Original poster: "resonance" <resonance@xxxxxxxxxxxx>Some experimenters use 1.6 x Cres value. I prefer 1.3 x Cres value for all my nst powered coils. The extra current makes the spark dance around more on the toroid and it looks more interesting. 1.6 x Cres will actually give you the longest possible spark but it's a strict capacitive type discharge and usually a single streamer where the 1.3 x Cres value makes multiple streamers of purple off the toroid and usually at 2-3 different discharge points. Less spark channel current but a lot more activity. The sparks rotate and seem to have an "it's alive!" type character. It's more of an art thing vs a long single streamer --- your choice.
Dr. Resonance
For your 30KHz transformer, the formulas that you want are these: Zout = Vs_oc_rms/Is_sc_rms Cres (nf) = 10^9 / (2*pi*freq*Zout)use 30KHz for your frequency. The assumption here is that the winding resistance is insignificant compared to the current limiting leakage reactance. If you can measure the primary and secondary winding resistance, we can verify the assumption.Gerry R.