• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 64 subscribers
  • Views 832 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.

LM3409 - Overheating MOSFET

user4216681
Prodigy 30 points
Other Parts Discussed in Thread: LM3409

Hello,

My design is based on LM3409 + LED (4A) - Vin = 12V. I have used Webbench to generate the schematics. The MOSFET selected has a fast recovery body diode.

The system works well but the MSOFET gets hot very fast.

I suspect that the driver might not drive the MOSFET in saturation as expected so that the resitance might be too high - hence a power dissipation through this resistance.

I have measured PGATE (between PGATE and the Gnd) and it goes from 5 to 12V (Vin). Looking at the block diagram (in the D/S), I would say that this is expected as the Vcc is generated from Vin and the PGATE should switch from Vcc to Vin.

question 1 - When I looked at the max rating condition, I am not sure tio understand the 2nd row. Should we read PGATE max = 7V or (Vin -PGATE) = 7V max ?

Vcc being generated internally, I do not see how PGATE could be below 7V

question 2 - do you have any idea why the PMOS gets hot? any experience to share here?

best regards, Agnès

  • 0
    TI__Expert 3495 points
    The PGATE max rating is Vin - PGATE = 7V. It won't exceed the max rating during normal operation. There are a number of reasons the MOSFET could get hot. It could be RDSON too high, switching fequency too high etc. Please tell what MOSFET you are using and the switching frequency and I can take a look.
    David Zhang
  • 0 in reply to David Zhang
    Prodigy 30 points

    Thanks David for your help.

    The Vgs switch from 5V to 12V at a freq = 471 kHz.

    The PMOS used is a Vishay Si7633. We are now checking what the result would be if the FDD5614 would be used instead - this is the one proposed by the Webbench tool. The idea here is to try to narrow down whether the issue is implementation related or in the MOS selection.

    I am also checking how the supply and ground lines are designed. 1A is huge and I have to make sure that the supply and the ground are "star" distributed (sorry I do not have the english translation for that) - I mean that the 12V line on Vin is different thna the 12V line on Rsns.

    In parallel, I'll aslo measure the current in the inductance in case it does saturate so that it rejects current in the MOS.

    Any other suggestions are welcome.

     

    best regards, Agnès

  • 0 in reply to user4216681
    TI__Expert 3495 points
    Agnes:
    Si7633 has high Qg. So the switch loss is high. FDD5614 has high RDSON130mOHM, the conduction loss is high. I suggest using Fairchild's NDB6030 which has Qg 26nC and RDSON 42mOHM.
    David Zhang
  • 0 in reply to David Zhang
    Prodigy 30 points

    Thanks David for your advices. We'll make the tests with the MOS you've suggested. I'll keep you posted.

    Best regards, Agnès

  • 0 in reply to user4216681
    Prodigy 40 points

    Hello Dear Agnès;

    Please can you share the result? What happened after you changed the mosfet?