VIC design session v38: reasoning

Lab note — VIC resonant sweep, 0.3–2mH gap bridge.

The gap didn't close the way I hoped. Every survivor pinned itself at 0.1mH per choke and f_r = 303.9kHz — the solver refused to buy real inductance in the 0.3–2mH band while holding a 250–300V film, so it kept the small choke and paid for voltage with turns and gauge instead. That confirms the frequency wall: to sit inside the practical 1–100kHz band at this C_eff (1.37nF, 8 cells parallel) I need far more L than 0.1mH, and the film voltage collapses when I add it. All five candidates VIOLATE the band. I'm noting that as the real finding, not a footnote.

The clean trade-off across #1–#5 is turns vs coil Q vs film drive. Going 200→600 turns (AWG18) at fixed 0.1mH bleeds coil Q from 355 down to 118 and cell voltage from 303V to 265V — more copper, more AC resistance, less magnification. Dropping to AWG22 (#5) craters coil Q to 79 and film to 243V/97Td. Current restriction tracks inversely: lower Q means lower series current (793→635mA), which is actually gentler on the cell but costs cold-chemistry Td.

I'm adopting #1: 200 turns AWG18, tank Q 25.2, 303V/121Td — deepest inside the 80–250 Td / 3–6 eV window, best displacement fraction, highest coil Q. Its only sin is frequency, shared by all.

Next session: stop fighting the band with L alone. Sweep C_eff down (fewer cells parallel, longer tube gap) to drag f_r toward 100kHz, then re-check whether 0.3–1mH chokes can hold 250V film. Also probe surface ×2–×4 to lift Td at lower cell voltage.