RE: Q Factor and Overall Efficiency

[Terry Fritz <twftesla-at-uswest-dot-net>]

Hi John,

At 12:14 PM 8/28/00 -0700, you wrote:
>
>Terry -
>
>I tried this type of test several years ago. It doesn't work because
>connecting anything to the secondary reduces the voltage an unknown amount
>and may cause sparkovers. Apparently you haven't tried the test I mentioned
>using a probe with no connection to the secondary. I don't have the
>instrumentation to do the probe test.

Your right!!  I ran some models of the resistor cap thing and it tells
nothing of importance.  In the CW case of this test, there is no
correlation to a working coil...  I'll just throw that resistor cap thing
in the trash ;-)

>
>As I recall you have made tests to determine the secondary voltage (Vs) of
>one of your coils. How did the test Vs compare with calcs using the energy
>equation and the sqrt(Cp/Cs) and sqrt(Ls/Lp) equations? The overall
>efficiency would then be
>     Eff = test Vs/energy Vs
>
>The operating Q factor would be
>      Q = test Vs/Vp
>Vp = Peak primary voltage

The equation  Vs = Q x Vp is wrong IMHO.  I use Vs = 1/2 x Vp x SQRT(Ls / Lp).

I know the peak voltage on my coil is about 280kV from numerous tests with
many types of equipment.  In my case:

Vs = 1/2 x 21000 x SQRT (75.4mH / 132uH) == 251 kV   About 10% low, but
pretty good!

This is basically the old Vs = Vp x SQRT(Ls/Lp) equation taking into
account that 1/2 the power is lost in the gap.

For CW coils Vs = Q x Vp may be far closer if one takes into account the
streamer load's effect on the Q.  But if you know that number, you probably
know the voltage anyway...  A CW coil's high Q is demonstrated by the
extreme sensitivity of those coils to frequency when driven by a generator
(most CW coils auto tune).  My CW coil is 350kHz and only 5 kHz makes a
dramatic difference in performance. 


So let's think about Vs = Q x Vp applied to disruptive coils...

In my case, The Q of my coil would have to be about 13 (280kV / 21kV).  My
coil runs at 83kHz so the bandwidth would be only 6kHz.  Thus, my output
would vary dramatically as I moved the primary tap over about +-1/2 turn.
However, it noticeably varies with +- 1.5 turns in actuality.  This
indicates that the Q is far less than 13.

I ran a computer study and produced a voltage vs. primary frequency graph
and posted it at:

http://63.225.104.218/test/TeslaCoils/Misc/BigLtr-VvF.gif