strange effect in mosfet (h-bridge)

[alfsch]

i built a simple half bridge with 2x irf730, v+ 60 volt.
for testing, no load, switching at 80khz (with about 1us deadtime).
so mosfets only switching...but they heat up! (just a little, about 10° or so).
(no rf seen on scope)
when coil (load) connected, mosfet stay cold !!

can anybody explain, why "free" switching mosfet, with lots of deadtime, can have power losses ??

 

[Pafi]

2 reasons:

- There is output capacitance.

- High dv/dt of hard switching can induce turn on impulse on opposit device.

"hard switching": high Vds when MOSFET starts to turn on.

"Soft switching", or "resonant switching": current of inductance makes voltage of drain-source alternate before MOSFET turns on.

 

[alfsch]

a: outp. cap. is also here with load, so this should make no heat up

b: High dv/dt of hard switching can induce turn on impulse , great idea, indeed there is a 5v spike at the "wrong" gate

thanks to your help, i can sleep good now 🙂

 

[IVX]

alfsch, some pics:

 

[Pafi]

outp. cap. is also here with load, so this should make no heat up

Cap is there, but with load it is charged and discharged by current of choke!

You're welcome!

 

[alfsch]

ok, the caps make a (small) part of additional loss, your right.

#ivx: the spike is in the gate signal , similar to your pics :-(

well, ill try at 300v with coil ... seems to work, without heating up.

switching-spike at gate is still there, but only 2v peak, so no unintended turn-on should happen