Re: first run of an MMC
Subject: Re: first run of an MMC
From: Terry Fritz <twftesla-at-uswest-dot-net>
Date: Sun, 24 Oct 1999 10:18:07 -0600
I think the polyester gotcha! Three strings is pretty risky in it's self,
but the dissipation of polyester rises exponentially around 100 kHz and
above. The Wima chart I have puts it at about 3% at 100kHz and probably 5
to 10% at 200kHz. Poly on the other hand is 0.12 and 0.2 at 100 and
200kHz. If you had poly caps, they would probably still be cold even with
three strings. If you get to about 10C rise in temperature, you are about
to blow them. Even though the others did not seem hot to the touch,
internally they may have been melting since the dielectric is such a good
thermal insulator too. The caps are not good at dissipating much internal
heat at all. You may find another bad one when you test them and may want
to cut it open. I bet it will show signs of melting the dielectric.
At least you gave the poly caps a good try. Now we know what they can't
do for sure.
At 11:44 PM 10/23/1999 -0400, you wrote:
>Last night I tried out my first attempt at an MMC. I've avoided
>describing it before, since I am aware that it contains two big no-no's;
>the dialectric is polyester and the individual cap value is .22µF. And I
>only had 100, which meant I could only use three strings...
>But the caps only cost me $0.20 a piece, so I thought if they failed, so
>Going by the cap rating, the whole thing should end up being about .027µF
>with 25 caps in each string.
>I also was trying out a new bifiliar-wound secondary, trying to make my
>system use less of the large 17 5/6 turn primary.
>Now, the fun begins..
>As the coil had been sitting around disassembled for while, I reset the
>saftey gap, hooked in the rotary, set the MMC on top of the SRSG's case
>and connected it, tapped into the primary at about six turns, selected
>two toroids,... ...switched on the SRSG, and ran up the variac. The gap
>was firing... hurah! The cap was working.
>I then decided to check the gap phasing, rather than fool around with
>tapping yet, since the saftey gap was firing frequently. In the mercury
>vapor light I was using to sync it, it was difficult to see, and I
>decided to change it while the motor was running... ...big mistake.
>The unpleasant sound of metal scraping metal at high speed and the top
>chunk of one of the fixed electrode supports went flying out of the open
>No one was hurt, and, only semi-daunted, I set the broken support with
>lexan glue and wired in my old RQ gap to replace the out of comission
>SRSG. I then tried to tune the coil, and ended up only being able to
>achieve smooth streamers out at only 3 1/2 turns, and they were really
>short till I started stacking on the terminal capacitance. That MMC
>certainly has a whole lot more capacitance than my SW cap bank ever did!
>I ended up with a enough Al ducting toroids stacked on the coil that the
>terminal was bigger in all demensions that the length ofthe secondary!
>And even at that, the streamers were smaller than what I'd achieved using
>the SW caps, and I was getting the best streamers at something like 2 or
>.027µF is about double resonant for a 12/60 NST like I was using, and I
>didn't expect the RQ gap to be able to really perform in such a system,
>and was looking forward to using the SRSG in the morning to see just how
>well the new system might work. But, saddly, at one point flames burst
>out of one of the caps in the MMC, and coiling had to be suspended.
>The destructed cap had one entire foil connection blown off, neatly
>severing the end from the rest of the cap. The entire bank was warm, but
>not worryingly much.
>Is this bank going to last me any time at all? That first cap blew...
>...Oh, that's right, it didn't blow untill after I tried ganging the
>12/60 in parallel with a 15/30 to get longer streamers... I'd actually
>forgotten that! But, the caps still blew after only about 10 minutes
>total run time. I have spares... ...just so long as this thing lasts
>long enough to perform at the MidOhio TeslaFest.
>So, this morning, rather than play with the big system, I threw together
>a bipolar coil from old components in ~5 minutes... ...6" streamers
>from a 9/30 NST. Sigh... ...I can still smell the ozone in my clothes.
>Get the Internet just the way you want it.
>Free software, free e-mail, and free Internet access for a month!
>Try Juno Web: http://dl.www.juno-dot-com/dynoget/tagj.