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

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

Original poster: "Barton B. Anderson by way of Terry Fritz <twftesla-at-qwest-dot-net>" <tesla123-at-pacbell-dot-net>

Hi Chris,
Terry just replied on MMC ratings, and it looks as though the geek ratings 
are up there, but apparently voltage hasn't been a problem. I know Terry 
"knows" his caps, so I would definitely trust his experience and results as 
well as the others who have run the geek group ratings without failures. 
Because I don't use the geek group mmc charts (easy enough to do the math 
myself), I was unaware that the geek ratings were low on the voltage end of 
the specs. Since that is the case, then we definitely would have seen more 
failures by now if it was a problem.

Funny, I think I was the first coiler to purchase MMC's from the Geek 
Group. I built my MMC's and derated them as I would with any other pulse 
cap in TC service. I don't think the MMC chart existed at that time.

So, for MMC's, Geek Group ratings have a low failure rate and should be 
fine. For any other pulse cap or untested MMC cap, I would derate the caps 
as mentioned previously (better safe then sorry). I now wonder if the cap 
that failed was just a lemon in a batch of apples. We may never know 
exactly what occurred.

Take care,
Bart

Tesla list wrote:

>Original poster: "Chris Roberts by way of Terry Fritz 
><twftesla-at-qwest-dot-net>" <quezacotl_14000000000000-at-yahoo-dot-com>
>
>
>So does that mean the geek group MMC ratings are off? If so, would 3 
>strings of 8 caps work? That approaches the original value and should have 
>a higher voltage rating. Sigh- thats quite a few more capacitors (plus the 
>dead one I have to replace). Almost makes me want to go back to bottle 
>caps (shudders at the thought).
>
>  Tesla list <tesla-at-pupman-dot-com> wrote:
>
>>Original poster: "Barton B. Anderson by way of Terry Fritz "
>>
>>Hi Chris,
>>
>> From your description, it sounds like the cap was arcing internally and
>>overvolting the resistors. Once the last resistor was removed, the cap
>>itself went into a runaway condition and failed from the same internal
>>arcing and all the fun dynamics that goes along with that. My guess is the
>>cap was originally damaged due to being overvolted by the transformer. The
>>NST at 9kV(rms) has a Vp of 12.7 kV. The cap bank was rated at 12kVp, so
>>you were definitely on the edge or slightly over.
>>
>>The rms current looks good and is the area where you will see reliability
>>ratings (it is also the typical area of concern), but the cap voltage must
>>also be considered. Maybe more emphasis on derating the cap should be
>>rolled into cap reliabil! ity. Often it's not and just assumed the user
>>should know this. Derating at 2.5 x Vp is a good rule of thumb.
>>
>>Take care,
>>Bart
>
>
>
>-Chris
>
>"Learn from the mistakes of others. You won't live long enough to make all 
>of them yourself."-unknown
>