PC board cap is a no-go

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

Original poster: "Sundog by way of Terry Fritz <twftesla-at-qwest-dot-net>" <sundog-at-timeship-dot-net>

Hi All,

   I was looking at etching the copper from PC boards and using them as 
elements of a flat stacked cap for the GU10A tube coil.  I *finally* got 
the chance today to set up a test.  12kv NST,  fullwave rectifier, and a 
variac on the NST input.

  It failed pretty miserably.  There was a lot of corona losses at the 
edges of the plates, even at ~11kvDC.  Puncture resistance was okay, but 
the sustained corona quickly weakened the dielectric enough for it to punch 
through after about 2 minutes of sitting at ~13kv DC (that'd be pretty hard 
on the GU10A).

   Yes, I could stack 2 plates together in series to halve the voltage 
across each one, but that would immediately quadruple the space needed, and 
also the cost.  Where before I'd need about 30 plates, I'd now need 
120.  At around $4 each, it's now real expen$ive.

  So I need an alternate plan here guys.  I'm up the creek without a 
suitable cap.  MMC's are out of the question.   Too much capacitance, not 
enough current handling ability.  transmitting caps are hard to find and 
expensive, and I don't know of any home-made cap that will take the raw 
current (much higher sustained current than a TC can deliver).   I get the 
feeling LDPE and aluminum flashing caps would get too hot and disentegrate 
internally, leading to a spectacular and probably fatal (to the tube) 
failure.   I'm open to suggestions.

Thanks all!

Shad Henderson
G3-1203