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.

TMS320C5535: Connect two eZdsp5535 via I2S

Part Number: TMS320C5535


Hello,

we connected two eZdsp5535 via I2S1 interface from expansion connector P2. The I2S1_RX is connected to I2S1_DX and vice versa. The AIC3204 of one board is the I2S clock master and generates the clock for the whole system. When we transfer audio to the other board, it sounds totally scrambled. When we work with a single board and output the mic audio directly to the headphones, audio quality is good. So my question is, are there any considerations when connecting the two boards via I2S1? Do we need serial resistors or buffers?

Thanks for your help

Marc

  • Marc,

    Not that I'm aware of.
    Please see this thread where I helped someone with a I2S-I2S connection between two C55x EVMs.
    e2e.ti.com/.../2386851

    Seems like something isn't configured correctly.
    Are you sampling at the same rate on both ends?

    Lali
  • Hello,

    yes we sample with 16Bit 16KHZ. I2S Clocks are synchronous on both boards. One AIC3204 is I2S master, the other AIC3204 operates as slave and generates its MCLK out of the I2S BCLK. You can see a system diagram here: https://e2e.ti.com/support/dsp/c5000/f/109/t/664992.

    When we work on one board and output the Mic data via headphone, everything works fine. But as soon as we connect the other board, the local audio data in and output doesn't work anymore (sound is scrambled). Even if we don't' transfer the data to the remote side. So running the same software with and without the second board connected via I2S1 makes a difference. If we transfer the data to the remote side, things are getting even worse.

    Both boards are powered via USB from the same PC. Perhaps there is a ground loop problem?!  Should we use 2 different PCs?

    Thank you

  • Marc,

    Yes, you can try with 2 PC's,but unsure what might be causing the noise you are observing.
    I have looped in a colleague to see if he has some suggestions.

    Lali
  • Hi,

    Have you shorted GNDs across the two boards?

    Have you looked at the signals with a scope to make sure there are not accidentally two masters both driving the lines? This would cause a scrambled sound. 

    You should be able to power them both from a single PC via USB.

    Hope this helps,
    Mark

  • Hi Mark,

    sorry for the late response. The signals look good now and sound can be transferred in both directions. We shortened the i2s connection cables between the two boards. Additionally we grounded the cables between the i2s signals/clocks. That solved our problem.

    Now I am working on integration of aec/aer algorithm as mentioned here https://e2e.ti.com/support/dsp/c5000/f/109/t/664533. Perhaps you can help me with this issue.

    So this thread can be closed.

    Thanks a lot for your support

    Marc

  • Great! Thanks for closing the loop on this thread with the confirmed solution.

    Please open a new thread for any other issues you encounter.

    Regards,
    Mark