Re: Streamer lengths

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

Original poster: "Gerry Reynolds" <gerryreynolds-at-earthlink-dot-net> 

Malcolm, John, and Bart,

Thankyou for your replies.  I think the "Controlled spark" is the way to go
for comparing one coil to another but I have another purpose in mind that I
wasn't very clear on.  What I'm trying to do with my small coil is determine
how well I'm doing with respecct to John Freau's empirical spark length
equation:

1.7 sqrt (power).

I'm tryind to measure performance, vary this and that, or try this or that
to find out what works best before I build my bigger coil.

I'm using a 4500V 22ma xformer,  multisegmented sparkgap using 1/2 x 3 inch
copper pipe with 2 100 CFM fans blowing down onto the sparkgap.  An inverted
conical primary made with 1/4 inch copper tubing  with 1/4 inch spacing and
sloped at 30 degrees.  The inner turn is two inches from the secondary.  The
secondary is 3.5 x 18 inches of 30 awg with about 1500 turns.  The top load
is 3 x 11 toroid mounted 4 inches above the top turn of the secondary.

I've set the safety gap to just fire with the unloaded xformer and the
variac at 140 VAC. In real operation with the variace set to 120 VAC, I
increase the main gap until the safety gap occasionally fires, scope the
peak spark gap voltage, and measure the spark length performance both to
free space and to a grounded rod.  I've started with a Cp that is 2.2 x Cres
and currently working down to resonance.  I limit the output voltage by the
safety spark gap setting (about 8000 Vpk)  and make sure  the same spark gap
voltages are reached (if possible) from one experiment to another.  In any
case, as I approach resonance, I expect the current to go much higher than
the 22ma short circuit output current of the xformer.  This will increase
the VA draw from the outlet to above what was measured (Vos and Iss) to
characterize the xformers thevenin output.


For John's Freau's equation, I think John said the power was watts (not VA)
from the power outlet. I guess this leads to  two questions:

1.  How did John measure his sparks when taking data for his equation?

2.  How did he determine the power?
      a.  Did he take the name plate numbers (4.5KV at 22ma in my case) and
derate this by a PF.
      b.  Did he measure the line current and phase relationship with line
voltage to determine the true power from the outlet at the operating point
he was using.


Anyway, I'm trying to determine if I got everything I can get from my
system.  My best result to date are with a Cp 1.5 * Cres (I haven't explored
with lower Cp yet).  This gives me 11 inch streamers to free space
(including spark growth).  Sparks to a gounded object have been
characterized two ways.   (1) a 15 inch separation where a corona is
beginning to developed from the grounded rod.  (2) a 9 inch separation where
a power arc will sometimes develope between the grounded rod and toroid.

If I only use the 4500v and 22ma values for power (~100 VA), I should get a
17 inch spark.  If I derate this by the power factor (I'm guessing 0.5 for a
static spark gap), my maximum spark would then be about 12 inches.  If I run
at resonance (for example) and use the actual line power at resonance,
should this be the bases for Johns Freau's equation?


 > Original poster: "John H. Couture" <couturejh-at-mgte-dot-com>
 >
 >
 > Gerry -
 >
 > I believe the "Controlled Spark" is a fair and reasonable method of
 > comparing Tesla coil output spark lengths. The controlled spark makes
 > engineering sense compared to the random streamer or spark that is
presently
 > used.
 >
 > The controlled spark is a constant load on the TC output. This means the
 > energy per spark is a constant and not a varying energy per spark. A
varying
 > energy per spark can give an unfair advantage to a coiler because the true
 > input energy for the spark is unknown. The controlled spark is a
horizontal
 > spark from the toroid to a ground point. It is determined by moving the
 > ground point until the sparking is constant and not intermittent.
 >
 > The controlled spark leaves a lot to be desired but so far no one has come
 > up with anything that makes more sense.
 >
 > John Couture
 >