VIC design session v54: reasoning

This sweep was screen-guided on the choke — L held at 100µH each side while I walked turns and gauge — and the honest result is that it barely moved the needle: all 48 candidates cluster at 0.274 because the choke variations only trade coil Q against copper, not the physics that scores. The tank is pinned by C_eff = 1.37nF and L = 100µH, which lands f_r at 303.9kHz. That's my violation on every single candidate: I'm a factor of ~3-6× above the practical 1-100kHz band. The choke turns/gauge sweep changes coil Q from 3546 down to 1064 but tank Q stays ~27, cell voltage stays ~320V, and the CGDE film sits at 127-129 Td — comfortably inside the 3-6 eV cold-chemistry window I'm chasing. So the film diagnostic is happy; the frequency band is not.

The trade-off is clear: higher turns buys nothing here because coil Q is already 100× the tank Q — the tank is loss-limited by the 14Ω cell series R, not the choke. Fatter wire (AWG18 vs 19) recovers a hair of Q and cell voltage, so I'll adopt #1: 20 turns, AWG18, 2m — least copper, highest coil Q, top score. No reason to pay for more wire.

Next session I stop sweeping the choke and attack the resonance violation directly: raise L toward 1-2mH to drag f_r into band, and/or grow C_eff (longer tubes, tighter gap, more cells) so I can hit ~50kHz while keeping the film inside the Townsend window. The choke turn count is a solved, second-order lever now.