mosfet self turn on

[mag]

Hi all,
in a two switch flyback power supply i have noticed something very strange.
It is a quasi resonant flyback with 24v / 40a using two mosfets stw18nm80 as primary switches
and a synchronous rectifier on the secondary side. The two mosfets are driven with a pulse transformer.
When i look at the waveforms vds and vgs when the mosfets are cold everything is normal.
If i let it warm up at full load even for less than 1 minute I see something very strange.
Removing the load I see the mosfets vds going and staying at 0v even when the gate voltage is
0V. The same waveforms are visible on the low side and on the high side.
The most strange thing is that it seems that the mosfet are really driven but the vgs is zero...
This increase the standby power a lot (>25W) and creates for sure other problems.

How is it possible that a mosfet decides to switch on by itself? If I replace the mosfets the problem disappears.
Is it possible that my stw18nm80 are fake or damaged in some way?

Tomorrow I will post some scope plots to better explain the situation.

Thank you very much

ciao

-marco

[nigelwright7557]

Are you exceeding the Vds maximum voltage ?
This would cause the mosfets to breakdown.

[mag]

No, my vds peak at 570V and the mosfet is rated 800V. i was suspecting the same thing because probably is a fake mosfet which vds is not 800V...

What happens if it beaks down? Its vds will be clamped at the breakdown voltage or will go to around zero due to parasitic bjt turn on?

[Andrew Eckhardt]

What do you get for ohms from gate to source on the bad parts?

[mag]

I don't know, but for sure they are >1k when measured them mounted in circuit there is a 1k resistor between gate and source and the ohmmeter was indicating something around 1k.
What i have normally seen is that sometimes when a mosfet fails the gate get shorted with the source with few ohms and that is not the case now.