Electrogravitics

Electrogravitics is the claim that a strong electric field couples to gravity, so that a charged asymmetric capacitor loses weight or gains thrust — antigravity by high voltage. The effect that is actually there is different and completely conventional: a high-voltage asymmetric electrode pair in air produces a corona discharge; ions accelerated by the field collide with neutral molecules and drag them along; the reaction is a real thrust on the electrode assembly. That is electrohydrodynamic or electroaerodynamic propulsion, and it requires air.

The whole brief is the gap between those two paragraphs, and the gap has been measured four times across thirty-three years, each time better than the last — 1991, twice in 2004, and 2024. This page owns the physics verdict: what the effect is, what has been measured, at what precision, in air and in vacuum, and what the current bounds exclude.

It does not own the episode, and the seam is stated in both directions because it is a live one. History of Electrogravitics owns who spent money between roughly 1948 and 1970, Agnew Bahnson's simultaneous patronage of a real physics institute and an antigravity bench programme, the Gravity Research Foundation, the 1956 industry report, and the five-step conversion of a public funding story into a suppression narrative. None of that is re-argued here. That brief defers the physics to this one in its own text; this one returns the compliment. Where a reader arrives asking but wasn't this taken seriously by industry?, the answer is one sentence and a link: it was, publicly, and the funding history is set out there.

The structural point that makes the page worth its length is that the name covers two things, one of which is real, flying and funded. That is why the technology tree records an industrial-capacity requirement rather than declining the question, and the requirement belongs to the real technology rather than to the antigravity claim.

Established 1991, US Air Force, Phillips Laboratory. R. L. Talley of Veritay Technology, in Twenty First Century Propulsion Concept, PL-TR-91-3009 for the Air Force Systems Command, tested six electrode configurations including ball-and-plate and symmetric variants, with and without dielectrics, at voltages up to 19 kV — attempts at 50 kV failed on breakdown — at pressures down to about 5 × 10⁻⁷ torr, with a minimum detectable force of about 2 × 10⁻⁹ N. The conclusion, verbatim: “direct experimental results of this investigation indicate that no detectable propulsive force was electrostatically induced by applying a static potential difference up to 19 kV between the electrodes of test devices under conditions in which electrical breakdowns did not occur.” Speculative The report notes only that “very limited test results” from pulsed fields near breakdown “may warrant further consideration in the future”, and section 3 returns to that sentence because nothing in the record closes it.

Established 2004, the identification of the mechanism, in the AIAA Journal. Tajmar's The Biefeld–Brown Effect: Misinterpretation of Corona Wind Phenomena used cylinder and ring aluminium electrodes 6 cm apart inside a 50 cm Faraday cage with laser displacement measurement, at a force accuracy of ±10.5 micronewtons, to a maximum of 38 kV, 0.6 mA and 22.8 W. Observed thrust was below 10 micronewtons; the calculated corona-wind thrust for the same configuration is 163 millinewtons. Established The conclusion is that corona-wind effects “were misinterpreted as a connection between gravitation and electromagnetism”, and that any hypothetical anomalous force must be at least five orders of magnitude below the electric wind effects — with the historical vacuum claims assessed as “most likely a corona wind triggered by not sufficient outgassing”.

Established 2004 again, NASA Glenn, a contractor report on the device itself. Canning, Melcher and Winet's Asymmetrical Capacitors for Propulsion, NASA/CR-2004-213312, ran four asymmetrical capacitor thruster designs on a rotating arm with laser and photodiode readout. In air, Device 4 reached 138 rpm at 46 kV, and the devices “generally only rotated in one direction, the direction so that the disk was in front”. In vacuum: “no device showed signs of rotation at a pressure less than 300 Torr, with one exception” — Device 2 turned about an eighth of a revolution when a large arc occurred at 5.5 × 10⁻⁵ Torr and 44 kV, attributed to material ablation, which is to say to expelled matter. Established Device 4's measured force came out at 77% of the prediction from a plain ion-drift model, and the report concludes that operation is “fully explained by a very simple theory that uses only electrostatic forces and the transfer of momentum by multiple collisions”, with the flat statement on new physics: “we find no evidence to support such a conclusion.”

Established 2024 is the current bound and it is much wider than a capacitor test. Tajmar, Kößling and Neunzig, in Scientific Reports 14:19427, built custom horizontal, vertical and rotation balances in vacuum with laser interferometry, reaching force resolution in the nano-newton range and below against about 100 micronewtons for a standard precision balance. Devices were tested at up to 40 kV: symmetrical disc capacitors with low- and high-permittivity dielectrics, asymmetrical capacitors, gradient dielectrics, trapezoidal geometries, and for the first time solenoids — toroidal coils with low- and high-permeability cores, split cores and crossed coils producing helical fields — and tunnelling currents from Zener diodes and varistor arrays. Established The result: no anomalous forces or torques down to the nano-newton or nano-newton-metre range, “providing new limits many orders of magnitude below previous assessments”.

Established Four named theory families were tested and closed out, which is what makes that paper the strongest source on this page. Biefeld–Brown electrogravitics itself: a capacitor-generated gravitational force, ruled out for all geometries. Ivanov's “root gravity”, predicting milligram-scale weight changes: contradicted by two to three orders of magnitude. Mbelek's modified Kaluza–Klein scalar force from solenoids, claimed to be unshieldable: null for coil pairs, and tested for the first time. Modified Maxwell extensions predicting force from crossed and toroidal fields: no evidence. Weber electrodynamics predicting mass change from dipole alignment: not observed.

Established The arc is the honest headline. The bound has improved from about 2 nano-newtons of force sensitivity on one geometry in 1991 to sub-nano-newton across capacitors, solenoids and tunnelling currents in 2024, and the answer has been the same every time. Established And note who did the 2024 work. It is the Dresden group — the same laboratory that built the balances to test the EmDrive and the Mach-effect thruster, whose chair has published on Biefeld–Brown since 2004, and whose entire research line is the search for breakthrough propulsion. A group funded to find anomalous propulsion effects, reporting for the fourth time that there are none. Interest runs squarely against the finding, which is what makes it heavy.

Established Now the other half of the name, which is real and has flown. Xu, He, Strobel, Gilmore, Kelley, Hennick, Sebastian, Woolston, Perreault and Barrett at MIT flew a five-metre-wingspan fixed-wing aeroplane ten times in steady level flight, carrying all batteries and power systems including a purpose-built ultralight 40-kilovolt converter, with no moving parts. Their own framing of the contribution: “we show that conventionally accepted limitations in thrust-to-power ratio and thrust density, which were previously thought to make electroaerodynamics unfeasible as a method of aeroplane propulsion, are surmountable.” Established It is explicitly a proof of concept, and the authors say so.

Established Why it does not scale is two numbers, and both come from the developers' own data. Thrust density: the measured envelope for single-stage electroaerodynamic thrusters is 2 to 3 newtons per square metre of frontal area and 5 to 15 newtons per cubic metre, which is about 50% and 10% of the corresponding one-dimensional space-charge limits. The space-charge limit is not an engineering shortfall — it is the physics of how much charge can be pushed through a gap before the field of the charge itself opposes you. A ten-stage ducted thruster later reached up to 26 newtons per square metre, an order of magnitude over the thruster in the first electroaerodynamic aircraft, at comparable thrust-to-power. Frontier Even 26 newtons per square metre is two to three orders of magnitude below a turbofan's frontal thrust density, and that comparison is this brief's arithmetic rather than the authors'.

Frontier The second number is worse and it is about the flight envelope. Thrust-to-power decreases as both altitude and flight speed increase. Electroaerodynamic propulsion is best exactly where aircraft are least useful: slow and low. Speculative An independent group measured electrohydrodynamic propeller efficiencies of 0.64 to 6.23 newtons per kilowatt against a turbojet at 2.5 — which is the standard optimistic framing, and it is true at zero airspeed, where thrust-to-power comparisons always flatter low-momentum devices. Established Where the improvements are actually coming from is electrode geometry: an independent 2025 paper reports a serrated single-ring emitter with a multi-ring collector, 17 grams, giving 164 millinewtons per metre of emitter at 40 kV across a 60 mm gap and 118.4 at 30 kV, with double the thrust-to-power ratio of a single-ring structure and a 28.2% thrust-density improvement. Real, incremental engineering.

Established And the binding constraint on the real technology is a power converter, stated by the people who built the flight hardware. He and Perreault's converter delivers 40 kV at 565 W from 491 grams — 1.15 kW/kg at 85% efficiency, switching at about 500 kHz. Their own sentence settles the adjudication: “no EAD UAV was ever flown prior to the work by the authors and colleagues … in large part due to the weight of the required power electronics.” Established The component limit is specific: high-voltage diodes are named as the key limitation in building a high-frequency high-voltage converter, forcing low-parasitic-capacitance parts and constraining the transformer secondary to under 7.5 kV, while above 8 kV ceramic through-hole capacitors are much lighter than screw-mount ceramic, mica or film alternatives; the voltage gain is distributed 2.5 times in the inverter, 15 in the transformer and 5.6 in the multiplier specifically to minimise magnetic mass.

Handwave Finally, the live claim in this territory as of 2026, because a brief that stopped at 2024 would be incomplete. Charles Buhler — a real NASA engineer who founded the Electrostatics and Surface Physics Laboratory at Kennedy — and Exodus Propulsion Technologies claim an “electrostatic pressure force”: about 10 millinewtons from a 30 to 40 gram device of Teflon, copper tape and foam in a custom vacuum chamber, with a claimed progression to 1.0 g of thrust by November 2023 and, by 2026, thrust said to persist after the power is switched off. Roughly 2,000 test articles are claimed; the work is presented at a conference and pursued through patents; there is no peer-reviewed publication, and the stated corroboration route is independent witnesses signing affidavits to a patent examiner. Established This belongs here rather than in the reactionless brief because it is an asymmetric-electrostatic-device claim in vacuum — exactly the class Talley, Canning and Tajmar tested — and the arithmetic is stark: 10 millinewtons is seven orders of magnitude above the 2024 nano-newton bound on comparable devices. That comparison assumes the geometries are comparable, which is the claimant's implicit assertion rather than a verified fact.

Established Hypothesis one is the mainstream and it is the position of every experiment in section 2: there is no non-Einsteinian electric–gravity coupling, and the Biefeld–Brown effect is corona wind. Established Hypothesis two is general relativity's own and it is worth stating because sceptics often overstate it: electric fields do gravitate, exactly as much as their stress-energy requires and no more. The energy density in a 40 kV/cm field is around 7 joules per cubic metre, whose gravitational effect is many tens of orders of magnitude below any laboratory sensitivity. The claim under test is never that electric fields do not gravitate; it is always a proposal for an additional non-Einsteinian coupling.

Speculative Hypothesis three is the residual proponent position and it is the one genuinely open item on this page. The claim is that the historical vacuum results were real and that modern tests miss the conditions — pulsed fields, operation near breakdown, specific dielectrics. Established Talley's own 1991 conclusion is conditional: no force “under conditions in which electrical breakdowns did not occur”, with the report itself flagging very limited pulsed-near-breakdown results as warranting further consideration. Speculative Thirty-four years later, this brief could find no source that closed that specific gap. It is almost certainly nothing. It is also, strictly, untested, and a page that claimed otherwise would be overclaiming in the same way the claims it criticises do.

Handwave Hypotheses four to six are the named theory families, and they were closed in one paper in 2024. Ivanov's “root gravity” predicted milligram-scale weight changes and is contradicted by two to three orders of magnitude. Mbelek's modified Kaluza–Klein predicts an unshieldable scalar force from solenoids — tested for the first time, with coil pairs, and null. Modified Maxwell extensions and Weber electrodynamics predict anomalous forces from crossed and toroidal fields and from dipole alignment respectively, and neither appeared. Established That these were tested at all is the notable part: a null on a general hypothesis class is worth far more than a null on one famous device.

Handwave Hypothesis seven is the live one: an unidentified electrostatic pressure force at the 10-millinewton scale in vacuum. Buhler and Exodus, patents rather than papers, seven orders of magnitude above the 2024 bound. Handwave The 2026 version, in which thrust persists after the power is switched off, is a free-energy claim rather than a propulsion claim, and this brief flags it as one rather than treating it as a stronger version of the same assertion.

Frontier Hypotheses eight and nine are about the real technology, they are both true, and they are not in conflict. Electroaerodynamic propulsion has a real niche — quiet, small, low-altitude, low-speed craft — and its binding constraint is power electronics; that is the MIT group's position and their own converter numbers support it. And electroaerodynamic propulsion will never scale to useful transport aircraft, because thrust density is space-charge-limited and thrust-to-power falls with speed and altitude; that is also supported by the same group's own thesis data. Established The niche is real and the ceiling is real, and this brief declares the tie rather than picking a side.

Speculative Hypothesis ten is about why this particular claim outlived its refutation: it is a naming accident. A real effect and a false theory share a label, and the real effect keeps lending the false theory credibility — every time a solid-state aeroplane flies, somebody reads it as partial vindication of antigravity. Nobody holds this as a stated position; it is this brief's reading of the persistence, and section 14 answers the inference directly.

Handwave For the gravitic claim there is no bottleneck, because there is no effect. The instruments are four orders of magnitude better than any claim on the table, the geometries have been swept, and three of the four proposed theory families were closed by a single paper that tested them for the first time. Established Nothing about this is waiting on money, sensitivity or access.

Established For the real technology the binding constraint is high-voltage power electronics, and it is an industrial constraint in the exact sense. The physics is settled, the demonstration exists, and what is missing is a component supply chain: high-voltage, high-frequency diodes with low parasitic capacitance, and ceramic capacitors above 8 kV at aviation power-to-weight, that nobody currently builds at volume. Established The developers' own statement is that no electroaerodynamic UAV had ever flown before their work “in large part due to the weight of the required power electronics”, which names the constraint about as directly as a primary source can.

Established The second constraint on the real technology is a physical limit rather than a supply-chain one, and it is the reason the niche is a niche. Thrust density is space-charge-limited: 2 to 3 newtons per square metre single-stage, 26 with ten stages and ducting, against the one-dimensional limit that the single-stage figure already reaches half of. Frontier An order of magnitude was found by staging; a second order of magnitude has no identified route.

Frontier And the third is the flight envelope, which no component improvement addresses. Thrust-to-power falls with both speed and altitude, so the performance is best in the regime where an aircraft carries the least useful payload the least useful distance. Speculative That shape — excellent at zero airspeed, worse everywhere a mission would want — is what makes the turbojet comparison at 2.5 newtons per kilowatt misleading even though the numbers in it are correct.

Established This brief records an industrial requirement rather than a brief-to-brief edge, and the requirement belongs to the real technology rather than to the antigravity claim. That distinction is the whole adjudication: the gravitic half waits on a coupling that four independent experiments have excluded and that nothing recordable stands behind; the electroaerodynamic half waits on components.

Established The upstream physics dependencies are all satisfied and none of them is a brief. Corona discharge, ion mobility and space-charge-limited current are textbook; the 2004 report's ion-drift model predicted the measured force to within 23% without any adjustable exotic term. Frontier That is an unusual state for a contested subject: the mundane explanation is not merely available, it is quantitatively accurate.

Established Two neighbouring briefs supply context this one deliberately does not restate. History of Electrogravitics supplies the funding episode and the suppression narrative's mechanism. Gravity Modification supplies the general question of whether any field can be made to gravitate to order, including the precision null tests and the superconductor claims, which are a separate claim with a separate failure mode. Established Exotic Materials for Propulsion owns the dielectric and electrode materials, and the passive components that the converter's mass budget turns on.

Established The experiment that settled the gravitic question has been run four times and the design is worth stating as a template. Sweep the geometry rather than testing one famous device; go to vacuum and vary the pressure; instrument to well below the claimed effect; and test the theory families' distinctive predictions rather than only the headline claim. Established The 2024 campaign is the completed version of that template — capacitors, dielectrics, gradient dielectrics, trapezoidal geometries, solenoids with three core types, crossed coils and tunnelling currents, all at up to 40 kV on balances resolving below a nano-newton.

Speculative The one experiment nobody has run is the one Talley flagged in 1991: pulsed high-voltage operation near breakdown in vacuum. His null is explicitly conditional on breakdowns not occurring, his report describes the pulsed near-breakdown data as very limited and worth revisiting, and this brief could find nothing that closed it. Frontier It should be reported as an open gap rather than folded into the general null, and it is a cheap experiment by the standards of everything else in this cluster.

Established For the real technology the outstanding experiments are engineering rather than settling. Multi-stage ducted thrusters beyond ten stages; electrode geometries beyond the serrated multi-ring; converters above 1.15 kW/kg; and a flight envelope characterised at altitude and speed rather than at the bench, because the thrust-to-power scaling is the thing that decides whether the niche is a product.

Handwave And the Exodus claim specifies its own resolving experiment, which is the same one Dresden already runs. A 30 to 40 gram device claiming 10 millinewtons on a balance that resolves below a nano-newton would produce a signal seven orders of magnitude above the noise floor. Established The instrument exists, in a laboratory whose stated purpose is finding exactly this kind of effect. Affidavits to a patent examiner are not a substitute, and the absence of any journal path after ten years is itself information.

Established The engineering requirements for the real technology are unusually well specified, because the flight hardware exists and its designers published the mass budget. The converter: 40 kV output, 565 W, 491 grams, 1.15 kW/kg, 85% efficient, switching at about 500 kHz. The airframe it flew: five-metre span, all batteries and power systems aboard, ten flights in steady level flight, no moving parts.

Established The component-level requirements are the interesting part and they name specific parts. High-voltage diodes with low parasitic capacitance are the key limitation on a high-frequency high-voltage converter, which constrains the transformer secondary to under 7.5 kV and forces the gain to be distributed across three stages — 2.5 times in the inverter, 15 in the transformer, 5.6 in the multiplier — specifically to minimise magnetic mass. Frontier Above 8 kV the capacitor choice is a mass decision: ceramic through-hole parts are much lighter than screw-mount ceramic, mica or film.

Established The thruster-side requirements are geometric and are being met incrementally. Emitter and collector geometry drives both thrust density and thrust-to-power: a serrated single-ring emitter with a multi-ring collector at 17 grams gives 164 millinewtons per metre of emitter at 40 kV with a 60 mm gap and doubles the thrust-to-power of a single-ring structure. Frontier Staging and ducting bought an order of magnitude in thrust density; the next order has no identified route, and the space-charge limit says why.

Handwave And for the gravitic claim there are no engineering requirements to write, because there is nothing to engineer. Established What the subject has produced as an engineering artefact is the measurement apparatus: horizontal, vertical and rotation balances in vacuum with laser interferometry at nano-newton resolution, against about 100 micronewtons for a standard precision balance. That is a real specification and it exists because someone kept testing the claim.

Established The seam with History of Electrogravitics is the sharpest boundary on this page and it is stated in both directions. That brief owns the episode: the 1948–1970 funding history, Bahnson's dual patronage, the Institute of Field Physics, the Gravity Research Foundation, the 1956 industry report, the aviation-press coverage, and the conversion of that public history into a suppression narrative. This brief states none of it. This brief owns the verdict: what the effect is, what has been measured, at what sensitivity, and what the current bounds exclude. Established One fact crosses in each direction and only one. That brief keeps “the government paid to test it twice and published the null both times” as a fact about funding; this brief keeps the same two reports as physics, with their voltages, pressures and detection thresholds. Same documents, different claims. Established The secret-aircraft myth is that brief's, and this one answers it only upstream: there is no effect to install.

Established The seam with Electrodynamic Propulsion Concepts is the other one that has to be divided explicitly, because both briefs are drawn to the same 2018 flight. This brief owns ion wind entirely — the Biefeld–Brown physics, the flight, thrust density, thrust-to-power, the 40 kV converter and its industrial constraint. That brief owns everything that works in vacuum — ion and Hall thrusters, magnetoplasmadynamic and plasma engines, electrodynamic tethers, magnetic sails — and carries one sentence saying ion wind needs air and pointing here. Established The clean formulation for both pages: that is the vacuum half of electrodynamic propulsion; this is the atmospheric half, plus the discredited gravitic claim that shares its name.

Established Three further boundaries are narrower but worth naming. Gravity Modification owns gravity-manipulation claims in general, including the superconductor and frame-dragging claims, which are a different claim with a different failure mode and are not evidence about this one. Reactionless Propulsion owns reactionless claims as a class; the Buhler claim appears in both, and the split is that the other brief treats it as a momentum claim while this one treats it as an electrostatic-device claim measured against the 2024 bound. Frontier Mach Effect Thrusters is not adjacent: an electrostatic claim and a gravitational-inertial one are different mechanisms with different literatures, and this brief's question is closed where that one's is not.

Established The institutional record here is unusually clean and it is the answer to the “nobody looked properly” objection. Four tests, three institutions, two governments, thirty-three years. The US Air Force paid a contractor to evaluate the concept and the contractor reported no detectable force. NASA paid a contractor to assess asymmetrical capacitor propulsion and the contractor reported that a simple ion-drift theory explained it. Established In both cases the interested party — paid to evaluate a concept its sponsor was curious about — published a null.

Established The 2024 result is the strongest form the evidence can take and the reason is structural rather than technical. The Dresden chair's research line is breakthrough propulsion; the balances were built to find effects like this; the same group tested the EmDrive and the Mach-effect thruster on the same instruments; and the chair has published on Biefeld–Brown since 2004. Frontier When the people looking hardest for an effect publish the tightest bound against it, that is interest running against the finding, and it should be weighted accordingly rather than treated as one null among four.

Frontier The real technology's institutional shape is completely different and worth the contrast. It lives in a university aeronautics department and a power-electronics laboratory, publishes in Nature and IEEE journals, has independent groups doing incremental electrode work in other countries, and states its own ceiling in its own thesis data. Established Two branches of one name, one of which publishes its limits and one of which publishes affidavits.

Frontier Which brings the sharpest available institutional test, and it belongs to a sympathiser rather than a critic. The Breakthrough Propulsion Physics selection rule — “reliability of assertions is more important than the implications”, with criteria multiplied rather than added so that a zero on credibility cannot be offset by a large payoff — applied to the Exodus claim gives an immediate answer, and applied to electroaerodynamic propulsion gives the opposite one. Established That programme's own published outturn was 16 tasks with about a third not viable, a quarter with clear sequels and the rest unresolved, which is the base rate any new claim in this space should be read against.

Speculative And one institutional observation about the patent route specifically. A claim pursued through patents and affidavits rather than through journals has chosen a forum in which the standard of proof is what somebody was prepared to assert with legal consequences, not what somebody measured. That is a different thing, and a brief that treats a patent as evidence of a working device has confused the two.

Established The principal hazard is that the real technology lends the false one credibility, and this is not hypothetical. Every solid-state flight is reported somewhere as partial vindication of antigravity. The two claims share a name, an electrode geometry and a voltage, and differ in the only respect that matters. Frontier Naming both clearly, in the same brief, is the only response that works — a page that covered only the refutation would leave the strongest evidence for the mundane explanation off the page.

Established Second, and specifically about overclaiming from the sceptical side. Talley's null is conditional on no breakdown, and his flagged pulsed-near-breakdown case appears never to have been closed. Established Reporting the record as though every configuration had been tested would be the same failure mode as the claims this brief criticises, on a smaller scale and in the more comfortable direction, and the honest posture is to state the gap and note that it is almost certainly nothing.

Handwave Third, the investment hazard is live in 2026 rather than historical. A claim of 10 millinewtons from a 40 gram device, from a founder with a genuine NASA laboratory pedigree, pursued through patents with affidavits as corroboration, is exactly the configuration that attracts money from people who cannot evaluate it. Established The relevant instrument exists, in a laboratory that wants the effect to be real, and the claim has not been taken to it.

Frontier And fourth, the aircraft myth has a specific cost that is not epistemic. It attaches a real defence programme to a nonexistent physics, which makes every subsequent conversation about that programme harder. The physics answer is upstream of the aircraft question and short: there is no effect to install; the narrative's mechanism belongs to History of Electrogravitics.

Established The realistic civilizational payoff is a niche, and stating it as a niche is the honest thing to do. Quiet, small, low-altitude, low-speed craft with no moving parts — sensing, monitoring, indoor and near-ground flight where noise and mechanical failure matter more than range. That is a real product category and it is where the space-charge limit and the thrust-to-power scaling both point.

Frontier The transport-aircraft version is not coming, and the reason is a physical limit rather than an engineering backlog. Two to three newtons per square metre of frontal area, an order of magnitude better with staging, still two to three orders below a turbofan, with thrust-to-power falling as you go faster and higher. Established A brief that promised electroaerodynamic airliners would be contradicting the developers' own thesis data.

Handwave The antigravity version, if it were real, would be the largest technological change in history — and that is precisely why the claim persists through four escalating nulls. Established The payoff is enormous, the evidence is a bound that keeps getting tighter, and the gap between those two facts is what keeps the subject alive rather than any measurement.

Frontier And the durable contribution of the whole episode is a template. A real effect was misattributed for decades, correctly identified by a corona-wind calculation, quantified to within 23% by a plain ion-drift model, engineered into a flying aircraft, and then bounded across an entire hypothesis class on instruments a thousand times better than the claim. Speculative The best answer to “electrogravitics” is not that there is nothing there. It is that there is something there, it has been identified, and it is on its third generation of hardware.

These horizons split, because the two things sharing this name are at completely different stages:

• 10 yr: Frontier On the real side, expect incremental electrode and staging work, converters above 1.15 kW/kg, and possibly a small commercial quiet-flight product; the enabling constraint is component supply rather than physics. Established On the gravitic side, expect nothing: the 2024 bound covers capacitors, solenoids and tunnelling currents at nano-newton resolution and nothing on the horizon improves it by an order of magnitude. Speculative The one item genuinely worth watching is whether anybody runs Talley's flagged pulsed-near-breakdown case, which is cheap and has been open for thirty-four years.
• 25 yr: Frontier Electroaerodynamic propulsion either becomes a real niche product line or remains a laboratory demonstration, and the discriminator is high-voltage power electronics at aviation power-to-weight rather than any thruster advance. Handwave Electrogravitic antigravity has no horizon at any scale, because there is no coupling and four independent campaigns have said so at improving precision. Speculative The claim itself will very likely still be circulating, since its persistence is a naming accident rather than an evidential dispute.
• 50 yr: Speculative The space-charge limit does not move, so the ceiling on thrust density is structural; what could change is the aircraft class rather than the physics, if very small quiet platforms become important for reasons unrelated to propulsion. Established Nothing in the gravitic branch is contingent on time. It is contingent on a coupling whose existence has been excluded, geometry by geometry and theory family by theory family.
• Any horizon: Handwave Electrogravitic antigravity: not on any timeline, absent an electric–gravity coupling that four independent experiments have now bounded below a nano-newton. Established The defensible statement is not that it is impossible but that every specific proposal — Biefeld–Brown, Ivanov, Mbelek, modified Maxwell, Weber — has been tested and excluded at a stated sensitivity, by a group that wanted at least one of them to work.

• Depends on The two halves of this name wait on completely different things, and only one of them waits on anything recordable. The gravitic claim waits on an electric–gravity coupling that does not exist: four independent campaigns from 1991 to 2024 have bounded it from about 2 nano-newtons on one geometry to sub-nano-newton across capacitors, solenoids and tunnelling currents at up to 40 kV, closing out the Biefeld–Brown, Ivanov, Mbelek modified Kaluza–Klein, modified-Maxwell and Weber families. No brief on this map produces what that half would need, and no typed edge is recorded for it. The real technology sharing the name waits on components, which is what the requirement below records. Adjacent physics is all satisfied: corona discharge, ion mobility and space-charge-limited current are textbook, and a plain ion-drift model predicted the measured force to within 23% in NASA's own 2004 assessment.
• Requires (not on this map) This token belongs to the real technology sharing this brief's name, not to the antigravity claim — the antigravity claim waits on a coupling that has been bounded below a nano-newton and nothing recordable stands behind it. The evidence is direct and comes from the people who built the flight hardware. He and Perreault's purpose-built converter delivers 40 kV at 565 W from 491 grams — 1.15 kW/kg at 85% efficiency, switching at about 500 kHz, and their own paper states that “no EAD UAV was ever flown prior to the work by the authors and colleagues … in large part due to the weight of the required power electronics.” The component limit is named: high-voltage diodes are the key limitation in a high-frequency high-voltage converter, forcing low-parasitic-capacitance parts and constraining the transformer secondary to under 7.5 kV, while above 8 kV ceramic through-hole capacitors are much lighter than the screw-mount ceramic, mica or film alternatives. That is an industrial-capacity constraint in the exact sense: the physics is settled, the demonstration exists — a five-metre aeroplane flew ten times on 40 kV with no moving parts — and what is missing is a supply chain for high-voltage, high-frequency diodes and capacitors at aviation power-to-weight that nobody currently builds at volume.
• Enables The real half enables a specific and modest class: quiet, small, low-altitude, low-speed craft with no moving parts, where noise and mechanical reliability matter more than range. It does not enable transport aircraft, on the developers' own data — thrust density is space-charge-limited at 2 to 3 newtons per square metre single-stage and 26 with ten stages and ducting, and thrust-to-power falls with both speed and altitude. The gravitic half enables nothing, and no edge is claimed for it. What the whole episode has enabled is a measurement capability: vacuum balances at nano-newton resolution with laser interferometry, against about 100 micronewtons for a standard precision balance.
• Adjacent Corona and plasma physics, high-voltage engineering and power electronics supply the mechanism and the constraint. Within this map: History of Electrogravitics owns the 1948–1970 funding episode and the suppression narrative and is not re-argued here; Electrodynamic Propulsion Concepts owns everything in this family that works in vacuum, while this brief owns the atmospheric member entirely; Gravity Modification owns gravity-manipulation claims in general, including the superconductor claims that are a separate failure mode; Reactionless Propulsion shares the Exodus claim and treats it as a momentum claim where this brief treats it as an electrostatic-device claim; and Exotic Materials for Propulsion owns the dielectrics, electrodes and passive components the converter's mass budget turns on.

Established “A charged capacitor loses weight.” In air it moves, because it makes wind. In vacuum it does not: not at 19 kV in 1991 down to 5 × 10⁻⁷ torr with a 2-nano-newton detection threshold, not at 44 kV on a rotating arm in 2004, and not at 40 kV on a nano-newton balance in 2024. Established Three governments-worth of instruments, thirty-three years, one answer.

Established “The effect was demonstrated in vacuum.” The 1991 Air Force test found nothing at about 5 × 10⁻⁷ torr. The 2004 assessment concludes the historical vacuum result was corona wind from insufficient outgassing. And the single vacuum motion in NASA's 2004 test — about an eighth of a revolution — occurred during a large arc and was attributed to material ablation, which is to say to expelled mass. Established The one apparent vacuum success in the entire record is a rocket.

Handwave “Ion wind is antigravity's respectable cousin, so the claim is half-vindicated.” No, and this is the inference the whole page exists to block. Ion wind is a reaction against expelled air. Established It is the refutation of the gravitic reading, not a softened version of it — the 2018 flight is evidence against electrogravitics, because it demonstrates that the whole observed effect has a mundane explanation that is now fully engineered and flying.

Established “Nobody has looked properly.” Four tests, three institutions, two governments, thirty-three years, and a sensitivity improvement of several orders of magnitude. Established The 2024 campaign did not test one device; it tested a hypothesis class — capacitors with low- and high-permittivity dielectrics, gradient dielectrics, trapezoidal geometries, solenoids with low- and high-permeability and split cores, crossed coils producing helical fields, and tunnelling currents from Zener diodes and varistor arrays — and closed out four named theory families at once.

Established “Electric fields do not gravitate at all.” They do. General relativity couples gravity to the stress-energy tensor and an electromagnetic field carries stress-energy, so there is a coupling — it is just fantastically weak, with a 40 kV/cm field carrying around 7 joules per cubic metre whose gravitational effect is tens of orders of magnitude below any laboratory sensitivity. Established The correct sceptical sentence is that fields gravitate exactly as much as their energy says and no more, and every electrogravitics theory is a proposal for an additional, non-Einsteinian coupling.

Handwave “The B-2, or some black programme, uses electrogravitics.” The physics answer is upstream of the aircraft question and it is one sentence: there is no effect to install. Established The narrative's origins, its named authors and the mechanism by which a public funding history became a suppression story belong to History of Electrogravitics, and this brief hands it over rather than arguing it.

Handwave “Exodus has found the effect that Brown was looking for.” The claim is about 10 millinewtons from a 30 to 40 gram device in vacuum, which is the exact device class that Talley, Canning and Tajmar tested. Established 10 millinewtons is seven orders of magnitude above the 2024 nano-newton bound on comparable geometries — that is arithmetic rather than opinion, and it assumes only what the claimant implicitly asserts, that the geometries are comparable. Handwave The 2026 version, in which thrust persists after power is removed, is a free-energy claim; and affidavits to a patent examiner are not replication.

Speculative “Every configuration has been ruled out.” Not quite, and this brief will not overclaim in its own favour. Talley's 1991 conclusion is explicitly conditional — no force “under conditions in which electrical breakdowns did not occur” — and his report flags very limited pulsed-near-breakdown results as warranting further consideration. Established No source this brief could obtain has closed that specific gap in thirty-four years. It is almost certainly nothing; it is also, strictly, untested, and it is reported here as an open item rather than folded into the general null.

Established “Electroaerodynamic propulsion will scale to real aircraft.” Not on the developers' own numbers. Thrust density is space-charge-limited at 2 to 3 newtons per square metre single-stage — already about half the one-dimensional limit — reaching 26 with ten stages and ducting, which is still two to three orders below a turbofan; and thrust-to-power falls with both speed and altitude. Frontier The niche is real and the ceiling is real, and both are supported by the same thesis.

Established And the framing itself. “Electrical fields can be coupled to gravity” fails, and it fails in the strongest available way: refuted at four escalating levels of precision by people who wanted it to be true, most recently across capacitors, solenoids and tunnelling currents at up to 40 kV on balances resolving below a nano-newton, with Ivanov's prediction excluded by two to three orders of magnitude. Frontier And the mundane explanation is not a consolation prize. A five-metre aeroplane flew ten times on 40 kV with no moving parts, and its performance is now being improved by ordinary electrode-geometry engineering. That is the brief's real content, and it is why this page is long rather than short.