• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 46 subscribers
  • Views 325 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.

ADS42JB49: Sinusoidal "noise" superimposed on input

Sharon Tseng
Prodigy 70 points
Part Number: ADS42JB49


I am seeing 1-4Vpp xMHz frequency sine wave superimposed onto all of my injected signals on my own test board. I have tried injecting a pulse of different widths at 5MHz, a 10MHz sine and square wave, and DC values 4V and 0.5V. The amplitude and frequency of the superimposed sine is consistent for each injected signal but varies across signals. I have 2 identical boards - one is able to see the DC values with the sinusoid imposed but one shows garbage for all DC values.

My set up is BNC connector - FDA - ADC. When grounded or left floating at the BNC connector, I see a full range sinusoid. Measuring the injected signal at the output of the FDA shows a clean signal. If I ground the lines between the FDA and ADC with a floating BNC cable, I don't see any noise.

The signal integrity of the JESD lines are good and I am sure I am interpreting the data from the ADC output correctly. All of the test signals (ADC, SERDES, link layer) and a stress toggle pattern for each data lane pass. Where could this be coming from?

  • 0
    TI__Mastermind 40630 points

    Hi Sharon,

    Please send along your schematic for this design. Looks like the amplifier might be unstable.

    Thanks,

    Rob

  • 0 in reply to Rob Reeder
    Prodigy 70 points

    FDA_ADC_pg.pdf

    Hello Rob,

    This is the FDA & ADC page. 

    To elaborate on my earlier point, I am measuring at test points TP45 and TP47, injecting into a BNC connected to CUR_IN+/CUR_IN-, and looking at the output of Channel A, which gets processed through an FPGA. These are my results:

    1. Floating BNC input
      1. Flatline at TPs
      2. Full range sinusoid at FPGA output
    2. Injected signal into BNC
      1. Injected signal with expected gain at TPs
      2. Injected signal with 2Vpp sinusoid at FPGA output
    3. Floating BNC input, GND at TPs
      1. Flatline at FPGA output 

    This leads me to believe the FDA is working appropriately but please let me know if you have other theories. Thank you!

  • 0 in reply to Sharon Tseng
    Prodigy 70 points

    Hi Rob,

    I disconnected the FDA and injected directly into the ADC inputs and everything is clean. I am still stumped on why I see a clean signal at the test points when injecting a signal into the BNC (#2a).

  • 0 in reply to Sharon Tseng
    TI__Mastermind 40630 points

    Hi Sharon,

    Try connecting the amplifier again and removing both C85/86. Probably seeing the cap charge and discharge.

    Regards,

    Rob