"raypsi" wrote in message
...
I built my own 7.5VACT 21amp filament xformer. I needed to supply
-500VDC screen voltage to the center tap.So a local ham tells me the
insulation better be able to withstand at least 500VDC.

I built my own hipot tester from a voltage doubler and a bank of eight
400WVDC caps. Each cap had a 330 volt charge so like 2640VDC. I
figured no problem. I looky up the specs and see if this thingy
passes.

I googled and can find no specs for hipot testing. I came across in a
usernet group: 2500VAC at 1ma for 1 minute. That didn't really tell me
much.

My results are 0,2 micro amps thru a 1 meg series resistor at 1980VDC
source voltage. This from either the core or the primary to secondary.
The current didn't change even when I went to a 10meg series resistor.

Is there nowhere that you can get specs without paying big bucks for a
UL or IEEE book on the topic?

I'm surprised that you couldn't find UL, IEEE or MIL-SPEC (or MIL-STD)
sources on the net through Google or any other search engine.
Just taking the snippet you did come across, I believe that it means that
you must maintain a leakage current less than 1 ma. for 1 minute with an
applied voltage of 2500VAC (about 3500V peak). This is so far above your
typical operating conditions, I'm tempted to say you passed. You do mention
that your results were from the primary OR core to the secondary instead of
primary AND core to the secondary, which is typical for this test, but I'm
not sure it matters for a home-built. I used to have to do a test like this
using 1500VDC with an automatic tester. If there was a problem with
insulation, it showed up quickly. If the room lights were dimmed, you could
see the place where it failed by the arc to ground.

-NM