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.

TPS23757: circuit design review

Part Number: TPS23757

Hi:

I have designed a circuit largely based upon the  TPS23757 EVM design but the circuit does not need to operate from any adapter power.  I would greatly appreciate a quick review of the attached design to ensure that I am handling the APD and PPD pins correctly and that I have not made any significant errors.

Thanks in advance,

Craig

hhc_poe.pdf

  • Hello Craig,

    I can review your circuit later this week. In the meantime, I have notified our PD expert to keep an eye on this thread.

    Thanks,
    Thomas A.
  • Hi Thomas:

    Have you had a chance to look over my schematic yet?  Would very much appreciate any feedback.

    Thanks,

    Craig

  • Hello Craig,

    Sorry for the delay due to training last week.

    Since you are not using an adapter:
    Pull PPD down to VSS
    Pull APD down to ARTN (not VSS)
    Pull APb down to ARTN
    10k resistor between APb & Vc is not needed

    Other differences:
    100nF bypass for VC is smaller than 220nF on EVM
    Missing compensation resistor & cap on secondary side of optical feedback (R33 & C30 on EVM)
    Double check termination on data side of Ethernet transformer (may be fine depending on PHY)

    Thanks,
    Thomas A.
  • Hi Thomas A.:

    Thanks for taking a look at the design and catching the incorrect ground returns for the APD and APb pins for adapter-less operation.  I had already found and corrected the wrong Vc cap and also fixed some bad PCB footprints.  I also had to change the MOSFET used for Q1 in the original reference design from the now obsolete Si4892DY to the new Q1 as a Si4134DY.  This looks fine to me but do you agree?

    I am still confused about the feedback you gave me for the original R33-C30 compensation network.  I removed this network because it was marked with a note (1) that said the parts were not used.  In addition, if the R33 were used as 0 ohms, that would simply parallel C30 with C29.  Are you saying that I should still include the series R-C in parallel with C29 because it may be required?  Please elaborate

    I have attached my latest schematic revision for clarity.

    Thank you again for your support,

    Craig

    5826.hhc_poe.pdf

    I have attached the latest revision of the circuit. 

  • Hi Craig,

    That newer FET looks fine to me.

    R33, C30 network: That was a mistake on my part. You do not need these components.

    Your new schematic looks great. All the changes look correct.

    Best of luck,
    Thomas A.