OL-DRSSTC - 12

["Tesla list" <tesla@xxxxxxxxxx>]

Original poster: Terry Fritz <vardin@xxxxxxxxxxxxxxxxxxxxxxx>

Hi All,

Today I tracked down the voltage droop problem.

I looked much into buss resonances and all that and there are some 
issues there.  You want to keep the leads and all real short and all that.

But the main issue turned out to be the high internal resistance of 
the electrolytic cap array.  I disconnected the poly buffer caps so 
the thing was running off the electrolytics only:

http://hot-streamer.com/temp/OL-DRSSTC-2005-10-14-002.gif

When the IGBTs turn on, the voltage across the caps drops to the 10 
volt IGBT rail.  The electrolytics contribute NO real current to the 
oscillation!!!  They are only good for charging the buffer caps.  The 
two 2.5uF buffers are supporting the entire primary loop current!

Here is the rail with the original 2.5uF 942 buffers:

http://hot-streamer.com/temp/OL-DRSSTC-2005-10-14-001.gif

I then replaced the caps with two 5 uF 940 series caps:

http://hot-streamer.com/temp/OL-DRSSTC-2005-10-14-003.gif

The old caps are shadowed in white and the blue area is the droop 
with the caps of twice the value.  The droop is reduced 
dramatically!!  Probably a little more than expected sue to some 
resonant effects.

The drive signal now looks very good so the new caps were just enough 
to do the trick:

http://hot-streamer.com/temp/OL-DRSSTC-2005-10-14-004.gif

The voltage right across the electrolytics is also fairly smooth now:

http://hot-streamer.com/temp/OL-DRSSTC-2005-10-14-005.gif

So it looks like everything is all worked out now :-))

Cheers,

	Terry