PRODUCTION DATA information is current as of publication date.
Products conform to specifications per the terms of the Texas
Instruments standard warranty. Production processing does not
necessarily include testing of all parameters.
192-kHz Sampling Rates
– Data Formats: 16-, 20-, or 24-Bit
Left-Justified, I2S, or Right-Justified Input
Data
– 64-Fs Bit-Clock Rate
– 128-, 192-, 256-, 384-, 512-, and 768-Fs
Master Clock Rates (Up to a Maximum of
50 MHz)
• Audio Processing
– 48-Bit Processing Architecture With 76 Bits
of Precision for Most Audio Processing
Features
– Volume Control Range 36 dB to –127 dB
•Master Volume Control Range of 18 dB to
–100 dB
•Eight Individual Channel Volume Control
Ranges of 18 dB to –127 dB
– Programmable Soft Volume and Mute
1
Please be aware that an important notice concerning availability, standard warranty, and use in critical applications of Texas
Instruments semiconductor products and disclaimers thereto appears at the end of this data sheet.
2PurePath Digital is a trademark of Texas Instruments.
3Matlab is a trademark of Math Works, Inc.
4All other trademarks are the property of their respective owners.
PRODUCTION DATA information is current as of publication date.
Products conform to specifications per the terms of the Texas
Instruments standard warranty. Production processing does not
necessarily include testing of all parameters.
SLES257–SEPTEMBER 2010
Update Rates
– Four Bass and Treble Tone Controls with
±18-dB Range, Selectable Corner
Frequencies, and Second-Order Slopes
•L, R, and C
•LS, RS
•LR, RR
•Sub
– Configurable Loudness Compensation
– Two Dynamic Range Compressors With Two
Thresholds, Two Offsets, and Three Slopes
– Seven Biquads Per Channel
– Full 8×8 Input Crossbar Mixer. Each
Signal-Processing Channel Input Can Be
Any Ratio of the Eight Input Channels.
– 8×2 Output Mixer – Channels 1–6. Each
Output Can Be Any Ratio of Any Two
Signal-Processed Channels.
– 8×3 Output Mixer – Channels 7 and 8. Each
Output Can Be Any Ratio of Any Three
Signal-Processed Channels.
– Three Coefficient Sets Stored on the Device
Can Be Selected Manually or Automatically
(Based on Specific Data Rates).
– DC Blocking Filters
– Able to Support a Variety of Bass
Management Algorithms
• PWM Processing
– 32-Bit Processing PWM Architecture With 40
Bits of Precision
– 8× Oversampling With Fifth-Order Noise
Shaping at 32 kHz–48 kHz, 4× Oversampling
at 88.2 kHz and 96 kHz, and 2× Oversampling
at 176.4 kHz and 192 kHz
– >102-dB Dynamic Range
– THD+N < 0.1%
– 20-Hz–20-kHz, Flat Noise Floor for 44.1-, 48-,
88.2-, 96-, 176.4-, and 192-kHz Data Rates
– Digital De-Emphasis for 32-, 44.1-, and
48-kHz Data Rates
– Flexible Automute Logic With Programmable
System Provides Clear AM Reception– Adjustable Modulation Limit
– Power-Supply Volume Control (PSVC)
1.2Overview
The TAS5508C is an 8-channel digital pulse-width modulator (PWM) that provides both advanced
performance and a high level of system integration. The TAS5508C is designed to interface seamlessly
with most audio digital signal processors. The TAS5508C automatically adjusts control configurations in
response to clock and data rate changes and idle conditions. This enables the TAS5508C to provide an
easy-to-use control interface with relaxed timing requirements.
The TAS5508C can drive eight channels of H-bridge power stages. Texas Instruments H-bridge parts
TAS5111, TAS5112, or TAS5182 with FETs are designed to work seamlessly with the TAS5508C. The
TAS5508C supports both single-ended or bridge-tied load configurations. The TAS5508C also provides a
high-performance, differential output to drive an external, differential-input, analog headphone amplifier
(such as the TPA112).
The TAS5508C uses AD modulation operating at a 384-kHz switching rate for 48-, 96-, and 192-kHz data.
The 8× oversampling combined with the fifth-order noise shaper provides a broad, flat noise floor and
excellent dynamic range from 20 Hz to 20 kHz.
The TAS5508C is a clocked slave-only device. The TAS5508C receives MCLK, SCLK, and LRCLK from
other system components. The TAS5508C accepts master clock rates of 128, 192, 256, 384, 512, and
768 Fs. The TAS5508C accepts a 64-Fs bit clock.
www.ti.com
The TAS5508C allows for extending the dynamic range by providing a power-supply volume control
(PSVC) output signal.
AVDD_PLL9P3.3-V analog power supply for PLL. This terminal can be connected to the same
AVSS5, 6PAnalog ground
AVSS_PLL8PAnalog ground for PLL. This terminal should reference the same ground as
BKND_ERR37DIPullupActive-low. A back-end error sequence is generated by applying logic low to this
DVDD15, 36P3.3-V digital power supply
DVDD_PWM54P3.3-V digital power supply for PWM
DVSS16, 34,PDigital ground
35, 38
DVSS_PWM53PDigital ground for PWM
HP_SEL12DI5 VPullupHeadphone in/out selector. When a logic low is applied, the headphone is
LRCLK26DI5 VSerial-audio data left/right clock (sampling-rate clock)
MCLK63DI5 VPulldownMCLK is a 3.3-V master clock input. The input frequency of this clock can range
MUTE14DI5 VPullupSoft mute of outputs, active-low (muted signal = a logic low, normal operation =
OSC_CAP18AOOscillator capacitor
PDN13DI5 VPullupPower down, active-low. PDN powers down all logic and stops all clocks
power source used to drive power terminal DVDD, but to achieve low PLL jitter,
this terminal should be bypassed to AVSS_PLL with a 0.1-mF low-ESR
capacitor.
terminal DVSS, but to achieve low PLL jitter, ground noise at this terminal must
be minimized. The availability of the AVSS terminal allows a designer to use
optimizing techniques such as star ground connections, separate ground planes,
or other quiet ground-distribution techniques to achieve a quiet ground reference
at this terminal.
terminal. The BKND_ERR results in no change to any system parameters, with
all H-bridge drive signals going to a hard-mute (M) state.
selected (speakers are off). When a logic high is applied, speakers are selected
(headphone is off).
from 4 MHz to 50 MHz.
a logic high). The mute control provides a noiseless volume ramp to silence.
Releasing mute provides a noiseless ramp to previous volume.
whenever a logic low is applied. The internal parameters are preserved through
a power-down cycle, as long as RESET is not active. The duration for system
recovery from power down is 100 ms.
DESCRIPTION
(1) Type: A = analog; D = 3.3-V digital; P = power/ground/decoupling; I = input; O = output
(2) All pullups are 200-mA weak pullups and all pulldowns are 200-mA weak pulldowns. The pullups and pulldowns are included to ensure
proper input logic levels if the terminals are left unconnected (pullups => logic-1 input; pulldowns => logic-0 input). Devices that drive
inputs with pullups must be able to sink 200 mA, while maintaining a logic-0 drive level. Devices that drive inputs with pulldowns must be
able to source 200 mA, while maintaining a logic-1 drive level.
23, 64
RESET11DI5 VPullupSystem reset input, active-low. A system reset is generated by applying a logic
SCL25DI5 V
SCLK27DI5 VSerial-audio data clock (shift clock) input
SDA24DIO5 V
SDIN131DI5 VPulldownSerial-audio data input 1 is one of the serial-data input ports. SDIN1 supports
SDIN230DI5 VPulldownSerial-audio data input 2 is one of the serial-data input ports. SDIN2 supports
SDIN329DI5 VPulldownSerial-audio data input 3 is one of the serial-data input ports. SDIN3 supports
SDIN428DI5 VPulldownSerial-audio data input 4 is one of the serial-data input ports. SDIN4 supports
VALID39DOOutput indicating validity of PWM outputs, active-high
VBGAP10PBand-gap voltage reference. A pinout of the internally regulated 1.2-V reference.
VR_DIG33PVoltage reference for 1.8-V digital core supply. A pinout of the internally
VR_DPLL17PVoltage reference for 1.8-V digital PLL supply. A pinout of the internally
VR_PWM48PVoltage reference for 1.8-V digital PWM core supply. A pinout of the internally
VRA_PLL1PVoltage reference for 1.8-V PLL analog supply. A pinout of the internally
VRD_PLL7PVoltage reference for 1.8-V PLL digital supply. A pinout of the internally
XTL_IN20AIXTL_OUT and XTL_IN are the only LVCMOS terminals on the device. They
XTL_OUT19AOXTL_OUT and XTL_IN are the only LVCMOS terminals on the device. They
TYPE
(1)
5-V
TOLERANT
TERMINATION
(2)
low to this terminal. RESET is an asynchronous control signal that restores the
TAS5508C to its default conditions, sets the valid output low, and places the
PWM in the hard mute (M) state. Master volume is immediately set to full
attenuation. On the release of RESET, if PDN is high, the system performs a 4to 5-ms device initialization and sets the volume at mute.
I2C serial-control clock input/output
I2C serial-control data-interface input/output
four discrete (stereo) data formats and is capable of inputting data at 64 Fs.
four discrete (stereo) data formats and is capable of inputting data at 64 Fs.
four discrete (stereo) data formats and is capable of inputting data at 64 Fs.
four discrete (stereo) data formats and is capable of inputting data at 64 Fs.
Typically has a 1-nF low-ESR capacitor between VBGAP and AVSS_PLL. This
terminal must not be used to power external devices.
regulated 1.8-V power used by digital core logic. A 4.7-mF low-ESR capacitor
should be connected between this terminal and DVSS. This terminal must not
be used to power external devices.
regulated 1.8-V power used by digital PLL logic. A 0.1-mF low-ESR capacitor
should be connected between this terminal and DVSS_CORE. This terminal
must not be used to power external devices.
regulated 1.8-V power used by digital PWM core logic. A 0.1-mF low-ESR
(3)
capacitor
terminal must not be used to power external devices.
regulated 1.8-V power used by PLL logic. A 0.1-mF low-ESR capacitor
be connected between this terminal and AVSS_PLL. This terminal must not be
used to power external devices.
regulated 1.8-V power used by PLL logic. A 0.1-mF low-ESR capacitor
be connected between this terminal and AVSS_PLL. This terminal must not be
used to power external devices.
provide a reference clock for the TAS5508C via use of an external
fundamental-mode crystal. XTL_IN is the 1.8-V input port for the oscillator
circuit. A 13.5-MHz crystal (HCM49) is recommended.
provide a reference clock for the TAS5508C via use of an external
fundamental-mode crystal. XTL_OUT is the 1.8-V output drive to the crystal. A
13.5-MHz crystal (HCM49) is recommended.
should be connected between this terminal and DVSS_PWM. This
DESCRIPTION
SLES257–SEPTEMBER 2010
(3)
should
(3)
should
(3) If desired, low-ESR capacitance values can be implemented by paralleling two or more ceramic capacitors of equal value. Paralleling
capacitors of equal value provides an extended high-frequency supply decoupling. This approach avoids the potential of producing
parallel resonance circuits that have been observed when paralleling capacitors of different values.
(3)
(3)
2.2TAS5508C Functional Description
Figure 2-1 shows the TAS5508C functional structure. The following sections describe the TAS5508C
The power-supply section contains supply regulators that provide analog and digital regulated power for
various sections of the TAS5508C. The analog supply supports the analog PLL, whereas digital supplies
support the digital PLL, the digital audio processor (DAP), the pulse-width modulator (PWM), and the
output control (reclocker). The regulators can also be turned off when terminals RESET and PDN are both
low.
2.2.2Clock, PLL, and Serial Data Interface
The TAS5508C is a clocked slave-only device that requires the use of an external 13.5-MHz crystal. It
accepts MCLK, SCLK, and LRCLK as inputs only.
The TAS5508C uses the external crystal to provide a time base for:
•Continuous data and clock error detection and management
•Automatic data-rate detection and configuration
•Automatic MCLK-rate detection and configuration (automatic bank switching)
•Supporting I2C operation/communication while MCLK is absent
The TAS5508C automatically handles clock errors, data-rate changes, and master-clock frequency
changes without requiring intervention from an external system controller. This feature significantly
reduces system complexity and design.
www.ti.com
2.2.2.1Serial Audio Interface
The TAS5508C operates as a slave-only/receive-only serial data interface in all modes. The TAS5508C
has four PCM serial data interfaces to permit eight channels of digital data to be received though the
SDIN1, SDIN2, SDIN3, and SDIN4 inputs. The serial audio data is in MSB-first, 2s-complement format.
The serial data input interface of the TAS5508C can be configured in right-justified, I2S, or left-justified
modes. The serial data interface format is specified using the I2C data-interface control register. The
supported formats and word lengths are shown in Table 2-1.
Serial data is input on SDIN1, SDIN2, SDIN3, and SDIN4. The TAS5508C accepts 16-, 20-, or 24-bit
serial data at 32, 38, 44.1, 48, 88.2, 96, 176.4, or 192 kHz in left-justified, I2S, or right-justified format.
Data is input using a 64-Fs SCLK clock and an MCLK rate of 128, 192, 256, 384, 512, or 768 Fs, up to a
maximum of 50 MHz. The clock speed and serial data format are I2C configurable.
2.2.3I2C Serial-Control Interface
The TAS5508C has an I2C serial-control slave interface (address 0x36) to receive commands from a
system controller. The serial-control interface supports both normal-speed (100 kHz) and high-speed (400
kHz) operations without wait states. Because the TAS5508C has a crystal time base, this interface
operates even when MCLK is absent.
The serial control interface supports both single-byte and multiple-byte read/write operations for status
registers and the general control registers associated with the PWM. However, for the DAP
data-processing registers, the serial control interface also supports multiple-byte (4-byte) write operations.
The I2C supports a special mode which permits I2C write operations to be broken up into multiple
data-write operations that are multiples of 4 data bytes. These are 6-byte, 10-byte, 14-byte, 18-byte, etc.,
write operations that are composed of a device address, read/write bit, subaddress, and any multiple of 4
bytes of data. This permits the system to incrementally write large register values without blocking other
I2C transactions. In order to use this feature, the first block of data is written to the target I2C address, and
each subsequent block of data is written to a special append register (0xFE) until all the data is written
and a stop bit is sent. An incremental read operation is not supported.
2.2.4Device Control
The TAS5508C control section provides the control and sequencing for the TAS5508C. The device control
provides both high- and low-level control for the serial control interface, clock and serial data interfaces,
digital audio processor, and pulse-width modulator sections.
2.2.5Digital Audio Processor (DAP)
The DAP arithmetic unit is used to implement all audio-processing functions: soft volume, loudness
compensation, bass and treble processing, dynamic range control, channel filtering, input and output
mixing. Figure 2-3 shows the TAS5508C DAP architecture.
The DAP accepts 24-bit data from the serial data interface and outputs 32-bit data to the PWM section.
The DAP supports two configurations, one for 32-kHz to 96-kHz data and one for 176.4-kHz to 192-kHz
data.
2.2.5.1TAS5508C Audio-Processing Configurations
The 32-kHz to 96-kHz configuration supports eight channels of data processing that can be configured
either as eight channels, or as six channels with two channels for separate stereo line outputs.
The 176.4-kHz to 192-kHz configuration supports three channels of signal processing with five channels
passed though (or derived from the three processed channels).
To support efficiently the processing requirements of both multichannel 32-kHz to 96-kHz data and the
2-channel 176.4-kHz and 192-kHz data, the TAS5508C has separate audio-processing features for
32-kHz to 96-kHz data rates and for 176.4 kHz and 192 kHz. See Table 2-2 for a summary of TAS5508C
processing feature sets.
2.2.5.2TAS5508C Audio Signal-Processing Functions
The DAP provides 10 primary signal-processing functions:
1. The data-processing input has a full 8×8 input crossbar mixer. This enables each input to be any ratio
of the eight input channels.
2. Two I2C programmable threshold detectors in each channel support automute.
3. Seven biquads per channel
4. Four soft bass and treble tone controls with ±18-dB range, programmable corner frequencies, and
second-order slopes. In 8-channel mode, bass and treble controls are normally configured as follows:
– Bass and treble 1: Channel 1 (left), channel 2 (right), and channel 7 (center)
– Bass and treble 2: Channel 3 (left surround) and channel 4 (right surround)
– Bass and treble 3: Channel 5 (left back surround) and channel 6 (right back surround)
– Bass and treble 4: Channel 8 (subwoofer)
5. Individual channel and master volume controls. Each control provides an adjustment range of 18 dB to
–127 dB. This permits a total volume device control range of 36 dB to –127 dB plus mute. The master
volume control can be configured to control six or eight channels. The DAP soft volume and mute
update interval is I2C programmable. The update is performed at a fixed rate regardless of the sample
rate.
6. Programmable loudness compensation that is controlled via the combination of the master and
individual volume settings.
7. Two dual-threshold dual-rate dynamic range compressors (DRCs). The volume gain values provided
are used as input parameters using the maximum RMS (master volume × individual channel volume).
8. 8×2 output mixer (channels 1–6). Each output can be any ratio of any two signal-processed channels.
9. 8×3 output mixer (channels 7 and 8). Each output can be any ratio of any three signal-processed
channels.
10. The DAP maintains three sets of coefficient banks that are used to maintain separate sets of
sample-rate-dependent parameters for the biquad, tone controls, loudness, and DRC in RAM. These
can be set to be automatically selected for one or more data sample rates or can be manually selected
under I2C program control. This feature enables coefficients for different sample rates to be stored in
the TAS5508C and then selected when needed.
Signal-processing channels86 + 23
Pass-through channelsN/A5
Master volume1 for 8 channels1 for 6 channels1 for 3 channels
Individual channel volume
controls
Four bass and treble tone controlsFour bass and treble tone controls
with ±18-dB range, programmablewith ±18-dB range, programmable
Bass and treble toneorder slopesorder slopes
controlsL, R, and C (Ch1, 2, and 7)L, R, and C (Ch1, 2, and 7)
Biquads5621
Dynamic rangeDRC1 for seven satellites andDRC1 for five satellites and DRC2DRC1 for two satellites and
compressorsDRC2 for subfor sub (Ch5 and 6 uncompressed)DRC2 for sub
Input/output mapping/
mixing
DC-blocking filters
(implemented in PWMEight channels
section)
Digital de-emphasis
(implemented in PWMN/A
section)
LoudnessEight channelsSix channelsThree channels
Number of coefficient sets
stored
corner frequencies, and second-corner frequencies, and second-
LS, RS (Ch3 and 4)LS, RS (Ch3 and 4)
LBS, RBS (Ch5 and 6)Sub (Ch8)
Sub (Ch8)Line L and R (Ch5 and 6)
Each of the eight signal-processing channel inputs can be any ratio of thebe any ratio of the eight input
eight input channels.channels.
Each of the eight outputs can be any ratio of any two processed channels. Each of the eight outputs can be
Eight channels for 32 kHz,Six channels for 32 kHz, 44.1 kHz,
44.1 kHz, and 48 kHzand 48 kHz
32 kHz–96 kHz32 kHz–96 kHz176.4- and 192-kHz
8-CHANNEL FEATURE SET6 + 2 LINEOUT FEATURE SETFEATURE SET
83
Two bass and treble tone
controls with ±18-dB range,
programmable corner
frequencies, and second-order
slopes
L and R (Ch1 and 2)
Sub (Ch8)
Each of the three signalprocessing channels or the five
pass-though channel inputs can
any ratio of any of the three
processed channels or five
bypass channels.
Three additional coefficient sets can be stored in memory.
2.3TAS5508C DAP Architecture
2.3.1TAS5508C DAP Architecture Diagrams
Figure 2-1 shows the TAS5508C DAP architecture for Fs = 96 kHz. Note the TAS5508C bass
management architecture shown in channels 1, 2, 7, and 8. Note that the I2C registers are shown to help
the designer configure the TAS5508C.
Figure 2-2 shows the TAS5508C architecture for Fs = 176.4 kHz or Fs = 192 kHz. Note that only channels
1, 2, and 8 contain all the features. Channels 3–7 are pass-through except for master volume control.
Figure 2-3 shows TAS5508C detailed channel processing. The output mixer is 8×2 for channels 1–6 and
The architecture of the TAS5508C is contained in ROM resources within the TAS5508C and cannot be
altered. However, mixer gain, level offset, and filter tap coefficients, which can be entered via the I2C bus
interface, provide a user with the flexibility to set the TAS5508C to a configuration that achieves
system-level goals.
The firmware is executed in a 48-bit, signed, fixed-point arithmetic machine. The most significant bit of the
48-bit data path is a sign bit, and the 47 lower bits are data bits. Mixer gain operations are implemented
by multiplying a 48-bit, signed data value by a 28-bit, signed gain coefficient. The 76-bit, signed output
product is then truncated to a signed, 48-bit number. Level offset operations are implemented by adding a
48-bit, signed offset coefficient to a 48-bit, signed data value. In most cases, if the addition results in
overflowing the 48-bit, signed number format, saturation logic is used. This means that if the summation
results in a positive number that is greater than 0x7FFF FFFF FFFF (the spaces are used to ease the
reading of the hexadecimal number), the number is set to 0x7FFF FFFF FFFF. If the summation results in
a negative number that is less than 0x80000000 0000, the number is set to 0x8000 0000 0000.
2.3.2.128-Bit 5.23 Number Format
All mixer gain coefficients are 28-bit coefficients using a 5.23 number format. Numbers formatted as 5.23
numbers have 5 bits to the left of the binary point and 23 bits to the right of the binary point. This is shown
in Figure 2-4.
(1 or 0) y 23 + (1 or 0) y 22 + … + (1 or 0) y 20 + (1 or 0) y 2−1 + … + (1 or 0) y 2−4 + … + (1 or 0) y 2
−23
23 Bit22 Bit20 Bit2−1 Bit2−4 Bit2
−23
Bit
M0008-01
u
Coefficient
Digit 8
u
u uS x x x
Coefficient
Digit 7
x.
x x x
Coefficient
Digit 6
x
x x x
Coefficient
Digit 5
x
x x x
Coefficient
Digit 4
x
x x x
Coefficient
Digit 3
x
x x x
Coefficient
Digit 2
x
x x x
Coefficient
Digit 1
Fraction
Digit 5
Sign
Bit
0
Fraction
Digit 6
Fraction
Digit 4
Fraction
Digit 3
Fraction
Digit 2
Fraction
Digit 1
Integer
Digit 1
u = unused or don’t care bits
Digit = hexadecimal digit
M0009-01
TAS5508C
www.ti.com
The decimal value of a 5.23 format number can be found by following the weighting shown in Figure 2-5. If
the most significant bit is logic 0, the number is a positive number, and the weighting shown yields the
correct number. If the most significant bit is a logic 1, then the number is a negative number. In this case,
every bit must be inverted, a 1 added to the result, and then the weighting shown in Figure 2-5 applied to
obtain the magnitude of the negative number.
SLES257–SEPTEMBER 2010
Figure 2-4. 5.23 Format
Figure 2-5. Conversion Weighting Factors—5.23 Format to Floating Point
Gain coefficients, entered via the I2C bus, must be entered as 32-bit binary numbers. The format of the
32-bit number (4-byte or 8-digit hexadecimal number) is shown in Figure 2-6.
Figure 2-6. Alignment of 5.23 Coefficient in 32-Bit I2C Word
As Figure 2-6 shows, the hexadecimal (hex) value of the integer part of the gain coefficient cannot be
(1 or 0) y 223 + (1 or 0) y 222 + … + (1 or 0) y 20 + (1 or 0) y 2−1 + … + (1 or 0) y 2
−23
223 Bit222 Bit20 Bit2−1 Bit2
−23
Bit
M0008-02
TAS5508C
SLES257–SEPTEMBER 2010
concatenated with the hex value of the fractional part of the gain coefficient to form the 32-bit I2C
coefficient. The reason is that the 28-bit coefficient contains 5 bits of integer, and thus the integer part of
the coefficient occupies all of one hex digit and the most significant bit of the second hex digit. In the same
way, the fractional part occupies the lower three bits of the second hex digit, and then occupies the other
five hex digits (with the eighth digit being the zero-valued most significant hex digit).
2.3.2.248-Bit 25.23 Number Format
All level adjustment and threshold coefficients are 48-bit coefficients using a 25.23 number format.
Numbers formatted as 25.23 numbers have 25 bits to the left of the decimal point and 23 bits to the right
of the decimal point. This is shown in Figure 2-7.
www.ti.com
Figure 2-7. 25.23 Format
Figure 2-8 shows the derivation of the decimal value of a 48-bit 25.23 format number.
Figure 2-8. Alignment of 5.23 Coefficient in 32-Bit I2C Word
Two 32-bit words must be sent over the I2C bus to download a level or threshold coefficient into the
TAS5508C. The alignment of the 48-bit, 25.23 formatted coefficient in the 8-byte (two 32-bit words) I2C
word is shown in Figure 2-9.
u = unused or don’t care bits
Digit = hexadecimal digit
M0009-02
TAS5508C
www.ti.com
SLES257–SEPTEMBER 2010
2.3.2.3TAS5508C Audio Processing
Figure 2-9. Alignment of 25.23 Coefficient in Two 32-Bit I2C Words
The TAS5508C digital audio processing is designed so that noise produced by filter operations is
maintained below the smallest signal amplitude of interest, as shown in Figure 2-10. The TAS5508C
achieves this low noise level by increasing the precision of the signal representation substantially above
the number of bits that are absolutely necessary to represent the input signal.
Similarly, the TAS5508C carries additional precision in the form of overflow bits to permit the value of
intermediate calculations to exceed the input precision without clipping. The TAS5508C advanced digital
audio processor achieves both of these important performance capabilities by using a high-performance
digital audio processing architecture with a 48-bit data path, 28-bit filter coefficients, and a 76-bit
accumulator.
The TAS5508C has a full 8×8 input crossbar mixer. This mixer permits each signal processing channel
input to be any ratio of any of the eight input channels, as shown in Figure 2-11. The control parameters
for the input crossbar mixer are programmable via the I2C interface. See the Input Mixer Registers(0x41–0x48, Channels 1–8), Section 7.16, for more information.
www.ti.com
Figure 2-10. TAS5508C Digital Audio Processing
2.5Biquad Filters
For 32-kHz to 96-kHz data, the TAS5508C provides 56 biquads across the eight channels (seven per
channel).
For 176.4-kHz and 192-kHz data, the TAS5508C has 21 biquads across the three channels (seven per
channel). All of the biquad filters are second-order direct form I structure.
The direct form I structure provides a separate delay element and mixer (gain coefficient) for each node in
the biquad filter. Each mixer output is a signed 76-bit product of a signed 48-bit data sample (25.23 format
number) and a signed 28-bit coefficient (5.23 format number), as shown in Figure 2-12. The 76-bit ALU in
the TAS5508C allows the 76-bit resolution to be retained when summing the mixer outputs (filter
products).
The five 28-bit coefficients for the each of the 56 biquads are programmable via the I2C interface. See
Table 2-3.
SLES257–SEPTEMBER 2010
Figure 2-12. Biquad Filter Structure
All five coefficients for one biquad filter structure are written to one I2C register containing 20 bytes (or five
32-bit words). The structure is the same for all biquads in the TAS5508C. Registers 0x51–0x88 show all
the biquads in the TAS5508C. Note that u[31:28] bits are unused and default to 0x0.
Table 2-3. Contents of One 20-Byte Biquad Filter Register (Default = All-Pass)
From 32-kHz to 96-kHz data, the TAS5508C has four bass and treble tone controls. Each control has a
±18-dB control range with selectable corner frequencies and second-order slopes. These controls operate
four channel groups:
•L, R, and C (channels 1, 2, and 7)
•LS, RS (channels 3 and 4)
•LBS, RBS (alternatively called L and R lineout) (channels 5 and 6)
•Sub (channel 8)
For 176.4-kHz and 192-kHz data, the TAS5508C has two bass and treble tone controls. Each control has
a ±18-dB I2C control range with selectable corner frequencies and second-order slopes. These controls
operate two channel groups:
The I2C registers that control bass and treble are:
•Bass and treble bypass register (0x89–0x90, channels 1–8)
•Bass and treble slew rates (0xD0)
•Bass filter sets 1–5 (0xDA)
•Bass filter index (0xDB)
•Treble filter sets 1–5 (0xDC)
•Treble filter index (0xDD)
2.7Volume, Automute, and Mute
The TAS5508C provides individual channel and master volume controls. Each control provides an
adjustment range of 18 dB to –100 dB in 0.25-dB increments. This permits a total volume device control
range of 36 dB to –100 dB plus mute. The master volume control can be configured to control six or eight
channels.
The TAS5508C has a master soft mute control that can be enabled by a terminal or I2C command. The
device also has individual channel soft mute controls that are enabled via I2C.
The soft volume and mute update rates are programmable. The soft adjustments are performed using a
soft-gain linear update with an I2C-programmable linear step size at a fixed temporal rate. The linear
soft-gain step size can be varied from 0.5 to 0.003906. Table 2-5 lists the linear gain step sizes.
Time to go from 36.124 db to –127 dB in ms10.6721.3342.6785.34170.67340.35682.701365.4
Time to go from 18.062 db to –127 dB in ms1.332.675.3310.6721.3342.6785.33170.67
Time to go from 0 db to –127 dB in ms0.170.330.671.332.675.3310.6721.33
2.8Automute and Mute
The TAS5508C has individual channel automute controls that are enabled via the I2C interface. Two
separate detectors can trigger the automute:
•Input automute: All channels are muted when all 8 inputs to the TAS5508C are less in magnitude than
the input threshold value for a programmable amount of time.