Re: Big Spark Induction Coil

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

Original poster: "Antonio Carlos M. de Queiroz by way of Terry Fritz <twftesla-at-qwest-dot-net>" <acmq-at-compuland-dot-com.br>

Tesla list wrote:
> 
> Original poster: "Kurt Schraner by way of Terry Fritz
<twftesla-at-qwest-dot-net>" <k.schraner-at-datacomm.ch>

> I probably don't get, in what sense "dependence on spark length":
> thought, the equations, leading to optimal ("magic"-) values of k,
> regarding sec-voltage, don't contain any spark-length parameters? (- But
> probably only some misunderstanding on my part.)

The magic values of k cause complete energy transfer to the
secondary capacitance, assuming no breakout. If a spark occurs
much before the energy transfer is complete, the behavior of the
system is different, with the spark being less energetic. For
maximum output energy, the secondary spark must start close
to the instant when the secondary voltage is maximum, and so
there is a well defined corresponding optimum spark length. With
the magic ks, at this instant the primary voltage and the
currents in the coils are all zero. This can be made to happen
in a Tesla coil or in an induction coil.

> Nice, you verified the same behavior. Of course I'm not shure, what's
> happening with the spark at "microscale", meaning if it's
> lighting/extinguishing every half-cycle of the f1 oscillation, which
> might seem plausible. 

I am thinking that the f1 time corresponds to the time taken by the
dissipation of the energy stored in the secondary coil, through a
decaying DC current through the spark, with a small oscillation
added to it. The spark ends when this current is not anymore enough
to keep the spark hot enough. See the post by Terry showing the
current waveform in the secondary of a Tesla coil when the
output is shorted. The same happens in an induction coil.

> The "light from the spark" experiment looks very
> promising for that reason. I wonder, if a carefully designed
> jacobs-ladder experiment, zooming the spark on a corresponding photo
> (and, of course, registering the BPS, as a time-reference), might reveal
> the same info? My current JL-pic's are just too coarse, to see anything
> more, than a certain blurr within the single 50Hz or 100Hz (BPS was not
> recorded) spark-strikes.

The sparks appear to last for a few ms, judging by the f1 times. This
is a long time for a spark, but too fast for regular video or cameras.
I started an experiment about observing the light of the spark with
a phototransistor. What I could see was just interference from the
spark pulse. The next attempt will include a Faraday cage...
Interestingly, the interference from the spark appears as a high pulse
at the start of the spark, containing a few cycles of oscillations
at many tens of MHz (the initial electrostatic spark), followed by
"silence" through the f1 oscillations, ending with a copy (~) of the
f2 oscillations. This is consistent with what is expected as the
secondary voltage.
 
> Obviously a resistor, as a replacement for a spark, is very primitive.
> I've tried to add a DC voltage source, in series to the 150k resistor
> with "sim" 24Ab, setting the DC voltage to 1kV - 1.2 - 1.5 - 2 - 10kV.
> While the primary voltage beeing distorted "somehow", I'm not able to
> reproduce the observed effect, by this means. 

Try to reduce the resistance of the spark. It really behaves as a
voltage source while there is current through it (the voltage 
polarity is always associated with the current, dissipating energy in 
the spark).

> I'm also not feeling very
> comfortable, giving a physical interpretation to this voltage source.

The vxi characteristic curve of a spark gap is similar to what
appears in a gas discharge lamp. They require a high voltage to
start conducting, but after this behave as a voltage source in
series with a small resistance. The voltages depend on the temperature
of the gap/lamp.

> May be, the current-depending spark resistance, as well as the
> rectifying effect of the gap, should be included somehow. I'm aware of
> spark model-examples with much higher complexity, but have not yet tried
> to implement one in any of my "sims" (i.e. http://205.147.9.162/nl50.htm
> , "Spark Gap Modelling" by Christophe Basso, Intusoft).
> 
> BTW: it would be nice to have a comparable "static spark gap" building
> block for Microsim 8/Schematics. Anybody on the list knows, if it could
> be included in the trial-version, and able to program it? (The starting
> complexity could probably be taken from the link above).

I think that I have seen this. I will check. I have some ideas about
how to model a spark that I would like to try.

> I'd really like to see the "photos of storage scoped waveforms of a TC
> connecting to grounded object", Malcolm mentioned. "Think" the strike of
> a TC to a grounded object would afford a different simple spark model,
> than Terry's successful simple "220k+5pF-per-foot" streamer model.

Appears to be a voltage source, as first approximation.

Antonio Carlos M. de Queiroz