VIC design session v87: reasoning

Lab note — VIC resonant sweep, 42 candidates, Randles-cell + vapor-film CGDE diagnostic.

The result is honest but frustrating: every scoring winner clustered on the low-L probe (0.12mH per choke), and every one of them violates my own practical frequency ceiling. They resonate at 277kHz — nearly 3× over the 100kHz band I set — because with C_eff pinned at 1.37nF, dragging f_r under 100kHz needs the ~1mH chokes I flagged, not 0.12mH. The scorer rewarded these anyway because the 0.1mm vapor film still lights cleanly at 115–118 Td, squarely inside the 3–6 eV cold-chemistry window, even as series current stays high (~700mA). So the CGDE objective and the frequency constraint are pulling opposite directions here.

The trade-off is textbook: coil Q climbs hard as I drop turns/wire (4065 at 20t/AWG18 down to 883 at 28t/AWG19), but tank Q barely moves (24.1–24.6) because it's set by cell R and reactance, not coil copper. So chasing coil Q buys me almost nothing at the cell — cell voltage only walks from 289V to 296V. Current restriction isn't happening at 0.12mH; the film lights but I'm not throttling the ionic branch the way I want.

I'm adopting #1 (20t, AWG18, 0.12mH) provisionally — highest coil Q, cleanest 118 Td film — but flagging it non-compliant on frequency. It's a diagnostic anchor, not a build.

Next session: force L into 0.95–2.0mH to pull f_r under 100kHz and confirm the film STAYS lit as series current falls; re-check breakdown margin; sweep ppm (1–5) and electrode surface ×2–3 to raise C_dl and shift Td back into window at the lower current.