RE: Failure Mode...Weird!

["John H. Couture" <couturejh-at-xxxxxxxxxxxxxxxx> (by way of Terry Fritz <twftesla-at-xxxxxxxxxx>)]

Sundog -

The main problem in selecting a volts/turn for the secondary wire insulation
is that the safety factor is difficult to determine. The insulation
breakdown voltages are listed in many tables and can be easily checked by
tests. However, a factor of safety must be applied to these breakdown values
and that is where the problem lies.

I have tested wire with enamel insulation and found the breakdown voltages
to vary from about 4000 (magnet wire) to 8000 (formvar) volts. For design I
use 500 volts per mil of enamel. This is a safety factor of over 10 and
works for small coils. With 1000 turns at 500 volts/turn the total output
volts are 500 KV. This will work with larger coils but higher voltage will
stress the insulation and reduce the operating life.

To test the insulation I used a 7500 volt NST with two 10 KV diodes from the
junk box. With a variac this gave me a DC test voltage up to 10 KV. Cut off
about 12 inches of wire and form it into a U shape. Scrape the insulation
off one end and insert the U about half way into salt water in a jar .
Connect one hot lead to the test wire and the other lead to a bare wire
inserted into the salt water. Both leads are hot. Connect a high voltage
voltmeter across the two leads.

Crank up the variac until the NST gives a grunt and the voltmeter drops to
zero. You may have to use a magnifying glass to see the tiny pinhole in the
insulation. This works with thin enamel insulations but thicker insulations
may not break down below 10 KV. I found some vinyl insulations did not break
down below the max 10 KV but lasted only a short time in TC operation
because of the high RF voltage stresses.

John Couture

------------------------------





-----Original Message-----
From: Tesla list [mailto:tesla-at-pupman-dot-com]
Sent: Tuesday, October 31, 2000 6:35 AM
To: tesla-at-pupman-dot-com
Subject: RE: Failure Mode...Weird!


Original poster: "sundog" <sundog-at-timeship-dot-net>

Hi all, John,

I was using formvar wire on it, 26ga.  I didn't give much thought to the
volts/turn of the coil.  Physically it was wound 22.2" (counted turns while
winding & measured), for 1282 turns.  The form is green PVC, with 2 baffles
in it.  At it's best I was getting ~4' streamers, and 4.5' strikes to
grounded objects (garage door, end of the lathe, shelves, halloween dummy
(promptly set it on fire!).  I was hoping for better than that, and think
with a larger topload I can hit 5'+.  As it is, I'm getting 2 to 3
streamers.  I'm also running an 80nf tank cap, which seems to help a lot.
(compared to 35nf)  Now, I got all the wire off of it, chocked it up in the
lathe, and did the "on-off switch dance", for lack of a variac :)  I got it
all smoothed out and 99.999% of the carbon off.  Then I used my 2.4hp DC
motor (run on 16v5A instead of 130v21A) to turn the form nice & medium
speed, while I wound more 26 ga on it.  Threw on a couple of coats of clear,
and I'll test it this afternoon (halloween) to be run tonight.  The "ole'
reliable" 4" primary and sec. coil is going also.  But i have to pull the
second half of the cap bank to run it, or else I tune in at ~1 1/3 turns (no
joke!), and the primary, gap, and tap get blistering hot in a second or two.
haven't worked out what the peak amps are, but it's something probably very
unhealthy for the caps.  And the wiring, for that matter.

 What's a safe figure to aim for for volts/turn?  I don't believe this was
an overcoupling issue, nor an out of tune.  Just a weird breakdown that I'd
rather not see again.  Pics of the tracks on the inside of the wire to come
after halloween when I get it all developed.
											Keep it safe,
 						and Happy Halloween!!!  Muhuhahahahahhaha!!!!!
												Sundog