[Prev][Next][Index][Thread]
Re: Power versus Spark Length
From: FutureT-at-aol-dot-com[SMTP:FutureT-at-aol-dot-com]
Sent: Wednesday, July 23, 1997 3:41 AM
To: tesla-at-pupman-dot-com
Subject: Re: Power versus Spark Length
>snip>
<< However, if you use averaging methods rather than one shot methods, you
> can measure the watts input for a number of sparks over time (energy) and
> use the following equation. In this setup the Cp voltage is averaged over a
> time period.
> Input W = .5 Cp Vp^2 BKS / Eff (Per second)
> The energy output would be represented by the length of horizontal
> continuous sparks (controlled sparks) from the secondary terminal to a
> ground point.
> I admit this still leaves alot to be desired but is better than what we are
> now doing.
---------------------------------------
>-GL
>
John C.
> >>
John C, All,
Your thoughts on TC analysis are greatly appreciated, even if we
often don't seem to agree. I find your insights important, although
I often come to a different conclusion when all is said and done.
Malcolm, Bert H, and others, in some previous work have shown that a
TC functions differently when the spark hits a ground. The ground strike
causes a change in the output impedance of the system which
reflects back to the primary tank and results mostly in trapped
energy in the gap. This energy is not available to the output spark
and therefore changes the performance of the coil. The gaps can
actually be seen to glow brighter during ground strikes due to this
wasted energy. Another unwanted effect is a worsening of the gap
quenching. All this leads to a reduction in the efficiency of the TC.
I think that it is important that we analyze our coils while they are
functioning in a normal, properly loaded condition. This is not the
case during controlled, steady ground strike conditions.
It is interesting to watch a slow motion video of TC sparks. The sparks
start out short, and start growing and reaching outwards with wavering
tongues. These tongues often ebb and flow, usually never growing
smoothly, rather the growth can be seen as a "tug of war" between the
forces of energy dissipation, and the forces of spark formation. When
conditions are right, the forces of growth exert the upper hand, and the
sparks succeed in reaching magnificent lengths. The spark can take
anywhere from 1/15 th to 1/4 second, in a particular coil, to reach its full
and final length. After reaching this ultimate length, the spark generally
pulls back greatly, seemingly exhausted by its efforts, before starting its
"growth-dance" anew.
It is clear from the above, that a particular gap firing does not produce a
particular output spark. Rather, what we see as an output spark, is
actually the result of a gradual spark growth over many gap firings. Each
gap firing adds energy and sometimes length to the partially formed
output spark.
The process described is in many ways an essential essence of what a
Tesla coil is, and how it functions. If our efforts to measure the coil in a
"scientific" manner interfere with this natural TC behavior, then IMO we
have failed totally in our TC analysis.
We must study our coils while they are operating normally. Whatever we
might gain by using a controlled spark length, we will lose even more by
interfering with one of the key, inherent aspects of TC operation -- that of
the unfettered growth of "hunting" sparks.
There must be some scientific branch that deals with just such
phenomena, perhaps fractals, or chaos theory? There are many such
parallels in science, such as the turbulence that leads to the great red
spot on the planet Jupiter. The spot changes continuously, yet science
has found a way to analyze and characterize the formation and behavior
of this manifestation.
If we can describe a mathematical model of spark formation, and relate
it to power input, break-rate, toroid size, etc., then we will be approaching
a complete analysis of TC behavior. This is the TRUE scientific approach.
Still, I suspect there must be an easier way that we're missing. It might
be possible to do some kind of computer simulation also?
Comments welcomed,
John Freau