# Re: Power vs. Spark Length

```
From: 	FutureT-at-aol-dot-com[SMTP:FutureT-at-aol-dot-com]
Sent: 	Friday, July 25, 1997 4:35 AM
To: 	tesla-at-pupman-dot-com
Subject: 	Re: Power vs. Spark Length

>snip>
<< When watts per foot of spark is used, a period of one second is
> implied. This means many sparks are involved in the one second period and
> should be averaged (controlled spark).
>>snip>

>  John Couture

John C, All,

I want to make sure I have a correct understanding of your controlled
spark idea.  The way I understand it, the spark should be of the max
length that will remain connected to a ground point indefinitely.

One TC I built gave 80" sparks measured by occasional strikes to
ground.  Sometimes the sparks hit a ground point that was only
35" away.  Yet, the sparks did not remain attached for long before
they detached and headed in a new direction.  I estimate that to
obtain sparks that *stay* attached, a strike distance of maybe 20"
or so would have to be used.  20" is so completely different than
the 80" occasional strike distance, that I question whether there can
be any reliable relationship between the two types of measurements.
In other words the ratio between the two may vary with break-rate,
coil size, etc.  Also this 20" connected spark would be of a fearsome
brightness.  Maybe some smaller coil could produce a 20" connected
spark of lower brightness, depending on break-rate, etc.  What I'm
suggesting is that a connected spark may give a false impression of a
coil's capabilities even as compared with other coils.  If two different
coils give a 20" connected spark length but one can give a free air 80"
spark, and the other can give only a 50" free air spark, then IMO the
controlled spark is giving us a misleading measurement.

You said we must obtain an average spark length over time to
agree with our average input power... yet we cannot easily measure
true efficiency anyway.  I guess it comes down
to what we are trying to accomplish.  You stated that it does not make
sense to measure an isolated spark because we don't know how much
power went into the formation of that spark.  So this is your goal, you
want to know how much power (energy really) went into a particular
spark.  Since this is not easily determined, you are using the averaging
method to get around that (in a sense).  That is a valid goal, if I can
acheive it without forcing the coil to work in a way that disrupts its
normal operation.

I want to see what our coil's maximum spark capabilities are.  By
measuring max spark length, you're right, I'll never know how much
power is in a certain spark...and I'm willing to accept that (for now).
I'm sure there are other mechanical or electrical systems which also
strive for max rather than average performance, i.e. the average is
ignored as unimportant, although I can't think of any at the moment.

As you said, we are making many compromises in using
spark length as a measurement.  Using
a controlled spark length...we are making more compromises.  By
using max spark length we are just swapping one compromise for
another...but at least the max spark length measurement allows our
coil to be working in its normal mode of operation, and allows for a
convenient way of comparing one coil with another.  Greg's idea of
looking for a certain number of spark hits per minute is a helpful
refinement. We can of course chose to measure our TC "efficiency"
(layman's usage) in "sections" (power supply, RF section, etc), and
this is a good idea to do.  Is this all scientific?  No.  Is it practical?
Yes.

Perhaps in the end, some of us will favor one method, or another for
our TC measurements and analysis, and all these methods will over
time gradually improve our understanding of these wonderful devices
which we all enjoy even as we debate these issues.  In any case, it's
time for buildin' and measurin'!