• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 11 subscribers
  • Views 296 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.

AFE7900EVM: Signal Quality Problem? And how to change sample rate?

Mike Neuman
Prodigy 150 points
Part Number: AFE7900EVM

I've followed TI204c-setup exactly. I'm producing a 1.83GHz signal, and getting the following result:

1.83GHz signal, 1800 NCO

This looks great and as expected.

Now, move the signal down to 1.5GHz:

1.5GHz signal, 1800 NCO

This has obviously gone awry. Something is happening at DC (although why it's flat isn't clear to me). Also several spurs.

If I move the signal down to 1.45GHz, the output is virtually indistinguishable:

1.45GHz signal, 1800 NCO

Same problem.

So, try again with the NCO at 1500:

1.55GHz signal, 1500 NCO

All sorts of problems here.

I'm guessing this has something to do with the ADC sample rate being around 3GS/s? Is this expected behavior? Is there a recommendation on how to sample in this frequency range?

Second question: Using latte, what are the steps to change the sample rate to something other than 737.28MS/s?

Thank you in advance!

  • 0
    TI__Genius 17891 points

    Hi Mike,

    The issue that you are seeing is that you are operating near the Nyquist boundary. You are seeing is the image in the 1st Nyquist boundary, which causes the HSDC Pro to see it as a spur or as the fundamental tone. 

    Ideally you would select the NCO and Fs such that the nyquist boundaries are not within the band of interest. 

    The data rate is dependent on the sampling rate and the decimation/interpolation rates. You can update these parameters in the AFE79xx GUI script to change the data rates. Then in the TI204c-IP you will have to update the SERDES lane rate in addition to the clock frequencies to match the script. 

    Regards,

    David Chaparro

  • 0 in reply to David Chaparro
    Prodigy 150 points

    Hi David,

    I understand the Nyquist zone, but at worst this should produce one or two tones, not all the noise seen at DC and surrounding the tone. Is the signal at half the ADC rate (and all the multiples) just not usable? Or is this a configuration or interference issue?

    On the other issue--are there any guides that step me through this? Otherwise it's a couple hours of guessing whether I got all the settings right. :-)

    Thanks!

  • 0 in reply to Mike Neuman
    TI__Genius 17891 points

    Mike,

    The other spurs should be caused by HSDC Pro. Please change the setting below and you should see them disappear. 

    Go to Test Options -> Notch Frequency bins and set the number of bins removed to 0. 

    Regards,

    David Chaparro