• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 47 subscribers
  • Views 411 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.

Reference design for AFE5801

John J Wu
Genius 3870 points
Other Parts Discussed in Thread: AFE5801, VCA2615

Hi, we are referring reference designs for AFE5801, the circuit for input channels in upper left corner on page 20 in the SLOU257A user guide, "AFE5801 8-Channel Variable Gain Amplifier (VGA) with Octal High-Speed ADC". We have some questions:

1)      Why do they bother with the bias connection? C56, C58 and C59 decouple it completely so it does nothing.

2)      If you ever installed C102 it would short out half the primary winding of T1. This would basically render the transformer non-functional.

3)      What’s the point of C57? Is this in case there’s some DC bias on the input? What would cause that?

4)      Given that the input is single-ended, why have the balun at all. According to the datasheet the AFE is fine with single-ended inputs.

 It looks like we should just terminate our coax with a 50R resistor and connect the signal to INP and connect INM to a 100nF to AVSS. Would you agree?

Thanks,

John

  • 0
    TI__Expert 6320 points

    John,

     

    1)  These components are most likely design to allow for either DC or AC coupling.  The user wouldn't need the bias point in the case of AC coupling, and wouldn't use the caps in the case of DC coupling.

    2) I am not sure why that is there.

    3) In the case of a test signal, there could be a DC offset.  Such as a signal generator.

    4) The input to the AFE is differential and can be AC or DC coupled, differential or single-ended.  I believe the noise performance is better with the differential connection, thus the balun. I can verify this.

    5)This is the simplest configuration, but I believe there is a performance loss.

    Thanks,

    Chuck Smyth

  • 0 in reply to Chuck Smyth
    Prodigy 50 points

    With regard to the AFE 5801 (and presumably that entire series), what is the low frequency response?  One can select DC coupling inside the IC (to the PGA?) but is there DC coupling all the way to the ADC? This part is of interest for a non-ultrasound application if there is decent low-frequency response down to at least a few kHz.

    Thank you,

    Larry Lucas

  • 0 in reply to Larry Lucas
    TI__Expert 6320 points

    Larry,


    There is not much info on the AFE5801  low end. I believe the performance will be comparable to the VCA2615, which also doesn't contain much info.  I will look into this and get back to you.


    Thanks,


    Chuck Smyth

  • 0 in reply to Chuck Smyth
    Prodigy 50 points

    Chuck,

    Thanks for the info. The VGA in the VCA2615 is internally AC coupled, so it is unlikely to have much response below ~50 kHz. If that is also true of the AFE5801, then it won't work for our project, although It is a nice part for traditional ultrasound.

    Larry

  • 0 in reply to Chuck Smyth
    Prodigy 50 points

    Chuck,

    After reading the AFE5801 data sheet very carefully several times, I conclude that DC coupling all the way from the input to the ADC is possible. There is even a register bit to set to put 0.1 V or 0.15 V DC on the buffer input to test the ADC output response. The bit also sets the internal AC coupling to off while doing the test. So there is still hope for the design that I was thinking about.

    Of course, none of that says anything about the uniformity of the low frequency response. I will try to test that with the EVM. Since the EVM comes with transformer coupling, that will have to be changed first, and the common-mode voltage will have to be matched. A nuisance, but it can be done.

    Any information that you can add will be greatly appreciated.

    Larry