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

Re: It Just Stopped



Original poster: Mddeming-at-aol-dot-com 

Hi Sean,

Since I ran each side of each NST to 6 kV RMS, that's 12kV RMS across the 
total output. I'm guessing that they're probably OK to 15kV
  I checked the dc resistance of the cap and it is >30Mohm (meter limit). 
It is possible that I toasted it. I have another plastic-cased cap, sans 
label, which is physically much bigger.

Maxwell 37667 Cap: 2"x3"x5" 30 nF 35kV w/ 1/4-20 thd mounting holes
"Unlabeled Cap" 3"x5"x9" 43.75 nF ??kV w/ 3/8-16 thd mounting holes

Based on the stud size and the difference in volume, 135 cu.in vs 30 cu.in. 
for the Maxwell, my first guess is that it is a lot "beefier"  45% more 
capacitance but 4.5X the volume. I may try it before buying a bunch of MMC 
caps.After all, "A bird in the hand..(can be very messy)'
;-)))
Matt D.

In a message dated 8/2/04 10:39:32 PM Eastern Daylight Time, 
tesla-at-pupman-dot-com writes:
Original poster: "Sean Taylor" <sstaylor-at-uiuc.edu>

Hi Matt,

Assuming both the NSTs are fine past 6000 Volts (IE no internal arcing that
starts at around 6,500 V :-) ), then I would guess your cap failed to a
short.  Even though you're measuring something at low voltage, there may be
carbon tracking that is shorting the cap out.  LCR meters are typically
designed to measure capacitances with a series or parallel resistance so
it's possible a ~10 kOhm resistance is effectively shorting your cap
causing the meter to read approximately correctly, but the cap to
effectively short your transformers.  Check the DC resistance of your cap
(after shorting it to make sure there is no charge), and the meter should
go to showing and open circuit.  However, the 0.5% change you show in the
capacitance isn't very much and could easily be the temperature affecting
the meter and/or capacitor.

I think most people would say that using that cap with a 15 kV transformer
(and at that current) is pushing it a bit (if that's one of the 37667 caps
- 35 kV, 30 nF).  I would build an MMC for that set up with 2 strings of 10
caps each of the 0.15uF, 2 kV CDE 942C series caps.  Unfortunately, since
The Geek Group is out of them right now, you'll have to find another source
such as Richardson Electronics - http://www.rell-dot-com.  I haven't ordered
from them in quite a while, but when I did, it took about 2 months to get
the caps, and there was a minimum order of 27 or so caps.

Before you order new caps though, you may want to use a string of diodes
and charge your cap up to 20 kV or so and see if it holds a charge, or
requires excessive current in to maintain that charge.

Good luck figuring it out!

Sean Taylor