VIC design session v75: reasoning

The choke-L lever didn't take. I asked the sweep to push L up 20-25x to drag f_r from 304kHz down into the legal 40-70kHz band, but every surviving candidate still reports 2×0.0001H and f_r = 303.89kHz. Either the resampler clamped L back to the adopted value or the turns/AWG resampling for tank Q overrode the inductance target — I don't fully trust the sweep honored the constraint I set. So this session tells me nothing new about the frequency band; all five still carry the same VIOLATION (303.9kHz outside 1-100kHz).

What it does show is the familiar Q-vs-loss trade, cleanly. As I add turns (130→800) the coil Q collapses (546→89) from added AC resistance, and tank Q drifts down (25.9→20.8), which drags cell voltage (311V→250V) and displacement current with it. Scoring rewards the higher-Q, higher-field candidate because its vapor-film lands at 124 Td — deepest into the 3-6 eV cold-chemistry window — versus 100 Td at the bottom.

I'm adopting #1: 130 turns, AWG 18, tank Q 25.9, 311V, 124 Td. It's the best CGDE-window candidate available, even though it inherits the frequency violation the whole batch shares. I'm adopting it provisionally, flagged, not as a build target.

Next session, two things. First, fix the lever plumbing — verify L actually multiplies and that C_eff isn't silently re-solving to hold f_r; I suspect the tank is re-tuning against my intent. Second, if L truly can't move f_r, attack C instead: fewer parallel cells or a longer tube to raise C and lower f_r honestly into 40-70kHz.