Re: magnifier questions

["Tesla list" <tesla@xxxxxxxxxx>]

Original poster: FIFTYGUY@xxxxxxx 

In a message dated 12/13/04 6:11:19 PM Eastern Standard Time, 
tesla@xxxxxxxxxx writes:
The problem with trying the optimum designs in a magnifier intended to
produce sparks in open air are two: The coupling between the primary and
secondary coils results quite tight,

    This seems to historically have been a problem, although Dr. R 
recommends an auto-transformer configuration which sounds interesting.

> and it's difficult to make the
> capacitance required across the secondary coil. Look for a mode >where
>the last multiplier is high, and the C2 capacitance results small enough
>to be a distributed capacitance.

    That's what I ran into with the designs I made using Antonio's 
programs. If you make the driver inductance L2 small to get the ratio of 
L3/L2 around the "conventional wisdom" of 1:8 or so, C2 goes way up. At 
least this is simply a practical construction problem. And since the 
voltages at C2 are not as high as at C3, perhaps we could use some 
very-high-voltage rated caps, instead of isotropic capacitances like 
toroids and transmission lines? 600 pF at 200kV? That's a string of 
100  .06uF 2000V caps. I suppose the safe place to ground them would be 
somewhere in the middle of the transmission line, though...

> Another problem is that it will be
>difficult to obtain adequate quenching in the very low modes. Maybe
>better to aim as something as 4:5:...
    Isn't one reason for using the magnifier topology that the energy 
transfers out of the primary tank quicker? Or does the increase in gap 
current in low modes offset the losses over longer ringup times with lower 
current in higher modes?

>Large toploads are a good idea. See what you can build without >reducing
>excessively the voltage gain. Try to keep the maximum output voltage
>somewhat above the breakout voltage of the terminal that you use, to
>avoid the need of breakout poits.
    FWIW, I'm not convinced the voltage gain has a proven effect on spark 
length, past the initial breakout before the streamer starts to grow. In 
fact, I'm wondering if too much voltage gain might be a BAD thing, in that 
it encourages extra breakouts (maybe even branching from the main streamer) 
that robs power. Pure conjecture on my part.

>At high power, after breakout (somewhere), it's difficult to predict
>what will happen.

    Gosh, this seems to be the real problem!

-Phil LaBudde