VIC design session v91: reasoning

Ran the 32-50 kHz seam hunt, but honesty first: the optimizer ignored my intent. I asked for 1-2 cells (C_eff ~1.37 nF) to push f_r up into the 32-50 kHz band; every surviving candidate instead parked at 8 cells parallel, C_eff 20.53 nF, and f_r 12.91 kHz. That's a solver constraint I need to run down — the 3.7 mH floor plus that capacitance pins the resonance at 12.9 kHz, well below the target band. So this sweep did NOT actually test the seam I wanted.

What it does show is the Q-vs-loss trade cleanly. All five share the same tank (Xl=Xc=600Ω, tank Q=1.2) because the water's series R (509Ω) dominates — coil Q ranges 150 down to 69 but barely moves the score. More turns and finer AWG only add copper resistance; #1 (800t AWG18, coil Q 150, 96 m wire) gives the highest coil Q for the least wire, so tank behavior is loss-limited by the cell, not the coil. Circulating current is a tame 0.23 A everywhere, so current restriction isn't binding here.

The damning number: 56 Td, film Paschen 719 V against 140 V cell voltage. Nothing lights the vapor film — we're an order of magnitude below the 80-250 Td cold-chemistry window. Score 0.273 is pure resonance credit, zero CGDE.

I'm adopting #1 as the reference build (lowest copper, highest coil Q) but only as a chassis, not an answer.

Next session: force the 1-2 cell / 1.37 nF case, raise L to 8-15 mH to hit 40 kHz, and drive cell voltage far higher — 700 V+ — to break the film open. Chasing Td, not Q.