Re: Magnifier conversion

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

Original poster: "Antonio Carlos M. de Queiroz by way of Terry Fritz <teslalist-at-qwest-dot-net>" <acmq-at-compuland-dot-com.br>

Tesla list wrote:
 >
 > Original poster: "Terry Fritz" <teslalist-at-qwest-dot-net>

 > The model says I get 3% more RMS voltage this way and the actual test gives
 > 6% more voltage!!
 > http://hot-streamer-dot-com/temp/NewMag/0503-02.gif
 > So th conventional coil gives 50.54Vrms
 > http://hot-streamer-dot-com/temp/NewMag/0502-05.gif
 > The 2,3,4 magnifier gives 49.55Vrms
 > http://hot-streamer-dot-com/temp/NewMag/0502-07.gif
 > And the no C2 magnifier gives 56.01Vrms
 > http://hot-streamer-dot-com/temp/NewMag/0503-01.gif
 > I wonder if rather than having a mode that gives the highest "peak
 > voltage", a mode that gives the highest top terminal RMS voltage is our 
goal??

I am not sure if these simulated RMS voltage are meaningful. They
apparently are calculated by squaring the voltage, low-pass filtering
the result and taking the square root. Since the signal is not
periodical, what is read depends on how the low-pass filter is.
And the filter appears to be slower in decaying than the signal.

When C2 is below the calculated, the system operates close to a
mode that admits elements similar to the originals, except for
C2. The magsim program can produce a table where possible candidates
can be observed. For my system, the best combinations above the
designed mode 3:4:5 are 6:7:14, with C2=24 pF and 7:8:21, with
C2=14 pF. The first agrees quite well with what I observe, that
is essentially a distorted mode 6:7, or a bit more, for two coils.
With C2:
http://www.coe.ufrj.br/~acmq/tesla/mag345v1l.jpg
Without (lumped) C2:
http://www.coe.ufrj.br/~acmq/tesla/mag345c2l.jpg
Losses in both cases look the same.
Note that the two coils system that I made with the same resonator
operated in mode 6:7 only with an Oudin coil assembly. Normal
operation was in mode 7:8 to 8:9. Even the distributed C2 still
contributes to accelerate the energy transfer.

 > Yes,  This is all clean laboratory stuff right now.  The truth is in the
 > sparks ;-))  I note that the models and the real system have a little
 > frequency error.  I should probably go back and check the distributed
 > capacitances to tune things in more between the model and the scope
 > readings.  But the "no C2" case seems to give much better results RMS
 > voltage wise...  That would also match what the other magnifier builders
 > have often reported.

C2 accelerates the energy transfer in any way, although just a
discrete set of values produce efficient energy transfer (actually
just one value produces perfect energy transfer, but other values
get close enough). What may happen is that eventually gap quenching
becomes difficult, and the rise in the maximum output voltage by
having the "best" C2 is not enough to compensate. A high-power test
can reveal what happens. In my system, what I saw was no quenching
and something eating all the energy quickly (I will still look at
this more carefully). Possibly the corona around the plates of my
C2, or just the many streamers in my pointy top load. Without the
lumped C2 the output was ringing for more time, producing better
streamer output.

Antonio Carlos M. de Queiroz