Re: Dead MMC. was: AAAA! My coil...

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "Terry Fritz" <twftesla-at-qwest-dot-net>

Hi Malcolm,

At 08:38 AM 12/2/2002 +1300, you wrote:
>Fellow MMC coilers,
>                      Why push the ratings of the caps?

It's to save money.  If one goes from a 5 x 10 array to say a 7 x 14 array 
that cost goes up 96%.  "Money" seems to be a big deal to most coilers and 
MMCs are often the most expensive part of a coil for many people.  Of 
course, to each their own...

>I suggest it
>will prove over time to be a false economy. I would rather spend
>triple and build a cap to last. It really goes without saying that
>pushing the voltage rating is asking for trouble (if only a
>significantly altered capacitance over time before self-healing gives
>up) and pushing the current ratings is a disaster just waiting to
>happen. I haven't had to spend a cent on caps for over four years now.

Of course, people can build there caps anyway they want.  "I" like them 
cheap (inexpensive).  But if I blow a cap, I just fix it easily and go on 
(I never have blown a cap unless I was trying too).  If some one has a 
fancy show coil, perhaps they would want to overdesign the cap for higher 
reliability or life.  However, MMC failures from over voltage are very 
rare.  Construction and current are far more important.

If one wants to make a super reliable cap, I would say to multiply the 
number of rows and strings by 1.5 in Mark's chart and use two bleed 
resistors per cap.  But it is far more important to read the tips at:

http://hot-streamer-dot-com/TeslaCoils/MMCInfo/mmcinfo.htm

Including the one about keeping the bleed resistors about 1/4 inch away 
from the cap body.  If the resistors are lying on the caps, the voltage can 
arc from the resistors though the thin cover on the caps to the plates 
inside.  That seems to be by far the biggest cause of MMC problems.  If one 
does not use Geek Group caps, for goodness sake, be sure to use metal 
"foil" caps and not all "metalized" caps that can't take nearly as much 
current.  I have a bag of those GE42L series caps people have sent me where 
the end caps blow off when the tiny metal layers burn away from the ends.

BTW - I now consider G.E. 42L series as "bad" even though they do work for 
many people.  The metalized layers still burn out too easily to be a "good" 
MMC cap.  I changed the list at:

http://hot-streamer-dot-com/TeslaCoils/MMCInfo/good-bad.txt

Only time will truly tell if stressed MMC caps last a long time, but so far 
things look great.  I think just about all modern coils are either MMC, 
commercial pulse caps, or saltwater these days.  I note no one has made any 
rolled poly caps for about 1 1/2 years now.  It's sort of strange not 
having those "how thick of poly sheeting should I use" questions anymore :-))

BTW - Fully pulverized poly caps that have the layers all shot up from 
voltage breakdown only loose about 5% of their capacitance.  But normally, 
MMCs should never arc internally.  The self-healing feature is only a nice 
bonus of MMCs but normally is not needed at all.  We certainly never "try" 
to arc them internally.  The Geek Group's caps popping internally at 5700 
volts (2.44 joules!) with practically no resistance or inductance in the 
path is LOUD!!  I credit CD with finding a neat simple trick that allows 
the caps to dissipate that much energy without harm  ;-))

Cheers,

         Terry


>Malcolm