• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 29 subscribers
  • Views 510 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

PCM5122: Charge Pump Failure

Theodore Salmon
Intellectual 350 points
Part Number: PCM5122
Other Parts Discussed in Thread: PCM5102

Hello,

I'm using a PCM5122 (and previously a PCM5102) in my design and I've had a couple of prototypes end up with the 3.3v rail shorted to ground. A little investigation has yielded that the CPVDD pin is bringing the rail near ground. Similarly, the VNEG pin has only 20k ohm resistance to ground. No other pins appear to be shorted. The outputs of the DAC connect directly to an onboard amp, as well as being tee'd off into a vehicle amplifier.

My question is this: What could cause this kind of damage to the DAC? I have not done anything outside the box with the implementation, other than change the output filter to supply a higher output load.

Here is my schematic:

Here is the layout for this part, including where one channel is input into a small audio amp.

Thanks!

-Ted

  • 0
    TI__Guru 52663 points

    Hi Ted,

    Are your capacitors polarized? Your schematic seems to indicate that they are.  If you are using polarized caps (tantalum, for example) you could have one flipped.  Specifically the flying cap C41 and the VNEG cap C40.  

    Otherwise, I would expect an assembly issue.  Did you confirm that your footprint and schematic symbol match the device datasheet?

    Thanks,

    Paul

  • 0 in reply to Paul_Frost
    Intellectual 350 points

    Paul,

    Thanks for the response! My capacitors are all ceramic :)

    If you look closely, DGND is floating (which I did not catch until today) and "VCOM" is floating as well. I have been using VREF mode this whole time, so am I good to simply tie VCOM to GND as well, or should I still decouple it?

    I would hypothesize that leaving DGND disconnected is likely what has fried my parts. Would you agree?

    Thanks!

    -Ted

  • 0 in reply to Theodore Salmon
    TI__Guru 52663 points

    Hi Ted,

    I suspect that the DGND being disconnected is what damaged your part (or the subsequent debugging).

    You should just tie VCOM to GND if you are using VREF mode.

    Thanks,

    Paul