• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 2 answers
  • Subscribers 47 subscribers
  • Views 141 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.

ADC34RF52: Input Clock Specifications

Joe Paolino
Intellectual 310 points
Part Number: ADC34RF52

Hello,

Would you clarify TI's definition of VID (differential input voltage) for the ADC34RF52 Clock input.  The spec indicated

Is VID measured from signal itself as VOD seen below:

or is it measured as what is seen by the ADC's difference amp which effectively doubles the output swing when the input signals are summed:

I'm trying to determine the ratio of the step up transformer I should use to increase LVDS to approach the max spec of 2.4 Vpp

Thank you,

Joe

  • 0
    TI__Mastermind 19555 points

    Hi Joe,

    VID will be measured as the difference between the positive and negative pair. The first image you posted shows this.

    For this device, the clock should be AC-coupled and is self biased at the ADC side, Each half of the differential clock input should swing a max of 1.2V around your shifted LVDS common mode. For instance, if you have an 0.8V common mode on each of the differential lines, each individual line should swing ±0.6V (0.2V through 1.4V). This creates a 1.2Vpp swing on the half of the differential signal. As each half of the differential pair is 180° out of phase, then the total differential voltage will be twice this, at 2.4Vpp. Does this help to clear up any confusion?

    Regards, Chase

  • 0 in reply to Chase W
    Intellectual 310 points

    Thanks, Chase.  So it comes down to each half of the differential clock signal must be equal or less than CLKVDD (1.2V) when measured peak to peak.  Is that right?

  • +1 in reply to Joe Paolino
    TI__Mastermind 23030 points

    Hi Joe,

    Yes, this is correct, if you go by the max input level of the clock.

    I would suggest the clock level being lower than 1.2V SE on each clock leg and more in line with typical 1Vpp DIFF voltage instead as per the datasheet.

    Thanks,

    Rob

  • 0 in reply to Rob Reeder
    Intellectual 310 points

    Hi Rob, the primary reason I'm interested in maximizing clock amplitude is from the TI article "Clock Jitter Analyzed in the Time Domain, Part 3".  Clock amplitude was shown to have a noticeable impact on SNR.

    In my case, I will be particularly sensitive to clock performance because I will be sampling 900 MHz at ~500MSps.

    Is there a potential problem if I push each half of the clock to the 1.2V SE max instead of the 0.5VSE (1Vpp Diff) voltage?

    Thanks,

    Joe

  • 0 in reply to Joe Paolino
    TI__Mastermind 23030 points

    Hi Joe,

    As long as you are below the abs max ratings on the clock input pins, it should be fine.

    Maximizing slew rate of the clock will help, but only if the clock is clean to begin with.

    I would try a few experiments and "dial in" the right clock amplitude level based on your performance expectations.

    Regards,

    Rob

  • 0 in reply to Rob Reeder
    Intellectual 310 points

    Sounds good.  Thank you, Rob.  Yes, it's very clean (25fs).