• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 2 answers
  • Subscribers 63 subscribers
  • Views 167 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.

TPS22992: Corrupted by unknown cause

TerryU
Prodigy 145 points
Part Number: TPS22992

Hi Experts,

We assembled three pieces of boards using TPS22992, and we have found one of designated TPS22992 corrupted on all of the three boards. Although we use two TPS22992's on the board, one of them was fine. The corrupted ones are used for simple 5V line ON/OFF switch. And those failed boards have been working fine after replacing TPS22992. All of TPS22992's, corrupted ones, fine ones and replaced ones, are exact same batch from one reel.

I attach the circuit diagram of the corrupted one, and damaged photo of the worst damaged one. Regarding other two board, there are no visible damage, but don't work. It looks burnt so the direct cause might be electrical stress rather than physical stress.

I have been thinking the cause of this issue, however there is no idea. Does anyone have any idea? Even any possibilities would be help for me.

circuit_diagram_TPS22992.pdfDamage_photo.xlsx

Regards,

Tetsuya

  • 0
    TI__Genius 16190 points

    Hi Tetsuya-san,

    The schematic looks ok - the images you sent show the QOD pin as the most burnt, which is strange as it is NC in your schematic. Is there any chance there could be a hard from VOUT to GND or QOD is shorted to a DC power source the first time the board is powered up?

    Otherwise, electrical stress/internal failure can happen due to physical stress during manufacture. When you replaced the parts, did you use a different method (i.e. hand soldering) than during manufacture? If so, maybe the manufacturing process is harsh towards specifically this part?

    Please let me know what you think of those theories.

    Thanks,

    Patrick

  • 0 in reply to Patrick Shalton
    Prodigy 145 points

    Hi patrick

    Thank you for rapid response.

    Regarding QOD. It is NC and no trace on the board and weak physically. So I think the pad was removed at component replacement.

    Assembly and replacement were done by our partner factory. So i think assemble process would be fine.

    regards

    Tetsuya

  • 0 in reply to TerryU
    TI__Genius 16190 points

    Hi Tetsuya-san,

    Understood about QOD and assembly. Are there any components nearby or directly on the other side of the board that can get very hot? What is the typical load current?

    Thanks,

    Patrick

  • 0 in reply to Patrick Shalton
    Prodigy 145 points

    Hi Patrick,

    On the maximum current situation, 5V line run with 4.5A, however I have not applied load current in this case. it failed before applying load current. Heat component, FPGA, is a bit far and the FPGA component is not so hot. 

    One concern is that there is a large through hole type connector near the corrupted TPS22992. I think when the large connector is assembled, it would apply hand soldering or spot flow after soldering reflow process. And there is large GND layer and the layer is common with TPS22992. So it might apply heat at the connector soldering process via GND layer. However it would be physical stress, not electrical one.

    The attached excel is photo of a part of the board as reference.

    Board_photo.xlsx

    Regards,

    Tetsuya

  • 0 in reply to TerryU
    TI__Genius 16190 points

    Hi Tetsuya-san,

    Ah, I see, that looks to be the cause of the issue. I see on the original photos you sent the GND pad on the layout looks the most damaged aside from QOD.

    In the next build you do, can you try either lowering the temperature of the through hole soldering or soldering the damaged TPS22992 after the through hole connector? If the heat from the through hole connector is actually the issue, we'll need to work out a way to mitigate that transfer.

    Thanks,

    Patrick

  • 0 in reply to Patrick Shalton
    Prodigy 145 points

    Hi Patrick

    Thank you following up. Yes, it might need any mitigate action. I will think about it.

    Regards,

    Tetsuya

  • 0 in reply to TerryU
    TI__Genius 16190 points

    Hi Tetsuya-san,

    Understood - please let me know if you have any additional questions following or during the corrective actions.

    Thanks,

    Patrick

  • 0 in reply to Patrick Shalton
    Prodigy 145 points

    Hi Patrick,

    I appreciate your so much support.

    Thanks,

    Tetsuya