Re: Variac arcing = kickback? (fwd)

---------- Forwarded message ----------
Date: Thu, 15 Jan 1998 14:15:32 -0700
From: Bill Lemieux <gomez-at-netherworld-dot-com>
To: Tesla List <tesla-at-pupman-dot-com>
Subject: Re: Variac arcing = kickback? (fwd)

Tesla List wrote:

> I decided to see what would happen if I added more transformer power to
> my project, and I got disturbing arcing in the variac.  The variac is an
> 18 general radio 120V variac, which was feeding the two 9kV 60mA
> transformers on my project without incident.  There is 8 feet of 14-3
> cable between the variac and the project, and the ground wire in this
> bundle is unused.

Why?  Surely you aren't leaving the cases of your NST's floating?

> When I added transformer number three and fired up the coil, I got power
> arcs between the variac's moving brush assembly and variac housing above
> it.  The arc is jumping about 3/16" so I'm pretty sure it's not just the
> normal 140V there!

First off, what sort of RF filtering and safety gaps do you have to
isolate your NST's from high voltage RF?

Second, is the bottom of your secondary _well_ grounded?  How well? Are
you sure it's a good ground?  Have you measured it?  Even a ground that
less than 1 ohm to some other ground may have significant impedance at

> What ever is sparking across this gap is followed by
> a power arc from the mains, which melted a bit from both parts of the
> variac.

I'll bet. Keep in mind that if HV can get back to your variac, it can
back to you.  Consider additional protection (small spark gaps, line
etc) between the control panel and the NST's.

> Somethings seems to be kicking back from the neons.  This only
> seems to happen when there is a breakout from the toroid.  I am naturally
> concerned for all appliances in the house, and I would like to correct
> this problem ASAP.

> What's going on here?

Sounds to me like you have a poor secondary ground.
You need to address:
1. secondary ground
2. NST isolation
3. protection between NST's and your control panel.
4. filtering between the control panel and the wall/power source.

I have a custom-built control panel that I use to run all of my
and "just for fun" coils.  It has a spark gap-HV interlock circuit, and
massive Filtron inductive/capacitive filters between the power input
and the rest of the control panel.

I don't usually use any filtering or protection between my HV
transformers and the panel, but then I've never observed the problems
you have.  I always make 
certain that my control panel is tied into a heavy (&short!) safety
that is _separate_ from the ground used for the secondary and HV parts.

I generally have either a ridiculously large air-core inductor or a very
powdered iron or similar toroid wound with HV wire as an RF blocker
the coil LC tank and the HV tranny.

As for dissipating power in iron-core filters in the HV side, only the
coils need to worry about that.  Most of us aren't running enough
current through the HV leads to the tank circuit to cause significant
heating of the core.

I measured the short circuit current from my power transformer with and
without a _solid_iron_ toroid (removed from an old 3amp variac), with
perhaps ten turns of HV wire on it, and observed less than a 5%
difference in current availability.  I'll sacrifice that current on the
altar of Transformer Preservation, thanks.  On the other hand, I've got
current to burn, since this unit will deliver15kV at 500mA before it
_starts_ to current limit.

For NST's on the other hand, where you don't have current to burn, you
will almost certainly have to use ferrite, not powdered iron or
laminated cores.

And on larger machines, where you're delivering more than a few hundred
mA of HV (Greg's machine must be delivering what, 9 _amps_ of HV?),
you'd saturate an iron core in a heartbeat.  But most people running
larger coils use harder-to-kill transformers like pole pigs or custom

Sheesh, I'd meant that to be a short little reply- I do go on sometimes,
don't I?


.                personal email: gomez-at-netherworld-dot-com                .
.          personal web page: http://www-dot-netherworld-dot-com/~gomez       .