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.

Unstable operation, hot MOSFET on TPS23753A following PMP8803

Expert 1600 points

Replies: 4

Views: 1313

Hi,

    I followed PMP8803 to design a 48Vin to 12Vout/1A PD. Below is the SCH. It is almost the same as PMP8803, except MOSFET is CSD19533Q5A and one output cap is a 25V/10uF electrolytic capacitor

With 48V input, I found the MOS is very hot if Io>100mA. Below is when Io=100mA, CH2 is Vds, CH3 is CS(current sense). There is big noise on CS, and the MOSFET is hot.

Below is When Io=200mA. It works unstably and there is a lower frequency oscillation. It looks overcurrent comes periodically.

I've changed the gate resistor R10 from 10ohm to 0 and used a 150V MOS, but no any improvement.

Which measure can I do to locate the cause?

Thanks.

4 Replies

  • Hi Di,

    Where is your 48V power supply being placed? Is the output regulating <100mA?

    Is C5 an electrolytic?

    The schematic looks okay (the class resistor should be 90.9 ohms and not k but this shouldn't explain the unusual waveforms with a 48V power supply input). There might be a misconnection or a different component value somewhere on the board itself. So that might be good to check.

    Do you also have a .brd file of the layout?

  • In reply to Darwin Fernandez:

    Thank you, Darwin.

                   The issue is the same with 48V over RTN and Vss. I don't think the output regulating with any load. I found sometimes the output was obvious brownout. I also found the board consume 0.5W in no load.

                   C5 is an electrolytic. I've attached the .brd file. Could you recommend any measure I could try?

                   Thanks.

                   6082.PD_TPS23753A-0930.brd

  • In reply to Darwin Fernandez:

    Cause might be the RCD circiut(D4 R2 C6).

    D4 is SS110, which is stated as a schottky, but the reverse recovery time is unknown.

    R2 is 0603 package and C6 is 0805. 

    I changed D4 to ES1D which is a superfast diode. The heat decreased a lot on MOSFET.

    Any recommendation on selection of RCD devices?

    Thanks!

  • In reply to Di Wang1:

    Hi Di,

    It's best to stick with the reference design components and values. This reference design is a known working solution.

    However, if you'd like to experiment with different components, I recommend you look at the app note below. It shows step by step design equations to the different circuits in the poe design. Page 9 discusses the RCD snubber.

    http://www.ti.com/general/docs/lit/getliterature.tsp?baseLiteratureNumber=slva305&fileType=pdf

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.