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

Re: cap specs or non-specs

Tesla List wrote:
> 
> >From PikeGreen-at-multipro-dot-comSat Jun 22 22:41:46 1996
> Date: Sat, 22 Jun 1996 20:31:04 -0500
> From: Pike Green <PikeGreen-at-multipro-dot-com>
> To: "'tesla-at-pupman-dot-com'" <tesla-at-pupman-dot-com>
> Subject: RE: cap specs or non-specs
> 
> Rob, Richard H., and All,
> 
> Rob wrote:
> ----------
> I have worked as both a product engineer and a test engineer, so I am
> proposing that this group form its own specification for caps.  This
> can't be that difficult, and might extend the state-of-the-art.  Who
> knows, the next "black budget" gov. project might reference telsa
> group cap specs for some exotic cap application.
> ----------
> 
> Richard Hull wrote:
> ----------
> I always spell out every spec to my cap maker and demand impulse
> tests prior to shipment.
> ----------
> 
> Perhaps Richard could also post his specs.
> 
> Regards,
> Pike

Pike,

My specs vary based on what I plan to do with the capacitor and therefore 
I can't publish specific specs.  I will note that the tighter and more 
severe the specs,(the more you constrain the builder), the more you will 
have to pay.

My object is to the let the cap maker go ape!!  I typically say up 
front..."I dont care about the case type, its size, its composition or 
the output leads as long as it is the bottom of their line and cheapest 
form."  I tell them to write this spec down well as it is the prime 
spec!!

 I now proceed to outline the key points about how I'm gonna try and 
destroy whatever they send me!!

The #1 spec is break rate or pulses per second.  Next, the fact that a 
60hz AC sine wave will be the source of power applied and that no DC will 
ever be applied! Also I tell them that major reversals of current can 
occur at any time.  Furthermore that the current peaks might exceed 
20,000 amps, but usually will be held below 2,000 amps (This usually 
sends a few shivers up their spines)  I next give the peak voltage 
expected in the sine.  I then state that whatever they figure for the 
voltage standoff in this proposed killer service, I want that safe and 
optimum design voltage doubled in the cap that they send to me or don't 
bother sending anything.

I tell them to assume a total average .00001% duty cycle, in service. 
(They usually freak out here)  This is like saying I want a 3,000 hp 
Allison engine , but I'll only run it 3 seconds a month.  This forces 
them to overdesign the thing and puts it into perspective that you are 
basically impulse testing there cap. (trying you best to blow it up)

The killer specs force overdesign (expensive, normally), but the latitude 
in case design and unlimited size allows them to design a large safety 
margin into the cap which might not otherwise be possible in a cramped 
case made of metal. (vastly cheapens the design)

I usually wind up with a large solid monolithic ugly gray plastic block 
with female 1/2", threaded, recessed terminals on it.  T'ain't purdy but 
lasts a long time in Tesla service.  I haven't killed a custom 
manufactured unit yet!

Richard Hull, TCBOR