TSSP: List Archives

From: Paul
Date: Thu, 09 May 2002 22:22:40 +0100
Subject: Re: [TSSP] Secondary voltage stress factor

Boris wrote:

> Once,I had been ready to think just like you but
> changed my mind after seeing real scope
> discharges of several units.

Well you have the advantage there because I've never 
had a close look at the discharge currents.  

Perhaps I'm thinking too much like the characteristics
of a primary gap arc.  Maybe the toroid-ground discharge
arcs are more fragile, in that the current path doesn't
survive the initial zero crossing, except for the HF
currents which you describe, whose ringdown will be short
compared to an Fres 1/4 cycle.

> problem being here that these violent
> transients occure at 5 ,10 ,20 Mhz or more and EM
> theory not yet imbeded in a software starts to play
> important role.

Yes, if we say that our upper frequency limit for the
quasi-static approximation is when the coil is 5% of
a free space wavelength, we are limited to considering
modes up to around 30Mhz.  The highest I've actually
used is 24 Mhz.  Therefore we may expect to come up
against features that we cannot reproduce with the model.

The arc discharge itself I'd guess would be dominated
by a ringdown at the resonance of the arc's self-inductance
and the toroid capacitance. For a 1 metre arc and a Ctop
of 50pF this would be around 20 Mhz, so this may also be
a factor.

I think we should just apply some simple discharge
models to begin with, accepting the frequency 
restrictions, ignoring the reactance of the arc discharge
loop, and just see how well these compare with observations.

When we see from the scope the character of the discharge,
we can put this into the model to generate a base current
prediction.  Then we can compare the base current to test
the model.  Then the model can be used to look for 
transients big enough to contribute to secondary voltage
stresses.  I guess that counts as a strategy.

Cheers All,
--
Paul Nicholson,
--


Maintainer Paul Nicholson, paul@abelian.demon.co.uk.