The IVO Quantum Drive’s Orbital Test & Quantised Inertia Physics

August 29, 2026
|In Field Effect
|By Tim Ventura

In March 2025, a SpaceX Falcon 9 carried Rogue Space Systems’ OTP-2 satellite into orbit with an unusually consequential passenger: IVO Ltd.’s Quantum Drive, a compact electrical propulsion device designed to generate thrust without expelling onboard propellant. Physicist Mike McCulloch, whose Quantised Inertia theory inspired the drive, described the mission as, as far as he knew, the first orbital test of a reactionless propulsion system and “a bit of a watershed.” The experiment didn’t deliver the dramatic orbit-raising maneuver its developers had hoped for. A problem involving the testing hardware and command-and-control system limited drive operation to less than about a minute. Yet the test wasn’t empty. The team reported apparent movement during a July activation, and McCulloch later found that the spacecraft lost approximately 600 meters less altitude than a comparison satellite over three months. That difference was close to the rough scale he calculated for intermittent drive operation—but it could also have resulted from atmospheric drag, spacecraft attitude, electromagnetic forces, another propulsion payload, or the limitations of public orbital data. The result sits in the most scientifically uncomfortable and productive category: intriguing, potentially important, and nowhere near conclusive.

From a Question About Inertia to a Spacecraft Payload

McCulloch’s route into unconventional propulsion began far from rocket laboratories. He earned a physics degree from the University of York in 1991 and a doctorate in ocean physics from the University of Liverpool in 1995. From 1998 to 2008, he worked as an ocean and wave-model scientist at the UK Met Office, later joining the University of Plymouth as a lecturer in geomatics—the mathematics of positioning and measurement in space. He’s now a Visiting Specialist at Plymouth. Over roughly two decades, McCulloch says he has produced 28 papers and four books on Quantised Inertia, along with a multiyear DARPA-supported effort to investigate whether the theory could produce measurable thrust.

The question driving that work sounds almost elementary: what physically causes inertia? Newtonian mechanics describes inertial behavior with extraordinary accuracy. Objects continue moving unless a force changes their motion, and their resistance to acceleration is proportional to inertial mass. But to McCulloch, that’s a description rather than a mechanism. While modeling waves at the Met Office, he imagined a speedboat accelerating across water. The boat encounters more waves in front than behind, producing a pressure imbalance that resists its acceleration. Far enough behind the accelerating boat lies a horizon beyond which waves can never catch up. McCulloch began asking whether an analogous process involving quantum fields and relativistic horizons could generate inertia itself.

The path from that theoretical question to IVO began in May 2019, when company CEO Richard Mansell contacted McCulloch. IVO had already experimented with high-voltage lifters, capacitors, and a wireless-power system known as Capacitive Based Aerial Transmission, or CBAT. The system treated a transmitter and remote receiver like the opposing sides of an enormous capacitor, with the surrounding air acting as a dielectric. Mansell found McCulloch’s Physics From the Edge blog while investigating unconventional gravitational and capacitor effects. McCulloch then connected IVO with engineer Frank Becker and collaborator A. Bhatt, who had reported force measurements from high-voltage capacitors and compared them with Quantised Inertia calculations.

IVO didn’t simply reproduce the earlier apparatus. Becker and Bhatt had placed capacitors on a digital balance, leaving the experiment potentially vulnerable to cable tension, electrostatic effects, vibration, heat, and electronic interference. IVO used a pendulum to reduce those problems. It replaced the Kapton plastic dielectric used in earlier work with naturally oxidized aluminium, allowing the metal to serve as the conductor while its oxide layer acted as a thin dielectric. The company also arranged multiple capacitors in a proprietary stack. McCulloch advised that the elements couldn’t be packed too closely because, within his model, their relevant Rindler horizons could interfere with one another. IVO later tested stacked hardware in vacuum below (10^{-5}) Torr at E-Labs in Fredericksburg, Virginia. Those company-reported results encouraged the team to attempt the much harder experiment: operating the device aboard a spacecraft.

Two Flights, Two Very Different Setbacks

The first IVO hardware reached orbit aboard Rogue Space Systems’ BARRY-1 satellite, launched on SpaceX’s Transporter-9 rideshare mission on November 11, 2023. BARRY-1 was Rogue’s first orbital platform and carried the Quantum Drive as a hosted payload alongside Rogue’s own computing technology. The plan was to commission the spacecraft, establish a stable orbital baseline, and then activate the IVO hardware to look for a measurable change in the satellite’s trajectory.

That test never happened. Rogue reported persistent spacecraft-bus power problems during the Launch and Early Orbit Phase. Although the company completed some testing of its Scalable Compute Platform, it lost communications with BARRY-1 after approximately two months and suspended active operations in February 2024. Rogue stated unambiguously that it “never had the opportunity to test the IVO drive.” BARRY-1 therefore wasn’t a negative propulsion result. The spacecraft entered orbit with IVO hardware aboard, but no controlled Quantum Drive campaign was completed. At most, the mission demonstrated partial flight heritage: the payload survived integration, launch, deployment, and some time in orbit.

Rogue offered IVO another flight, this time aboard the larger OTP-2 platform. SpaceX launched OTP-2 at 11:43 p.m. Pacific time on March 14, 2025—March 15 in UTC—from Vandenberg Space Force Base on Transporter-13. Rogue now lists the 8U spacecraft as active in sun-synchronous orbit. OTP-2 hosts Rogue’s second-generation onboard computing platform as well as several customer payloads, including IVO’s Quantum Drive and a separate water-based ion thruster. That multi-payload architecture expanded what could be tested, but it also created additional variables that would later complicate any attempt to attribute a small orbital change to one device.

The hoped-for IVO demonstration was not subtle. The team wanted to operate the drive long enough to produce an orbital rise approaching 100 kilometers—an effect far beyond ordinary tracking uncertainty or minor changes in drag. Instead, the activation campaign was curtailed. McCulloch said, “Movement was seen and it caused some excitement,” but the drive could operate for less than approximately one minute before a technical problem interrupted the test. In a statement relayed through McCulloch, Mansell attributed the difficulty to the surrounding test system: “It isn’t the Drives themselves that were acting up, but something with the command & control mechanism.” That distinction may be correct, but it doesn’t establish that the movement came from the drive. It means only that the planned decisive experiment was interrupted before it could produce a decisive answer.

The 600-Meter Clue

Without access to all of IVO’s internal telemetry, McCulloch began monitoring publicly available tracking data. He followed OTP-2 under NORAD catalog number 63235 and compared it with object 63220, another satellite deployed by the same Falcon 9 into a broadly similar orbit. His reasoning was straightforward: if two spacecraft began at comparable altitudes and experienced the same broad atmospheric and solar conditions, their relative rates of orbital decay might reveal an additional force acting on one of them.

Between September 30 and December 30, 2025, object 63235—the IVO spacecraft—fell from approximately 506.30 kilometers to 501.42 kilometers, a loss of 4.88 kilometers. Over the same period, object 63220 fell from roughly 504.98 kilometers to 499.50 kilometers, a loss of 5.48 kilometers. The difference was approximately 600 meters over 90 days, or about 6.6 meters per day. McCulloch published the comparison under the title “A Tale of Two Satellites” and argued that the smaller decline of the IVO spacecraft was at least compatible with intermittent thrust.

His back-of-the-envelope calculation used a proposed force of 1.75 millinewtons and a spacecraft mass of 20 kilograms, producing an acceleration of (8.75 \times 10^{-5}) meters per second squared. A 60-second activation would add approximately 0.00525 meters per second to the spacecraft’s velocity. Under his simplified orbital calculation, one such activation would increase orbital height by about 9.5 meters. If the drive operated for one minute every other day, the average effect would be roughly 4.75 meters per day—close in scale to the observed 6.6-meter daily difference. That numerical resemblance is why McCulloch considers the result encouraging.

But almost every important input remains uncertain. The assumed firing schedule wasn’t a publicly released command log. McCulloch used 20 kilograms in his calculation, while describing the spacecraft in the interview as approximately 11 kilograms. The comparison satellite was lower in the atmosphere, which could increase drag, while the IVO spacecraft reportedly had a larger cross-sectional area—about 0.05 square meters versus 0.04—which could increase its drag instead. Their attitudes, tumbling rates, thermal behavior, ballistic coefficients, and operational histories weren’t demonstrated to be identical. Public orbital elements are fitted estimates rather than precision onboard navigation records. Atmospheric density changes with solar activity, electrical currents can interact with Earth’s magnetic field, and OTP-2 carried another propulsion experiment. The 600-meter difference is therefore a clue worth investigating, not a measurement that uniquely identifies Quantum Drive thrust.

The careful conclusion is that the orbital campaign produced an interesting but statistically weak anomaly. It may be consistent with IVO’s expected force, but consistency isn’t causation. It doesn’t yet establish that a non-drag force acted on the spacecraft, that the IVO device produced that force, or that Quantised Inertia explains it. McCulloch himself acknowledged the ambiguity, noting that IVO had considered a Lorentz-force explanation for at least part of the apparent movement. He estimated that the Lorentz force was too small, but he didn’t have the detailed circuit geometry and current data required to settle the issue. The mission gave researchers a reason to run a better experiment—not permission to skip one.

What Quantised Inertia Is Trying to Explain

Quantised Inertia begins with McCulloch’s claim that inertial mass isn’t an intrinsic label permanently attached to matter. Instead, he proposes that inertia emerges from an object’s acceleration and its relationship with the surrounding quantum field. In the speedboat analogy, waves pile up unevenly around an accelerating vessel. In the quantum version, an accelerating object experiences Unruh radiation, while a Rindler horizon forms behind it. McCulloch argues that the horizon suppresses some available quantum modes, leaving more radiation pressure in front of the object than behind it. The resulting imbalance resists acceleration and appears macroscopically as inertia.

Several ingredients in that picture belong to established theoretical physics. Rindler horizons arise in the description of accelerating observers. The Unruh effect predicts that an accelerating observer perceives a thermal field where an inertial observer sees a vacuum. The Casimir effect demonstrates that physical boundaries can alter the allowed modes of a quantum field. Quantised Inertia adds the controversial steps: that Unruh radiation is the physical cause of inertial resistance, that horizons can function like mode-restricting boundaries, and that engineered cavities can manipulate those restrictions to generate net force. Acceleration-induced thermality has been investigated experimentally, but direct detection and interpretation of the Unruh effect remain debated.

A commonly presented Quantised Inertia relation takes the form

[
m’ = m\left(1-\frac{2c^2}{a\Theta}\right),
]

where (m’) is the modified inertial mass, (m) is conventional mass, (c) is the speed of light, (a) is acceleration, and (\Theta) is a cosmic-scale horizon dimension. At ordinary accelerations, the correction is negligible and familiar Newtonian behavior emerges. At extremely low accelerations, the correction becomes significant and inertial mass is predicted to decline. McCulloch applies that limit to stars at the edges of galaxies. Instead of adding unseen dark matter to keep fast-moving stars gravitationally bound, he argues that their centrifugal response is weaker because their inertial mass is lower. He says the model reproduces galaxy rotation using visible matter, the speed of light, and a cosmic horizon scale, without adjustable dark-matter halos.

The proposed drive mechanism applies the same concept at a radically smaller scale. Inside a capacitor, electrons undergo Fowler–Nordheim tunneling through a strong electric-field barrier. McCulloch argues that their effective acceleration can reach approximately (10^{22}) meters per second squared, making their associated Unruh wavelengths short enough to interact with micrometer-scale plate spacing. The conducting plates suppress modes inside the capacitor more strongly than outside it, reversing the ordinary radiation gradient. In his model, the electron receives an extra impulse toward the anode and transfers that momentum to the structure. That’s why “propellantless” is the cleaner term for the IVO concept. “Reactionless” implies no momentum exchange at all, while Quantised Inertia proposes an open-system interaction with horizons and quantum fields—even though that mechanism remains unconfirmed.

The Long Experimental Trail

The modern capacitor experiments began with Becker and Bhatt, who reported tests using small high-voltage capacitors with plate separations from roughly 10 to 60 micrometers and diameters around 2.5 centimeters. They observed forces reaching approximately 6 millinewtons as the plate separation decreased. McCulloch calculated a Quantised Inertia prediction for the force-versus-separation trend and said it matched the reported values without tunable parameters. That agreement helped redirect his DARPA-supported work toward capacitors, but the original balance-based setup remained susceptible to cable forces, electrostatic coupling, thermal drift, vibration, and electronic artifacts.

McCulloch and engineer Richard Arundal later built a more systematic apparatus at Plymouth University’s Mayflower Labs. Their test stand placed a foil capacitor on an ADAM LPB 223i digital balance with 0.1-milligram sensitivity and a 120-gram capacity. A calibrated Vitrek V71 high-voltage tester charged the device through a Galinstan liquid-metal connection intended to isolate mechanical cable forces, while a surrounding shield cage reduced electrical and environmental interference. The team spent approximately 16 months developing the stand and conducted at least 182 numbered test runs while refining shielding, capacitor fabrication, and data collection.

The Plymouth team reported anodeward weight changes as large as 25 milligrams, equivalent to about 0.25 millinewtons. Successful signals appeared when capacitors were heated above 50 degrees Celsius—sometimes to 64 degrees—and operated within approximately 5 to 10 percent of dielectric breakdown. Turning a capacitor over reportedly reversed the direction of the measured force. Their working prediction was

[
F=\frac{0.00014IA}{d^2},
]

where (I) is leakage current, (A) is plate area, and (d) is plate separation. Because force varies inversely with the square of separation, halving the dielectric thickness should produce four times the force. Selected tests were reported to follow that prediction. Yet about half the runs produced no measurable mass change, and those null events weren’t included in the predicted-versus-observed plot. Reliability therefore remains one of the central unsolved problems.

IVO’s program added a pendulum, oxidized-aluminium dielectrics, proprietary stacking, and thermal-vacuum operation. McCulloch’s presentation describes one first stacked test using a 0.4-kilogram device, a 15-by-15-centimeter active area, a 17.8-micrometer separation, and an observed force around 0.9 millinewtons. A separate summary in the same presentation lists an IVO vacuum result near (6 \pm 2) millinewtons compared with a predicted 5.25 millinewtons, apparently representing a different configuration or campaign. McCulloch says several groups have reported forces in the range predicted by QI, but much of that record remains unpublished, presented at conferences, awaiting formal publication, or dependent on incomplete technical information. The body of evidence is broader than a single experiment, but it’s not yet the kind of independent, repeatable literature required for a new propulsion technology.

The Experiment That Could Settle It

A decisive orbital test would need to be designed around attribution. The cleanest mission would use a dedicated spacecraft with no second propulsion payload, precision GNSS orbit determination, calibrated onboard accelerometers, continuous attitude data, magnetometers, temperature sensors, electric-field monitoring, and complete current and voltage telemetry. An active drive should be paired with a dummy device drawing the same electrical power but designed not to produce thrust. The spacecraft should execute randomized forward, reverse, and off sequences, with the command schedule hidden from at least one independent analysis team until data processing is complete.

The result would need to be synchronized and reversible. Acceleration should begin when the drive is energized, stop when it’s switched off, and reverse when its polarity or orientation is reversed. The signal should follow a numerical prediction made before the data are examined. A matched dummy load should produce no acceleration. Long activation periods should create a velocity change far larger than the uncertainties from drag, tracking, thermal recoil, radiation pressure, magnetic interaction, outgassing, attitude control, and orbital fitting. Raw telemetry and the analysis protocol should be released so independent researchers can reproduce the result.

McCulloch also proposes a laboratory discriminator that he believes would be specific to Quantised Inertia: resonance. If electron acceleration and cavity dimensions are tuned so that particular Unruh wavelengths fit—or are excluded—between the capacitor boundaries, the measured force should rise to a predicted peak and then fall as the system moves away from resonance. He argues that conventional thermal, electrostatic, or magnetic artifacts wouldn’t naturally produce a peak at the geometry specified by QI. He has applied for funding for an unambiguous test associated with the UK’s National Physical Laboratory, while pursuing numerical optimization, new papers, another textbook, and a possible future American space collaboration.

The practical reward, if repeatable thrust were confirmed, would begin with satellites rather than starships. Spacecraft often reach the end of their useful lives because they exhaust station-keeping propellant even while their electronics and solar arrays remain functional. A compact electric drive with no propellant tank could maintain orbits, reposition satellites, reduce launch mass, and operate without an exhaust plume. McCulloch contrasts the concept with the Dawn spacecraft’s ion propulsion, which generated roughly 90 millinewtons but required substantial xenon propellant and about 2,100 watts of power. The proposed IVO unit produced far less force—about 1.75 millinewtons—but its advocates argue that eliminating hundreds of kilograms of spacecraft and propellant changes the acceleration equation. As McCulloch summarized it, “Finally, we can travel light.”

Beyond that lie possibilities that remain speculative even if the basic effect proves real. McCulloch estimates that a very small continuously accelerating probe could reach the predicted inner region of the Oort Cloud, roughly 3,000 astronomical units away, in about a year. He has also calculated a 10-to-15-year journey to Proxima Centauri by accelerating to a substantial fraction of light speed, reversing the drive, and decelerating into the destination system. He discusses nanoscale “quantum launch,” levitating materials, and possible energy-generation concepts as longer-range consequences of the same physics. None is an engineering forecast, and none follows automatically from a millinewton-scale anomaly. But that’s why the IVO mission matters despite its inconclusive outcome. It carried Quantised Inertia out of papers and laboratory balances, placed a physical device in orbit, exposed the weaknesses of the existing experiment, and showed exactly how much better the next one must be.

References

Podcast and Background

IVO Quantum Drive and the Orbital Missions

Quantised Inertia: Foundational Papers

Capacitor-Thrust Experiments and Theory

General Explanation and Further Reading