• State Resolved
  • Locked Locked
  • Replies 12 replies
  • Answers 2 answers
  • Subscribers 31 subscribers
  • Views 1790 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

PCM5102A: Errant Noise when IC is shut down

Theodore Salmon
Intellectual 350 points
Part Number: PCM5102A
Other Parts Discussed in Thread: PCM5122

Hello,

I'm using a PCM5102A with a BC127 BT module to output audio to an amplifier. Everything works well, but when the IC goes into shutdown due to lack of I2S data, there is audible noise output through the amp and into the speakers. I can get rid of this noise by connecting the R/L channels to AGND.

My question is this: Is there a way to prevent the PCM5102A from sleeping, or a way to connect R/L to AGND when there is no analog audio being output? Perhaps a Monostable?

  • 0
    TI__Guru 55113 points
    Hi Theodore,

    The click noise is the device entering a low power mode as you said. Really the only way to prevent this on the PCM5102A is to provide non-zero data to the device. Some BT modules allow this. You can consider the PCM5122 device as well, which can have the auto mute disabled. You can also see if your BT module has a 'zero data' flag, and use this an analog mute signal on for the amplifier.

    Thanks,
    Paul
  • 0 in reply to Paul_Frost
    Intellectual 350 points
    Paul,

    Thanks for the response! Unfortunately, the BC127 doesn't have a way to continue providing data to the DAC.

    I actually plan on tying AGND to the analog outputs with a FET. When the Charge Pump goes low, I'll sink those channels to AGND. Does that sound acceptable?

    Thanks!
    -Ted
  • 0 in reply to Paul_Frost
    Intellectual 350 points
    Paul,

    Another thought: What If I just tie AGND to L+ and R+ via a high value resistor? Something like 100K ohms.

    Thanks,
    -Ted
  • 0 in reply to Theodore Salmon
    TI__Guru 55113 points
    Hi Ted,

    Basically there is a short offset calibration that occurs when device receives data for the first time. When it enters auto mute, this offset calibration stops, and the output returns to its un-calibrated value, usually a few mV. I do not think connecting the output to GND will really solve this. You could look at a more conservative output filter, with a RC cutoff frequency closer to the edge of the audible band, maybe 15kHz. That might attenuate the click a bit. Otherwise you can look at muting the output amplifier you are using.

    Thanks,
    Paul
  • 0 in reply to Paul_Frost
    Intellectual 350 points
    Paul,

    Thanks again! The problem isn't so much the DACs power-on or off, but rather existing noise making its way over the wires and into the amp. So, the entire time that the DAC is off and not driving the L+/R+ lines, there are random clicks and noise. When I tie the R+/L+ to AGND directly, the noise goes away.

    I'll try it a bit later, but would the resistor trick cause any issues with signal quality? So far, in my quick test, it doesn't, but I'm not able to use a scope to verify.

    Thanks!
    -Ted
  • 0 in reply to Theodore Salmon
    TI__Guru 55113 points
    Hi Ted,

    The 100kΩ will not have any impact on the quality, you should be okay. Can you share the schematic?

    Thanks,
    Paul
  • 0 in reply to Paul_Frost
    Intellectual 350 points

    Paul,

    Neither 10k nor 100k resistors had any impact on quality or the issue, unfortunately.

    From my testing, it seems the charge pump shuts down after about 12 seconds after pausing and the CAPP pin goes down to about 80mV.

    I think the next best choice is to couple two NFETs from AGND to the +/- and connect their base to CAPP, so they sink AGND to those speaker lines when CAPP goes "low".

    Here's my schematic:

    Thanks!

    -Ted

  • 0 in reply to Theodore Salmon
    TI__Guru 55113 points
    Hi Ted,

    The method you recommend is the surest way to prevent clicks or pops on the output, as you are implementing an external analog mute. I believe there are small ICs available for this purpose, but they require some controller to know when there is zero data present. Does your I2S source have the functionality?

    Thanks,
    Paul
  • 0 in reply to Paul_Frost
    Intellectual 350 points

    Paul,

    My I2S source doesn't have this ability, unfortunately.

    Am I correct in assuming, then, that connecting to the CAPP pin on the PCM5102A is not a good choice for driving the analog mute circuit?

    Thanks,

    -Ted

  • 0 in reply to Theodore Salmon
    TI__Mastermind 46290 points
    Ted,

    Paul is out of office today.

    I'm not totally clear on what your question is.
  • 0 in reply to Kevin Duke
    Intellectual 350 points

    Kevin,

    I was just wondering if the Charge Pump's Positive Pin could be used to drive an analog mute circuit, since the pump goes high when the DAC is working, and low when it goes into power-off mode.

    However, I have found that the BC127 has a configurable GPIO pin that will go high when streaming audio and low when not. I think this will be my best bet to drive the analog mute.

    Thanks!

    -Ted

  • 0 in reply to Theodore Salmon
    TI__Mastermind 46290 points
    Ted,

    I see. Thanks for clarifying.

    I think I would agree with that being the preferred approach since CAPP isn't really fully described in the datasheet to assess the reliability / behavior of the alternative approach.