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.

PCMD3140: PCMD3140, 4 PDM Mics, CH 3/4 is 6 dB more than CH 1/2

Part Number: PCMD3140

Tool/software:

I am able to make rather high SNR 4-mic recordings using the EVM (verified with my algorithm, as a first smoke test).  Audio sounds good when listening to the WAV.

However, CH 3/4 amplitudes are always 6 dB more than CH 1/2.  I have tried swapping PDM mics CH1<->CH3, same result: it is unique to CH 3/4, past the PDM mic input.

To test reset volume values being wrong, I set the 4 volumes to the same level (in I2C app script), same 6 dB difference.

I then set CH 3/4 volumes to 6 dB above CH 1/2 volumes, then the WAV recordings appear almost identical in amplitude.

Ideas?  Thanks, Chris

>>> Side Note: I think there is a typo on Page 94 of the DS:

# Configure CH2_INSRC as Digital PDM Input by I2C write into P0_R65

w 9C 41 40

When I use this in my script, it disables CH2.

  • Hi Chris,

    Can you share the configuration scripts that is used in your PDM testing?

    Thanks and Regards,

    Lakshmi Narasimhan

  • Here is the script I activate in the I2C app: (a blend of TI examples)

    ### 4 PDM mics
    #
    # Power up IOVDD and AVDD power supplies
    # Wait for IOVDD and AVDD power supplies to settle to steady state operating voltage range
    d 100
    #
    w 9C 02 01 # Wake up device, external 1.8V AREG, must connect ARED to AVDD
    d 10 # Wait for at least 1 ms to allow the device to complete the internal wake-up sequence
    #
    w 9C 01 01 # Software Reset
    d 100
    #
    w 9C 02 01 # Wake it up again
    d 10
    #
    # no need to program ASI_CH1-CH4, as they default to Slot 0,1,2,3 (normal config that we want)
    ### w 9C 20 80 # edge for PDMDAT. the choice is arbitrary (I think) as it just changes channel order within a pair.
    w 9C 22 41 # Set GPO1=PDCMLK output (Default 3.072MHz)
    w 9C 2B 45 # Set GPI1=PDMDAT1, GPI2=PDMDAT2
    w 9C 3B 70 # MICBIAS configured as GPI2
    w 9C 3E E1 # CH1 volume 12 dB
    w 9C 43 E1 # CH2 volume 12 dB
    w 9C 48 D5 # CH3 volume 6 dB
    w 9C 4D D5 # CH4 volume 6 dB
    ### What is this? w 9C 41 40 # Configure CH2_INSRC as Digital PDM Input
    w 9C 73 F0 # Enable CH 1,2,3,4
    w 9C 74 F0 # Enable ASI Output Channel 1
    w 9C 75 60 # Power-up ADC, PLL

    Additional recording setup data:

    - Goldwave 6.80 windows program to do the recording, Options->FileFormat->PCM Signed 32 bit (4 ch, 3.1). File->New->4ch/3.1, 16000 sampling, Record.  Goldwave is a great tool, have successfully used it before to make 8 channel ADC6140 recordings.

    - TI main EVM board = AC-MB DC058 Rev C, 3.3V, J5 on, S1/0=On/On

    - TI daughter board = PCMD3140EVM-PDK, DC133 C, J11 1.8V, J14 on, J9 on, all other jumpers off.  4 PDM mics on custom board, then soldered to test EVM points.  100 ohms on PDMCLK at earliest access point (IN2M red post), 100 ohms on each of the 4 PDM mics close to mic signal start.  mics powered by AVDD.

    Thanks, Chris

  • Hi Chris,

    Thanks for sharing the script. Let me have a look at this and get back to you within a couple of days.

    Thanks and Regards,

    Lakshmi Narasimhan

  • Hi Chris,

    I tested the above script on an EVM (commenting out the volume control registers so that all PDM channels are at 0dB DVOL), and I observed that the signal levels were same for all 4 PDM channels.

    Can you confirm whether the discrepancy is seen with or without the DVOL settings from the above script?

    Thanks and Regards,

    Lakshmi Narasimhan