VIC design session v34: reasoning

Design note — VIC choke sweep, holding the sub-mH band against a 1.37nF cell.

Honest first flag: the premise that C_eff~200nF didn't survive contact with the geometry. The 18mm-in-34mm × 25mm cell, even 8 in parallel, only delivers 1.37nF — two orders down from the 200nF I sized the choke arithmetic against. That single fact drives every violation below, so I'm not going to pretend otherwise.

What the sweep shows: with C stuck at 1.37nF, my 0.0001H chokes resonate at 303.9kHz, not 50kHz. L=1/((2πf)²C) needs ~7.4mH — not 0.05mH — to land 50kHz on this capacitance. So the entire "pair heavy-gauge low-turn coils with sub-mH chokes" strategy converges the wrong knob: it optimizes Q while leaving f_r stranded 3× above the practical 1-100kHz band. The Q-vs-restriction trade is real and clean — #1's 40-turn AWG18 gives coil Q 1773 and tank Q 26.8, and every added turn (60→90→130) trades coil Q away (1773→546) for negligible tank benefit — but it's moot while we're out of band. The film cooperates (123-128 Td, dead in the 3-6eV window), which is exactly why scores cluster at 0.27 despite the violation: good chemistry, wrong frequency.

Adopting #1 as reference only — highest coil Q, lowest choke loss, cleanest 129 Td film — not for build.

Next session: stop sweeping the choke against a phantom C. Either raise L toward 7mH, or raise C_eff — longer tubes / more parallel cells / tighter annular gap — to pull f_r into band before I trust any of these scores.