Urgent help on IGBT

Currently I am working on a welding system and I need to use IGBT to repeatedly cut off the welding current. During the trials, I meet some
big problems and kept burning IGBT's. The IGBT I am using is CM300HA-12H, the gate driver is M57962L, and a DC-to-DC converter M57145L-01, all manufactured by Powerex. I soldered the circuit exactly as what the data sheet tells me, i.e., the gate drive board BG2B. Though this is for dual mode IGBT's, and what I am using is a single mode IGBT, I used only half of the schematic, i.e., the part for one IGBT. The electronical components are chosen exactly as those specified in CM300HA-12H datasheet. My purpose is to chop the welding current, 80A and above, every 80ms, with 1ms "on" time and 79ms "off" time. After implemented for about 10 second, the IGBT failed, as well as the M57145L. I used a cooling fan and heat sink. So the thermal runout should not be a concern. From the welding machine, the average voltage was 70V, though I had no means to test the transient voltage. Since this is the third IGBT I have burned, I really dare not use it before I can find out the problem maker. Your help will be highly apprecated from the bottom of my heart.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Some of the electronics newsgroups might know more about this kind of thing than we do.
However, I do know: A welding arc is the ultimate broadband noise source. It is surprising how a solid state device capable of handling a hundred amps can be destroyed by microamps of energy above its rated voltage. When all else fails, try reverse engineering someone else's design. You can find welding machine schematics on the web in electronic owner's manuals from Miller, Hobart, Lincoln, etc., although they usually only go down to the board level. Note the filtering components in the output circuit in this design, see page 29 of the manual at:
http://www.millerwelds.com/om/o360d_mil.pdf
snipped-for-privacy@engr.uky.edu (Wei Lu) wrote in message

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
A HEXFET drive circuit maybe better?

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Thanks. Though the welding arc is one of the noise sources. I think the inductor inside the welding machine may be a bigger problem maker, especially when turning on and off the welding current. Any idea about eliminate it in case no modification is made inside the welding machine?
snipped-for-privacy@control-tech.com (Charles Erskine) wrote in message

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Since the driver circuit blew, check for a damaged gate protection circuit being failed, like a transzorb or snubber or mov.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.