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.

LM5116 webench simulation changes?

Other Parts Discussed in Thread: CSD18504Q5A, CSD18537NQ5A, LM25116, LM5116

Hi,

I had a design went through below for a 24V to 5V@15A buck a few days back.

Design : 4561043/188 LM5116MHX/NOPB

LM5116MHX/NOPB 22.0V-26.0V to 5.00V @ 15.0A

Now I ran the same settings with webench and only 4-5 options to choose. This design is no longer there due to no FET to select. Actually what I am trying to do is to replace the Infineon FETs with CSD18504Q5A and CSD18537NQ5A to see if they works better in webench simulations.

If you can't pull the old design I can attach pdf here.

Any help or comments are appreciated.

  • FZ37,

    When I look at the design I am seeing some anomalous behavior at about 13.5A output current. The design appears to be working correctly other than the issue I saw. Let me check with the developers to see if anything has changed recently that would impact the design. WEBENCH® does not allow me to change the FETs either.

  • Britt,

    Thank you for looking into it.
    Another thing I noticed is the RDSON were all zeros for M1 and M2 in the old design.

    I ran a new one with 5V@12.5A output.
    Design : 4561043/203 LM25116MHX/NOPB
    LM25116MHX/NOPB 22.0V-26.0V to 5.00V @ 12.5A
    And the RDSON numbers are not zeros. But it still could not take 15A load for LM25116 or LM5116.
  • FZ37,

    Yes, I noticed that too. We did a fix to resolve an issue with FET selection and I believe that Is why you cannot find this solution now.

    I have not heard back yet, however, as soon as I have some information, I will post a follow-up.

  • Hi FZ37,

    Thank you for using WEBENCH. I sincerely apologize for the confusion and inconvenience caused by Rdson'0' bug. If you have created your original design between July25th-August12th, you must have have run into this bug along with inconsistencies in design creation between July 29th-12th. The fixes were released on August12th. It does seem like you re-launched your original design at 12.5A it should have looked as expected as you confirmed.

    The second part of the problem you mentioned was design creation at 15A using LM25116. You are right, WEBENCH fails the designs saying suitable FETs not found at 15A. WEBENCH designs are conservative, as we use worst case scenario to qualify FETs for a design. For instance, we use Max Rdsons, Max ThetaJA to estimate temperature rise at the junction, if the calculations indicate overheating with these MAX values we fail the FETs.

    With that in mind, I see that you started off by reducing load to 12.5A and were able to create a design. I was also able to create the below design at 13A at 75kHz and 30C. TO test if the design will startup in atleast Simulation, I used the "Schematic Edit" feature-> then changed the Rload resistance to 0.333Ohms for a 5V/15A load. And the start up results look fine with the recommended FET pair. To ensure the design works, you may have to build this board and bench test it under 15A load and stress test to ensure it works under all temperature conditions. We cannot guarantee this design, but may be a good place to start.

    Please let us know if we can assist further.

    Best Regards,

    Pavani Jella

    WEBENCH Applications

  • Hi FZ37,

    We noticed that there was an issue with our FET selection process. We were over estimating RMS currents in the high and low side switches, causing most of the eligible FETs to get filtered out. This is fixed now. Please notice that the below link takes you to a 15A design succesfuly. Please test it out and let us know if you have any concerns.


    Regards,

    Pavani 

    WEBENCH Applications

    https://webench.ti.com/webench5/power/webench5.cgi?VinMin=22&VinMax=26&O1V=5.00&O1I=15&base_pn=LM25116&AppType=None&op_TA=30&lang_chosen=en_US&origin=pf_panel&optfactor=3 

  • Pavani,

    Thank you so much for looking into the issue. It seems working fine and showing better results.

    Regards,
    FZ