Spark Dissipation

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

Original poster: "Gary Johnson by way of Terry Fritz <twftesla-at-uswest-dot-net>" <gjohnson-at-ksu.edu>

Today I looked at the effects of ion dissipation on the formation of the
next spark.  I applied a 5 or 10 ms burst at the base of my 14 ga coil at
about 300 V, getting about a 6 inch spark from a half inch diameter brass
breakout point. This was followed by a second burst after a delay of between
2 and 30 ms. This doublet repeated at intervals of about once per second.
The first spark is at maximum length. The question is whether the second
spark will try to follow the path of the first one, and if so, will it show
an increase in length?

It appears that the second spark might try to follow the path of the first
one for a short distance (an inch or less) but then finds its own path. Two
distinct sparks were evident most of the time, even at a spacing of 2 ms. 

The residual ions make it easier for the second spark to break out, so the
toroid does not rise to as high a potential before breakout, so the second
spark is actually shorter than the first one. The effect is small, and would
be difficult to see. In this system, the peak power just before breakout
drops by about a factor of two from one spark per second to the CW case.
Since spark length is proportional to the square root of this peak power,
this means that spark length drops from about 6 inches at one spark per
second to about 4 inches at CW. At 50 sparks per second (10 ms burst and 10
ms spacing) the spark length computed from the measured current would drop
about 4.5 percent from the 1 spark per second value, a drop impossible to
visually observe. At 125 sparks per second (5 ms burst and 3 ms spacing) the
spark length drops about 12 percent from the 1 spark per second case. There
are enough other things happening in a conventional Tesla coil (resonance in
the power supply, heating effects, etc.) that this probably would not be
obvious either. The point is that one should use whatever break rate gives
the most pleasing visual appearance, without expecting that going from 60 to
120 to 240 sparks per second will make a significant difference in spark
length.

Gary Johnson