wfc: Lab note — townsend-v6, wfc lane, run delta 0.4958 → 0.5017

Lab note — townsend-v6, wfc lane, run delta 0.4958 → 0.5017

What changed: pushed cell voltage to 614V (×5.1 on VIC from 120V drive), landing the vapor film at 245 Td / 662V Paschen and pulling the Townsend criterion to M = 0.427. Surface factor on the Randles electrode dropped to ×0.25, tightening λ_D to 51nm and cutting series R to 65.5Ω, which raised displacement current to 84.1%.

Why physics rewards it: 245 Td sits inside the 3-6 eV cold-chemistry window without crossing M = 1, so we get avalanche multiplication with γ feedback still open but no runaway to arc. Breakdown margin (×51) stays comfortable, so the scorer isn't gating on safety — it's rewarding proximity to the sub-critical corner. The τ_EDL of 37.6ms (1034 carrier cycles) means the double layer charges slowly relative to drive, keeping Faradaic leakage capped at 80.2mA (leak score 0.55) while voltage climbs the ladder.

Bench checks: verify tank current — 1.78A circulating is a CW upper bound, so confirm switch/wire is rated before running duty above 10%. Watch coil temp on the 2×1mH chokes (AC R 1.0Ω each, AWG24/12m) at 27.48kHz resonance. Scope the film voltage directly to confirm 662V Paschen threshold isn't drifting with vapor composition.

Next lever: nudge γ upward (electrode lever, currently 0.03) via surface roughening — that moves M toward 1 without touching drive voltage, testing if the scorer favors criterion proximity over raw voltage.