• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 29 subscribers
  • Views 426 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.

PCM1863 Gain Issues

Ed Arrington
Prodigy 90 points
Other Parts Discussed in Thread: PCM1863

I am using the PCM1863 to capture audio signals, and getting some very strange results when I set the gain levels in the positive range at 6dB and above.

The graph below shows some of my findings.  The input in all cases is an identical 300Hz signal.  The shift in phase only has to do with my capture method, and is not an issue.  The change in amplitude is the issue.

The blue line shows the signal at 0dB.  This signal looks great.  However, the orange signal, which sets the gain to +6dB shows an increase in amplitude in the positive side, and some kind of clipping on the negative side.  The amplitude on the positive side is not quite the 2x you would expect, and the negative side is obviously clipped.

Increasing the gain to +7dB generates the grey line.  The resulting amplitude on the positive side is closer to 0dB, and on the negative side is very clipped.

Pushing the gain to +8dB results in the yellow line, which shows almost no amplitude.  Zooming in on the graph, there is an amplitude change, but it is obviously small.

I am capturing this data immediately after it comes off the ADC, with no processing.

I assume the PCM1863 is not intended to work this way.  I have tried this in both differential and single-ended mode with similar results.

Is there a setting I'm missing somewhere?  How come the gain doesn't work as expected?

Thanks,

  • 0
    TI__Mastermind 28585 points

    Hi Ed,

    Is this your own design or are you testing an EVM? What is the signal source and amplitude?

  • 0 in reply to Converter Apps
    Prodigy 90 points

    Thanks for the reply.  This is our own design, we are not using the EVM.  The audio source is a sine-wave generator on my iPhone.  As near as we can tell, the signal reaching the PCM1863 is the same in every case, as the only change between the graphs is a software change to the PGA registers (0x1 and 0x2) for both left and right.  We are sampling at approximately 48kHz.

    Thanks for any wisdom you can add!

  • 0 in reply to Ed Arrington
    TI__Mastermind 28585 points

    Is this with iOS 8.0? :-)

    We would probably want to start by looking at your schematics just to rule out any hardware issues.

    It looks like some DC offset is causing clipping.

  • 0 in reply to Converter Apps
    Prodigy 90 points

    Turns out it is IOS8 :-). but I've had the app since IOS7.

    Thank you again for the response.  I'm happy to share the appropriate portion of our schematic, but I'm a little confused as to why that would make a difference.  The analog audio going into the PCM1863 is the same in every case I outlined above.  Again, the *only* change between the graphs to the gain control inside the ADC.  Everything outside the ADC is exactly identical in all cases, right?

    How would a change in gain inside the PCM1863 be affected by DC offsets in the analog domain ahead of the ADC?  That would make for a very unpredictable part :-).

    Perhaps reference voltages?  Our schematic looks just like Fig 51 in the documentation, except with 1.0uF and 1.1uF caps on the VREF and LDO respectively.

    Thanks,

  • 0 in reply to Ed Arrington
    Prodigy 90 points

    Just pinging again.  Are there any suggestions on resolving this gain issue?  Is anyone else having trouble with gain on the PCM1863?