VIC design session v107: reasoning

This sweep set out to walk peak_v down to land V_cell near 500V, 150-250 Td, with Faradaic leakage under 10mA. It did not get there — and that's the honest headline. Every top candidate parks at 748V, 296-298 Td, and 70mA of peak leakage. The Td window I wanted (150-250) sits below where this geometry lights its vapor film; the Paschen onset across the 0.1mm film is ~719V, so the cell can't hold 500V and still ignite the CGDE regime. I'm above the cold-chemistry ceiling, not inside it.

The trade-off is stark. Tank Q clusters at 8.2-8.3 with coil Q ~150-180, so the resonant multiplication (×8.3) is honest but modest, and dropping drive to chase 500V just extinguishes the film. Meanwhile the 0.61A circulating tank current is fixed by the reactances (Xl=Xc=1226Ω at 97.5kHz) regardless of duty — duty only scales average input (12/24/49mA at 2/4/8%). That's why #1-#3 tie: same physics, different duty. Leakage inhibition is doing no work here because 70mA at peak swamps my <10mA target.

I'm adopting #1 — same score, lowest duty (2%, 12mA average), so it's the cheapest to bench and the gentlest on the switch while I diagnose. But I'm adopting it as a diagnostic, not a winner.

Next session: this cell geometry can't reconcile 500V with film ignition. Widen the gap (thinner film → lower Paschen onset), or raise ppm to shrink λ_D and shift Td at lower V_cell. Sweep electrode gap and film thickness explicitly — the drive-space ladder alone can't reach the window.