In the spring of 1952, Collier’s Weekly came out with a cover on which Chesley Bonestell—the man who painted the sky for Gone with the Wind—had drawn a spinning wheel above Earth: Wernher von Braun’s orbital station, seventy-six meters across, three revolutions per minute, a crew of fifty. A hub station, a refueling platform for Mars expeditions. Just sixteen years later, that wheel attained immortality in 2001: A Space Odyssey, and ever since, the rotating station has been the most recognizable visual image of spaceflight: from Interstellar to The Martian.
And here’s the funny part: in all the seventy-four years since, not a single one has been built. Not “never made it up there,” not “didn’t work out”—just zero. Humanity has put tens of thousands of objects into orbit, learned to land rocket stages on a barge, gone to the Moon nine times, and is building a trip to Mars, but artificial gravity exists in only three dimensions: movies, concept art, and investor presentations. Humanity’s cumulative flight experience with artificial gravity amounts to two astronauts, about four hours, and 0.0005 g. One line in the entire history of spaceflight.
That record has stood for sixty years. Over the last month, this story has moved off dead center for the first time in a very long while: a company has emerged with artificial gravity right there on its roadmap, complete with dates, and an engineering plan has been published for testing how humans react to rotation without building anything new at all—using what already flies. I’m digging into what the recipe looks like, what kept us from baking it for three-quarters of a century, and why the only place it has finally begun to take shape is private capital, not space agencies.
A morning tech digest tossed this my way: The Space Review had published a major article by Joe Carroll—a veteran of space tethers, a man whose systems actually deployed in orbit in the nineties—on how to run an artificial-gravity test using Crew Dragon and a Falcon 9 second stage. The article is long, meticulous, full of kilonewtons and center-of-mass calculations—and yet its central claim sounds faintly indecent: everything needed for the first meaningful artificial-gravity test in sixty years is already sitting on the launch pad at Kennedy Space Center. No station needs to be built. You just have to not separate the things that already fly together.
What caught me wasn’t Carroll’s resourcefulness, but the suspended state of the whole field: on one side, a serious company called Vast promises commercial gravity-equipped stations as its end goal; on the other, humanity’s seventy-four-year record consists of one tethered flight test thirty-six meters long, producing five hundredths of one percent of Earth gravity. Somewhere between those facts lies a decades-long story about how the most conceptually mature technology in spaceflight also became the most overlooked.
Let’s start with the physics, because the recipe is insultingly simple. Artificial gravity by rotation is not exotic—it is a formula from school physics: centripetal acceleration, a = ω²r. Spin a capsule around its center of mass, and everything inside gets pressed against the “floor,” which becomes the outer wall. Pick any two of three numbers—radius, revolutions, desired gravity—and you get the third. For one g at a comfortable two revolutions per minute, you need a radius of about 224 meters. For Martian 0.38 g at the same speed—about eighty-five meters. That’s it.
The idea isn’t new, either. Tsiolkovsky was writing about rotation to create gravity at the turn of the century; in 1929, the Slovenian Hermann Potočnik, under the pseudonym Noordung, described a rotating wheel station in detail in a book about the problem of spaceflight; von Braun turned it into an image every American schoolchild in the fifties knew through Collier’s. NASA even built a full-scale prototype in 1960–1962—an inflatable twenty-four-foot torus called the Erectable Torus Manned Space Laboratory—and then shut the project down. Film picked it up faster than engineering did: 2001: A Space Odyssey showed a rotating wheel in 1968—after the real wheel project had already been buried.
A funny detail I calculated myself while working through von Braun’s numbers: his wheel—seventy-six meters in diameter, three revolutions per minute—was supposed, by his own estimates, to provide “about one-third g.” If you actually run the formula a = ω²r, it comes out closer to 0.38—almost exactly Martian gravity. The father of American spaceflight, apparently without stressing the point himself, drew the perfect Martian training rig fourteen years before a human stood on the Moon.
If the recipe is so simple, why zero flights? This is where the second paragraph usually brings up the “witchy” complexities of balancing, and they are real—but the root is deeper and more humiliating: the human inner ear.
Inside our cochlea and vestibular system, endolymph sloshes around—a fluid the body uses to read turns and head tilts. In a rapidly rotating environment, it starts lying: turning your head with the direction of rotation and against it produces different sensations, your vision “swims,” the brain receives conflicting signals from the eyes and the ear—and gives a person the whole package: dizziness, nausea, disorientation. This is not weakness of character, and you cannot Spartan-train your way through it: it is physiology, shared by almost everyone.
Classic Navy and NASA experiments with “slow rotation rooms”—two weeks at three revolutions per minute—showed that people adapt, but not for free. Siegel, writing for Big Think, offers a harsh estimate: for rotation not to torment most people, the structure needs to span more than a kilometer and a half, ideally a full five. For a full g at those speeds, you need a radius of more than two hundred meters—and there’s your answer as to why every piece of concept art shows such a huge station. Right: that’s why the station is so large in every drawing. Not because artists love grandeur, but because two numbers—“comfortable revolutions” and “useful gravity”—dislike each other quadratically: radius grows with the square of the rotation rate.
To be fair, the science of comfort limits is the shakiest part of the story. Three revolutions per minute are agonizing for some test subjects; Lackner and DiZio argued that even ten rpm can be okay with proper adaptation. Carroll adds a subtlety enthusiasts usually miss: all our data on rotation tolerance comes from rotating rooms on Earth, where the axis of rotation coincides with the gravity vector—while in real artificial gravity they are perpendicular, and the sensory experience is different. Humans do not know what they have not tested. And they do not test it because they do not know.
In 1966, NASA and Gemini finally got around to the only in-flight artificial-gravity test in history. On Gemini 11, Pete Conrad and Richard Gordon docked with an Agena target vehicle; during a spacewalk, Gordon tied them together with a thirty-six-meter Dacron line, and the pair began slowly rotating around their shared center of mass—about 0.15 revolutions per minute, one turn roughly every seven minutes. At a radius of around nineteen meters, the Gemini capsule lived under 0.0005 g of artificial gravity—two thousand times weaker than Earth’s. I checked the formula: it matches to the third decimal place. Physics works.
What were the astronauts doing at the time? Nothing noticeable: they felt no physiological effect whatsoever. But they did leave behind the best line in the history of spaceflight. After a camera placed on the panel drifted in a straight line toward the rear wall of the capsule, the crew described the experiment as “interesting and puzzling.” Sixty years later, that is still humanity’s best in-flight result in artificial gravity: “interesting and puzzling.”
The Red Army had a chance to answer in style: in 1965–1966, Sergei Korolev was planning a Voskhod mission tethered to the final stage of the launch vehicle and spun up for an extended period—twenty days of flight, three or four days in artificial gravity, a doctor aboard. Korolev’s death in January 1966 put an end to the project; in savage irony, the experiment was cut first, and then the flight itself was cancelled.
And after that—nothing. Not because anyone signed off on incompetence. History simply turned another way: the Moon consumed all the money and attention, rotating rooms stayed on the ground in laboratories, and in space people began to create artificial gravity by... running. On Skylab, astronauts ran laps through the large compartment—like the jogger in 2001: A Space Odyssey—generating force against the wall. Yes, it helped them warm up, and in that near-prototype of a gravity gradient indoors there was no point: you accelerate, press against the wall, push off.
Six decades later, the first high-level plan for gravity-equipped stations did not emerge at NASA or Roscosmos, but at a private company. Vast—a startup founded by Jed McCaleb, the man who made a fortune on Stellar and earlier walked away from Ripple—was established in 2021 with the mission statement “to develop habitable stations with artificial gravity”—not as a dream, but as the final line of its roadmap.
Proof this is not a presentation bubble: the company has already flown. In November 2025, the Haven Demo technology demonstrator went into orbit (550 kilograms, three months of operation, a controlled deorbit in February 2026)—a perfect display case for the paradigm of “crumb first, cake later.” The launch date for the full Haven-1 station has already slipped: the 2021 release said May 2026, while the company’s current roadmap says 2027. I immediately see a new era of honesty in that: Vast now has flight history, and no image-polishing promises for “last week.” Haven-2 is a modular structure the company explicitly positions as the ISS’s successor: the first module in 2028, four units by 2030, 500 cubic meters of pressurized volume.
But the last line of the roadmap is the one that made me latch onto the company in that digest: by the mid-thirties, a station with artificial gravity, rotating... end over end, nose first, at 3.5 revolutions per minute. Notice how elegantly that number has been chosen: 3.5 rpm is a vestibular-comfort compromise—slightly above the “warm” two revolutions, noticeably below the frightening six. Haven-2 modules join into a dumbbell and spin around their common center of mass—not a wheel anymore, but the classic configuration from the reader’s question to Siegel and Carroll’s plan. I ran the numbers: to get a full g at 3.5 rpm, the dumbbell needs to be about 146 meters long. That is a lot, but by orbital-assembly standards it is a transportable figure, requiring almost nothing new. Martian gravity needs only 55–57 meters—and that is plain orbital mechanics, not science fiction.
Now, the article that started it all. Carroll’s thesis: you do not have to wait for a station to get the first data. After stage separation, Crew Dragon is already close to the situation of “two bodies connected by a structure”: the spacecraft itself plus the spent Falcon 9 second stage, whose masses are in a ratio of roughly 3.5:1. Do not detach the stage. Instead, spin the pair around their common center of mass—and there you have artificial gravity in the cabin where the crew is sitting.
His parameters are calculated scrupulously, and that is the most valuable part. With the stage and spacecraft connected by a rigid structure (no tether in the first stage!), the crew sits four meters from the pair’s center of mass—at three revolutions per minute, that yields about 0.04 g, a weak but steady “push.” With an inflatable compartment in place of the nose cone, the radius grows to seven meters. On a long Dyneema tape tether (150 meters, safety factor of six, only about 30 kilograms of mass), the pull reaches 0.38 g—full Martian conditions. The standard Draco thrusters provide the spin-up thrust; the rotation axis is pointed at the Sun so the solar panels work “by the book.” The test is reversible at any moment: detach the stage, stop the rotation, carry on to the ISS as scheduled.
The simplicity that sets this plan apart from the Gemini story: no docking, no spacewalk, no separate launch—no cost-driving events whatsoever. Any flight with enough RCS margin and spare time—private, Fram2, Axiom, potentially Polaris—can run it as an “add it to the mission” option. And then there is his “survey” at the end of the article: Carroll publicly lists the open engineering questions (will the cabin ceiling hold, where is the solar-panel balance point, will one hundred kilograms of RCS be enough), working simultaneously as an engineer and a PR man for his own field.
What would this test buy first? Not “gravity”—knowledge of vestibular tolerance specifically in an artificial-gravity configuration, where the axis of rotation is perpendicular to the familiar sensory environment of Earth. Thresholds of discomfort, adaptation speed, differences among people. This is precisely the missing parameter without which investors will not believe in stations—and which no rotating room on Earth can provide.
I cross-examined Carroll’s long-range plan in my head, and it does have weak points. First, a rotating pair is sensitive to any movement: crew shifting around inside the cabin changes the moment of inertia and rocks the rotation. Siegel compares it to an unbalanced car wheel, which in orbit becomes a problem for the entire structure. Second, RCS: propellant consumption for holding the pair together, when all these maneuvers also have to avoid wrecking the orbit (one hundred kilograms is an optimistic estimate; ISS phasing is ticking away). Third, solar panels: Dragon’s panels need to remain in sunlight for the entire test—and it is not known in advance whether the rotating pair will keep the panels on one side of the axis, or whether they will cross the plane of rotation. Fourth, and most important: economics. A crew is expensive, mission flight time is expensive, and the buyer for those four hours of rotation is, for now, effectively one entity: NASA research programs with a weakening budget. If private astronauts do not want to go in circles for science, the plan has no customer.
But notice the asymmetry of the risks. If Carroll’s plan fails, we lose—nothing: the mission proceeds as a normal one. If it works, even partially, humanity gets its first flight data on rotation tolerance beyond Earth laboratories. An option with limited downside and structural upside—exactly the kind of mechanism that has been enough to move the history of technology forward, from reusable rockets to gigabit satellite networks.
Finally, the common denominator of this whole story—and one that will be unpleasant for Star Wars fans, where every ship has smart gravity. Of the 74 years since the recipe for artificial gravity became known, humanity spent about 60 simply... not working on it. The reasons were always banal: priorities succeeding one another (the Moon, shuttles, the ISS, “back to the Moon”), the nonlinearly boring fundamental science of the vestibular system (it helps neither ⚔ change the world nor bomb it), and the absence of a screaming, demanding customer.
The catalyst of the past two years is not a scientific breakthrough—there has been none—but the emergence of a business structure for which a rotating station is not research, but a product: a place where people live well rather than survive. Vast is inventing nothing new—the idea of a rotating home was described by Potočnik back in 1929. Let’s repeat that: inventing nothing. It is simply the first person in seventy-four years whose commercial interests require “down” in orbit to be down.
Three things. First: most concepts that seem to us like “technologies of the future” are actually just technologies without a sponsor. Cleanly drawing the line between “technically impossible” and “nobody needed it badly enough to pay for it” may be the most useful mental tool for reading prophecies about the future.
Second: every abandoned technology has its own Gemini 11—some frustrating little island where they tested one layer of the recipe and moved on, untempted. Five hundredths of a percent of g, sixty years before all the current plans—that is funny and historically credible at the same time.
And third: the most inexplicable economic force in the history of technology is one rich person’s vision, binding a company to turn a dream into a line item. Jed McCaleb wanted to live in space with a normal spine—and for the first time in six decades, the most boring of orbital physics has a date on the roadmap and a predecessor that has already come back from orbit. The recipe is 74 years old. At last, there are cooks.