• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 64 subscribers
  • Views 232 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

Original question:

LMG3410-HB-EVM：Switch temperature is too high

LMG3410-HB-EVM: Upper Arm Switch Overheat Problem

Brong Chen
Prodigy 40 points
Part Number: LMG3410-HB-EVM

Tool/software:

I use LMG3410-HB-EVM as the high freq. arm of my totem-pole PFC. It was broken three weeks ago, and I have fixed it many times, the heat of the Upper arm switch still higher then expected, especially when I supply 5V for this board, and the problem gets worse and worse. The FAULT of Upper arm is low level yesterday when i fixed it last time, before then, it is high level and I didn't detect any obvious short or signal abnormality. I think whether I broken a layer of board when I re-solder the switches.I use full bridge rectification to test whether the circuit is corrected, if i don't supply 5V for this board, the positive half cycle of input current is higher then negative one for about 0.1mA. Otherwise, there're only positive half cycle of input current. By the way, the specification of the circuit is: 230Vrms for Input Voltage; 400V for Output Voltage; Power is 12kW; and Switch Freq. is 65kHz.

  • +1
    TI__Expert 7583 points

    Hi Brong,

    Can you please let me know what your operating conditions and drain current are? What is the temp rise of the upper arm? Is it 4degC higher than lower arm?

    Thanks.

    Best,

    Pratik A

  • 0 in reply to Pratik Adibatla
    Prodigy 40 points

    Hi Pratik,

    The test conditions are: input voltage of 2.5 Vrms and input current of 0.8 A. The temperature of the upper arm is 35.5 °C when 5 V is supplied to the GaN board, and 32.5 °C when 5 V is not supplied. The upper arm is 7 °C higher than the lower arm.

    The temperature issue has been solved. It was caused by insufficient solder on the pad of the GaN FET.

    However, another problem still remains. It is unclear whether the issue is due to a damaged inductor in the GaN FET buck-boost converter. When testing full-bridge rectification of the totem-pole PFC without supplying 5 V to the GaN board, the input voltage is 230 Vrms, the input current is 3.21 A, and the temperatures of both the upper and lower arms are 32.5 °C. But when 5 V is supplied to the GaN board, the negative half-cycle of the input current becomes 0 A.

    I measured the voltages at all test points on the GaN board without applying input voltage and only supplying 5 V. There is no abnormality on the lower arm, while the FAULT, RDRV, and VNEG pins of the upper arm are all around 0 V. Based on the operation analysis of the full-bridge rectification in the totem-pole PFC, it appears that the lower arm is the faulty one.

    Thanks.

    Best,

    Brong

  • +1 in reply to Brong Chen
    TI__Intellectual 1150 points

    Hi Brong! 

    Two things:

    Please confirm if you are following the bias supply recommendation in your circuit: 

    If yes, could you please probe the VDD pin of Q1 (high side GaNFET) and make sure that the pin gets 12 V.

    Thanks,

    Best,

    Adhvaidh

  • 0 in reply to Adhvaidh Aravind
    Prodigy 40 points

    Hi Adhvaidh,

    It's about 14V for VDD pin of both Q1 and Q2

    Thanks.

    Best,

    Brong

  • +1 in reply to Brong Chen
    TI__Intellectual 1150 points

    Hi,

    Thanks for confirming!

    1) Could you also check if the LDO_5V pin gets 5 V. 

    - If no, then the FAULT being zero indicates the UVLO at the LDO5V pin

    - If yes, then, there is some issue on the buck-boost side.

    2) In your actual question, you mentioned that you did not supply 5 V, and observed 3A of input current. Can you elaborate more on the 5V supply? Because, if 5 V is not being supplied, the board won't work, and there'll be no current drawn!

    3) Could you please send a picture of your setup? It would be great if you could annotate 

    Thanks

    Best,

    Adhvaidh

  • 0 in reply to Adhvaidh Aravind
    Prodigy 40 points

    Hi,

    The LDO_5V pin receives 5 V, so the issue appears to be on the buck-boost side. As mentioned earlier, we initially suspected the problem was caused by the buck-boost inductor. After replacing the inductor, VNEG now shows a negative voltage and FAULT is at a high level.

    I would appreciate your support on this matter.

    Best regards,
    Brong

  • 0 in reply to Brong Chen
    TI__Intellectual 1150 points

    Hi Brong,

    I am glad that your circuit is working properly.

    Best,

    Adhvaidh