Eugene Podkletnov’s Four Experiments in Gravity Control

September 23, 2026

What began as a quality-control test for superconducting ceramics led Dr. Eugene Podkletnov into a series of experiments involving weight loss, gold-coated rotating disks, high-voltage impulses, and coils that briefly lift themselves that’s spanned roughly 35 years. In a recent interview with podcast host Tim Ventura, the chemist and materials scientist outlined four distinct approaches, including reports of bent metal, altered laser intensity, and an impulse traveling at 64 times the speed of light. Taken together, the experiments describe a changing research strategy: moving from spinning materials toward controllable electromagnetic fields, while trying to determine what’s actually producing the reported forces.

A Gravity Experiment That Began With Materials Testing

Podkletnov says the original experiment wasn’t intended to investigate gravity. He was assessing ceramic targets for radio-frequency magnetron sputtering, a process used to prepare thin films. The disks he described in this interview were eight inches across and half an inch thick, and their quality mattered: porosity or patches that weren’t superconducting could compromise their usefulness. Working with a disk above a magnet in liquid-nitrogen vapor, his team scanned its surface with an ordinary nail suspended from a string connected to a balance. Some readings, he recalled, didn’t behave as expected.

To investigate whether magnetic interactions were responsible, the team replaced the nail with a pencil, a glass stick, and a rubber stick. Podkletnov says the apparent weight changes persisted, increased with rotation speed, and became larger when samples were positioned closer to the disk’s outer edge. The apparatus evolved from a solid disk to an annular one, with windings passing through the central opening to support higher-speed rotation. He described efforts to reduce vibration and airflow, including covering the cryostat with glass and making measurements above the enclosure rather than directly over an exposed cold surface.

The numbers that became associated with this work were modest in percentage terms but extraordinary in interpretation: apparent weight reductions of approximately 0.3–0.5%, with transient peaks approaching 2%. Podkletnov reaffirmed those figures in the interview and added another observation: he says air pressure decreased within the disk’s upward projection, extending toward the ceiling, by corresponding percentages. He regards that pressure response as supporting the weight measurements. The interview, however, doesn’t provide the pressure-instrument specifications or records needed to examine that proposed correspondence independently.

The research entered the literature in 1992, when Podkletnov and R. Nieminen published a paper in Physica C. Its original language emphasized gravitational shielding, but Podkletnov’s interpretation later changed. Introducing that distinction in a 2025 interview with Ventura, he explained, “we no longer use the term gravitational shielding.” He instead describes modification of the local gravitational field near the apparatus. His collaborations with Ning Li and Giovanni Modanese supplied theoretical perspectives, but the central distinction remains important: a change in a balance reading is the reported observation; identifying its cause as a change in gravity is the interpretation.

The Role of the Gold Coating

The room-temperature experiment takes the story away from cryogenic ceramics. In the latest interview, Podkletnov described a disk approximately 20 centimeters in diameter made from doped aluminum, rotating inside a vacuum chamber at 8,000–15,000 revolutions per minute. He emphasized unbalanced rotation—the deliberate wobble described in earlier interviews—as part of this configuration. Samples weighing approximately 20–30 grams, he says, experienced repulsion when brought near the disk’s surface. He describes the interactions as torsion fields, which he associates with rotating systems and regards as similar in some respects to gravity. This is a different apparatus from the original YBCO experiment, despite their shared use of rapid mechanical rotation.

Earlier accounts provide a more specific picture of how those samples were positioned. In a 2020 interview, Ventura described videos showing a sample attached to a movable armature, with a restraining string in one orientation to prevent contact with the rotating disk. Podkletnov reported testing materials including glass, wood, paper, and metal, with some samples weighing 30–50 grams. He described movement as far as five to seven centimeters from the disk at approximately 8,000–12,000 RPM. Those details describe supported samples allowed to move, not necessarily objects floating freely without any attachment.

The most useful clarification in the latest interview concerns the gold itself. Podkletnov said the uncoated rotating system produced forces in both directions, above and below the disk. The very thin gold layer, in his explanation, was added to favor one direction. He also discussed implantation treatments as a more costly, labor-intensive part of some versions. That’s a more specific role than simply labeling the apparatus a room-temperature superconductor: his latest explanation emphasizes directing the reported force, while his earlier descriptions had also associated thin gold layers with superconducting behavior at room temperature.

He says a reproduction was carried out in Prague with Professor Jan Rak approximately six or seven years before this interview, and he believes qualified laboratories could reproduce the effects with a suitable chamber, motor, and correctly prepared disk. Those are his accounts of reproducibility, not an independent replication report supplied with the interview. The coating thickness, dopant recipe, degree of imbalance, and complete measurement records aren’t specified in this conversation. What’s clear is the direction of the research: he’s trying to preserve the reported force while moving away from the cooling and fabrication demands of bulk YBCO.

Rotating the Magnetic Field Instead of the Machine

Podkletnov’s next step removes the rotating disk altogether. Explaining the transition during the interview, he put it simply: “It is enough to rotate the magnetic field.” His proposed apparatus contains an ordinary inner coil enclosed by an outer toroidal winding. In the drawing corrections he supplied for the slideshow, he emphasized that the outer turns wrap through the central opening and around the ring’s cross-section, as in a conventional toroidal transformer—not along its circumference. The assembly remains circular in plan, but its flattened profile extends outward toward a sharp, tapered rim.

The rotating field comes from sequential electrical activation rather than moving hardware. Podkletnov offered an example with 12 winding sectors. One sector could be energized and then the neighboring sector, making the active region advance around the ring. Alternatively, several sectors could be active together: his example was sectors 1, 6, and 9, followed by their neighbors. Advancing each by one position would give 2, 7, and 10. These numbers illustrate his switching concept; they aren’t a complete disclosed drive sequence. The important feature is that the electromagnetic pattern moves while the coil remains mechanically stationary.

He also described a carrier frequency and two additional frequencies differing by a factor of 10.6, while declining to disclose their absolute values. In his 2025 conversation with Ventura, he said the apparatus consumed several kilowatts and described assemblies weighing from approximately half a kilogram to three or four kilograms. The claimed result was self-lift of the coil assembly—not simply the separate disk pictured above it in some illustrative slides. In his earlier explanation, the device creates a thin attractive region above its windings and carries that region upward with it. That proposed “gravity well” remains his interpretation, not a measured field map presented here.

The practical limitations are substantial. Podkletnov was candid about the motion: “I can’t say that it is very stable in the air, no.” One side can rise first, and the assembly can overturn. Current is driven close to what the laboratory winding can tolerate, leading to overheating and damaged insulation. Earlier descriptions put individual levitation episodes at roughly 15–23 seconds before electrical failure. The latest account again describes only several seconds of operation. Even on its own terms, this is a report of brief, unstable laboratory motion, not sustained flight by an engineered propulsion system.

Inside the High-Voltage Impulse Generator

The impulse generator is a separate branch of the program: its superconducting emitter doesn’t rotate. Podkletnov says the motivation was to investigate high-voltage discharges through superconductors, including the behavior associated with Josephson junctions inside the material. Early arrangements used a Van de Graaff generator; later versions employed Marx capacitor banks. In the configuration he described during the latest interview, a cryogenically cooled emitter faced a copper O-ring across a gap of approximately half a meter, with a voltage difference ranging from 500,000 volts to two million volts.

The earlier papers document other configurations, and their dimensions shouldn’t be merged indiscriminately with the latest account. In their later discharge paper, Podkletnov and Modanese describe a quartz chamber approximately one meter in diameter and 1.5 meters long. Its target was a solid copper disk 100 millimeters across and 15 millimeters thick, with an adjustable electrode separation of 15–40 centimeters. The current interview’s half-meter spacing and O-ring target therefore belong to a different described configuration. That distinction matters when interpreting either the drawings or the sequence of reported results.

The published apparatus used an external solenoid producing approximately 0.9 tesla, while another magnetic arrangement helped trap flux in the superconducting emitter during cooling. The chamber was evacuated to about one pascal, and the reported anomalous pulses were associated with emitter temperatures of approximately 50–70 kelvin. Above roughly 500 kilovolts, the authors described a change from an ordinary spark to a flat, glowing discharge originating across the emitter’s face. In that published configuration, peak current was on the order of 10,000 amperes, and the luminous discharge lasted roughly 10–100 microseconds.

The visible discharge and the claimed impulse beyond the chamber must also be kept separate. The former is an electrical event between electrodes; the latter was inferred from mechanical responses and other detector readings along the projected axis. The luminous cyan beam used in the interview’s illustrations is a visual aid, not a photograph of radiation traveling through the laboratory. Nor does a voltage figure alone tell readers the apparatus’s power, transferred momentum, or efficiency. Those require additional measurements, and the accounts describe several evolving configurations rather than one fixed machine with a single performance specification.

Why the Emitter’s Crystal Structure Matters

At the center of the apparatus is a deliberately engineered bilayer ceramic. One layer is a normal conductor whose composition closely resembles that of the superconductor, but which doesn’t become superconducting under the relevant operating conditions. The other is a dense superconducting layer facing the target. The research papers describe emitters approximately 80–120 millimeters in diameter and 7–15 millimeters thick, using YBCO-based compositions. Podkletnov’s latest explanation focuses on managing the interface between charge transport in the normal layer and transport by paired electrons in the superconducting material.

The face isn’t supposed to resemble a cluster of crystals protruding into the chamber. Podkletnov described seeds placed on the material’s surface that grow larger crystalline regions during heat treatment. The published fabrication account specifies seeds approximately one cubic millimeter in size, spaced about 15 millimeters apart, followed by controlled crystal growth. A surface layer approximately 0.3 millimeter thick was subsequently removed with diamond tooling. His interview description emphasizes that the emitting face was cut and polished. That preparation produces a flat, finished surface with crystalline texture, not exposed crystal spikes.

There’s reported experimental evidence for the material’s importance within the authors’ own program. Their discharge paper says conventional emitters made from copper, aluminum, steel, and chromites didn’t produce the unusual effects. It also says removing the normal-conducting layer made discharges less regular and the anomalous radiation weaker. The authors didn’t claim a settled explanation for that layer’s function: they discussed possible effects on the metal–superconductor interface and on discharge current. Podkletnov now stresses the interface as a major condition for success, but that emphasis shouldn’t erase the uncertainty explicitly recorded in the earlier paper.

Comparisons between two emitters add another concrete detail. The superconducting layer was four millimeters thick in one and eight millimeters in the other; the thicker version could retain a stronger magnetic field and produced larger pendulum responses. The later paper also reports that trapped magnetic flux increased impulse strength by approximately 25%. These observations are central to Podkletnov’s explanation of why simply discharging high voltage through a commercially available superconductor might not reproduce his results. His claim depends on a particular material structure and magnetic state, not merely on the presence of superconductivity.

From Pendulum Deflections to Bent Metal

The flat metal target provides one of the interview’s clearest accounts of how an unexpected observation changed the apparatus. Podkletnov says the initial target was severely bent during discharges, prompting the change to a copper O-ring. He also described knocking over a standing book several meters away and damaging part of a brick wall at two million volts. Older interviews include accounts of permanent deformation in metal plates several inches thick. These are reports spanning different tests and operating conditions; they shouldn’t be treated as interchangeable demonstrations or assigned automatically to every version of the generator.

The 2001 paper provides a more quantitative record using an 18.5-gram rubber pendulum bob. For the stronger of two emitters, reported horizontal displacement increased from 56.5 millimeters at 500 kilovolts to 142 millimeters at two million volts. The corresponding rise in the bob’s height increased from 2.0 to 12.7 millimeters, giving calculated potential-energy changes of approximately 0.36 and 2.31 millijoules. Each table entry averaged 12 discharges, with a stated standard deviation of five to seven percent for individual measurements. Those numbers describe the response of a small test mass, not the energy content of the entire reported beam.

The authors also tested metal, glass, ceramic, wood, rubber, and plastic pendulums, with sample masses spanning approximately 10–50 grams. They reported similar deflections at fixed voltage and an impulse approximately proportional to sample mass within their measurement variability. Detectors were placed near the apparatus and 150 meters away, with intervening shielding that included a 30-centimeter brick wall and a 25-millimeter steel plate; the farther arrangement included additional brick walls totaling 80 centimeters. They reported comparable responses at the measurement locations and interpreted the material independence and transmission through shielding as reasons to describe the effect as gravity-like.

The latest interview also includes a human-exposure anecdote that deserves caution rather than imitation. Podkletnov said some participants put a hand in the projected path during tests at approximately 500 kilovolts. Describing the sensation, he said, “It’s like somebody just slaps you over the hand.” He also reported momentary displacement of the skin on the palm. The story helps explain why he regarded the effect as mechanical, but it isn’t a calibrated force measurement or a safety study. Nothing in that anecdote establishes a safe exposure level, and the high-voltage or human-exposure experiments shouldn’t be attempted.

Testing Laser Light—and Timing the Impulse

To investigate whether the impulse affected light, Podkletnov’s team arranged a laser beam at a shallow angle to the reported impulse path. In the latest account, the two paths overlapped for approximately 50–60 meters. A sensor monitored the laser’s intensity while the generator discharged, and he reported a temporary decrease of roughly seven percent, varying with voltage. He interprets the decrease as photons being carried or pushed out of the laser beam. The measured quantity, however, was received optical intensity; that reading alone doesn’t directly track individual photons or establish the interaction mechanism.

The earlier discharge paper gives more specific optical parameters: a crossing angle near 0.1 degree, an interaction length around 57 meters, a 694-nanometer ruby laser, and a second blue laser operating at 473 nanometers. It reported intensity decreases of seven to ten percent followed by a rapid return toward baseline. Importantly, the authors called the measurements preliminary and discussed possible interference from air-density changes and refraction. They argued that the observed recovery time weighed against that explanation, but also identified measurements in vacuum as desirable. The authors therefore treated the optical findings as observations requiring further investigation, rather than a fully resolved interaction with light.

The propagation-speed experiment produced the most provocative number in the interview. Podkletnov described a roughly two-year investigation using two rubidium clocks and impact sensors across a 1,200-meter measurement distance. Earlier accounts identify piezoelectric sensors as the mechanical detectors. He reported a propagation speed of 64c after repeated tests and described using laser alignment to locate the projected impact area outside the building. For scale, the reported speed would correspond mathematically to about 63 nanoseconds across 1,200 meters, compared with approximately four microseconds for light. Those times illustrate the claimed measurement’s scale; they aren’t additional timing results supplied in the transcript.

There’s an important chronological boundary here. The earlier discharge paper explicitly stated that the propagation speed was still unknown and proposed future timing experiments. The 64c result appears in later accounts, including Podkletnov’s discussion of work with Modanese published in a 2012 book chapter. It therefore shouldn’t be attributed to the original discharge paper or presented as an established measurement of gravity in general. Podkletnov also left the underlying entity unresolved in the latest interview: it might involve gravitons, he suggested, or a wave in the physical vacuum. Describing those alternatives, he added that “further experiments will prove what we are dealing with.”

The Unexpected Effect Behind the Generator

Not all the reported activity was in front of the emitter. In the latest interview, Podkletnov offered a more detailed explanation of the rear-side emission he’d mentioned previously. He said specialists in high-voltage discharges described a returning electrical pulse after the forward discharge reached its target, with approximately four to seven percent of the initial energy traveling back. In his interpretation, that returning discharge passes through the emitter and produces a separate effect behind the apparatus. The percentage concerns the return-discharge account; the transcript doesn’t establish it as a measured efficiency for the anomalous rear emission itself.

His most striking example involved a technician holding an electrical tester. Podkletnov said the technician noticed that his fingers had “submerged into the plastic” of the instrument. He didn’t provide material-analysis results, a temperature record, or a detailed description of what remained afterward. He did describe an operational response: the team subsequently kept people away from the rear of the device. That change is significant because it treats the observation as a potential hazard rather than an entertaining side effect. The anecdote remains an anecdote, but it shaped how he says the laboratory was operated.

In a 2013 interview with Ventura, Podkletnov described rear-side radiation with harmonics whose frequencies were difficult to determine, effects extending several meters, and declining intensity with distance. He also claimed interactions that joined metal and plastic. Those descriptions differ from the sharply directed, longer-range forward impulse, and the distinction shouldn’t disappear under the general label “gravity beam.” The discharge papers’ argument about the forward impulse’s unusual energy–momentum relationship doesn’t, by itself, identify the rear emission. Its spectrum, physical carrier, and material-interaction mechanism remain unspecified in the accounts available here.

During the interview, Podkletnov compared the reported material effects with observations associated with John Hutchison. Ventura described having handled and photographed unusual samples attributed to Hutchison, including wood embedded in aluminum, while noting that he hadn’t watched those samples being produced. That’s an important evidentiary boundary: inspecting an unusual object isn’t the same as observing and documenting the process that created it. The comparison may suggest questions about transient material behavior, but it doesn’t demonstrate that the two sets of claims share a mechanism—or establish that either involves gravitational modification.

From Laboratory Reports to an Operating Prototype

One purpose of Ventura’s slideshow and interactive models was to make these distinctions visible. The original experiment rotates a cryogenic superconducting disk; the room-temperature version rotates a prepared metal disk; the impulse apparatus uses a stationary superconducting emitter; and the toroidal system switches fields around stationary windings. Podkletnov’s corrections to the drawings—especially the enclosed inner coil, perpendicular outer winding, sharply tapered rim, and polished emitter face—helped prevent those configurations from collapsing into one generic “antigravity machine.” The models can explain geometry, but their animation isn’t experimental data or a validated simulation of gravity.

The documentation is uneven across the four approaches. The impulse papers contain fabrication descriptions, detector arrangements, material comparisons, and numerical results. The newer rotating-field discussion supplies a geometry concept, an example of sector activation, a frequency ratio, and descriptions of short-lived self-lift, while withholding key operating frequencies. The room-temperature accounts add dimensions, rotational speeds, sample masses, and coating functions, but not a complete reproducible specification. Recognizing those differences makes the research easier to evaluate: a detailed measurement report, a proposed mechanism, and a laboratory recollection aren’t the same kind of evidence.

Podkletnov’s stated objective is to assemble the people and resources needed to turn this work into an operating prototype for use in space, in the atmosphere, and underwater. In the 2025 interview, he envisioned an international engineering laboratory and described the need for patient funding over five to seven years. The latest interview returns to a possible foundation or think tank and a team that includes both theoretical and practical specialists. Those are development ambitions, not demonstrated applications. For the coil approach in particular, heat, insulation failure, unstable motion, and short operating duration are already identifiable problems within his own account.

Closing the latest interview, Podkletnov described where he wants to concentrate next: rotating magnetic fields. Explaining the personnel needed, he said, “we need theoreticians, we need practical people.” It’s a fitting place to leave a story that began with a nail, a balance, and an unexpected reading during materials testing. Across roughly 35 years, the accounts have progressed from fractional weight changes to powerful impulses and whole-assembly lift, but the central question hasn’t changed: what is producing the motion? The next chapter is the effort to build a prototype—and make the connection between those reported forces and gravity clear enough for other laboratories to test.