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.

(Split) PCM5141 96K not working

Other Parts Discussed in Thread: PCM5142, PCM5141

Hi Matt,
I see the original thread has been closed related to PCM5142/5141. I would like to continue with some more questions in relation to this....
We are having problems with the PCM5141 and 96khz config (havent attempted 192k yet).
What works: Dumping the pps reg values generated from GDE for 48kHz. The PCM5141 outputs audio at 44.1 and 48k no problem as expected.
What doesn't work: Dumping the pps reg values generated from GDE for 96kHz. These is no audio output when using 96k I2S input.

Perhaps you have some pointers? I know there are some limitations to the default configuration, however we are using our own stripped back generation of the pps (5band eq and volume only).
We are starting to test this on the eval board now also.
Cheers,

  • Hi Aran,

    The default configuration is limited to 44.1/48 KHz. To operate at a higher sampling rate you must change the ROM program being used by the part, this is detailed on page 42 of the data sheet. Also are you using the EVM for this test?

    Justin
  • Hi Justin, yes we are aware of the ROM limitation... more info below: 

    I am having problems with the PCM5141 and 96khz (havent attempted 192k yet but this needs to work also). 
    What works: Dumping the pps reg values generated from GDE for 48kHz. The PCM5141 outputs audio at 44.1 and 48k no problem.
    What doesnt: Dumping the pps reg values generated from GDE for 96kHz. These is no audio output from 96K I2S.

    Our design directs away from ROM and is just with 5band Eq + Volume only (as PIC attached).

    We can not get 96 or 192 working on the 5141 eval board either (even with using SB-AudioEVM0354_AnyRate.BIN). We did find the situation where moving to 96K over SPDIF input on the eval did work,however it seems that with SPDIF input the 96Khz input works only becuase the SRC IC down-converts to 48Khz... therefore we have not been able to test or get 96K tested directly into the 5141 on the eval board.

    For our application we are running SPDIF into WM8805 and I2S into the PCM5142.

    If you have any help hope to get 96Khz and 192Khz sample rates operational that would be appreciated.

    Thanks

  • Hi Aran,

    Try using the Int2xOutROM_1 instead of the Int8xOutROM_1, the Int2xOutROM_1 is meant to support 176/192 kHz.

    Justin
  • Hi Justin, still no joy - Im focusing on 96K first b4 I tackle 192k.... im sending the I2C regs from REG_Section_program and then miniDSP_D_reg_values from base_main_Rate96_pps_driver.h (generated in GDE).

    Ive tested our 96K SPDIF source on other equipment to ensure its valid...I2S looks ok going into 5141.... pfw attached...

    Any thoughts?

    le2596_DSP_ProcessFlow_v2.zip

  • Hi Justin - any news? We are about to launch product - this is one of the final hurdles...

    We have more testing results with still no success: 
    I created a very simple project in GDE with just I2S_In_TI_v1 --> Int2xOutROM_TI_v1 (I2S input direct to DAC output).... the result for 96K is that the audio output is decimated and very grainy - severely distorted.

    I am eagerly waiting your advice of how to proceed.

    Thanks, Aran.

  • Hi Again, more testing results with still no success:
    I created a very simple project in GDE with just I2S_In_TI_v1 --> Int2xOutROM_TI_v1 (I2S input direct to DAC output).... the result for 96K is that the audio output is decimated and very grainy - severely distorted.

    I am eagerly waiting your advice of how to proceed.

    Thanks, Aran.

  • Hi Aran,

    When in operations could you tell me what value you see in register 34 on page 0, register 37 page 0 and register 91 on page 0? This will show what rate was set and then what clocks are being detected. This part is able to operate at 192 KHz and has been demonstrated before.

    Justin
  • Hi Justin, at 48KHz registers read: 34(0x00) 37(0x00) 40(0x02) 91(0x54)

    Something is wrong... I measure MCLK=12.288Mhz, BCLK = 3.072Mhz, LRCLK = 48Khz on the IC pins (256fs).... seems the 5141 is detecting the ratio wrong as 192khz @ 128fs (rather than 96K @ 256fs)...

    Any thoughts?

    Cheers,

    Aran.

  • Split from original thread due to new issues: e2e.ti.com/.../374787
  • Hi Aran,

    We are going to run through this here on an EVM on Monday and get back with you on our setup for the registers.

    Justin
  • Hi Aran,

    I was able to get the eval board working at a sample rate of 96kHz. The file below is a detailed description of my process:

    2844.PCM5142 Evaluation Module at 96kHz.docx

    The issue with the eval board comes from the SRC not the PCM5142. Purepath Studio does not properly set the registers of the SRC. After setting the registers with the following file (I used Purepath Console) the PCM5142 had no issues with a sampling rate at 96kHz.

    To get the eval board working set the registers of the SRC with the following file.

    0827.SRC_96kHz.cfg

    Regards,

    Mike