Because of the holidays, TI E2E™ design support forum responses will be delayed from Dec. 25 through Jan. 2. Thank you for your patience.

  • State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 46 subscribers
  • Views 643 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.

ADC12DS080: Driving differential inputs

Gonzalo Alcaraz34
Prodigy 70 points
Part Number: ADC12DS080
Other Parts Discussed in Thread: ADC12DS105

Hi all, and thanks in advance for your responses.

I'm designing a board that uses this particular IC to measure some fully differential signals. What kind of circuit should I use at the input to drive the ADC properly? The input circuitry suggested in the datasheet is made up of baluns to transform common mode signals to fully differential and so is of no use to me. 


Also, I will probably be oversampling the signals and using as much as I can get of this part's full scale bandwidth (which is 1GHz), so I have to take this into account. I've been doing some reading on various application notes on using ADC drivers, but am not sure what (if anything), I need. I've seen this part used succesfully right on the ouput of an IQ Demodulator (LTC5586) with only an antialias filter in between, but since I will be measuring different signals off of different boards, I'm unsure as to what I have to take into account.

Again, thanks for your responses, any and all help you can give me on the matter is much appreciated.

  • 0
    TI__Mastermind 44560 points
    Hi Gonzalo
    What is the reason that baluns are not a good approach for you?
    Do you need to DC-couple your input signal, or do you just need to add gain and an AC-coupled amplifier would also work?
    If you need gain, is fixed gain or adjustable needed?
    Once we have the additional detail one of our ADC experts will provide a recommendation.
    Best regards,
    Jim B
  • 0 in reply to Jim Brinkhurst1
    Prodigy 70 points

    Hi Jim, and thanks for your help.

    Baluns, as I understand it, are used to convert a single signal to differential, and since I'll probably only be measuring fully differential signals, I'd like to take advantage of that, rather than first converting diff to single and then backwards.

    One thing I know I will use this board for is to measure the output of an antialias filter fed by the output of an LTC5586, so I do need it to be DC-coupled. For this application, I don't really need gain.

    Bottom line is that what I'm trying to achieve is a board that's as flexible as possible. Do I actually need to have an ADC driver? Could I get away with an analog front end that's just an SMA connector and an impedance matched trace, and leave the rest on other boards as befits the particular needs of what I'll be measuring? 

    Best regards,

    Gonzalo Alcaraz

  • 0 in reply to Gonzalo Alcaraz34
    TI__Mastermind 44560 points
    Hi Gonzalo
    The nominal Vcm (input common mode voltage) of the ADC12DS080 is 1.5V.
    Since the output is already differential, I would recommend trying the circuit shown in Figure 14 of the LTC5586 datasheet (simplified schematic of the IF amplifier output with Anti-Alias Filter) configured for 1.5V common mode voltage. At frequencies approaching 1 GHz the best performance will only be achieved with the LTC5586 and ADC12DS080 on the same circuit board.
    You mentioned you want to use as much of the full scale bandwidth (1 GHz) as possible. Please note that since the maximum sample clock rate of the ADC is 80 MHz, the maximum Nyquist bandwidth for each channel of the ADC is only 40 MHz. You will be able to digitize frequencies above 40 or 80 MHz, but will need to do careful frequency planning of your down-converter RF, LO and resulting IF frequencies to prevent aliasing problems.
    I hope this is helpful.
    Best regards,
    Jim B
  • 0 in reply to Jim Brinkhurst1
    Prodigy 70 points

    Thank you Jim, this did clear up most of my doubts. 

    We're planning to do undersampling of the measured signals. We currently have a setup using ADC12DS105 evaluation board and an FPGA and it works quite well.

    Best regards,

    Gonzalo Alcaraz