WFC design session v42: reasoning

This sweep set out to drop peak_v until V_cell landed in the 500–600V / ~180 Td CGDE window, and it did NOT get there — the top three all sit at 837V across the film, 334 Td, squarely in Townsend-avalanche territory, above my target cold-chemistry band (80–250 Td / 3–6 eV). So the honest headline: attacking peak_v alone, at the 10mm/12mm×50mm geometry and Q≈70, the tank still magnifies 12V to 837V before the leakage score degrades. I could not both hold resonance and starve the field into the window purely by trimming drive amplitude — the ×69.7 magnification is the floor here.

The trade-off is clean. Q≈70 buys the voltage but also the 0.68A circulating current (rate the wire/switch for that, not the 14mA average). Faradaic leakage stays quenched — 17.9mA at peak, leak score 0.87, displacement current 99.1% — because the ultra-low duty (2%) keeps τ_EDL charging incomplete; that's the lever working. The 50→55mm variants (#4/#5) trade 15 Td and a little Q for slightly more capacitance and a marginally better breakdown margin (×84 vs ×80), but score lower — length isn't the win.

I'm adopting #1: 50mm cell, 2% duty, 14mA average input — lowest duty at top score, least Faradaic exposure, best breakdown margin among the top tier.

Next session: I can't reach 180 Td by amplitude alone, so lower Q deliberately — detune slightly or add series damping — and widen the film gap in the Paschen calc. Vary Q (40–70) and film thickness (0.05–0.15mm) together; that's the real path into the window.