From: boris petkovic
Date: Wed, 30 Oct 2002 11:35:25 -0800 (PST)
Subject: Re: [TSSP] Top Voltage
Hi Bert,All,
Bert Hickman wrote:
> The initial formation of corona at the surface of an
> electrode in a
> highly nonuniform field occurs at a surface E-field
> of ~25-30 kv/cm.
> This is a very low current discharge (microamperes).
---
Yeah,but here is the catch in case of Terry's 38 kHz
system!I was playing with characteristic values needed
to preserve breakout at quite a long time before field
falls from cca 55 to 28 kV/cm and.. I'm far from being
satisfied with results I estimated from a literature
given conductive current density J needed for 1.3"
ball.For Q=94 there are about 9 cycles involved before
field goes below 30 kV.According to some statistical
formative time lags data in papers, that should be
more than enough time to avalanche and than streamer
form . "Crazy" assumptions would be that
a) after-Townsend area could be wider or
b) avalaches due to some effect cannot form into
streamers.
Something is definitely strange.I'm brainstorming what
could cause this.Mechanism should beheave in
extraordinary fashion where avalanches in one moment
increase local field but after that decrease it and
alltogather without conductive formed streamer.It
would have sense (maybe?) if streamer requires some
minimum lenght necessary for formation.Than (maybe
again)there's link between gradE* over area close to
small electrode surface with respect to previous
assumption of minimum streamer lenght in previously
-/+perturbed field by space charge layers ( heavy
ions?).And all that to happen with very low energy
expenditures.If a streamer forms, Terry should see
indications of it in a complete dark,shouldn't he?At
least something else is also indicative:Condenser
connected to the bottom of secondary doesn't build any
trace of voltage across it if visible breakout doesn't
form.
---
> Initial corona
> breakout preferentially occurs when the electrode
> has a negative
> polarity.
---
Formative times shorter for negative polarity?
Reference I have gives same minimum field for both
coronas(~30 kV/cm).
---
Formation of streamers ("cold" filamentary
> discharges that
> extend a significant distance from the electrode)
> will not occur until
> the electrode is driven to a significantly higher
> potential.
---
Is this more expressed,in terms of relative potential
percentages, when smooth electrode is small than when
it is big (same electrode geometry )?
---
These are
> very brief (10-50 nS), higher current discharges
> (peaks reaching ampere
> level). The hot discharges that Tesla Coilers are
> interested in are
> leaders, and these hotter conductive plasma channels
> behave more like an
> arc and typically carry peak currents of amperes to
> perhaps tens of
> amperes. Leaders are "fed" by all the combined
> streamer currents
> converging at the the tip of the leader.
---
I think there's basic diference between leaders
generated by VP generators,by DC source like in
lightning,and these formed by TCs.Latter ones just
need to be formed ,most of the rest of the job
performs fast oscilating TC ringing.Heat ,remanent
ionisation and charge debris from previous shots plays
important role in botling up channels more and more as
we know.
Propagation of leader's sequence doesn't necessary
mean dv/dt is rising.
---
Leader and
> streamer propagation
> require a rising d|v|/dt.
---
Bert,what do you think of minimum voltage (single
shot) necessary for formation of TC leader(s)?
(Long gap conditions assumed).
---
>
> Unlike initial corona breakout, "long" sparks (for
> DC, Low Frequency AC,
> and unipolar impulses) seem to propagate more
> efficiently (at a lower
> electrode voltage) when the active electrode is
> positively polarized and
> streamers and leaders are "cathode-directed".
> However, it's not
> presently known if the presence of space charges
> (from previous partial
> discharges) significantly alters the positive
> polarity preference in
> "long", low frequency RF air discharges.
---
This a good point.What do you mean by low frequncy RF
:Below 20 kHz.10 kHz?I guess ,as you go to 10 kHz and
down there must be some lower limit of voltage and
rise time of waveform not to go beneath if one wants
still to get good TC spark prolongation effects.
---
There's
> simply not very much
> about long RF discharges in the literature, and
> interpreting the few
> existing measurements taken via the OLTC is subject
> to somewhat
> conflicting interpretations - there are simply too
> many variables at
> present.
---
Tell me about it..
Regards,
Boris
__________________________________________________
Do you Yahoo!?
HotJobs - Search new jobs daily now
http://hotjobs.yahoo.com/
Maintainer Paul Nicholson, paul@abelian.demon.co.uk.