• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 30 subscribers
  • Views 223 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.

PCM3168A: part failures - clock lines at 1V

Andrew Ostler
Prodigy 60 points
Part Number: PCM3168A


We have a product using PCM3168A which has been in production for over a year. Recently we've been getting a number of failures (in the field, after sale).

The symptoms are that the generated audio clocks are at the wrong frequency, and only 1V in amplitude.

We run the codec with the ADC as master, set to 256x fs, and the DAC as slave. The system clock is 24.576MHz, so we expect a 96kHz LRCLK.

When the part fails, we get a 64kHz LRCLK, and as I say the output LRCLK and BCLK signals are only 1V.

Is this a known issue?

  • 0
    TI__Mastermind 29765 points

    Hi,

    No, this is not a known issue. Is the system clock still 24.576MHz when Fs is 64KHz? Is this 64KHz clock stable and clean?

    You might want to probe these clocks frequencies along with the BCLK for consistency, also check your IOVDD, DVDD levels.

    Regards.

    Peter

  • 0 in reply to pdjuandi
    Prodigy 60 points

    Yes, the system clock is unchanged. The 64kHz clock appears stable but of course it's no use because at 1V the rest of the system doesn't recognise it. Power supplies are still fine (3.3V/5V).

    Is there anything that could cause the chip to output clocks at less than VDD, other than simple failure?

  • 0 in reply to Andrew Ostler
    TI__Mastermind 29765 points

    Hi,

    When failed is it in slave or master mode? Is this possibly driven by external?

    Is the digital data level OK? Sounds like either driven from outside or some kind of loading on these 2 clocks.

    On frequency, you might want to check BCLK correlates with this WCLK and bit depth.

    64KHz is odd sampling rate, some setting or clock has changed.

    Regards.

    Peter

  • 0 in reply to pdjuandi
    Prodigy 60 points

    It's in master mode. Nothing else is generating clocks - the 64kHz LRCLK originates in the PCM3168A.

    64kHz corresponds to a 384 fs multiplier for a 24.576MHz system clock, so is in theory supported by the codec.

    Just to reiterate - this is something that happens without firmware or other changes. A system goes from 96kHz 3.3V clocks on one day to 64kHz 1V clocks the next. It can only be hardware failure. I'm trying to establish whether this is a known issue - for example if there has been a bad batch of parts.

  • 0 in reply to Andrew Ostler
    TI__Mastermind 29765 points

    So the clock just change by itself and is it recoverable, meaning will it go back to 3.3V and 96KHz? This is at room temp and with the same test condition and setup. Is it possible to capture on the scope the MCLK, WCLK, BCLK and DATA when it fails? Can you provide your settings as well?

    Thanks.