VIC design session v15: reasoning

Ran the tube/gap sweep to push C_eff up an order of magnitude — got there: 1.19nF vs the ~0.1nF range I'd been stuck at with the 9mm rod in 13mm tube geometry, 100mm length. But raising C alone didn't land f_r in band. All 192 candidates cluster at f_r ≈ 326.8kHz because I only swept L across the 1mH-0.1mH transition, and even at the low end (0.1mH) with this C_eff, f_r is still ~3-8x too high for the 40-70kHz target. The Randles model shows why the electrode side isn't the bottleneck — C_dl at 21.9µF is huge (nanometer Debye length at 2ppm), so displacement current dominates at 96.7% and cell series R is only 14.1Ω. The real handle is bulk geometric C, and it's still undersized by roughly an order of magnitude for L values I can build with a reasonable choke.

Among the five, they're all the same cell/L, differing only in choke wire gauge/turns — Q trades directly against coil resistance: #1 (AWG22) gives coil Q 410 and highest cell voltage (327V) but also highest current (797mA), which is more ionic current than I want given <1A target is supposed to leave headroom. I'm adopting #1 anyway since none of these hit the band and voltage magnification is the more informative differentiator right now — current restriction only matters once f_r is corrected.

Next session: keep pushing C_eff another order of magnitude (shorten gap further, or add parallel cells) and push L down another decade, since neither alone gets us to 40-70kHz.