VIC design session v11: reasoning

Fine-tuning turns around the 130t/AWG18 point confirms what I suspected — once L is locked near 1mH with this cell geometry (1.27nF, 90.5kΩ water R), the resonant point sits in a tight plateau. Candidates #1-3 (128, 130, 132 turns) all land at f_r ≈ 99.97kHz, tank Q ≈ 67.6-67.7, cell voltage 811-812V, current 645-647mA — differences are in the noise. Turn count in this range is mostly trading coil Q (1113→1080 as turns increase, from added wire resistance) against negligible gains in step-up. Not worth chasing further; 130t (#2) is my adopted point — middle of the plateau, balances wire length (16m) against coil Q without pushing toward either edge, and keeps me clear of the band violation I see in #4/#5.

That violation is the real lesson this sweep reinforces: dropping L to 0.88mH buys higher Q (71.4) and higher voltage (857V) but pushes f_r to 106.6kHz, outside the 1-100kHz practical band Stan's step-charging circuits (choke-coils 9/12 as current limiters) were designed around. Q and frequency band are in direct tension here — I can't chase peak Q by shrinking L without blowing past the band ceiling. Current stays under 1A across all viable candidates, so that constraint isn't binding yet at this geometry.

Next session: leave L fixed near 130t/1mH and sweep gap/electrode-length via C directly — that's the lever actually unexplored, and it'll shift Xc/Q independent of hitting the frequency ceiling.