Re: MMC cap bank

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

Original poster: "Finn Hammer by way of Terry Fritz <teslalist-at-qwest-dot-net>" <f-h-at-c.dk>

Gang:

MKP383 types will not work for long, no matter which rating you adhere
to.
They are of the metalized plates type, which means they lack the massive
contact surface area to the outside, supplied only by metal foil types.
(this being the cross section of the whole length of foil, as opposed to
metallized dielectric, fused to the sprayed metal endcaps)
It is a strange thing that they are rated as pulse caps at all.
http://www.bccomponents-dot-com/Uploads/Datasheets/mmkp383.pdf
These caps will blow the endcaps off in a matter of minutes, no doubt.
A better alternative from BC-Components, now that they have discontinuet
the fabulous KP/MMKP 376 types, would be the KP/MKP 375 which are
similar to the venerable panasonic caps which started the whole MMC
thing.

Cheers, and sorry!, Finn Hammer



Tesla list skriver:
 >
 > Original poster: "Lau, Gary by way of Terry Fritz <teslalist-at-qwest-dot-net>" 
<gary.lau-at-hp-dot-com>
 >
 > I'm not sure I agree with this dismal assessment.  MMC caps are typically
 > designed with voltage ratings governed by the cap's DC rating, not the AC
 > rating.  The proposed cap's string rating was 25kV, a reasonable margin
 > above the peak voltage from a 14.4kV power supply.  The difference between
 > a pig and an NST application is the RMS current.  An MMC designed for a pig
 > must handle higher current, so more parallel strings must be
 > used.  Designing longer strings won't improve the ability to handle higher
 > currents.  Longer strings of the same cap size will result in lower
 > capacitance per string, so more strings will be required to achieve the
 > desired final capacitance, but it would be more economical to just use
 > shorter strings of smaller caps.
 >
 > That said, I don't know if the proposed BC caps have a proven suitability
 > for MMC's.  The price does seem attractive though!  I think Terry commented
 > that they appear somewhat small compared to other similarly rated units?
 >
 > Gary Lau
 > MA, USA
 >
 >  >Original poster: "Dr. Resonance by way of Terry Fritz
 > <teslalist-at-qwest-dot-net>" <resonance-at-jvlnet-dot-com>
 >  >
 >  >The rating you specified alone should tell you it won't work for a long
 >  >time --- perhaps short time until you start blowing them.
 >  >
 >  >The pig is 14.4 kV and you intend to run the cap string at 900V x 10 which
 >  >only equals 9 kV.  Failure is certain --- the only question is when.
 >  >
 >  >With NST's we use 13-14 caps per series.  With pole xmfrs I would 
recommend
 >  >around 16 caps per series as an absolute min.
 >  >
 >  >Less than this will eventually begin to failure.
 >
 >  >> Hello all,  I'm constructing my first mmc bank. Cap's are .056 uf. each,
 >  >> 2500 vdc. 900 vac. each. 10 per string .0056 total per string, a 
total of
 >  >> 10 strings max. .056 possible, probably tap out a smaller value at first
 >  >> although. Cap's are digi-key # BC2048-ND, AC and pulse double metallized
 >  >> polypropylene film. Bleeders are 10 megohm 1/2 watt. Using 14.4 kv. 
10 kva.
 >  >> pig, MOT. ballast on primary, asynchronous gap, speed is ajustable so it
 >  >> may be in synch. at some point. Will it last?  Comments? 
Thanks    Arcing
 >  >> in Wisconsin.......