WFC design session v40: reasoning

Lab note — WFC resonant sweep, surface-treatment as the fresh lever.

The finer grid confirms the 10mm-in-12mm × 60mm cell as the sweet spot: C_eff 1.58nF resonating my 0.9mH chokes at f_r = 94.45kHz, tank Q 56.5, coil Q 126.8. Driving surface_factor down to ×0.25 on cold (4°C), 0.1ppm water did exactly what I hoped for displacement current — C_dl rides at 0.9µF, so 99.0% of the cell current is EDL charging and only ~19mA leaks Faradaically at the 734V peak (leak score 0.86). Breakdown margin ×91. That is a clean, leakage-inhibited climb up the ladder, and it's the whole point of the 4,798,661 control discipline.

But the vapor-film diagnostic is honest with me and I'll be honest here: at 293 Td the 0.1mm film lights in the townsend-avalanche regime, ABOVE my 80–250 Td / 3–6 eV cold-chemistry target. I resonate beautifully but I overshoot the CGDE window — the field per unit density is too hot for the super-Faradaic chemistry I'm chasing. Raising peak_v (#4/#5, 791V → 315 Td) only makes it worse. The scores cluster at 0.744 because duty (2/3/5%) barely moves average input; it's a near-flat ridge.

I'm adopting #1: lowest peak_v (734V), 2% duty, 14mA average in — least energy, softest film, closest to target.

Next session: pull peak_v BELOW 734V (drop drive/Q) to walk Td down from 293 into 150–200, OR thin the assumed vapor film — Td scales with E/N, so a lower cell voltage at this same geometry is the direct knob. Hold gap/length/count fixed.

Caveat: the 0.1mm film thickness is an assumption, not measured — it dominates the Td result and needs bench verification.