They do self-heal...

To: tesla-at-pupman-dot-com
Subject: They do self-heal...
From: Terry Fritz <twftesla-at-uswest-dot-net>
Date: Tue, 06 Jul 1999 21:58:17 -0600
Approved: twftesla-at-uswest-dot-net
Delivered-To: fixup-tesla-at-pupman-dot-com-at-fixme

Hi All,

	Some may remember my EEMMC cap.  This was a super under rated cap I made
from 8 Panasonic caps.  It put out 2 foot arcs and I deliberately blew it
up a few weeks ago.  I set it aside but tonight I took it out and cut one
of the blown caps out of it to open up.  The blown cap had a broken epoxy
case indicating that it had arced internally.  This cap was failed in true
Tesla coil operation apparently from over voltage.

	The first surprise was when I cut into it with a razor knife.  The cap had
a 10 meg resistor across it for the last two weeks but the metal knife hit
the internal floating layer I think and ZAPPP!!  I got quite a jolt :-))
These darn things just never give up!

	No doubt about it.  There was a nice blown area from over voltage about
1/4 inch in diameter.  Looks like it was well contained an the cap healed
as it should have.  Deeper into the cap I found a second site that also
looked like it self healed.  There was a third place (and many others) were
it didn't blow but it looked like it was about to.  The poly was very
cloudy.  Still deeper in there was a very large blown area that blow the
layer over it but it still looked like it healed up.  So it looks like
metalized poly caps are easily able to self heal if they get hit in Tesla
coil service!! 

	The end metal and lead area looked fine.

	I then looked for the voiding that I saw in the cap that I removed from my
EMMC but I only found a few very slight areas.  However, some of the places
that looked like they were about to blow through, but hadn't yet, had the
same voided and cloudy look to an extreme degree.  Perhaps the damage I saw
in the EMMC cap is just the beginning of it going really bad :-(  This cap
(EEMMC) had only about a minute on it compared to the ~20 minutes on the
EMMC.  This indicates that the voiding problem may be caused by prolonged
over voltage rather than a sudden spike.  This would make sense...

On the bleeding edge of MMC technology...

	Terry