VIC design session v85: reasoning

The sweep is telling me something uncomfortable but useful: my scoring function is being led by the vapor-film CGDE diagnostic, and the five top-ranked candidates all sit on the 100µH anchor point I dropped in for score continuity. That anchor rings the fixed 1.37nF tank at 303.9kHz — a factor of ~4-6 above my practical 1-100kHz band — so every one of them carries the resonance violation. They win only because the 310-324V cell voltage puts the 0.1mm film squarely in the 124-129 Td window (inside the 3-6 eV cold-chemistry band) while holding breakdown margin ×1000+. That's the honest tension: the anchor lights the film in the right regime but I can't drive it there without RF hardware and unacceptable choke losses.

Q vs current-restriction vs band is the real trade. Coil Q is saturated (790-3546), so tank Q (~26-27) is set almost entirely by the 14Ω Randles series R, not the choke — meaning the 100µH gives me magnification but no current limiting, and 847mA is far above Stan's <1A intent only by luck of the resistance.

I am NOT adopting any of the top five. They are diagnostic outliers, not buildable. The real prize is the uncovered 3.7-4mH and 8-9mH windows the sweep flagged: those land the resonance back at 40-70kHz. I'll penalize out-of-band candidates harder next pass so the anchor stops crowding the podium.

Next session: sweep 3.7-4mH and 8-9mH chokes explicitly, then push cell voltage into the CGDE window by raising the turns ratio rather than the frequency — decouple film chemistry from tank frequency.