The Podkletnov
Replication Trail
Bilayer YBCO and the experiment nobody funded
A thirty-three year argument over whether spinning a supercooled ceramic disc can modify gravity. Most replication attempts used the wrong geometry. The one that didn't came up null. The clean test that would actually settle the question hasn't been run.
The desktop-scale version of the impossible
A Boeing 777 with 239 people aboard vanished from every tracking system on Earth on March 8, 2014. In the alleged surveillance footage that surfaced years later, three luminous orbs hold a coordinated triangular formation around the aircraft for about a minute, then a sudden terminal event, and the plane is gone. No wreckage trail. No distress signal. Nothing in publicly known physics that explains it.
The 4Orbs investigation argues those orbs are plasma reactors built around fast-rotating superconducting components. The terminal event in the footage shows them converge on the Boeing 777, and the aircraft physically disappear from the sky mid-flight, leaving no wreckage trail across the Indian Ocean. Doing that would require fields capable of displacing 200 tons of matter in seconds. That's a strong claim.
If it's right, though, the same physics has to work on a bench as it does in the sky. The orbs are converging on a 200-ton object and removing it from where it is. Mass is what gravity acts on, and the gravitational pull on that mass is what holds the aircraft to the Earth. So whatever the orbs are doing has to be acting on gravity, on mass, or on the relationship between the two. Spin a much smaller version of the same superconducting setup on a laboratory table, and you should see some small but measurable change in the weight of nearby objects.
In 1992, a Russian materials scientist named Eugene Podkletnov claimed he'd found it. He cooled a flat ceramic disc to roughly liquid-nitrogen temperatures, which turned it into a superconductor. He spun the disc at high speed using magnets. Then he placed test objects on a balance directly above the spinning disc, and they got lighter. The effect was small but consistent: between 0.05% and 0.3% weight reduction, depending on rotation speed. Five years later, in a follow-up paper, he claimed up to 2%. The scientific establishment has been arguing about whether the effect is real for thirty-three years.
A young American chemist named Amy Eskridge spent her career trying to break that argument open. She had a specific technical claim about why most attempted replications had failed, and she had it on tape three months before her death in 2022. The claim is still testable. The experiment that would settle it has still not been done.
This is what suppression of advanced propulsion research looks like in 2026: an experiment nobody has funded.
What Podkletnov actually claimed
Eugene Podkletnov was a doctoral candidate at Tampere University of Technology in Finland when he and Risto Nieminen published their paper in Physica C in 1992 (volume 203, pages 441–444). The setup was simple enough to describe in a sentence: a 14.5-centimetre disc of YBCO (a ceramic that becomes a superconductor when cold enough), levitated above a magnetic coil, spun at high speed by a rotating magnetic field. Test bodies of various materials placed above the spinning disc lost a small but consistent fraction of their weight, depending on rotation speed.
The paper passed peer review. That's the foundation of the whole controversy.
Then 1996 turned ugly. Podkletnov submitted a follow-up paper to Journal of Physics D claiming a much larger effect, up to 2.1% weight reduction over a 27-centimetre disc of a different design. The Sunday Telegraph got hold of it and ran a front-page story headlined "Anti-gravity device gives science a lift" on September 1, 1996. Within days, his co-author Petri Vuorinen disavowed prior knowledge of the paper, claiming his name had been added without his consent. Tampere's laboratory director said Podkletnov had been working entirely on his own. Podkletnov withdrew the paper after it had already been accepted. The university expelled him in 1997. The trigger was the publicity around the withdrawn paper. No university body issued any finding that the experiments had been falsified, and the 1992 paper, which had passed peer review, was never retracted.
So the bigger-effect paper that anchors most of the modern debate has never been formally published. It exists as a preprint, and that preprint is the protocol that subsequent replication attempts have tried to follow, with mixed and largely null results. The 1992 paper had a smaller effect on a simpler disc. The 1997 preprint had a larger effect on a more elaborate disc. They describe two different physical systems, and most of the field has treated them as a single claim.
That collapse is where the trouble starts.
The geometry that matters
The 1997 disc had a structure the 1992 disc didn't. It was a Bilayer: two ceramic layers fused together, with the lower layer in the superconducting state at 70 K and the upper layer composition-matched but a normal conductor at the same temperature. Two regimes of matter, joined at an interface.
Why does that matter? Because most of the strange and useful behaviour in condensed-matter physics happens at interfaces, not in bulk. Solar cells live on the boundary between two doped semiconductors. The transistor is a sandwich. Josephson Junctions are sandwiches. If a gravity-modifying effect existed in a rotating superconductor, the physical case for expecting it at the boundary between superconducting and normal-conducting layers is at least defensible. It's the kind of argument a competent solid-state physicist would make.
So when subsequent replication attempts used the simpler 1992-style disc and reported a null, the geometry of the test was already different from the geometry that produced the larger claimed effect. That gap is what Eskridge's argument turns on.
The replication trail
Replications fall into two camps: people who tried to follow the 1997 protocol exactly, and people who built adjacent setups and called the null result a refutation.
| Year | Team | Disc geometry | Result | Matches 1997 protocol? |
|---|---|---|---|---|
| 1992 | Podkletnov & Nieminen (Tampere) | 14.5 cm single-layer YBCO | Positive (0.05–0.3%) | Pre-protocol |
| 1997 | Podkletnov (preprint, withdrawn) | 27 cm bilayer YBCO | Positive (up to 2.1%) | Yes (defines the protocol) |
| 1997 | Li, Lucas, Hassen, Levi & Torr (UAH/NASA) | Small commercial single-layer disc | Null | No |
| 1999–01 | NASA Marshall (Koczor & Noever) | 30 cm single-layer YBCO | Null | No |
| 2001–02 | NASA Marshall + Podkletnov (consultant) | Bilayer to 1997 spec | Null | Yes |
| ~2002 | University of Toronto (Podkletnov consulting) | YBCO disc, geometry incomplete | Null | Partial |
| 2002 | Boeing GRASP / Phantom Works | Not publicly disclosed | Unpublished | Unknown |
| 2003 | Hathaway, Cleveland & Bao | Custom YBCO + RF field | Null | Partial |
| 2006–07 | Tajmar et al. (ARC / ESA) | Niobium & YBCO rings, not discs | Different effect (GMLM) | No |
| 2022 | Fetisov (preprint, arXiv 2209.03332) | Dispersed YBCO + RF, room temp | Positive | No (different methodology) |
Most "replication failures" tested the simpler 1992 disc. The one strong test of the bilayer geometry happened at NASA Marshall, with Podkletnov in the room, using the geometry he specified. It still came up null. The 1997 protocol has been tested rigorously by an independent team exactly once.
That single bilayer test is a serious problem for the bilayer-explains-everything position. It doesn't fully kill the claim, since parameter space is wide and Marshall did only one bilayer test, but it's a load-bearing data point that any honest treatment has to include.
The Boeing GRASP project is the wildcard. Boeing officially denied funding the project with company money but acknowledged it could not comment on classified work. No public results were ever published. The silence is consistent with either a quiet null or a classified positive result, and from the outside we don't know which.
What does the trail tell you? That most "replication failures" of Podkletnov used the simpler 1992 disc, that the one strong test of the bilayer geometry came up null at NASA Marshall, and that the field has not been allowed to converge on a clean answer either way.
The Estes Park confrontation
In December 2018, at a small breakthrough propulsion workshop in Estes Park, Colorado, Amy Eskridge walked up to Martin Tajmar at the buffet table with a stack of his own papers in her backpack. She introduced herself, told him she'd been reading his work since high school, and opened one of the printouts to the methods section.
In her own words, from a 2020 interview with Jeremy Rys and Mark Sokol:
Eugene Podkletnov uses a bilayer of YBCO. We had Martin's paper right there in our backpack. Right there in the protocol it says you used a monolayer, not a bilayer, and you reported no results. How do you expect to get the same result not a bilayer like Eugene did?
The most interesting phenomena in the world happen across gradients, across junctures of delta differences. If you use a monolayer and not a bilayer, your results might be different, and therefore your null results using a YBCO monolayer are in fact irrelevant to Eugene's results using a YBCO bilayer.
According to Eskridge, Tajmar's response was to invite her to send samples for testing.
Then she went further. In the same interview she alleged that "Martin Tajmar is funded by the German Air Force to publish null results."
That allegation doesn't survive verification. Tajmar's TU Dresden Chair of Space Systems is funded by the European Space Agency, by the German Research Foundation (DFG), by the Federal Ministry for Economic Affairs and Energy (BMWi), and by the German Aerospace Center, DLR. DLR is the civilian counterpart to NASA. The Luftwaffe (Germany's actual air force, part of the Bundeswehr) has no public record of funding Tajmar's lab. The conflation appears to come from "Luft" appearing in both "DLR (Deutsches Zentrum für Luft- und Raumfahrt)" and "Luftwaffe."
We note Eskridge made the allegation. We don't repeat it as fact.
Two things are happening here at once. Eskridge's broader point about geometry, that the field tested mostly the simpler disc and called the null a refutation of the more elaborate one, is technically defensible. Her specific attack on Tajmar is geometrically off, because Tajmar wasn't trying to replicate the 1997 protocol. He was running a different experiment on a different effect. Both can be true, and a fair treatment of the controversy has to hold both.
The honest reading of the confrontation: a 28-year-old chemist made a legitimate point about a structural gap in the replication literature and overshot in characterising one specific researcher's motives.
Fetisov reopens the file
In September 2022, A. V. Fetisov posted a preprint on arXiv (2209.03332) reporting a Podkletnov-like effect at room temperature, achieved through an entirely different methodology. Instead of a rotating bulk superconducting disc, Fetisov used dispersed YBCO particles mixed with a crystal hydrate (potassium carbonate sesquihydrate, K₂CO₃·1.5H₂O) in gas-tight containers, exposed to a 50 MHz radio-frequency field for roughly half an hour. Test bodies showed sharp weight loss. The effect correlated with external surface area and oxygen content in the YBCO, with anomalies aligned to the so-called superconducting "60-K plateau."
Fetisov's setup is a new experiment, possibly probing the same underlying physics, definitely not a clean Podkletnov replication. The preprint hasn't been independently reproduced. The methodology is unconventional enough that the appropriate mainstream response is "interesting, show us more rigorous data."
What the Fetisov result does mean for our purposes: the file isn't closed. A peer-reviewed-pending paper from 2022 reports a positive weight-loss effect on YBCO. The replication literature has not converged on a definitive null. That's a load-bearing fact for a field that's spent thirty years assuming it has.
What a clean test would look like
Here's the honest state of the field after thirty-three years. The original Podkletnov claim has not been definitively confirmed. It has not been definitively refuted. Most replications used setups that don't match the 1997 protocol. The one strong replication that did match (NASA Marshall's bilayer disc with Podkletnov consulting) produced null results, but it was a single test in a wide parameter space.
So what would actually settle the question?
A pre-registered experiment with the disc fabrication procedure, layering geometry, magnetic field configuration, rotation speed, temperature stability, balance mechanics, and shielding setup all specified in advance. Built by an independent team. Run with raw data published. Witnessed by skeptics and proponents both. Done without consultancy fees on either side that could bias the framing.
Cost isn't the obstacle. A few hundred thousand dollars is well within reach for any serious materials-science group. The obstacle is that the question has been pushed out of mainstream physics by the 1996 controversy, the awkwardness of the claim, and the absence of obvious institutional incentive to settle it.
Why would a topic worth a clean experimental test get the silent treatment for thirty years?
Why this matters
Amy Eskridge died on June 11, 2022, in Huntsville, Alabama. She was 34. Her death was ruled a self-inflicted gunshot wound. In multiple recorded statements before her death, she said: "If you see any report that I killed myself, I most definitely did not."
We don't speculate on the circumstances of her death. Her father Richard Eskridge, a retired NASA Marshall engineer, has publicly told reporters there was "nothing suspicious" about it. We note for the record that one of the few people in the field who knew the bilayer-versus-monolayer technical literature in detail, who had family ties to NASA Marshall's advanced propulsion work through her father, and who was actively trying to bring antigravity research into the open, is no longer here to argue the case.
The Podkletnov question doesn't go away because she did. The bilayer geometry is still in the published record. The replication trail is still incomplete. And the 1997 protocol has been tested rigorously by an independent team exactly once.
Here's why this matters to the broader 4Orbs investigation. If the orbs in the alleged MH370 surveillance footage are FRC plasma reactors built around rotating superconducting components, as more than 600 video analyses on this site have argued, then the question of whether spinning a small piece of supercooled ceramic can modify the weight of objects nearby is the desktop-scale version of the same physics. A 0.3% effect on a bench-top disc and a 200-ton aircraft removed from the sky over the Indian Ocean would be the same family of phenomena, separated by eight or nine orders of magnitude in scale.
The bench-top experiment has been run partially, dismissed prematurely, and starved of follow-up funding. The question of what flew over the Indian Ocean on March 8, 2014, depends in part on whether anyone is willing to fund the follow-up.
Related Topics
Navy patents on inertial mass reduction, room-temperature superconductors, and high-frequency gravitational wave generation. Adjacent physics, different protocol.
Quantum sensing claims allegedly used to track human signatures from orbit. The same Lockheed-NASA technical lineage that built the bilayer testbed.
How Forbes connects field-reversed plasma reactors to the three luminous objects in the MH370 footage.