At 02.51 2011.12.28, you wrote: >On Wed, 28 Dec 2011 01:15:40 +0100 >Electron wrote: > >> But I still can't get the damn thing to work well, in practice I get=20 >MHz (or GHz??) full output swing awful noise at the gate. I can=20 >provide screenshots if it may be useful. > >I've seen ferrite pearls on the gate lead of the FETs to avoid >oscillations (as near to the device as possible). This lowers the gain >at high frequencies. Thank You very much, putting a SMD ferrite bead inbetween definitely fixed it. :-) But the datasheet is the cu= lprit here: it says to put the driver output pin as close as possible to the MOSFET/IGBT gate input pin, and in case the distance is more than 2 inches, to put a 10 ohm resistor inbetween. For this prototype, I had soldered the driver output pin directly on the gate of the IGBT, and *this* caused the oscillations!! If I had put some wire inbetween, thus some inductance, maybe it would have not self-oscillated (a close look with the oscilloscope revealed that what appeared as noise was the driver turning ON/OFF continuosly with respectively 100ns and 50ns times). The SMD ferrite bead that I've now put inbetween, as per Your suggestion, fixed it and the turn ON and OFF of the transistor still looks very fast, which is suprising for me, as at those frequences the ferrite bead must have at least 100 ohms of impedance. There must be some magic in it (magic is whatever I don't current understan= d). Of course the ferrite bead is still preferable to e.g. a 10 ohm resistor, right? Thanks a lot again. Cheers, MarI/O > >John >--=20 >http://www.piclist.com PIC/SX FAQ & list archive >View/change your membership options at >http://mailman.mit.edu/mailman/listinfo/piclist --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .