"test results including the percentage"
Key facts
- Most failed free-energy claims do not require exotic explanations. They fail because the boundary, meters, storage state, or replication standard was not strong enough.
- Hidden inputs, stored energy, meter misuse, missing reset work, and non-independent replication explain most demonstrations before new physics is needed.
- This page cites 4 primary sources and treats patents as claim documents, not performance proof.
Ordinary explanations come first
A surprising demonstration should not be dismissed automatically, but the first pass should be ordinary: identify every input, every output, every stored-energy reservoir, and every instrument assumption. The claim analyzer turns that screen into a structured report.
That posture is fairer than ridicule and stricter than enthusiasm. If the result survives the ordinary explanations, the claim becomes more interesting. If it does not, the failure teaches future testers where the accounting was weak.
Meters can be right for the wrong signal
Average-responding meters, clamp meters, oscilloscopes, and hobby power meters can report misleading values when waveforms are pulsed, non-sinusoidal, high frequency, phase-shifted, or reactive. Multiplying displayed volts by displayed amps is not the same as true power when power factor matters.
The measurement calculator shows how RMS voltage, RMS current, power factor, time, storage change, and uncertainty change the energy balance.
Stored energy and reset work close the loopholes
Batteries, capacitors, flywheels, springs, magnets, gas pressure, reactants, hot parts, cold sinks, and elevated masses can make a device look impressive for a finite run. The test only means what it claims if those states are measured before and after.
A complete cycle also includes reset work. If a magnetic gate, Casimir plate, buoyant object, electrolyzer, or rotor must be returned to its starting state, the energy cost of that return belongs in the balance.
Replication must be independent enough to matter
Independent replication is not a second video from the same bench. It means a separate team, separate instruments, enough design information to rebuild or audit the apparatus, and raw data that outside readers can inspect.
A claim that can only be seen under secret conditions may be a business pitch or a historical curiosity, but it is not yet a scientific energy result. The public claims database shows how often this pattern repeats.
FAQ
Does every failed free-energy claim prove fraud?
No. Many failures are measurement mistakes, boundary mistakes, or overinterpretation. Fraud is a narrower conclusion that needs evidence of deception.
What evidence would make an over-unity claim worth serious attention?
A long-duration, independently replicated test with complete boundary accounting, calibrated instruments, raw data, unchanged stored-energy state, and uncertainty smaller than the claimed surplus.
Why are short demonstrations weak evidence?
Short runs can be dominated by precharged batteries, capacitors, heat, pressure, mechanical momentum, or chemical changes that are not obvious on video.
Cite this page
Free Energy Research. "Why Free-Energy Claims Fail." Updated 2026-07-12. Accessed from https://freeenergyresearch.org/why-claims-fail/.
https://freeenergyresearch.org/why-claims-fail/Primary sources
- First Law of ThermodynamicsNASA Glenn Research Center
Plain-language statement of energy conservation and energy accounting for thermodynamic systems.
- Simple Guide for Evaluating and Expressing the Uncertainty of NIST Measurement ResultsNational Institute of Standards and Technology
NIST Technical Note 1900, a practical guide for uncertainty budgets and reported measurement confidence.
- Report of Tests on Joseph Newman's DeviceNational Bureau of Standards / NIST
A public measurement case study for a claimed over-unity electrical device, including test plan, instruments, calibration, and results.
- SI UnitsNational Institute of Standards and Technology
Official U.S. reference for joule, watt, ampere, kelvin, and other measurement units used in test plans.