Re: OLTC update - primary IGBT loss

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "Paul Nicholson by way of Terry Fritz <twftesla-at-qwest-dot-net>" <paul-at-abelian.demon.co.uk>

Terry wrote:
> As it turns out, I apparently do not need more top capacitance but
> rather more inductance for higher Q and better streamer power
> transfer.

I think that such a conclusion may be premature, although not
necessarily wrong.  Perhaps you should hold off committing to a new
secondary until after a closer look at the breakout behaviour of the
present coil.  It may be that during closer investigation it will 
become clear that either you are running short of charge, or have
insufficient topvolts.  I think we understand the linear response of
your coil, but we've a lot more info to wring out of its interaction
with the load.

> Bandwidth is a problem.  I can go to
> 40MHz on the streamer current and 20MHz on I base.

That ought to do for a start. We'll notice if substantial components
lie above the probe cutoff because when we calculate streamer
dissipation from the measured current, it will fall short of the
amount required to reproduce the observed ringdown.  Hmm, that
statement assumes the streamer load is a significant part of the total
dissipation, which may not be the case (:

> It appears that the SonoTube will not be transmuted into gold
> bullion from this exercise ;-))

I don't think we've unequivocally established that sonotube is lossy,
although it certainly has given us cause for suspicion.  In this case
I think that proximity effect is the main culprit.

> So I ordered up some threaded brass balls from MSC.  1, 1+3/8, and
> 1+7/8 inch diameter. 

Those put us in the right ballpark for breakout at voltages in reach
of your Jennings probe.  Once you've settled on a terminal geometry,
let me know and I'll try to predict the firing voltage at which
breakout should start.
--
Paul Nicholson
--