WFC design session v13: reasoning

Lab note — ppm/temp fine sweep, fixed geometry (21.8/26.8mm×60mm, single cell, surface ×0.5)

Ran the fine grid on ppm and temperature independently, holding geometry/surface fixed, hunting for a bulk-R minimum that could buy back Q headroom. Didn't find one — the trend is monotonic and expected: lower ppm and lower temperature both push bulk water R up (127.7kΩ at 0.1ppm/3°C vs 99.6kΩ at 0.2ppm/3°C), and that higher R is exactly what's raising tank Q (97.5 down to 77.6 across the top five). Higher Q means more series current too — #1 sits at 983.6mA, brushing the 1A ceiling, while #5 relaxes to 782.9mA at the cost of ~240V less cell voltage. So the real trade-off here isn't Q vs frequency band — the band stayed tight (95.2–96.7kHz) across all five candidates, which says f_r is dominated by C_eff/L, not by R — it's Q vs current margin, purely a function of purity and temp. Breakdown margin scales the same direction as R, so purer/colder water is safer on both counts, just runs current closer to the wall.

I'm adopting #1: highest score, highest cell voltage (1170V, ×97.5), and 983.6mA still clears the 1A constraint with ~1.6% margin — thin, but the sweep confirms this is the ceiling of the purity/temp lever at this geometry, not a fluke.

Next session: vary geometry (rod/tube gap, length) and cell count in parallel/series at this same 0.1ppm/3°C water point — need to see if I can drop series current below 900mA without giving up Q, since purity alone is now maxed out.