On Sat, 26 Jul 2008 14:12:40 -0700, "Chris F."
put finger to keyboard and composed:

I forgot to mention that the customer had brought in two of these units,
one being a bit newer than the other. The other unit happened to be in
working order, so I opened it up and found a similar design with another
strangely-marked cap. But since it was working, all I had to do was measure
the operating voltage and capacitance. The voltage across it was about
320VDC, and the capacitance was just 1.18uF. This cap, unlike the one in the
other unit, appeared to be a non-polarized metal-film type. So I installed a
2.2uF, 400V metal-film cap (the closest thing I could find) in the other
unit and it appears to work. Both units create a good thick spark of about
1/8". So I guess the problem is solved. BTW these units are not terribly
old, both have mfr'd dates in the early 90's.

If the capacitor forms part of a tuned circuit, then doubling its
value may change the resonant frequency and peak voltage in this
circuit. Unless you've already done so, I would confirm that the
voltage across the replacement cap does not exceed its rating.

However, if the design of your electric fence controller is anything
like my Silicon Chip kit, then it works by charging a dump capacitor
and then discharging its stored energy into the fence. The April 1999
kit uses a 7uF 250VAC capacitor which develops a DC voltage of 340V
across it. This amounts to a stored energy of ...

E = 0.5 x C x V^2 = 0.40 joules

If yours is a dump cap, then doubling its capacitance would double the
controller's energy output. I'd make sure you are not contravening any
local safety regulations by doing so. Having said that, it would still
appear that your controller puts out less than one third of the energy
that my kit does. shrug

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.