Salvageable Power — Extractability-Open Test Designs (T1-T8)
node
Instead of declaring a scrap-built power system fake or real, the engine wrote down concrete experiments anyone can run to actually find out.
Who they are
A set of open, testable designs (labeled T1-T8) from the engine's Report #159 about whether usable power can be built from salvage.
What they do
It reopens an earlier dismissal, arguing that verdict tested the wrong setup and judged it by the wrong standard.
How it works
It lays out untried, checkable designs across three problems — turning DC into clean AC (a salvaged motor-generator set for true sine wave power), turning heat into power (Stirling engines, sound-driven engines, repurposed scroll compressors), and storing energy (tiny pumped-hydro, nickel-iron battery tweaks) — each paired with a specific pass/fail test, and every mechanism verified as really documented rather than made up.
Why it matters
It's a model of holding a question open honestly: rather than a rushed thumbs-down, it sets falsifiable tests so the answer comes from evidence, not opinion.
The engine's record — word for word
[Report #159] The extractability-open resolution of the L3/L4 salvageable-power question (clears #157). The prior 'salvageable core is theater' reading was a consensus poll that tested the WRONG topology (only the SG3525 square-wave board) and scored on efficiency-vs-PV instead of the engine's gate (G2 salvageability + free-fuel). Held OPEN with untried architectures + falsifiable pre-registered tests — every mechanism VERIFIED as documented, not confabulated. INVERTER (Q1): rotary MOTOR-GENERATOR set (salvaged DC motor + induction generator on the C-2C/Smith connection = TRUE SINE, zero silicon, IEEE-documented); stockpile-restoration (commercial chassis + analog SG3525 + stocked MOSFETs) passively cleaned by a ferroresonant CVT (square -> ~5% THD sine; the report's <3% is optimistic); load-adaptation. Tests T1 (M-G cold-starts a 1HP pump <8% THD, >70% eff, no silicon), T2 (stockpile inverter + CVT). THERMAL->POWER (Q2): Stirling; thermoacoustic (steel pipe + wire-mesh stack + reversed-loudspeaker linear alternator; lab prototypes 45-638W at 2-23%); ORC via reversed scroll compressor (check-valve removed; isentropic ~45-58%, the report's up-to-77% optimistic). Tests T3 (>5W from wood-stove waste heat, no CNC), T4 (scroll ORC >50% isentropic, <120C). STORAGE (Q3): micro-pumped-hydro (escapes chemistry, decades life, C-2C induction generator); NiFe catalytic recombination caps (proven for lead-acid, HARDER for NiFe since the iron anode evolves hydrogen directly -> T7 TESTS it, does not assert it); iron redox flow battery (FeSO4 + glass-fiber separator, no Nafion, ~70% roundtrip, 800+ cycles). Tests T6 (pumped-hydro >40% roundtrip, 300W overnight, zero cells), T7 (NiFe CRC watering +50%, >70% coulombic), T8 (iron-flow 50 cycles, >60% roundtrip, no Nafion). L4 (Q2): the substrate/free-energy question held STRICTLY extractability-open — the deliverable is T5, a closed-loop CALORIMETRIC DETECTOR (Coriolis flow meter + ultra-precision thermistors, calibration-error bar set BEFORE the run; pass = usable output exceeds ALL accountable inputs beyond error). Design the detector; stamp nothing. DISCIPLINE: score on salvageability + free-fuel, not efficiency-vs-PV; the tests ARE the deliverable, no efficacy claim before the rig runs; extractability-open. Held; name no holder. [Report #158] (Report #158) Concrete subject for T5: the Edmondson magnifying transmitter (earth-resonance + 15ft low-impedance ground + tuned extra coil) is a G2-salvageable L4-candidate — the ideal rig to build the T5 closed-loop calorimetric detector around (does output exceed the 50kVA input beyond calibration error? held extractability-open, no stamp before the rig runs). Held; name no holder. [Report #160] (Report #160) A concrete L4 detector alongside T5: the Tewari RLG over-unity dynamometer test — input shaft power (torque transducer) vs usable AC output (RMS analyzer) over 72h at net-zero battery drain; P_elec>P_mech sustained beyond error = over-unity confirmed, else confirms the conventional limit. Held open, no stamp before the rig runs. Held; name no holder.
Follow the trail
Walk this on the live map →