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.

TPS40422: temperature effect on the loading performance

Part Number: TPS40422

Hi,

I have a TPS40422+CSD87350 two phase design, the target loading is 60A. Attached is the schematic.

when power up with 40A loading, the board works fine, but failed to work with 60A loading.

I tried a lot check with component, soldering issue, but no improve. then I found after I heat the board to a level like above 40 degree Celsius. the board works.

So, what is the root cause for the board fail? and why the pre-heat can make it work? is there any issue with my schematic design?Power Sch.pdf

  • Hello,

    The schematic looks fine.
    Can you use TI Fusion to check which fault is declared when issue happens?
    Also, please check if temperature readout and current readout are normal during operation.

    In PMBus, IOUT_CAL_GAIN must be equal to inductor DCR value, otherwise current readout and OC limit will not be accurate.

    Thanks
    Qian
  • Hi Qian,

    I found the TI Fusion SW tool, How can I connect the board to this tool? Is there any debugger to use?

    Thanks

  • Where can I buy USB I/O interface?
  • Hi,

    The USB I/O adapter is available at TI store.

    Thanks

    Qian

  • The adaptor has a 10-pin connnector.
    Connect adaptor Pin 9 to TPS40422 pin 12 (CLK)
    Connect adaptor Pin 10 to TPS40422 pin 11 (DATA)
    Connect adaptor Pin 6 to board ground.

    I guess OC happens because PMBus command IOUT_CAL_GAIN is not progrommaned be equal to inductor DCR.
    Also, the OC threshold per phase is only 30A in default, it is too low for your application. PMBus command IOUT_OC_FAULT_LIMIT is used to set OC threshold.

    What's the L and DCR being used on board? For temporary solution, you can add a resistor in parallel with C563 and C565 to attenuate current sense signal, hence to solve OC problme.

    Thanks
    Qian
  • Hi Qian,

    Yeah, I got the adaptor, and there is OC fault.

    But actually, we have mass production last month, and there is no issue with the default setting (we already have 1000 pcs board shipped with default setting). I checked the component, found that the failed board installed TPS40422 with data 2016 Oct. and the good board is new chip in 2018.

    So seems that the part is too old, right? now we have 500 failed board, and I want to repair the board with minimum effort, do you have any suggetion?

    "For temporary solution, you can add a resistor in parallel with C563 and C565 to attenuate current sense signal, hence to solve OC problem." What is resistor value to aprallel with C563?

    Thanks

    Mingsen

  • HI Qian,

    Latest update: I change R623 and R624 from 2.2K Ohm to 4.7 K Ohm. the board works!
    Could you please check if this modification is OK? is there any risk?

    Thanks
    Mingsen
  • Hi Mingsen,

    Can you let me know the inductance and DCR value? What's the OC threshold you expect?

    If changing R623 and R624 value is not enough, are you able to program the controller on those 500 failed boards?

    Thanks

    Qian

  • Hi Mingsen,

    I just sent an message with my email addrss.
    Let's continue the discussion by email.

    Thanks
    Qian