
Data Manual
2001 Digital Audio: Digital Speakers

Texas Instruments and its subsidiaries (TI) reserve the right to make changes to their products or to discontinue any product or service without notice, and advise customers to obtain the latest version of relevant information to verify, before placing orders, that information being relied on is current and complete. All products are sold subject to the terms and conditions of sale supplied at the time of order acknowledgment, including those pertaining to warranty, patent infringement, and limitation of liability.
TI warrants performance of its products to the specifications applicable at the time of sale in accordance with TI’s standard warranty. Testing and other quality control techniques are utilized to the extent TI deems necessary to support this warranty. Specific testing of all parameters of each device is not necessarily performed, except those mandated by government requirements.
Customers are responsible for their applications using TI components.
In order to minimize risks associated with the customer’s applications, adequate design and operating safeguards must be provided by the customer to minimize inherent or procedural hazards.
TI assumes no liability for applications assistance or customer product design. TI does not warrant or represent that any license, either express or implied, is granted under any patent right, copyright, mask work right, or other intellectual property right of TI covering or relating to any combination, machine, or process in which such products or services might be or are used. TI’s publication of information regarding any third party’s products or services does not constitute TI’s approval, license, warranty or endorsement thereof.
Reproduction of information in TI data books or data sheets is permissible only if reproduction is without alteration and is accompanied by all associated warranties, conditions, limitations and notices. Representation or reproduction of this information with alteration voids all warranties provided for an associated TI product or service, is an unfair and deceptive business practice, and TI is not responsible nor liable for any such use.
Resale of TI’s products or services with statements different from or beyond the parameters stated by TI for that product or service voids all express and any implied warranties for the associated TI product or service, is an unfair and deceptive business practice, and TI is not responsible nor liable for any such use.
Mailing Address:
Texas Instruments Post Office Box 655303 Dallas, Texas 75265
Copyright 2001, Texas Instruments Incorporated

The TAS3002 device is a system-on-a-chip that replaces conventional analog equalization to perform digital parametric equalization, dynamic range compression, and loudness contour. Additionally, this device provides high-quality, soft digital volume, bass, and treble control. All control parameters are uploaded from an outside MCU through the I2C slave port or from an external EEPROM through the I2C master port.
The TAS3002 device also has an integrated 24-bit stereo codec with two I2C-selectable, single-ended inputs per channel.
The digital parametric equalization consists of seven cascaded, independent biquad filters per channel. Each biquad filter has five 24-bit coefficients that can be configured into many different filter functions (such as band-pass, high-pass, and low-pass).
The internal loudness contour algorithm can be controlled and programmed with an I2C command.
Dynamic range compression/expansion (DRCE) is programmable through the I2C port. The system designer can set the threshold, energy estimation time constant, compression ratio, and attack and decay time constants.
The TAS3002 device supports 13 serial interface formats (I2S, left justified, right justified) with data word lengths of 16, 18, 20, or 24 bits. The sampling frequency (fS) may be set to 32 kHz, 44.1 kHz, or 48 kHz. The 13 serial interface formats are listed and described in Section 2.1.
The TAS3002 device uses a system clock generated by the internal phase-locked loop (PLL). The reference clock for the PLL is provided by an external master clock (MCLK) of 256fS or 512fS, or a 256fS crystal.
The TAS3002 device has six internally configurable general-purpose input (GPI) terminals that control volume, bass, treble, and equalization. Each GPI terminal has a debounce algorithm that is programmed into the TAS3002 internal microcontroller.
− Downloading of control data to the device externally from the EEPROM or an I2C master
1−1

Figure 1−1 is a block diagram showing the major functions of the TAS3002.
1−2

VREFM
VREFP
VRFILT
AVDD
AVSS
DVDD
DVSS

VCOM
AOUTL
AOUTR
L+R
SDOUT2
SDOUT1
Figure 1−1. TAS3002 Block Diagram
1−3

Figure 1−2 shows the terminal locations on the package outline, along with the signal name assigned to each terminal.
REFM
VREFP
AOUTL RINA RINB
AOUTR VCOM
AINLP
V AINLM
AINRM
AINRP
LINB

NC AVDD NC GPI5 GPI4 GPI3 GPI2 GPI1 GPI0 ALLPASS SDOUT1 SDOUT0
Figure 1−2. TAS3002 Terminal Assignments
1.5 Terminal Functions
Table 1−1 lists the terminals in alphanumeric order by signal name, along with the terminal number, terminal type, and a description of the terminal function.

1−4

1−5

1−6

The TAS3002 device works in master or slave mode.
In the master mode, terminal 21 (IFM/S) is tied high. This activates the master clock (MCLK) circuitry. A crystal can be connected across terminals 13 (XTALI/MCLK) and 14 (XTALO), or an external, TTL-compatible MCLK can be connected to XTALI/MCLK. In that case, MCLK is outputs on terminal 12 (MCLKO), with terminals 19 (LRCLK/O) and 20 (SCLK/O) becoming outputs to drive slave devices.
In the slave mode, IFM/S is tied low. LRCLK/O and SCLK/O are inputs and the interface operates as a slave device requiring externally supplied MCLK, LRCLK (left/right clock), and SCLK (shift clock) inputs. There are two options for selecting the clock rates. If the 512fS MCLK rate is selected, terminal 11 (CLKSEL) is tied high and an MCLK rate of 512fS must be supplied. If the 256fS MCLK is selected, CLKSEL is tied low and an MCLK of 256fS must be supplied. In both cases, an LRCLK of 64SCLK must be supplied.
The TAS3002 device is compatible with 13 different serial interfaces. Available interface options are I2S, right justified, and left justified. Table 2−1 indicates how the 13 options are selected using the I2C bus and the main control register (MCR, I2C address 01h). All serial interface options at either 16, 18, 20, or 24 bits operate with SCLK at 64fS. Additionally, the 16-bit mode operates at 32fS.
Figure 2−1 through Figure 2−3 illustrate the relationship between the SCLK, LRCLK, and the serial data I/O for the different interface protocols.
2−1

The digital output modes (SDOUT1, SDOUT2, SDOUT0) are described in Sections 2.2.1 through 2.2.3.
The normal output mode for the MSB-first, right-justified, serial-interface format is for 16, 18, 20, or 24 bits. Figure 2−1 shows the following characteristics of this protocol:
MSB LSB MSB LSB … … SDIN
MSB LSB … … … … MSB LSB … …
Left Channel Right Channel
LRCLK = fS · 2−2

The normal output mode for the I2S serial-interface format is for 16, 18, 20, or 24 bits.
X LSB SDIN
MSB
… …
…
X LSB
MSB
… …
…
SDOUT
X LSB
MSB
… …
…
X LSB
MSB
… …
…
Left Channel Right Channel
LRCLK = fS · 2−3

The normal output mode for the MSB-left-justified, serial-interface format is for 16, 18, 20, or 24 bits.
MSB LSB … … MSB LSB … … SDIN
… …
… …
SDOUT
MSB LSB … … MSB LSB … …
… …
… …
Left Channel Right Channel
LRCLK = fS · 2−4


td(SDOUT) SDOUT valid from SCLK falling edge (see Note 1) (1/256fS) + 10 ns
tsu(SDIN) SDIN setup before SCLK rising edge 20 ns
th(SDIN) SDIN hold after SCLK rising edge 100 ns

LRCLK
SDOUT1 SDOUT2 SDOUT0
SDIN1 SDIN2
Figure 2−4. For Right-/Left-Justified and I 2 S Serial Protocols
2−5

2−6

The TAS3002 device contains a stereo 24-bit ADC with two single-ended inputs per channel. Selection of the A or B analog input is accomplished by setting a bit in the analog control register (ACR) by an I2C command. Additionally, the TAS3002 device has a stereo 24-bit digital-to-analog converter (DAC).
Figure 3−1 shows the technique and components required for analog input to the TAS3002 device. The maximum input signal must not exceed 0.7 Vrms. Selection of the above component values gives a frequency response from 20 Hz to 20 kHz at a sampling frequency of 48 kHz without alias frequency problems.
AINRM
0.47 µF
RINB
0.47 µF
AINRM
AINLP
AINLM
0.47 µF
LINA
LINB
AINLP
0.47 µF
1 Analog Inputs − Use 0.47 µF for 20-Hz Cutoff
Input Select Command From Internal Controller
2 Anti-Alias Capacitors for fS = 48 kHz
3 Tie unused analog inputs to analog ground through 0.1-µF capacitors.
3.2 Analog Output
3.2.1 Direct Analog Output
The full scale analog output from the TAS3002 device is 0.707 Vrms. It is referenced to VCOM which is approximately 1.5 Vdc. VCOM must be decoupled with the network shown in Figure 3−2.

Analog Output (Adjust Capacitors for Desired
VCOM
0.1 µF
AGND
Figure 3−2. VCOM Decoupling Network
3.2.2 Analog Output With Gain
Because the maximum analog output from the TAS3002 device is 0.707 Vrms, the output level can be increased by using an external amplifier. The circuit shown in Figure 3−3 boosts the output level to 1 Vrms (when it has a gain of 1.414) and provides improved signal-to-noise ratio (SNR). Since this circuit lowers the noise floor, THD + N is improved also.
Analog Output (Adjust Capacitors for Desired
VCOM
0.1 µF
AGND
C1 = C2 = C3
C4 = C5
Figure 3−3. Analog Output With External Amplifier

Figure 3−4 shows the TAS3002 reference voltage filter.
0.1 µF
15 µF +
1 µF +
0.1 µF
0.1 µF
4 2 3 45
AVSS(REF)
VRFILT
VREFM
44 VREFP
TAS3002
Figure 3−4. TAS3002 Reference Voltage Filter
3−3

3−4

The TAS3002 device implements a TI proprietary soft volume update. This feature allows a smooth and pleasant-sounding change from one volume level to another over the entire range of volume control (18 dB to mute).
The volume is adjustable by downloading a gain coefficient through the I2C interface in 4.16 format—4 bits for the integer and 16 bits for the fractional part. NO TAG lists the 4.16 coefficients converted into dB for the range of –70 dB to 18 dB with 0.5-dB step resolution.
Right and left channel volumes can be unganged and set to different values. This feature implements a balance control.
Volume is changed by writing the desired value into the volume control registers. This is done by asserting the volume-up or volume-down GPI terminal (see Section 7.6.1) for a limited range of volume control. Alternatively, volume control settings can be sent to the TAS3002 device over the I2C bus.
Soft mute is implemented by loading all zeros in the volume control register. This causes the volume to ramp down over a duration of 2048fS samples to a final output of 0 (− infinity dB).
Soft mute can be enabled by either asserting the mute GPI terminal (see Section 7.6.1) or sending a mute command over the I2C bus. Subsequent assertions of the mute GPI terminal toggle soft mute off and on.
The TAS3002 device is capable of mixing and multiplexing three channels (SDIN1, SDIN2, and the ADC output) of serial audio data. The mixing is controlled through three mixer control registers. This is accomplished by loading values into the corresponding bytes of the mixer left gain (07h) and mixer right gain (08h) control registers. See Figure 4−1 for a functional block diagram of the input mixer.
The values loaded into these registers are in 4.20 format—4 bits for the integer and 20 bits for the fractional part. NO TAG lists the 4.20 numbers converted into dB for the range of –70 dB to 18 dB, although any positive 4.20 number may be used.
To mute any of the channels, 0s are loaded into the respective mixer control register.
Mixer controls are updated instantly and can cause audible artifacts for large changes in setting when updated dynamically outside of the fast load mode; therefore, it is desirable to use fast load in conjunction with the soft-volume mode.
SDIN1, SDIN2, and the ADC output can be mixed with a user-selectable gain for each channel. The gain control registers are represented in 4.20 format.
4−1

Left Channel Mix Coefficients I2C Register Address 08h
= (3) 24-Bit Left Mix Coefficient
SDIN2_L 7 Biquad Filters Tone
DRCE
ADC_L
SDOUT1
SDIN1_R
SDIN2_R
7 Biquad Filters Tone
ADC_R
R_SUM
SDOUT2 1/2 L + R_SUM
Right Channel Mix Coefficients I2C Register Address 07h SDIN1 ^ SDIN2 ^ ADC
= (3) 24-Bit Right Mix Coefficient
4.4 Mono Mixer Control
The TAS3002 device contains a second mixer that performs the function of mixing left and right channel digital audio data from the input mixer in order to derive a monaural channel. This mixer has a fixed gain of −6 dB so that full scale inputs on L_sum and R_sum do not produce clipping on the resulting L+R_sum.
The output of this mixer is present on terminal 24 (SDOUT2) and is generally used for a digitally-mixed subwoofer or center channel application.
The treble gain level may be adjusted within the range of 15 dB to –15 dB with 0.5-dB step resolution. The level changes are accomplished by downloading treble codes (shown in NO TAG) into the treble gain register. Alternatively, a limited range of treble control is available by asserting the treble-up or treble-down GPI terminal (see Section 7.6.1).
The treble control has a corner frequency of 6 kHz at a 48-kHz sample rate.
The gain values for treble control can be found in Section NO TAG.

The bass gain level can be adjusted within the range of 15 dB to −15 dB with 0.5-dB step resolution. The level changes are accomplished by downloading bass codes (shown in NO TAG) into the bass frequency control register. Alternatively, a limited range of bass control is available by asserting the bass-up or bass-down GPI terminal (see Section 7.6.1).
Bass control is a shelf filter with a corner frequency of 250 Hz at a 48-kHz sample rate.
The gain values for bass control can be found in Section NO TAG.
De-emphasis is implemented in the DAC and is software controlled. De-emphasis is valid at 44.1 kHz and 48 kHz.
To enable de-emphasis, values are written into the analog control register via the I2C command. See Section 4.8 for analog control register operation.
Figure 4−2 illustrates the frequency response of the de-emphasis mode.
3.18 (50 µs)
10.6 (15 µs)

The analog control register (ACR) allows control of de-emphasis, selection of the analog input channel to the ADC, and analog power down.
An I2C master is required to write the appropriate command into the ACR. The ACR subaddress is 40h.
Bit 7 6 5 4 3 2 1 0
Type R/W
Default 0

5−4 Reserved R/W Reserved. Bits 5 and 4 return 0s when read.
00 = De-emphasis off (initial condition after reset) 01 = 48 kHz sample rate de-emphasis selected 10 = 44.1 kHz sample rate de-emphasis selected 11 = Reserved
0 = LINA and RINA selected (initial condition after reset) 1 = LINB and RINB selected
0 = Normal operation (initial condition after reset) 1 = Power down
4−4

The necessity for applying loudness compensation to playback systems to compensate for the fact that the ear perceives bass and treble less audibly at low levels than at high ones has been established since the first data was published by Fletcher and Munson in 1933.
There are many equal-loudness contours in publication, like Steven’s contours, Robinson and Dadson contours. Some have even reached the acceptance level of ISO recommendation.
The TAS3002 device has a simplified loudness contour algorithm that diminishes the effect of weak bass at low listening levels. Since contour has volume level dependency, the user must define the relation between the gain of the contour circuit and the volume level.
Figure 4−3 is a block diagram of this circuit.
The loudness contour is activated by sending an activation command via I2C from an external device. Optionally, a contour gain command can be sent by an external device to provide tracking with the system volume control.
Loudness biquad filters for the left and right channels are independently programmable via I2C. Their subaddresses are 21h and 22h, respectively. The digital filters are written as five 24-bit (4.20) hex coefficients for each channel.
Loudness gain values for the left and right channels are independently programmable via I2C. Their subaddresses are 23h and 24h, respectively. The gain values are written as one 4.20 hex coefficient for each channel.
When the frequency of the loudness contour is determined, a digital filter must be developed. Then, the gain of the filter is determined. These values are placed in the storage area of the system controller (microcontroller) and sent to the TAS3002 device when it is desired to activate the loudness contour.
If it is necessary to change the frequency or gain of the contour, new gain and filter coefficients are sent by the system controller. This function is performed normally when the volume control is changed (that is, more volume, less contour). The gain of the loudness contour filter then tracks the volume control.
The loudness contour biquad filters are provided in addition to the seven equalization biquad filters.
See Section NO TAG for programming instructions.
Volume · Biquad Gain · 4−5

The TAS3002 device provides the user with the ability to manage the dynamic range of the audio system. The DRCE receives data, and affects scaling after the volume/loudness block. As shown in Figure 4−4, the DRCE is applied after the volume/loudness control block as a DRCE scale factor. The DRCE must be adjusted such that the signal does not reach the hard limit value. However, if the signal does reach the maximum digital value, the saturation logic serves as a hard limiter that does not allow the signal to extend beyond the available range.
(7) 2nd Order IIR Filters
SDIN2_L
ANALOGIN_L
(Analog in From ADC)
RIGHT_SUM SDIN1_R
(7) 2nd Order IIR Filters
SDIN2_R
Figure 4−4. TAS3002 Digital Signal Processing Block Diagram
The DRCE instruction consists of eight bytes that must be sent each time in the order shown in the example code of NO TAG. Each instruction downloaded must be eight bytes. If only one byte is changed, all eight bytes must be transmitted. The first two bytes remain the same for every instruction, however the last six bytes can be programmed using hexadecimal values from the corresponding tables referred to in Section NO TAG.
With high compression ratios and fast attack times available, this function is suited for a commercial killer in a television set application.
This function is enabled by setting terminal 27 (ALLPASS) on the TAS3002 device to 1. When asserted, the internal equalization filters are set into AllPass (flat) mode. When this terminal is reset to 0, the equalization filters are returned to the equalization that was in use before the terminal was asserted.
In AllPass mode, the bass and treble controls are still functional.
This function is frequently used for headphones. When the headphone plug is inserted into its jack, a switched contact in the jack enables the AllPass function.
The AllPass function also can be activated by writing a 1 to bit 2 of the analog control register.

The TAS3002 device contains two main control registers: main control register 1 (MCR1) and main control register 2 (MCR2). The MCR1 register contains the bits associated with load speed, SCLK frequency, serial-port mode, and serial-port word length. It is accessed via I2C with the address 01h.
Bit b7 b6 b5 b4 b3 b2 b1 b0
Type R/W R R R/W R/W
Default 1 X

0 = Normal operation mode 1 = Fast -load mode (default)
0 = SCLK is 32 fS. 1 = SCLK is 64 fS.
00 = Left justified 01 = Right justified 10 = I2S 11 = Reserved
00 = 16-bit 01 = 18-bit 10 = 20-bit 11 = 24-bit
The TAS3002 device contains two main control registers: main control register 1 (MCR1) and main control register 2 (MCR2). The MCR2 register contains the bits associated with the AllPass function and the download of bass and treble control information, and it is accessed via I2C with the address 43h.
Bit 7 6 5 4 3 2 1 0
Type R/W R/W R
Default 0 0 0 x x x 0 0

7 Reserved R/W 0 = Normal operation (initial condition after reset) 1 = Download bass and treble
6−5 Reserved R Reserved. Bits 6 and 5 return 0s when read.
1 DM(1−0) R/W 0 = Normal operation (initial condition after reset) 1 = AllPass mode (bass and treble are still functional)
0 INP R Reserved. Bit 0 returns 0 when read.
4−7

4−8

The biquad block consists of seven digital biquad filters per channel organized in a cascade structure, as shown in Figure 5−1. Each of these biquad filters has five downloadable 24-bit (4.20) coefficients. Each stereo channel has independent coefficients.
Biquad 1 ... Biquad 0 Biquad 6
5.1.1 Filter Coefficients
The filter coefficients for the TAS3002 device are downloaded through the I2C port and loaded into the biquad memory space. Each biquad filter memory space has an independent address. Digital audio data coming into the device is processed by the biquad block and then converted into analog waveforms by the DAC. Alternatively, filters can be loaded by asserting terminals on the GPI port.
The biquad structure that is used for the parametric equalization filters is as follows:
H(z)
a0 a1z 1 a2z 2
(1)
The coefficients for these filters are represented in 4.20 format—4 bits for the integer part and 20 bits for the fractional part. In order to transmit them over I2C, it is necessary to separate each coefficient into three bytes. The upper 4 bits of byte 2 comprise the integer part; the lower 4 bytes of byte 2 plus byte 1 and byte 0 comprise the fractional part.
The filters can be designed using the automatic loudspeaker equalization program (ALE) or a script running under MatLab named Filtermaker. Both of these tools are available from Texas Instruments.
5−1

5−2

Control parameters for the TAS3002 device can be loaded from an I2C serial EEPROM by using the TAS3002 master interface mode. If no EEPROM is found, the TAS3002 device becomes a slave device and loads from another I2C master interface. Information loaded into the TAS3002 registers is defined in Appendix A.
The I2C bus uses terminals 16 (SDA for data) and 15 (SCL for clock) to communicate between integrated circuits in a system. These devices can be addressed by sending a unique 7-bit slave address plus R/W bit (1 byte). All compatible devices share the same terminals via a bidirectional bus using a wired-AND connection. An external pullup resistor must be used to set the high level on the bus. The TAS3002 device operates in standard mode up to 100 kbps with as many devices on the bus as desired up to the capacitance load limit of 400 pF.
Furthermore, the TAS3002 device supports a subset of the SMBus protocol. When it is attached to the SMBus, then byte, word, and block transfers are supported. The SMBus NAK function is not supported and care must be taken with the sequence of the instructions sent to the TAS3002 device.
Additionally, the TAS3002 device operates in either master or slave mode; therefore, at least one device connected to the I2C bus must operate in master mode.
The bus standard uses transitions on SDA while the clock is high to indicate start and stop conditions. A high-to-low transition on SDA indicates a start and a low-to-high transition indicates a stop. Normal data bit transitions must occur within the low time of the clock period. Figure 6−1 shows these conditions. These start and stop conditions for the I2C bus are required by standard protocol to be generated by the master. The master must also generate the 7-bit slave address and the read/write (R/W) bit to open communication with another device and then wait for an acknowledge condition. The slave holds SDA low during acknowledge clock period to indicate an acknowledgment. When this occurs, the master transmits the next byte of the sequence.
After each 8-bit word, an acknowledgment must be transmitted by the receiving device. There is no limit on the number of bytes that can be transmitted between start and stop conditions. When the last word transfers, the master generates a stop condition to release the bus. Figure 6−1 shows a generic data transfer sequence.
8-Bit Register Data for Address (N)
8-Bit Register Data for Address (N+1)
8-Bit Register Data for Address (N+2)
A
A
A
A
SCL
Figure 6−1. Typical I 2 C Data Transfer Sequence

Table 6−1 lists the definitions used by the I2C protocol.

Master The device that initiates a transfer, generates clock signals, and terminates the transfer
Multimaster More than one master can attempt to control the bus at the same time without corrupting the message.
Arbitration Procedure to ensure the message is not corrupted when two masters attempt to control the bus.
Synchronization Procedure to synchronize the clock signals of two or more devices
The 7-bit address for the TAS3002 device is 0110 10X R/W where X is a programmable address bit, set by terminal 7 (CS1). Combining CS1 and the R/W bit, the TAS3002 device can respond to four different I2C addresses (two read and two write). These two addresses are licensed I2C addresses that do not conflict with other licensed I2C audio devices. In addition to the 7-bit device address, subaddresses direct communication to the proper memory location within the device. A complete table of subaddresses and control registers is provided in Appendix A. For example, to change bass to 10-dB gain, Section 6.3.1 shows the data that is written to the I2C port:

I2C ADDRESS BYTE A6−A1 CS1 (A0) R/W

Start Slave Address R/W A Subaddress A Data A Stop
Start Start condition as defined in I2C
A Acknowledgement as defined in I2C (slave)
Stop Stop condition as defined in I2C
Whenever writing to a subaddress, the correct number of data bytes must follow in order to complete the write cycle. For example, if the volume control register with subaddress 04h is written to, six bytes of data must follow; otherwise, the cycle is incomplete and errors occur.
6−2

The TAS3002 saves in a stack or first-in first-out (FIFO) buffer the last 7 bytes that were sent to it. When an I2C read command is sent to the device (LSB=high), it answers by popping the first byte off the stack. The TAS3002 then expects either a Send Ack command or an I2C Stop command from the host. If a Send Ack command is sent from the host then the TAS3002 pops another byte off the stack. If an I2C Stop is sent then the TAS3002 ends this transaction. The proper sequence for reading is described as follows:
I2C Start
Send I2C address byte with read bit set to 1 (LSB set equal to 1)
Receive Byte 6 (if an ACK is sent after byte 6 it locks up the TAS3002)
I2C Stop
The TAS3002 device performs interpolation algorithms for its volume and tone controls. If a volume or tone change is sent to the part via I2C, the command sent after the volume or tone (bass and treble) change causes an I2C wait state to occur. This wait state lasts from 41 ms to 231 ms, depending on the system clock rate, the command sent, and, in the case of bass or treble, the amount of the change.
Secondly, if a long series of commands is sent to the TAS3002 device, it may occasionally create a short wait state on the order of 150 µs to 300 µs while it loads and processes the commands.
When a sample rate of 32 kHz is used, longer wait states can occur, occasionally up to 15 ms.
The preferred way to take care of wait states is to use an I2C controller that recognizes wait states. During the wait state period, it stops sending data over I2C. If this function is not available on the system controller, fixed delays can be implemented in the system software to ensure that the controller is not trying to send more data while the TAS3002 device is busy. Sending I2C data while the TAS3002 device is busy causes errors and locks up the device, which must then be reset.

Table 6−3 gives typical values of the wait states that can be expected with the various functions of the part:

Volume 62 ms 49 ms 41 ms Not dependent on size of change
Bass 231 ms 167 ms 153 ms 0 to −18 dB
Treble 231 ms 167 ms 153 ms 0 to −18 dB
DRC on 300 µs 300 µs 300 µs
Equalization 15 ms 190 µs 300 µs Can occur with each filter
The TAS3002 device supports a subset of the SMBus protocol. With proper programming techniques, it is possible to use the SMBus to set up the TAS3002 device.
The TAS3002 device supports the block write protocol that allows up to 32 bytes to be sent as a block. To send a command using this format, the most significant bit (MSB) of the TAS3002 subaddress must be set high and the subaddress (also with MSB set high) must be programmed into the SMBus command byte. This operation signals the TAS3002 device that the next byte is the SMBus byte-count byte. The next byte after the byte count is then entered into the device as the first byte of data.
68h 8rh xx dd dd dd
Subaddress (r = subaddress)
Byte Count (Don’t Care)
The TAS3002 device also supports the SMBus write byte protocol. Writing to the main control register (MCR), bass, and treble registers requires using the byte write protocol. To send a command using this protocol, the most significant bit (MSB) of the TAS3002 subaddress must be set high and the subaddress (also with MSB set high) must be programmed into the SMBus command byte. The next byte after the command byte is then entered into the device as the first byte of data.
Subaddress (r = subaddress)

If separate I2C/SMBus commands are sent too frequently, the TAS3002 device can generate a bus wait state. This happens when the device is busy while performing smoothing operations and changing volume, bass, and treble. The wait occurs after the bus acknowledge on the first data byte and can exceed the maximum allowable time allowed according to the SMBus specification (worst case 200 ms).
CODE Start 68 84 06 01 00 00 01 00 00 Stop
ACTUAL Start 68 84 06 01 Wait† 00 00 01 00 00 Stop † If the master does not recognize bus waiting or if the master times out on a long wait, the master must not send consecutive I2C/SMBus commands
without a time interval of 200 ms between transactions.
The TAS3002 device supports a subset of SMBus readback. When an SMBus read command is sent to the device (LSB = high), it answers with the subaddress and the last six bytes written.
SENT Start 69h xxh 07h Stop
RECEIVED Start 07h aah ddh Stop
xxh = Command byte. It is a don’t care because the response contains only the subaddress and the last six bytes of data written to the TAS3002 device. aah = The last subaddress accessed in the device ddh = Data bytes from the TAS3002 device NOTE: Use read sequence defined in 6.3.2

6−6

The TAS3002 device contains an internal microcontroller programmed by Texas Instruments to perform housekeeping and interface functions. Additionally, it handles I2C communication and general purpose input functions.
The microcontroller uses a 256fS system clock and can access up to 8K bytes of memory. It interfaces with the digital audio interface I2C master/slave for downloading data and coefficients. It also interfaces with two internal DSPs for transferring coefficients and other information.
The TAS3002 coefficients are loaded through I2C in the master or slave mode. Standard audio processing functions (volume, bass, and treble) can be controlled/activated through external switches connected to the six GPI terminals. Upon reset, the internal microcontroller sets all coefficients and audio parameters to the default values. See Section 7.2.2 for default values.
If the TAS3002 address is 68h (ADDR_SEL=0), it becomes the bus master device and attempts to load parameters and coefficients from the external EEPROM. If no EEPROM is present, the TAS3002 device remains in its default condition. If addresses other than 68h/69h are set, the TAS3002 device only operates as an I2C slave device.
If the microcontroller determines the TAS3002 device has an I2C address of 68h/69h and the EEPROM is present, the microcontroller downloads coefficients from the EEPROM. Once the download is complete, it enables the serial audio in the mode defined by an I2C write to the MCR to transfer data into and out of the device. Before reading the EEPROM, the serial audio port defaults to I2S mode.
The TAS3002 device allows the user to update volume, bass, and treble dynamically by an I2C slave command or by a simple GPI input. The GPI can select volume up and down, bass/treble up and down, or digital equalizations. Up to five different equalizations (that is, flat, jazz, rock, voice, etc.) can be stored in the external EEPROM. Also, DRCE, MCR1, MCR2, and loudness contour are enabled and disabled by I2C.
When the TAS3002 device operates in the I2C master mode, it echoes changes to all of its functions to other I2C addresses that are defined in its external EEPROM. If no addresses are defined, it does not echo.
7.2.1 Power-Up Sequence
An active low on terminal 6 (RESET) while MCLK is running resets the internal microcontroller and DSPs. RESET synchronizes internally and can be asserted asynchronously or with the simple RC circuit in Figure 7−1. On reset, SCL and SDA go to a high-impedance state. If the I2C address is set to 68h, approximately 400 µs after RESET returns to a 1, the device sends a one-byte query via I2C to look for an EEPROM. If an EEPROM is found, the TAS3002 becomes an I2C master; otherwise, it becomes an I2C slave. When using address 68h in the slave mode, an external master must wait until after the EEPROM query or else bus contention and improper operation occur.
I2C address x6Ah does not query the bus for an EEPROM. The address for the EEPROM is A0h.
The TAS3002 device has an asynchronous reset terminal (RESET). This reset is synchronized with various clocks used in this device to generate a synchronous internal reset. Upon reset, the TAS3002 device goes through the following process:
7−1

Because the TAS3002 device has an internal power-on reset (POR), in many cases, additional components are not needed to reset the device. It resets internally at approximately 80% of VDD.
In the case where the system power supplies are slow in reaching their final voltage or where there is a difference in the time the system power supplies take to become stable, the TAS3002 reset can be delayed by a simple RC circuit.
TAS3002
RESET
0.1 µF
Figure 7−1. TAS3002 Reset Circuit
The reset delay for the above circuit can be calculated by the simple equation:
C = Value of the capacitance from RESET (pin 6) to DVSS R = Value of the resistance from RESET (pin 6) to DVDD The circuit described in Figure 7−1 delays the start-up of the TAS3002 device approximately 1.2 ms.
When it is necessary to control the reset of the TAS3002 device with an external device, such as a microcontroller, RESET (pin 6) can be treated as a logic signal. It then brings the device out of reset when the voltage on RESET reaches VDD/2.
While in fast load mode—FL bit (bit 7 of main control register 1) = 0—it is possible to update the parametric equalization without any audio processing delay. The audio processor pauses while the RAM is updated in this mode.
10 kΩ · 7−2

Bass and treble cannot download in this mode. Mixer1 and Mixer2 registers can download in this mode or normal mode (FL bit = 0).
Once the download is complete, the fast load bit must be cleared by writing a 0 into bit 7 of main control register 1 (MCR1). This puts the TAS3002 device into normal mode.
During initialization, the output of the codec is disabled. Throughout reset and initialization, the output of the DAC is muted to prevent extraneous noise being sent to the system output.
Data from the ADC and other internal processing is purged so that when reset/initialization is complete, only valid inputs are sent to the system output.
The TAS3002 device has an asynchronous power-down mode. In the power-down mode, the internal control registers and equalization programming of the device are stored in the device.
1. Assert the power-down control signal (1)
2. Set the serial audio input clocks to 0
The TAS3002 device goes into power-down mode.
1. Assert RESET (logic 0)
2. Restart the serial audio clocks
5. Negate RESET (logic 1)
The device then returns to the state it was in before power down (resumes normal operation).
7−3

RESET
MCLK
SCLK
LRCLK
SDATA
Figure 7−2. Power-Down Timing Sequence
In power-down mode, the TAS3002 device typically consumes less than 1 mA.
Terminal 9 (TEST) is tied low in normal operation. This function is reserved for factory test and must not be asserted.
Figure 7−3 shows the flow chart of the interface between the microcontroller and its peripheral blocks.
During initialization, the microcontroller fetches a control byte from its EEPROM or receives a command from I2C.
The six GPI terminals are programmed to operate as indicated in Table 7−1.


Initially (after reset), the TAS3002 GPI is set to control volume, bass, and treble. Simultaneously setting GPI bits 1 and 5 low for 1 second changes the function of the GPI terminals to control mute and equalization.
To return to volume, bass, and treble control, simultaneously set GPI terminals 2 and 3 low for 1 second.
When a GPI terminal is activated, the TAS3002 device echoes its function over I2C to a TAS3001 device mapped to address 6Ah. Therefore, a system with two audio equalization chips can be implemented without the need for a microcontroller.
The GPI provides simple but flexible input port to activate the input parameters. Each terminal input is an active logic low.
7−5

Power Up
Restore Volume
Initialize Default
and MCR
Initialize TAS3002
EEPROM
TAS3001
Load Parameters
Slave Write
and Coefficients
to DSP
Volume/Bass/Treble Up/Down
GPI
Echo to TAS3001
Switch BQ Set
Save Volume, Mute
Power Down
Save PWR_DN
Stop PLL
Stop DRC_OFF
DRC
Figure 7−3. Internal Interface Flow Chart
7−6

Table 7−2 through Table 7−5 show the 512-byte and 2048-byte EEPROM memory maps.

015h−01Ah 2 DRC (ratio, threshold, energyα, attackα, decayα)
7−7


015h−01Ah 6 DRC (ratio, threshold, energyα, attackα, decayα)
04Fh−05Dh 15 Biquad 2 TAS3002 05Eh−06Ch 15 Biquad 3
1A4h−1B2h 15 Biquad 2 TAS3001 1B3h−1C1h 15 Biquad 3
1F6h−1F7h 2 DRC (ratio, threshold, energyα, attackα, decayα)
Bytes are in the same order as they appear in the I2C register map. The EEPROM address is A0h.
7−8


001h 1 1 0 1 0
015h−019h 6/2 DRC (ratio, threshold, energyα, attackα, decayα) 1F6h−1F7h
7−9


TAS3002 ADDRESS NUMBER OF BYTES FUNCTION CATEGORY TAS3001 ADDRESS
001h 1 1 0 1
015h−019h 6/2 DRC (ratio, threshold, energyα, attackα, decayα) 1F6h−1F7h
7−10

Supply voltage range: AVDD −0.3 V to 3.6 V DVDD −0.3 V to 3.6 V Analog input voltage range: −0.3 to AVDD + 0.3 V Digital input voltage range: −0.3 to DVDD + 0.3 V Operating free-air temperature, TA 0°C to 70°C Storage temperature range, Tstg −65°C to 150°C Case temperature for 10 seconds +122°C Lead temperature from case for 10 seconds +97.8°C Electrostatic discharge (see Note 1) 2000 V
† Stresses beyond those listed under absolute maximum ratings may cause permanent damage to the device. These are stress ratings only, and
functional operation of the device at these or any other conditions beyond those indicated under recommended operating conditions is not implied. Exposure to absolute-maximum-rated conditions for extended periods may affect device reliability. NOTE 1: Human body model per Method 3015.2 of MIL-STD-833B.
TA = 25°C, AVDD = 3.3 V, DVDD = 3.3 V Voltages at analog inputs and outputs and at AVDD are with respect to ground.

Operating 34 mA Supply current, analog Power down (see Note 2) 88 µA
Operating 47 mA Supply current, digital Power down (see Note 2) 942 µA
TA = 25°C, AVDD = 3.3 V, DVDD = 3.3 V

VOH High-level output voltage IO = −1 mA 2.4 V
VOL Low-level output voltage IO = +4 mA 0.4 V
8−1

TA = 25°C, AVDD = 3.3 V, DVDD = 3.3 V, fS = 48 kHz, 20-bit I2S mode All terms characterized by frequency are scaled with the chosen sampling frequency, fS. See Figure 8−1 through Figure 8−4 for performance curves of the ADC digital filter.
ADC decimation filter (LPF) Pass band 0.0 20.0 kHz C dec at o te ( )
Pass band ripple ±0.01 dB
Stop band 24.1 kHz
Stop band attenuation 80 dB
Group delay 720 µs
ADC high-pass filter (HPF) Pass band (−3 dB) 0.87 Hz C g pass te ( )
Deviation from linear phase 20 Hz to 20 kHz 1.23 degrees
f − Frequency − Hz 0 2 fs 4 fs 6 fs 8 fs 10 fs 12 fs
0 f − Frequency − Hz 0.2 fs 0.4 fs 0.6 fs 0.8 fs 1 fs


−0.002
0 f − Frequency − Hz 0.1 fs 0.2 fs 0.3 fs 0.4 fs 0.5 fs

−0.2
−0.4
−0.6
−0.8
0 f − Frequency − Hz 1 fs 2 fs 3 fs 4 fs
8.5 Analog-to-Digital Converter
TA = 25°C, AVDD = 3.3 V, DVDD = 3.3 V, fS = 48 kHz, 20-bit I2S mode All terms characterized by frequency are scaled with the chosen sampling frequency, fS.

Signal to (noise + distortion) ratio −1 dB, 1 kHz, 20 Hz to 20 kHz 82 dB
Power supply rejection ratio 1 kHz (see Note 3) 50 dB
Overall ADC frequency response 20 Hz to 20 kHz ±0.1 dB

TA = 25°C, AVDD = 3.3 V, DVDD = 3.3 V, fS = 48 kHz, 20-bit I2S mode

TA = 25°C, AVDD = 3.3 V, DVDD = 3.3 V, fS = 48 kHz, 20-bit I2S mode All terms characterized by frequency are scaled with the normal mode sampling frequency, fS. See Figure 8−5 and Figure 8−6 for performance curves of the DAC digital filter.

Stop-band attenuation 28.8 kHz to 3 MHz 75 dB
0 fs/2 1 fs 2 fs 3 fs 4 fs 5 fs
f − Frequency − Hz

−0.05
−0.1
0 f − Frequency − Hz 0.1 fs 0.2 fs 0.3 fs 0.4 fs 0.5 fs
Amplitude − dB · 8−4

TA = 25°C, AVDD = 3.3 V, DVDD = 3.3 V, fS = 48 kHz, input = 0 dB-fS sine wave at 1 kHz

Dynamic range −60 dB, 1 kHz 92 96 dB
Signal to (noise + distortion) ratio 0 dB, 1 kHz, 20 Hz to 20 kHz 83 dB
Full scale, single-ended, output voltage range 1.9 VPP DC offset −7.0 7.0 mV
TA = 25°C, AVDD = 3.3 V, DVDD = 3.3 V The output load resistance is connected through a dc blocking capacitor.

VCOM internal resistance (see Note 4) 1 kΩ
VRFILT internal resistance (see Note 5) 1 kΩ
8−5


t(buf) Bus free time between start and stop 4.7 µs
t(low) Low period of SCL clock 4.7 µs
t(high) High period of SCL clock 4.0 µs
tsu(sta) Setup time repeated start 4.7 20 µs
th(dat) Data hold time (See Note 6) 0 µs
tr Rise time for SDA and SCL 1000 ns
tf Fall time for SDA and SCL 300 ns
tsu(sto) Setup time for stop condition 4.0 µs
C(b) Capacitive load for each bus line 400 pF
P

SCL
t(high) is measured from the end of tr to the beginning of tf.
tsu(sto) · 8−6

Figure 9−1 and Figure 9−2 show the TAS3002 stereo and 2.1-channel applications, respectively.
RESET
TAS3002
SPDIF
I 2 S
EEPROM I 2 C

RESET
Analog Out (To Satellite Amplifiers)
SPDIF
I 2 S
EEPROM I 2 C
I 2 S_OUT
SDOUT2
I 2 S PCM1744
TAS3001

The TAS3002 device is packaged in a 48-terminal PFB package. The following illustration shows the mechanical dimensions for the PFB package.

10−1

10−2
Owners and our team usually reply within a day — and your question helps the next person too.
Be the first to ask — other owners or our team will answer.