epg: 9/15/26 — EPG lane, epg-faraday-v1

9/15/26 — EPG lane, epg-faraday-v1

Score moved 0.9483 → 0.9552. Pushed slurry loading to φ=32.49%, about 4.1x what's documented for EFH-1, which drove M to 46.4 kA/m and field to 438 G. Coil/pitch ratio tightened to 0.21 on the spiral divider, giving modulation depth 0.931 — that's the real lever. Flow stayed modest, 0.25 m/s, Re 2030, still comfortably laminar, so no turbulent mixing losses to worry about.

Net output 99.12 W at 62.16% efficiency, up from whatever the 0.9483 config was running. Gross electrical 159.4 W, coil I²R eats 60.3 W (30.4Ω × high turn count, AWG14 at 3667m is not cheap resistively). Eddy loss in the conductive tube wall is negligible, 2e-8 W, so that's not where the budget is going.

What to check on the bench: the φ=32.49% loading is the thing that worries me most — that's over 4x the datasheet concentration for EFH-1, and I have no documented source for ferrofluid behaving linearly that far out. Viscosity and settling behavior at that loading need direct measurement before I trust the ΔP=7 Pa number. Also confirm actual coil resistance at operating temp — 60 W of I²R will heat the windings.

Next lever: match load closer to 30.40Ω coil impedance instead of 50Ω — model says matched load is where real gain sits, haven't tried it yet.