• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Subscribers 81 subscribers
  • Views 388 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.

TMS320F2812: Using signal generator and built-in ADC of TMS320F2812, the amplitude falls as the frequency increases

larry fischer
Prodigy 100 points
Part Number: TMS320F2812

I am sampling a signal on ADC channel 1. When I generate a sin wave, using a signal generator, the max amplitude of the sampled signal, falls off as frequency increases.  If I test with a scope, the scope shows the amplitude to be consistent.  The frequencies tested are 100Khz, 500Khz, 1Mhz, 5Mhz, and 9Mhz.  The amplitude falls off consistently.

I have tried many different ADC clock frequencies - no change.  I have increased the S&H time.  No change.  

Any obvious reasons why this might be happening?

Thank you,
larry

  • 0
    TI__Guru 62975 points

    Larry,

    Can you describe how the signal is conditioned? Is there a high-bandwidth op-amp buffering the signal to the ADC pin?

    Are you able to use the same ADC configuration to correctly sample the signal generator in DC mode?

    -Tommy

  • 0 in reply to tlee
    Prodigy 100 points

    At this second, Im bypassing the incoming circuitry and plugging the Signal Generator straight into the ADC line.

    Im not sure what you mean by this: "Are you able to use the same ADC configuration to correctly sample the signal generator in DC mode?"  

    Larry

  • 0 in reply to larry fischer
    TI__Guru 62975 points

    Larry,

    I would recommend using a high-bandwidth buffering circuit placed near the ADC pin. It does sound like there's a low-pass characteristic in the path from the signal generator to the ADC.

    Converting a static DC level from the signal generator can help to gauge the quality of the signal path. A good path will produce a series of accurate conversions that are tightly clustered for a single channel. If this is confirmed, a good stress test would be to configure a sequence that alternates conversions between the static DC level channel and another channel on the same S/H capacitor that is sampling another voltage (like ground).

    -Tommy

  • 0 in reply to tlee
    Prodigy 100 points

    But the signal generator is directly connected to the ADC input pin...

  • 0 in reply to larry fischer
    TI__Guru 62975 points

    larry fischer said:
    But the signal generator is directly connected to the ADC input pin...

    Can you elaborate on this comment? I'm not sure how to respond.

  • 0 in reply to tlee
    TI__Guru 71755 points

    Larry,

    When you mentioned that the the scope capture of the signal looks correct, i.e. the amplitude is consistent over increasing signal frequency, I want to make sure that you are measuring at the ADC input pin and not back at the signal source.  I think both what Tommy is driving at, and here is well, is that we want to make sure there is not an un-intentional LP filter in the signal chain.

    Assuming this is true, we also have to consider that at 9MHz we are violating Nyquist (12.5MSPS/2) for recreation of the waveform, and 5MHz is getting pretty close.  With this in mind, you're not going to get alot of data each period to recreate the waveform.  My thought here is that we are simply missing the peak amplitudes the faster the input signal becomes.  

    Best,

    Matthew

  • 0 in reply to MatthewPate
    Prodigy 100 points

    Ahhh - I see.  Thank you for emphasizing that point.  I will try that.  

    Sorry for the delay - I work on multiple projects and got buried.  I will try this in the next  day or so.

  • 0 in reply to larry fischer
    Prodigy 100 points

    Uhh - if you guys are in the SF bay area, I owe you a beer or two!

    Yes - your recommendation was brilliant.  Hooking the signal generator and the scope to the DUT, shows the amplitude truly falling off.  When the scope measures just the signal generator, the amplitude is steady.

    Thank you!!!!!

    Larry