[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: New DRSSTC Benchmark



Original poster: "Antonio Carlos M. de Queiroz" <acmdq@xxxxxxxxxx>

Tesla list wrote:
Original poster: "D.C. Cox" <resonance@xxxxxxxxxx>

Does this "multiplier" also apply to the SQR of the sec and pri inductances? How large of a multiplier value are we talking about?
Does this relate to autotransformer action or some other mechanism?

Yes. For excitation between the resonances, squared sinusoid input and the primary and secondary circuits tuned so: Lb/La = Ca(1-k^2)/Cb (1) The ideal voltage gain for square wave input is: (8/(k Pi)) sqrt(Lb/La) (2) Assuming low losses, of course. The voltage gain is inversely proportional to the coupling coefficient. La, Ca = primary; Lb, Cb = secondary. For excitation at the resonances, the gain depends on how long the driver operates, and for the same number of cycles the gain is lower and the input current higher.

I have usually used Vi x 70% x SQR (Ls/Lp) which seems to relate very close to measuring the peak potential on an o-scope and firing the pri once every 10 seconds through a diode-resistive charger (around 10 meg). I calibrate the scopes small metal metal plate pickup (2 x 2 in) by using a 200 kV DC power supply on the toroid. This gives me a reference signal level to compare the peak TC potential out to (waveform is of course independent of peak potential --- HV Engineering, 2nd ed. Naidu & Kamaraju).

You can probably use the same technique, with the 70% of (1) to account for some losses and nonlinear voltage profile along the secondary.

Antonio Carlos M. de Queiroz