[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: IGBT's
- To: tesla@xxxxxxxxxx
- Subject: Re: IGBT's
- From: "Tesla list" <tesla@xxxxxxxxxx>
- Date: Thu, 01 Sep 2005 12:09:42 -0600
- Delivered-to: testla@pupman.com
- Delivered-to: tesla@pupman.com
- Old-return-path: <vardin@twfpowerelectronics.com>
- Resent-date: Thu, 1 Sep 2005 12:10:50 -0600 (MDT)
- Resent-from: tesla@xxxxxxxxxx
- Resent-message-id: <0VdTj.A.F7G.qQ0FDB@poodle>
- Resent-sender: tesla-request@xxxxxxxxxx
Original poster: Jimmy Hynes <jphynes@xxxxxxxxx>
The main
source of overvoltage is from hard-switching some current causing
voltage transients across the IGBTs. The best way to cure this is to
1) install TVS (transient voltage supressors) rated for say 880V
across the C-E of each IGBT, and 2) install low inductance capacitors
across the power rails of the bridge. I cannot stress how crucial a
low inductance setup is (the voltage spikes will be proportional to
the inductance... V=-L*di/dt).
While those are important steps, I think the most important thing is
to keep it from hard switching in the first place. Hard turnoffs are
really bad, and can be avoided by using primary feedback, and some
way of assuring a soft turn off at the end of the burst.
You still need to do what Steve was talking about to deal with the
diodes reverse recovery spike due to hard turn ons though