Re: Freq Shift, was Second Commandment

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

Original poster: "by way of Terry Fritz <twftesla-at-uswest-dot-net>" <FutureT-at-aol-dot-com>

In a message dated 3/4/01 7:58:59 PM Eastern Standard Time, tesla-at-pupman-dot-com 
writes:

> I don't understand "a point must be used on the coil" in your comment
>  below.  Also, you say "pulls itself into tune" but I wonder if that's
>  really the case:  Would it not be that secondary-circuit Q diminishes so
>  drastically after break-out that Fr matters hardly at all? 

Ken, all,

I agree that the bandwidth should be wide enough after breakout
to give full spark length despite a small 10kHz shift in freq.  Some
aspects of my tube coil seem to be changing the issue though.

I seem to be seeing a 10kHz shift in freq when the spark breaks out
and grows.  The coil begins at 373kHz, then shifts down to 363kHz
or so.  Terry has determined that sparks have about 0.8pf
per foot I believe, and based on the total 14pf of my secondary/
toroid combo, such a freq shift would seem reasonable.  Either
that or there's something going on with my coil since it's a vttc
with tuned primary and secondary.  Splitting of frequencies
may be confusing the issue.  The calc'ed primary freq is 404kHz, 
and the measured freq is 400kHz with
the secondary removed.  At a coupling of 0.22 = k, a lower
split-response at about 362kHz would be expected.  This matches
very closely with the 363kHz that the coil seems to be running at
when loaded by spark at full power with 13" spark output.  I am
now picking off the signal across a 10 ohm, 10 watt resistor in
series with the base of the secondary which seems to reduce
the spark length about 1/2" or so.  I suppose I'm now seeing
only the secondary freq on the scope.

To obtain break-out from a 1/2" ball, I had to re-tune the coil
radically to about 480kHz calced, which may be about 435kHz
actual.  This results in a short spark.  This need for re-tuning
seems to be a part of the problem.  I'm not sure why it needs
this radical re-tuning.   Another part of the problem
with using a ball on a typical vttc running on raw ac, is that the
breakout has to wait until peak of sine is reached, then only
the downsloping part of the 60Hz sine is left, which creates
short branched sparks.  

When I do tune the coil for best breakout and longest sparks
from a ball, I then see a 20kHz shift when the sparks break
out, even though the sparks are shorter than with the point.
This suggests to me there are other factors at work, such
as the freq splitting, loading, impedance matching, etc, which
are confusing the issue.

Ken, I suppose that based on the size of your coil, the
size of the sparks, and the freq of operation, I would expect
about a 5kHz drop in frequency.  Since the sparks do have
some capacitance, wouldn't some sort of freq shift be 
expected?

John Freau