This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TLV320AIC3256: ADC Data Out?

Part Number: TLV320AIC3256

I am working with the TLV320AIC3256 both on the evaluation board and on a custom board.  We are able to play audio (in both implementations) through the I2S interface, through the headphone and line out ports with no problem.  However, we are unable to receive any data output from the part. 

On the evaluation board, I have tried simply setting the pre-defined "High Performance Stereo Recording (IN1)" mode using the following dialog window in the configuration application.

I have also made sure I configure the clock settings by pressing the "Program USB Audio Settings" in the Clocks/Interface dialog window:

I think this is where I am not following how the evaluation module is supposed to work.  I am assuming that at this point, I should be able to feed audio (line-level audio) into the J2 port of the board and be able to hear that audio through the PC's speakers by listening to the "Line-In" source created by the EVM.  When I listen to that source, I don't hear anything and the level meter shows there is no signal.  I have checked the MicPGA gains, ADC Gains, and the audio input routing.  It all appears correct.

I can load some of the miniDSP apps such as the "Stereo AGC" app which seems to route the audio input through some DSP operations and back out the headphone jack (J5).  That works.  I can also connect the input jack to the headphone amplifiers through the mixing amplifiers and can hear the audio input.  So I know the audio input path works.

Is my understanding of how the ADC input works correct?  I have found a few forum posts that talk about clock settings must be correct, but I am not sure what is happening on the EVM and if I need to change the clock settings or not.  It seems as if those clock settings are pretty much defined by the TAS1020 device. 

  • Hi Patrick,

    Since the input path seems to work, the issue might be your audio interface configuration. Double check that you're receiving the clocks from USB and sending out word data. You can even probe the DOUT before it makes it to the TAS to see if you're getting any output at all.

    Another workaround would be to use the miniDSP as a sort of bypass: Decimation filter from ADC in -> I2S out

    Best regards,


  • Thank you Jeff.  Monitoring DOUT with a scope does show activity, but when we feed that back to DIN via our host controller (just loop it back), we just hear static on the output (when we use our custom board).  We also used the internal loopback mode (Page 0, Register 29) and hear the same static on both the custom board and the EVM.  I am not sure why we don't hear the same static when we try to listen to the EVM line-in source on the PC.  There appears to be no data coming back from the TAS (maybe because it is static?).

    I did try a simple decimation filter -> I2S output in the miniDSP as well.  Static as well from DOUT.  I did verify that I configured the DSP correctly by also splitting the decimation filter output to the headphone output, and that worked as expected.  I believe that means I have a working path between ADC and DAC.

    Am I correct to say that if I am not using the miniDSP (just using a signal processing block such as PRB_R1 for example), that there are no other filter registers that must be initialized prior to being able to receive valid data from DOUT?  ...that there is nothing else in the data path between the ADC output and the I2S DOUT interface?

  • Hi,

    I tested with the EVM by loading either the low power or high performance recording script, generate a tone from Audacity with headphone and using Goldwave to record from USB-AudioEVM and able to capture the sine tone as shown.

    My connection is from PC headphone output to J2 of EVM and USB to PC.


  • Hi,

    I repeated the exact same process, but still do not have audio from the codec back through the PC.  When I select the EVM as the input (recording) source, there is just no data.  I was able to confirm using the internal audio loopback of the CODEC that the ADC is operating.  It just doesn't seem to make it back through the I2S port, through the TAS1020 and back to the PC.  I suspect something may be physically wrong in the DOUT path on the EVM...maybe a bad solder joint.

  • Is it possible to check the I2S signals with external instrument like AP-PSIA or Flex?