Re: Cap size for pig?

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

Original poster: "Gerry Reynolds" <gerryreynolds-at-earthlink-dot-net> 


Hi Mark,

I've been doing a lot of spice simulations for static gaps using all sorts
of transformers, from plate transformers, to 12/30, 12/60, 15/30, 15/60.
15/120, and PDT's (ballested).  I believe I have found a unifying equation
for all these trannies that have resonable efficiency.

Even the PDT follows the same curve when the effect the ballast has on
current limiting is simulated.  The same equations used for NST's to
calculate Cres can be applied to a PDT.

XL = Vs_rms_oc / Is_rms_sc                   (very little winding resistance
in a PDT)

Cres (nf) = 10^9 / (2*pi*line_freq*XL)

The normalizing equation is for power factor vs Cp/Cres.  Given Cp/Cres, PF
is calculated assuming that the gap voltage is set to Vs_peak_oc.  There is
a scaling adjustment for different Vgap settings that is reasonably accurate
for STR values that I will include below.

Anyway, from PF and the VA rating of the tranny, power delivered thru the
spark gap can be determined (power flow thru the gap was the basis for this
curve so power loss in the tranny need not be factored out).  From this
power and the energy per bang, the BPS can then be calculated using the
following relationship:

Power (watts) = Bang energy (joules) * BPS

The base line PF equation (for Vgap set to Vs_peak_oc) is:

PF = 0.78  -  0.27 * Cp/Cres   (PF flattens out at 0.6 and of course does
not actually go negative)

If the equation gives a PF larger than 0.6, use 0.6 instead.
If the equation gives a PF smaller than zero, use zero instead

Simulations were only done out to a Cp/Cres ratio of 2.0 so behavior beyond
this is unknown.

For other Vgap settings, the simulated PF was very close to a scaled version
of the base line PF in the STR region.   You can scale the above results by
Vgap/Vs_peak_oc.  In the simulations with Vgap larger than Vs_peak_oc, the
transient response would fire the gap for a considerable ratio above
Vs_peak_oc until large Cp/Cres ratios were done.  Of course, once the gap
stops firing the PF goes to zero.

Regards,
Gerry R




 > Original poster: Mark Broker <mbroker-at-thegeekgroup-dot-org>
 >
 > Well, first, we didn't include MOTS, PTs, or pole mount transformers (PDT)
 > because those are highly variable, whereas NSTs are pretty well
 > defined.
 >