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.

AM62L: question

Part Number: AM62L
Other Parts Discussed in Thread: SYSCONFIG

Tool/software:

Hi, TI

  I am using AM62L-LINUX-RT-SDK Version: 11.00.15.05.

  • Hi

    any update here?Our project is approaching the deadline

  • Hello Yu Chen

    It looks like this thread was not assigned properly.  I have reassigned. 

    However it appears you are using a non TI codec for your implementation. Please expect our support to be very limited for this from a hardware/software perspective. 

    Any reason you were not able to use a TI codec to simplify your development?

    Regards

    Mukul 

  • Hi, Mukul

    Thank you for your reply.

  • Hi Yu,

    From a hardware perspective:

    • The MCLK should not be routed to pin A12 (MCASP0_ACLKR), but rather routed to an AUDIO_EXT_REFCLK pin.
    • Given the symptoms you've described, please make sure the frame settings line up. Specifically, expected frame sync width match between the Codec and McASP.

    Please attach your SysConfig file to help me further assist you. In the meantime, the McASP Design Guide is a good resource to help with guidance on wiring up McASP and the different modes it offers.

    Regards,
    Karam

  • Hi,Karam

  • Hi Yu,

    My understanding is you have probed the output of McASP to the codec and saw no output on the right channel. To find the root of the issue, please probe:

    • The analog input signals at the ES8388 codec input pins to verify if the right channel analog signal is present
    • The codec's digital output going to McASP to confirm the codec is properly converting the right channel analog to digital

    This will help understand and isolate where the right channel is being lost.

    Regards,
    Karam

  • Hello Yu,

    Given the symptoms you have described, it seems like this is a possible frame settings mismatch between your codec and McASP. When the frame-sync settings on the codec and the McASP don't match, one half-frame can be interpreted incorrectly, so one channel appears silent even though stereo data is being fed to McASP.

    Please verify the frame-sync polarity and width settings align on both McASP and your codec. Refer to section 14.7.1.1.2.27 MCASP_AFSXCTL Register in the AM62L TRM for register values (FXWID, FSXP, etc.).

    Please let me know if this helps!

    Regards,
    Karam

  • Sorry for the delay.

  • Hi Yu,

    Thank you for the updates.

    1. I'm a bit unclear about what is being probed in the waveforms. Could you please probe the following so we can narrow down our focus:

    • The analog input signals at the ES8388 codec input pins to verify if the right channel analog signal is present
    • The codec's digital output going to McASP to confirm the codec is properly converting the right channel analog to digital


    2. It may also be possible that your receive frame sync width is not correctly configured. In I2S mode, it should be set to 'single word', but by default it may be set to 'single bit'. Please confirm your configuration by checking the FRWID field: AM62L TRM Section 14.7.1.1.2.15 MCASP_AFSRCTL Register.


    Please let me know if this helps!

    Regards,
    Karam

  • Hello Yi,

    Thank you for the waveform.

    • On ASDOUT (Codec -> McASP), the right words appear almost constant with little toggling activity, while left words appear to toggle normally.
    • This might indicate the right data coming out the codec to McASP is possibly not valid.

    To find where the issue starts, I need the other waveform I was requesting with an oscilloscope:

    • Probe the codec's right analog input pins (RIN1/2) 
    • For reference, also probe the left analog input pins (LIN1/2)


    Regards,
    Karam