Middle Byte
SPI1DAT1L08R/W8SPI 1 Data 1, Low Byte
SPI1DAT2H09R/W8SPI 1 Data 2, High Byte
SPI1DAT2M0AR/W8SPI 1 Data 2, Middle Byte
SPI1DAT2L0BR/W8SPI 1 Data 2, Low Byte
The CS5376A digital filter registers control hardware peripherals and filtering functions.
NameAddr.Type# BitsDescription
CONFIG00R/W24Hardware Configuration
RESERVED01-0DR/W24Reserved
GPCFG00ER/W24GPIO[7:0] Direction, Pull-Up Enable, and Data
GPCFG10FR/W24GPIO[11:8] Direction, Pull-Up Enable, and Data
SPI2CTRL10R/W24SPI2 Control
SPI2CMD11R/W16SPI2 Command
SPI2DAT12R/W24SPI2 Data
RESERVED13-1FR/W24Reserved
FILTCFG20R/W24Digital Filter Configuration
GAIN121R/W24Gain Correction Channel 1
GAIN222R/W24Gain Correction Channel 2
GAIN323R/W24Gain Correction Channel 3
GAIN424R/W24Gain Correction Channel 4
OFFSET125R/W24Offset Correction Channel 1
OFFSET226R/W24Offset Correction Channel 2
OFFSET327R/W24Offset Correction Channel 3
OFFSET428R/W24Offset Correction Channel 4
TIMEBRK29R/W24Time Break Delay
TBSCFG2AR/W24Test Bit Stream Configuratio n
TBSGAIN2BR/W24Test Bit Stream Gain
SYSTEM12CR/W24User Defined System Register 1
SYSTEM22DR/W24User Defined System Register 2
VERSION2ER/W24Hardware Version ID
SELFTEST2FR/W24Self-Test Result Code