Tesla Roadster 2: How Elon Musk Made It Fly

On May 9, 2026, Elon Musk offered the clearest explanation yet of how the Tesla Roadster 2 will be able to fly. The optional "Rocket" package, as Tesla calls it, doesn't rely on conventional combustion. Instead, ten SpaceX-built cold-gas thrusters will use 600 liters of compressed air to shove the car into the air and blast it from 0 to 60 mph in about one second. It's not a full-flying machine in the sci-fi sense, but it is a real, functioning hover-capable supercar.

The trick is Newton's Third Law. For every action there's an equal and opposite reaction. The Roadster's thrusters release high-pressure air downward or backward. The car moves upward or forward in response. Because there's no combustion, there's no flame and no hot exhaust. The thrusters are basically high-power versions of the small maneuvering jets used on satellites and spacecraft.

Cold-gas thrusters have long been used in aerospace, mainly for precise attitude control on satellites. In that role, they only need to nudge a few hundred kilograms or less. Putting them on a car is a much bigger challenge. To lift a vehicle that weighs over 1,700 kg, even for a few seconds, the tanks need to feed a huge volume of compressed air through specialized nozzles. SpaceX's own RCS systems aren't exactly off-the-shelf parts, but the design principle is proven.

Where does the compressed air come from? Tesla hasn't published a full spec sheet, but the hint is 600 liters. That's a large but manageable tank array. At high pressure, that air could produce enough thrust for a short hover, a few low-speed hops, and a significant boost during launch. Beyond that, the Roadster's main electric drivetrain takes over. The thrusters are best understood as a launch assist system, not a flight control system for cruising in the sky.

And that's the key detail many people overlook. The Roadster 2 can "fly" in the sense that it can leave the ground and move through the air for a short period, but it's not designed to climb to altitude, navigate, or land like an airplane. Musk has called it "a car that can cover the distance of a small gap" and "hover at relatively high heights." In practice, owners would use it for dramatic pull-aways, vertical launches, or clearing obstacles on special tracks. It's not a commuter aircraft.

Still, the engineering itself is remarkable. Cold-gas thrusters are specifically appealing because they are simple and safe. No fuel storage, no heat shields, and no risk of a pyroclastic explosion. Pressurized air is inert, and even a catastrophic leak would not produce a fireball. That makes the Rocket package easier to certify than a real jet engine, though the certification process for a car that can briefly lift off is still uncharted territory.

Regulators are watching. The NHTSA and the FAA have not published specific rules for ground-effect, short-hop cars. Tesla has not applied for an airworthiness certificate, and it may not need one if the Roadster stays below a certain altitude or remains under a certain number of occupants. But if the car can jump, law enforcement agencies are likely to ask where it is legal to use that feature. Musk has acknowledged that "a lot of legal nuance" remains.

There's also a practical question: price. The original Roadster was announced with a $200,000 price tag and a $250,000 Founders Series. The Rocket package, if ever offered, is likely to be a six-figure upgrade by itself. That would place the car in the same price range as a private hovercraft, but with far less range. The Roadster's battery, meanwhile, is still expected to deliver around 620 miles on a single charge, which is more than enough for a day at a track or a road trip.

Does this make the Roadster a "flying car" in the way most people imagine? Not really. There is no wing deployment, no rotor, and no long-duration lift. The phrase "flying car" is doing a lot of marketing work here. But the technology is real, and the physics is sound. A heavily loaded compressed-air system can absolutely move a vehicle off the ground. Whether that system makes sense for production, insurance, and everyday driving is another matter entirely.

For now, Tesla's official Roadster page still simply says "Coming Soon." The production timeline has been pushed back multiple times since the original unveiling in 2017. Musk has said the Roadster 2 is designed as the fastest car ever built, and the Rocket package is the reason. "No one has ever made a car like this," he said during the latest presentation. That much is true.

What remains to be seen is whether the flying Roadster is a commercial product or a halo experiment. Building a car that can briefly leave the ground is hard. Building ten thousand of them, with the same reliability and safety standards as a normal car, is harder. Cold-gas thrusters are durable and predictable, but they need servicing, monitoring, and a network of high-pressure filling stations, none of which exists for consumers.

Aerospace engineers who work with cold-gas systems point out that the technology is not new. The novelty here is scale. A typical spacecraft RCS thruster produces a few newtons to a few hundred newtons. A Roadster would need hundreds of kilo-newtons of thrust, even for a quick hop. That requires extreme tank pressure and fast-acting valves. The fracture mechanics alone — keeping a 600-liter pressure vessel safe at thousands of psi inside a crash structure — is a serious engineering hurdle.

Tesla has solved harder problems before. The Roadster's battery pack, for example, is designed to survive severe impacts. The difference is that a pressure vessel stores mechanical energy, and if it fails, the release of that energy is violent. Musk's team will need to prove that the system can survive a crash without creating a projectile or shrapnel hazard. That will be a central question in any anti-explosion safety review.

There is also the question of thrust vectoring. To control the car's orientation while airborne, the 10 thrusters have to be arranged carefully. Some fire downward to lift, some backward to accelerate, and some forward to brake. Musk's description of "10 SpaceX thrusters" suggests a panel of nozzles around the vehicle's underbody and rear diffuser area. Steering in the air, even for a split second, requires an on-board computer with gyroscopes and fast-response valves. That computer, of course, is not unlike a spacecraft's reaction control system.

The Roadster 2, then, is a perfect illustration of Musk's approach: take aerospace hardware, bolt it onto an electric car, and let the spectacle handle the publicity. It may not lead to air travel for everyone, but it could push regulators, insurers, and engineers to take short-duration vehicle flight more seriously.

In the end, the most honest answer to "How did Elon Musk make a Tesla car fly?" is: he used pressurized air and Newton's Third Law, and he kept the flying to a few seconds. It's a gimmick, for now. It's also an engineering proof of concept. The first time a Roadster owner launches off a driveway and lands 20 feet away, the question will no longer be whether it can happen, but whether it's legal.

That moment, by all signs, is still at least a full production cycle away. But the wait just got shorter.