mosfets keep failing

Hi all,

simple half bridge setup, mosfet power rail running at 70VDC (200V max mosfet spec), loaded with approx 50ma (about 20amps mosfet max spec)... frequency of approx 100khz, running into a dummy load resistor (hence the

50ma draw).

Problem is it works fine for about 20 seconds then the lowside mosfet fails with S-D short. I have had the high side fail also, but seems to be the low side failing at the moment. Mosfet gate voltage 12V. All IC driven, highside with charge pump.

Everything works fine, then blows the fuse after 20 seconds. The current remains constant at 50ma, then for no reason at all it jumps to 3.5amps and blows the fuse. Mosfet always fails S-D short. The gate never seems to be effected.

I can't imagine any reason for this whatsoever. The only thing I can think of is maybe the gate has a massive overshoot of like 50V or something and is blowing the mosfet gate, though I would have thought the gate would short to Source or Drain at least, but it never does.

If the mosfets were overloading, due to deadtime problems or something, then they would get pretty hot and normally fail totally short, in extreme case explode. Though I am nowhere near the spec limits of the mosfet this time.

I "don't have the time" to take any scope readings before the mosfets fail either, I can only think that the gate has a massive overshoot and is killing the mosfet, though I do not understand why it would fail S-D short each time. Also can't see how it could spike that high from a direct drive chip on 12V. Even with some ringing or overshoot it would still have to get up to like 20V to blow the gate, ive never seen anything *that* bad before.

Can anyone suggest why my mosfets always fail S-D short after about 20 seconds ?

Cheers, Chris

Reply to
exxos
Loading thread data ...

Your explanation suggests that somewhere you have an inductive load and probably on top of the drain.

Not considering any wiring errors or wrong soldering or hairline shorts, could it be the output inductance of your power supply??

Just a wild guess..

Mike

exxos wrote:

Reply to
siliconmike

Any chance that you are either overvolting the S-D (inc body-diode), or the b-d is fwd-conducting and being snapped OFF very sharply?

I have always had a total distrust of the body-diodes in power mosfets.... using TVS devices to limit the voltage across them, and Schottky power diodes to stop them from forward-conducting.

--
Tony Williams.
Reply to
Tony Williams

Overvoltage caused by the mark/space ratio not being 1:1. Overcurrent due to not enough dead time on drive waveforms. Get yourself a storage scope and capure the waveforms, or power it up on a low duty cyle.

Reply to
cbarn24050

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.