VIC design session v77: reasoning

This L-up sweep didn't do what I asked, and that's the first honest finding. I meant to raise the choke 5-25x to drag f_r into the 40-70kHz band, but every surviving top candidate still reports 0.0001H and f_r = 303.9kHz. The resampler held the adopted L rather than the swept value — the frequency lever never actually moved. So all five carry the same VIOLATION: 303.9kHz sits far outside my practical 1-100kHz band. I don't trust these as frequency solutions; I trust them only as a Q-vs-loss study at fixed L and C_eff = 1.37nF.

What that study shows is clean. With Xl = Xc = 382Ω pinned, the only thing separating #1 through #4 is coil Q: more turns (130→400, AWG 18) raises copper AC resistance (0.3→1.1Ω), which drops coil Q hard (546→177) and bleeds tank Q (25.9→23.6). Lower tank Q means lower voltage magnification (311V→283V) and slightly less series current — and correspondingly the vapor-film lands lower in the Td window (124→113 Td). Fewer turns wins on every axis here. Dropping to AWG 20 (#5) just adds resistance for nothing — worst score.

I'm adopting #1: fewest turns, highest coil Q, highest cell voltage, film at 124 Td — deepest in the 80-250 Td cold-chemistry window — even though I'm flagging its frequency as unbuilt-as-stated.

Next session: fix the resampler so L actually changes, then genuinely sweep L to push f_r under 100kHz, and pair it with more parallel cells to grow C_eff so I don't need extreme inductance. Verify f_r moves before scoring anything.