• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 81 subscribers
  • Views 783 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.

F2837xD ADC to DAC bypass

Aymeric Dutruel
Prodigy 50 points

Hello!

I'm currently working on the F28377D EVM and I'm doing a bypass between the ADC C and D (in 16bit resolution) and the DAC A and C in order to evaluate it and to implement, in the future, a numeric filter for audio application.

But, I have a problem when I put a sine in the ADC, I observe at the output of the DAC the same signal than the input but with a noisy 400kHz signal (which doesn't depend of the input signal frequency). Do you know what is the source of it?

Moreover, this problem disapears when the ADC is turned into the 12 bit resolution.

Regards,

  • 0
    TI__Guru 62975 points

    The behavior that you describe is not expected.

    What are you using for the DAC reference voltage?  (VREFHI vs VDAC).  Do you see any noise on the reference voltage?

    What is the magnitude of the 400kHz noise?

    I assume that you are right-shifting the 16-bit ADC results when programming the DAC?

  • 0 in reply to tlee
    Prodigy 50 points

    Hi tlee,

    I am using for both ADC and DAC the internal reference and i don't see any noise on it.

    The magnitude of the noise varies in function of the input magnitude but, it is in average, 300mV.

    You are right, I have scaled the 16 bit ADC result for a 12-bit DAC.

  • 0 in reply to Aymeric Dutruel
    TI__Guru 62975 points

    Aymeric,

    300mV sounds like a very large magnitude for noise...

    Can you try to isolate whether the corruption is at the ADC, DAC, or somewhere in the transfer?

    For example, if you capture a large set of ADC samples and plot the data, does the waveform look correct?  And conversely, if you manually output the same set of samples on the DAC (using something like a for-loop), is it able to reproduce the data as expected?

    The sample size should be large enough to include a few instances of corruption from the 400kHz signal.

    -Tommy

  • 0 in reply to tlee
    Prodigy 50 points

    Ok thank, you I will try what you said.

    I know that it's very large for a noise it seems like a modulation.

  • 0 in reply to Aymeric Dutruel
    TI__Guru 62975 points

    Aymeric Dutruel said:

    Hi tlee,

    I am using for both ADC and DAC the internal reference and i don't see any noise on it.

    The magnitude of the noise varies in function of the input magnitude but, it is in average, 300mV.

    You are right, I have scaled the 16 bit ADC result for a 12-bit DAC.

    It was just brought to my attention that you are using internal reference for the ADC.  The internal reference mode on this device is currently not supported because of performance issues.  I would not expect these issues to introduce 300mV worth of noise, but officially, the internal reference mode should not be used.