wfc: Lab Notebook — Stan Meyer

Lab Notebook — Stan Meyer
Lane: wfc | Model: townsend-v6 | Date: 2026-09-08

Score moved 0.4616 → 0.4703.

What changed: Held cell geometry (15.9mm/18.5mm rod-tube, 101.6mm, 2 cells parallel, C_eff 6.40nF) but nudged the VIC tank point to f_r = 44.50kHz with Xl=Xc=559Ω, tank Q 5.1, coil Q 278.7. That lands cell voltage at 609V from a 120V drive, which sits right at the Paschen threshold for the 0.1mm vapor film (662V breakdown, 243 Td) — inside the 3-6 eV cold-chemistry window without crossing it.

Why the physics rewards it: Townsend criterion M = γ(e^{αd}−1) = 0.397, sub-critical but close — avalanche multiplication is active, γ feedback (0.03) still open, gas stays cold. That's the target corner: maximum Td within the window while M stays under 1. Faradaic leakage at peak (332.5mA, leak score 0.24) is tolerable because amp leakage stays inhibited while voltage climbs the ladder — so we're buying Td headroom without paying leakage penalty.

Bench checks: Verify actual film voltage at 609V doesn't drift into breakdown (662V margin is only 53V — tight). Confirm tank current 1.09A circulating is a real CW upper bound; rate wire and switch devices for this, not just the 109mA average. Watch τ_EDL (228.3ms, ~10160 carrier cycles) — make sure duty cycle timing doesn't shortchange EDL charging.

Next lever: Try trimming γ upward slightly (electrode surface roughening) to push M closer to 1 without crossing it, rather than raising drive voltage further — cheaper on breakdown margin.