RE: Best cap size for a sync gap

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

Original poster: "Lau, Gary" <gary.lau-at-hp-dot-com> 

Hi Gerry:

 >Original poster: "Gerry Reynolds" <gerryreynolds-at-earthlink-dot-net>
 >
 >mV???? as in millivolts?
 >Shouldn't the bang voltage be in the order of 10's of KV or is this the
 >uncalibrated optic probe voltage?
 >
 >Also, did your safeties ever fire (especially at .02uf).  I would expect the
 >maximum voltage to be at 20nf (greater than the 21KVpk for a 15KVrms source)


 >>FWIW, if I scope just the unloaded NST secondary -at-120VAC input, I
 >>get 598mV p-p.

 >Ahh!  This is your calibration.

Yup.  Sorry, I should have applied the inferred calibration to my figures 
to avoid confusion.


 >Also, did your safeties ever fire (especially at .02uf).  I would expect the
 >maximum voltage to be at 20nf (greater than the 21KVpk for a 15KVrms source)

As the cap size was considerably LTR in all cases, the NST/cap tends not to 
ring up severely.  When I advanced the timing just beyond the sweet spot, 
the circuit lapses into a state where the gap presentations occur at or 
near the zero crossings and the gap does not fire at all (scary!).  I 
measured what this voltage was, but my notes are at home and I'll post them 
tomorrow.  But they were not considerably in excess of the open-circuit NST 
voltage.


 >> If I calculate the relative bang size with a simple scale-less formula of
 >> C*V*V (mV*uF*uF), I get

 >I think you mean (uF*mV*mV) ??

Yup.  But the figures I presented were correctly calculated.


 >I've done your experiment but only on a computer.  I'm wondering if the
 >halogen load is getting the Cp completely discharged in an equivalent amount
 >of time as the energy transfer time in a real TC.  I would expect the
 >voltage to drop off more rapidly as Cp was increased from .02 to .05 uf
 >
 >Seems like it might be possible to repeat the experiment with a real TC
 >since you are optically coupled.
 >
 >Gerry R

The dummy load and .04uF cap have a time constant of only 3.5usec, so it's 
all over pretty fast.  Scoping so you can see one entire 16msec cycle shows 
the bang being just a vertical line to zero.  But I would expect the 
discharge to be slower, not faster, as Cp increases.

I prefer to experiment in my basement if I can - just easier with 
everything there, rather than hauling it all out to the garage where 5 ft 
streamers are OK.  I suppose I could swap in the primary inductor and place 
an aluminum pan with water on top of the coil as an eddy-current dummy 
load, rather than a secondary (I saw this on one of the solid state guy's 
site).  But even if the dummy load discharges in an interval 10X greater or 
smaller than a true Tesla coil, I would not expect the overall result of 
this experiment to change, as the 60/120Hz charging behavior is not 
significantly altered.

Regards, Gary Lau
MA, USA