[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: 11 foot arc right off the bat

Original poster: Bert Hickman <bert.hickman@xxxxxxxxxx>

Hmm... probably neither. I suspect the cap was either a misapplied low duty cycle cap or that it was overvolted.

Maxwell (now General Atomics) makes 1 PPS (Mylar) and high repetition rate 500-1000 PPS (Polypropylene) single-ended plastic cased capacitors. Most of Maxwell's single-ended style 31XXX series capacitors are low duty cycle Mylar caps, while their series 37XXX caps are polypropylene. Some known exceptions include models 31393, 31981, and 31583 which are all polypropylene. If there's any question, a quick call or note to the General Atomics folks will resolve it.

When low energy density pulse caps explode in Tesla Coil applications, it's very often because a low duty cycle pulse capacitor was inadvertently misapplied in a high energy RF application. Because Mylar is so lossy under HV RF, the inner portions of the capacitor rolls rapidly overheat, outgas, and electrically fail (not always in that order). Increased internal pressure causes the welded polypropylene case to bulge (and eventually) rupture.

However, it's also possible to destroy polypropylene pulse caps through overvolting. Unfortunately, many amateur coilers have destroyed otherwise excellent polypropylene pulse caps by unknowingly overvolting them. If your cap's faceplate DC voltage rating is less than 3X the RMS voltage rating of your HV source, it WILL be overvolted by the oscillatory voltages seen in TC use. Since individual internal capacitor rolls are connected in series (or series/parallel groups), failure of one section then leads to cascading failures in the remaining sections. Internal arcing damage then causes outgassing sufficient to bulge or rupture the case. These failures can be quite exciting(!) in higher charging current/pig-driven systems, and are particularly spectacular when combined with less "forgiving" cylindrical cases (such as Condenser Products older TC caps). These cases simply explode (rather violently) with no advance warning.

I must disagree with Terry on current killing these caps - unless you were running a very low inductance primary at high break rates, it's quite unlikely that you destroyed these caps via excessive current (at least within a disruptive system). Mawell's single-ended caps are rated at 25 amps RMS (at 20% voltage reversal) for most 31XXX series caps, and 25 - 50 amps RMS for the 37XXX series caps. That's all day long folks, and shorter runs can take advantage of their significant thermal mass. They are also rated at 25 - 50 kA (non-repetitive) peak current - they ARE, after all, low inductance pulse caps.

The bottom line:
I suspect the caps either had the wrong dielectric system for TC application or that accidental overvolting killed them. We would need to know more about your HV supply's peak voltage, capacitor model, tank capacitance (and configuration, if made up of multiple caps), primary inductance, and estimated break rate in order properly assess the problem.

When properly applied, Maxwell caps are virtually bullet-proof. When misapplied, YMMV...


Tesla list wrote:
Original poster: "C. Sibley" <a37chevy@xxxxxxxxx>
That Maxwell looks just like mine, it blew the same way last week! What a mess! Anyone have any thoughts if this failure mode is due to over-voltage or over-current?

----- Original Message ----
From: Tesla list <tesla@xxxxxxxxxx>
To: tesla@xxxxxxxxxx
Sent: Sunday, May 14, 2006 12:53:22 PM
Subject: Re: 11 foot arc right off the bat

Original poster: "David Rieben" <drieben@xxxxxxxxxxxxxxx>
Was the crack of the case of the 50 KV Maxwell
caused by internal failure or a drop on the concrete?
David Rieben
----- Original Message ----- From: "Tesla list" <tesla@xxxxxxxxxx>
To: <tesla@xxxxxxxxxx>
Sent: Sunday, May 14, 2006 1:43 PM
Subject: 11 foot arc right off the bat

 >Original poster: "miles waldron" <mileswaldron@xxxxxxxxxxx>
 >I have become Ken's publicist. Check out the White Knight, kenscoil.com, a
 >work in progress. We are trying to come up with a practical, nice way, to
>document our work and reasoning. Any suggestions are appreciated, but please
 >be polite. We are serious coilers.