What “Free Energy” Claims Usually Mean (and Why the Phrase Is a Trap)
In maker forums and product pitches, “free energy” is often used to mean “a device that produces useful power without a fuel bill.” In thermodynamics, the word “free” has a very specific meaning (for example, a measure of how much energy is available to do useful work under defined conditions). In scams and misconceptions, “free” usually means “unpaid for,” “unlimited,” or “from nothing.” Those are very different ideas.
When you evaluate a claim, translate the marketing language into a concrete statement you can test: What is the device’s claimed output power (watts)? For how long? Under what operating conditions? What are all the inputs (electrical, chemical, thermal, mechanical, environmental)? If the claim avoids numbers, that is already a warning sign.
Many impossible claims rely on one of these moves: (1) hiding an input, (2) confusing energy with power, (3) confusing peak output with average output, (4) counting stored energy as “generated,” (5) using measurement errors or instrument limitations, or (6) mixing up heat flow direction and the need for a temperature difference.
Two Buckets of Impossible Machines: First-Law vs Second-Law Violations
Bucket A: “Overunity” (First-Law Violation)
These claims say the device outputs more energy than it receives, over a complete operating cycle, without consuming stored energy. Typical phrases: “over 100% efficient,” “self-charging generator,” “runs forever once started,” “outputs 2 kW from a 200 W input.” If the device is not drawing from a fuel, battery, spring, elevated weight, compressed gas, or some environmental gradient, then the claim implies energy creation. That is the core error.
Bucket B: “Perfect conversion” or “single-reservoir heat engine” (Second-Law Violation)
These claims may not explicitly say “more energy out than in.” Instead they promise converting ambient heat into work with no other effect, or achieving 100% conversion of heat to work, or moving heat from cold to hot with no input. Typical phrases: “extracts power from room temperature,” “runs on the heat in the air,” “cooling without power,” “turns waste heat into electricity at 100%.” The hidden issue is that even if energy is conserved, the direction and quality of energy matter. Work extraction from heat requires a temperature difference; moving heat “uphill” requires work input.
Continue in our app.
You can listen to the audiobook with the screen off, receive a free certificate for this course, and also have access to 5,000 other free online courses.
Or continue reading below...
Download the app
Common Misconceptions Makers Encounter (and How to Spot Them)
Misconception 1: “Magnets provide energy”
Permanent magnets can exert forces and enable energy conversion, but they are not a fuel. A magnet can store some energy in its magnetic field and in its magnetization state, but in normal use it does not supply continuous net energy. If a device claims to “run on magnets,” ask: what is the energy source? If the device cycles mechanical motion, any net output must come from an input: a battery, gravity, a wound spring, compressed air, or an external drive. Magnets can reduce friction in bearings or shape forces, but they do not create net work per cycle in a closed system.
Practical check: if the device is a rotor with magnets, measure torque vs angle over a full rotation. If you integrate torque over one full cycle and get net positive work, you should be able to reproduce it with a simple dynamometer and see continuous acceleration under load. In practice, you will find regions of assistance and regions of opposition that cancel, plus losses.
Misconception 2: “Resonance multiplies energy”
Resonance can produce large amplitudes with small driving forces, but it does not create energy. It stores energy alternately in different forms (for example, kinetic and elastic) and can reduce the required input power for a given amplitude when losses are low. If someone shows a resonant system with big motion and claims “free power,” ask for the input power measurement and the output power under load. The moment you extract significant power, the resonance amplitude drops unless the input power increases.
Misconception 3: “High voltage, low current means low power”
Power is not judged by voltage or current alone. A device can be dangerous and deliver significant power at high voltage and low current, or at low voltage and high current. Many demonstrations use meters incorrectly (for example, measuring peak voltage on a pulsed waveform and multiplying by average current). Always compute real power with correct instruments and methods for the waveform type.
Misconception 4: “Batteries recharge themselves during operation”
Some circuits can appear to “charge” a battery due to measurement artifacts, recovery effects, or because the battery voltage rises when load is removed. A battery’s terminal voltage is not a direct measure of stored energy. If a claim says “the battery recharges while powering the load,” demand a full energy accounting: watt-hours out to the load versus watt-hours removed from the battery, including any external inputs (mains ground, RF pickup, solar, thermal gradients, etc.).
Misconception 5: “Ambient heat is an unlimited fuel”
Ambient heat is energy, but at uniform temperature it is not readily convertible to work without a colder sink. If the device is in a room at steady temperature and claims to produce continuous work from “heat in the air,” it is claiming a single-temperature heat engine. In reality, to extract work from heat you need heat to flow from hot to cold. If there is no maintained temperature difference, there is no sustained heat flow to tap.
A Practical Evaluation Workflow for Suspect Claims
Use this workflow whenever you see a “free energy” device, a “self-running generator,” or a “COP infinity” heat gadget. The goal is not to argue; it is to force the claim into measurable terms and close the common loopholes.
Step 1: Define the system boundary and list every possible input
Draw a boundary around the device and include everything physically connected: wires, hoses, shafts, mounts, and even the table if it could transmit vibration. List inputs: electrical (mains, battery, ground paths), mechanical (hand cranks, vibration), thermal (hot/cold sources), chemical (fuel, reactive metals), pressure (compressed air), radiation (sunlight, RF), and gravitational potential (raised weights).
- If there is a wire, assume power can flow through it until proven otherwise.
- If there is a metal chassis, consider unintended return paths through ground.
- If there is a fluid line, consider enthalpy flow (hot fluid in/out) even if it “feels” small.
Step 2: Convert the claim into power and energy numbers
Ask for output power (W) and duration (s or h). Energy output is power multiplied by time. If the claim is “it runs a 100 W bulb,” verify whether it is actually 100 W at rated voltage, or a dim glow at lower power. If the claim is “it charges a phone,” measure watt-hours delivered to the phone battery, not just that the phone icon shows charging.
Step 3: Measure output power correctly (especially for AC, pulsed, or non-sinusoidal waveforms)
Many “overunity” demonstrations rely on incorrect electrical measurements. For DC, power is P = V × I, but you must measure both at the same time under load. For AC or pulsed signals, you need real power measurement (true RMS voltage and current plus power factor, or a wattmeter/power analyzer). A multimeter reading “RMS” may not be true RMS, and an oscilloscope reading peak voltage is not RMS.
- Use a resistive load you can characterize (power resistor, heater element) and measure voltage across it and current through it.
- For pulsed systems, record waveforms and compute average of instantaneous power p(t) = v(t) × i(t).
- Beware of reactive power: high current can circulate without net energy transfer.
Step 4: Measure input power with the same rigor
Measure the input at the source: wall outlet with a true power meter; battery with coulomb counting or watt-hour logging; mechanical input with torque and RPM; compressed air with pressure, flow, and temperature. If the device claims “no input,” verify there is no hidden battery, no inductive pickup, no ground current, and no thermal gradient being exploited.
Step 5: Run a time test long enough to exceed plausible stored energy
Many devices can output impressive power briefly by draining stored energy (capacitors, flywheels, batteries, compressed gas). A short demo is not evidence of generation. Estimate maximum stored energy and require a run time that would exceed it by a large factor.
- Capacitors: even large capacitor banks usually store modest energy relative to minutes of kilowatt output.
- Flywheels: energy scales with rotational speed squared; if RPM drops noticeably, you are spending stored energy.
- Batteries: measure mass/volume; if the claimed output is large, the battery would drain quickly unless it is huge.
Step 6: Check for “environmental inputs” that are real but misrepresented
Some “free energy” devices are actually harvesting energy from the environment: solar panels hidden in lighting, vibration harvesters on machinery, thermoelectric generators across a hot pipe, or RF energy harvesters near transmitters. These can be legitimate, but they are not “from nothing.” If the device is near a window, a hot motor, a radiator, or a strong RF source, treat that as an input and quantify it.
Measurement Traps That Create Fake Overunity
Trap 1: Confusing peak, RMS, and average
A pulsed waveform can have a high peak voltage but low average power. If someone multiplies peak voltage by average current, they can get a number that looks impressive but is not real power. The correct approach is to compute average of v(t)i(t) over time or use a power analyzer designed for non-sinusoidal signals.
Trap 2: Using the wrong meter bandwidth or sampling
Cheap meters can misread high-frequency or pulsed currents. A clamp meter may be inaccurate on DC or on non-sinusoidal waveforms. An oscilloscope probe may load the circuit or miss common-mode currents. If the device uses switching electronics, assume measurement complexity and demand instruments appropriate to the frequency content.
Trap 3: Hidden ground return paths
A device can draw power through an unexpected path: earth ground, shield braid, water pipes, or another connected instrument. If the demo uses multiple instruments connected to mains earth, you can accidentally create a power path that bypasses the measured input. Isolation transformers, battery-powered instruments, and careful single-point grounding help reveal this.
Trap 4: Battery “recovery” and voltage rebound
After a load is removed, a battery’s terminal voltage can rise due to chemistry and internal resistance effects. This can look like “it recharged,” especially if the demo only shows voltage. Energy is what matters: integrate current over time and track watt-hours.
Trap 5: Thermal lag and “mystery cooling/heating”
A device may appear to cool or heat without input because of thermal mass and time delay. For example, a cold plate warms slowly and can be misinterpreted as “absorbing heat and making power.” Use calorimetry or steady-state measurements: wait until temperatures stabilize and measure continuous power flows.
Red-Flag Claims and the Specific Error Behind Each
“100% efficient generator”
Error: ignores losses and implies no heat generation. Real generators have copper losses, core losses, friction, windage, and power electronics losses. If someone claims 100%, ask for a heat measurement at rated load and a full efficiency curve.
“Motor runs cooler under load”
Error: mismeasurement or changed operating point. A motor can run cooler at a different speed or with better airflow, but adding load generally increases losses for a given supply. Verify with controlled speed, controlled airflow, and measured electrical input.
“Self-running: output powers the input”
Error: missing energy storage or hidden input. A looped system can run briefly on stored energy. Require a long-duration test with sealed enclosure, independent measurement of all power paths, and a load that removes significant energy continuously.
“Extracts energy from the vacuum / zero-point energy”
Error: uses a real physics term to imply an engineering power source. Even if vacuum fluctuations exist, turning them into macroscopic continuous power is not something a bench-top device can do without an external gradient or resource. Treat it like any other claim: show measured input/output energy with controlled conditions and independent replication.
“Runs on ambient temperature difference” (but shows no difference)
Error: claims a gradient without providing one. If both sides of a thermoelectric module are at the same temperature, output must be near zero. If it outputs power, there is a hidden gradient (hot pipe, sunlight, airflow). Measure temperatures at both sides with good contact and insulation to confirm.
Step-by-Step: How to Audit a “Free Energy” Demo on Your Bench
1) Recreate the setup with a known load
Replace vague loads (LED strips, phone chargers, “mystery boxes”) with a resistive load you can measure: a power resistor bank or incandescent lamp with known behavior. Measure voltage across the load and current through it.
2) Instrument the input with isolation and logging
If the input is mains, use a plug-in power meter that reads real power (W) and energy (Wh). If the input is a battery, use a watt-hour meter inline. If the claim is “no input,” power the measurement instruments from batteries and keep the device physically isolated from building ground where possible.
3) Do an energy balance over time
Log input energy (Wh in) and output energy (Wh out) over a meaningful duration. Short bursts are not enough. Choose a duration so that output energy is far larger than any plausible hidden storage (for example, run long enough that even a large hidden battery would be obvious by mass/size).
4) Control for environmental energy
Move the device away from sunlight, hot equipment, and strong RF sources. If it claims to use ambient heat, place it in a well-mixed environment and measure any temperature differences it relies on. If it uses vibration, place it on a heavy, damped surface and see if output changes.
5) Seal and inspect
For a serious audit, seal the device enclosure, weigh it before and after, and inspect for batteries, supercapacitors, chemical reactants, compressed gas cartridges, or hidden wiring. Weight change can reveal consumed fuel or evaporated coolant. This is not about distrust; it is about closing the common loopholes that make demos misleading.
How Legitimate Technologies Get Misrepresented as “Free Energy”
Energy harvesting (real, but limited)
Small devices can harvest energy from light, vibration, temperature gradients, or RF. These are useful for sensors and low-power electronics. The misconception is scaling: harvesting milliwatts does not become kilowatts by clever circuitry. If someone claims household-scale power from a tiny harvester, ask for the available environmental power density and the collection area/gradient.
Heat recovery (real, but not magic)
Devices can recover some useful energy from waste heat when there is a temperature difference and a suitable converter. The misconception is assuming “waste heat” is automatically convertible to large amounts of work. In practice, the available fraction depends strongly on temperatures and losses. If a claim says “turns exhaust heat into electricity with huge output,” demand measured heat flow, temperatures, and electrical output under steady conditions.
Regenerative braking and energy recapture (real, but not perpetual)
Regeneration can return some energy to a battery when slowing down. The misconception is thinking you can loop it to run forever. Regeneration reduces net losses; it does not eliminate them. Any closed loop that includes real friction, electrical resistance, and switching losses will decay without an external energy source.
Quick Checklist: Questions That Collapse Most Impossible Claims
- What is the continuous output power in watts, and for how many hours?
- What are all inputs across the boundary (wires, shafts, fluids, heat sources, radiation)?
- Is the measurement real power (average v(t)i(t)) or a misleading combination of peak/RMS/average?
- Does the run time exceed plausible stored energy by a large factor?
- Can the device be tested in a controlled environment with isolation from ground and external sources?
- Can an independent person reproduce the measurements with their own instruments and loads?
Worked Example: “Overunity Inverter” Demonstration Audit
Suppose someone shows an inverter powering a 60 W lamp while the input meter shows only 10 W from a battery. Here is how you would dissect it.
Step 1: Verify the lamp power
Replace the lamp with a resistive load and measure voltage and current at the output. If it is AC, use a true power meter. Many inverters output a modified waveform; a cheap meter can misread it. If the lamp is dim, it may be consuming far less than 60 W.
Step 2: Verify battery input energy
Use a watt-hour meter inline with the battery. Log Wh over time. If the inverter draws pulsed current, a simple ammeter may under-read. Ensure the meter bandwidth is adequate or use a shunt and scope to compute average power.
Step 3: Eliminate alternate inputs
Check for any connection to mains earth or hidden charging path. If the inverter chassis is grounded through test equipment, it might be receiving power through that route. Run the entire setup isolated: battery-powered instruments, no earth-referenced scope unless isolated, and no other cables.
Step 4: Time test
Run long enough that, if the lamp were truly 60 W, the energy would be obvious. For example, 1 hour at 60 W is 60 Wh. If the battery only “supplied” 10 Wh by your measurement, but the load truly received 60 Wh, you have either a measurement error or a hidden input. In real audits, the mismatch almost always disappears when instruments and loads are corrected.
Language Patterns That Signal You Should Demand Better Evidence
- “It’s too advanced to explain” (but no data is provided).
- “Mainstream science doesn’t want you to know” (instead of showing repeatable measurements).
- “It works, but only when observed by the inventor” (lack of reproducibility).
- “The meter can’t measure it because it’s special energy” (measurement avoidance).
- “Efficiency over 100% because of resonance/quantum/magnets” (category error).
Practical Maker Takeaways: How to Stay Curious Without Getting Fooled
You can be open to unusual designs while still applying disciplined testing. The key habit is to treat every impressive demo as a measurement problem: define the boundary, measure real power in and out, run long enough to beat stored energy, and control environmental inputs. When a claim survives those steps, it stops being a “free energy” story and becomes a real engineering result that others can replicate and build on.
Bench audit minimum kit (typical): 1) True power meter (mains) 2) DC watt-hour meter (battery) 3) Known resistive loads 4) Shunt resistor + scope (for pulsed power) 5) Thermometer/IR camera for loss sanity checks 6) Isolation transformer or battery-powered instruments to avoid ground-path tricks
