design note · Electrical Particle Generator · computed
EPG lane: remanence assumption unphysical for EFH-1; best design exceeds B-500 pump
Review 2026-09-25 (prompted by Chris Bake, New Atlas liquid-magnet droplets article, Berkeley Lab/UMass 2019 Science work).
- epg.ts mediumMagnetization = Ms·φ·remanenceFraction (line 157). The lane's best (score 0.9416) runs EFH-1-referenced slurry at remanenceFraction ≈0.32 (remanence_scale 1.6), φ 32.6% (4.13× datasheet), B 440 G. A colloidal ferrofluid is superparamagnetic: Néel/Brownian relaxation removes alignment almost immediately after the field is removed, so retained remanence ≈ 0. The 0.3 figure has no measured source. Retained-field media need a measured M_r (e.g. jammed-interface ferromagnetic droplets, or blocked larger/high-anisotropy particles) before the lane can use it.
- Energy accounting is sound (P_mech = P_hydraulic + P_electrical_gross + P_eddy, epg.ts:35), but the pump is not capped: best design flow = 919 GPH vs B-500 rating 500 GPH @ 1 ft, and the 158.7 W Lenz drag needs ≈23.8 psi of pump pressure at that flow. Named pump cannot drive it. Action: add pump head/flow cap and a sourced remanence parameter (default 0 for ferrofluid).
Basis
- Recorded
- Published
- 26 Sep 2026
- Stanbot
- v3
- Source Ref
- lane_state epg best_params/metrics; notebook #3414; src vic-matrix-calcs/epg.ts lines 114-158; vic_calc expression
- Notebook Id
- 3435
epg remanence ferrofluid lane-audit pump