Newsgroups: comp.robotics
Path: brunix!uunet!europa.eng.gtefsd.com!darwin.sura.net!math.ohio-state.edu!magnus.acs.ohio-state.edu!usenet.ins.cwru.edu!agate!headwall.Stanford.EDU!Csli!pkahn
From: pkahn@Csli.Stanford.EDU (Philip Kahn)
Subject: dealing with high voltage spikes upon cable disconnect
Message-ID: <1993Jun12.034757.10470@Csli.Stanford.EDU>
Keywords: inductive spike, voltage clamping, snubber
Reply-To: pkahn@teleos.com
Organization: Teleos Research, 576 Middlefield Road, Palo Alto, CA 94301
Date: Sat, 12 Jun 1993 03:47:57 GMT
Lines: 24

I am driving a stepper using a constant current PWM bipolar driver (a
dual full H-bridge). Motor current is 1.2A, 3.3Ohm resistence. Driver
chip is rated to 1.75A continuous (though as usual, the device is
primarily heat dissipation limited, not power carrying limited).

When the cable is disconnected, never a great idea for a high current
inductive load, the driver is overloaded (i.e., vaporized). I want to
avoid this. I have several candidate ways to fix this, but I am still
looking for a book or source that discusses load disconnect in
specific. All power books I looked at talk about transients on
switching, not removal of inductive load and techniques for
bomb-proofing. Only foolproof technique I have so far involves 5
diodes (i.e., bypassing each bridge transistor). Not good. Transorbs
at 1600W peak ain't cheap. A bidirectional xener with a series
resistor across the load seems like the most useful approach to me,
but I'd like to see a discussion somewhere in a book to see guidelines
for part rating and if there are any hidden surprises. Favorite
techniques are also invited. I'll post the answer, since we all
usually have to deal with cable disconnects on powered inducive
devices. 

thanks,
phil...

