Re: NST and GFI ?
From: spam_proof-at-worldnet.att-dot-net <spam_proof-at-worldnet.att-dot-net>
To: Tesla List <tesla-at-pupman-dot-com>
Cc: James.P.Lux-at-jpl.nasa.gov <James.P.Lux-at-jpl.nasa.gov>
Date: Wednesday, February 16, 2000 4:47 PM
Subject: Re: NST and GFI ?
>On second thought, you're right. There shouldn't be any inherent
>current imbalance because the spark is an induced current. What goes
>in on the hot side of the primary should come out on the neutral side
>of the primary regardless of arcing (unless, of course, you have
>arcing to either of the hot or neutral lines). I've had problems running
>Tesla coils off of GFI outlets (they trip). I have a prototype GFI which
>I could measure a current imbalance that exceeded the trip threshold (as
>opposed to RF noise or something).
>One thing comes to mind... My transformer was center tapped (but I didn't
>ground the case--at least not deliberately). I was using a safety ground
>(third prong) as the ground for my secondary. Suppose for the moment that
>my transformer case wasn't as isolated as I thought. It is concievable
>that there could have been a current (e.g. from the ground to the
>transformer case) that was causing the imbalance I was reading and trip
>the GFCI. There wasn't anything obvious (such as an arc striking the
>transformer--internal arcing in a transformer is usually, but not always
>pretty obvious). I don't suppose any other ideas come to mind as to why I
>could measure a current imbalance.
But a secondary arc inside the transformer would still be a balanced load on
the primary side. However, what if there was a leakage path from the
primary winding to the case(ground). It could just be capacitance (which is
significant), and then, say, a pulsed RF signal (which is what the TC
radiates) couples better to one side of the primary than the other (if for
no other reason than the load impedance is different on the Neutral side).
Take a look at GIF image at
The path from the secondary, through C1 and the Line wire back through the
distribution transformer and then to ground is higher impedance than the
path from the secondary, through C2 and the neutral, through the
neutral/ground bond at the service entrance. So, you might get more RF
current flowing on the Neutral than the Line through the GFCI's current
sensor. The TC is pulsed at 120 Hz, so if the GFCI detector detects the RF
current imbalance, then rectifies and filters it (which is possible), it
might trip. This is all sort of sketchy, though...
Actually, the more likely leakage path is back through the NST from the HV
winding capacitance to the primary.