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.

LM5143-Q1: When EN enable, MOS will damage

Part Number: LM5143-Q1

hi team,

now customer use LM5143-Q1,meet a problem, this case is very urgent.

device work condition: VIN =24V, Vout=14.4v, the typ Iout=6A.

when customer enable the EN, the mos of Vout2 will breakdown.

of course customer test when remove MOS of Vout1 or Vout2, this circuit can work normal.

the attached file is schematic, could you help support this problem.G3_SCH_0305(3).pdf

  • Hello Colt 

    • Please check the dead-time if it is not enough, please try zero ohm gate resistor at R265012.
    • Does the damage happen at no load? If it doesn't happen at no load condition, please increase the load slowly while monitoring the temperature of the MOSFETs. 

    -EL 

  • Hello Ericlee,

    Thank you for your feedback.

    Now one phase circuit is Ok,no matter Vout1 and Vout2,but when I tried to use two phase circuit such as attached schematic will have a big current at no load and  damage the mosfet.

    The test happened at no load.

    I can still hear the sound of inductance saturation,I will make a significant improvement by changing the inductance to 4.7u, and the current will also decrease.

    Thanks

    BR Vincent

  • Hello Vincent 

    If you hear an audible noise, the issue might be caused by a loop instability. Please try R265011 = 12k C265005= 5.6nF combination. This change will reduce your crossover frequency while keeping the compensator zero frequency as is. 

    -EL

  • Hello Vincent 

    Also, please check these 6 components and their connections. If current sense inputs are not balanced in dual-phase single-output configuration, there can be a situation 'A' channel sourcing current, while 'B' channel sinking current. 

    -EL

  • Hi EricLee,

    Based on your suggestion, I tried to modify the parameters of R265011 = 12k C265005= 5.6nF, but the impact was minimal and the noise from the inductance could still be heard. Afterwards, I made several changes:

    1. Remove the C265024 capacitor and add a 10U/1210 capacitor near the layout of R265005.

    2. Change C265030 to 10U/1210.

    3. Remove C265008, connect 32pin RES to VDD, and disable hiccup mode.

    The circuit has become normal, and I can hardly hear the noise of the inductance,and I tested some parameter such as jitter, ripple were OK.

    Regarding the changes in 1 and 2, I have tried using small-sized capacitors (0805 and 0603) that have no effect on improving inductance noise,
    Next week, I will investigate which one or two of these changes are effective.
    Do you have any proposals, thank you.

    BR Vincent
  • Hello  Vincent 

    Good to hear your circuit bocome normal now.  

    I think it is valuable to double check your componenet selections (especially loop compensation componenets)  using Qucik Start https://www.ti.com/lit/zip/SNVC229 . Please don't forget entering the DERATED output capacitor values. 

    -EL 

  • Hi EricLee,

    My PCBA was left in the laboratory for two days over the weekend. When it was powered on Yesterday, there was a high current. I tried several times and it was the same phenomenon. When I measured it with oscilloscope, it became normal again. I left PCBA for a day to reproduce the phenomenon, but it did not appear when it was powered on today, and the current was normal.

    BR Vincent

  • Hello  Vincent 

    As I mentioned, I think it is valuable to double check your componenet selections (especially loop compensation componenets)  using Qucik Start https://www.ti.com/lit/zip/SNVC229 . There is a possibility that your system stability is at the boundary. 

    -EL