Could Liquid Air Refill the Roadster’s Thrusters in 15 Minutes?

I absolutely love Tesla. I love Elon Musk's willingness to look at something everyone else considers settled and ask, ‘But what if we did something completely ridiculous instead?’ The Roadster having literal rocket thrusters is perhaps the purest expression of that philosophy. There are sensible electric sports cars. And then there's a car that apparently wants to become briefly airborne.

So when I read that the Roadster's cold-gas thrusters could take hours to refill, and that Tesla was aiming to get that down to around 30 minutes, my brain supplied a profoundly unqualified suggestion: What if you just poured liquid air into it?

To establish my credentials: I build SaaS products. I have precisely zero qualifications in rocket propulsion, cryogenics or high-pressure storage systems. Tesla has actual rocket scientists. I strongly suspect someone at SpaceX thought of this roughly seven seconds after the concept of thrusters came up. But that has never stopped an amateur armed with a calculator and dangerous levels of enthusiasm.

A quarter of a tonne of air. In a car. Naturally.

The reported Roadster Rocket configuration has two pressure vessels totaling around 600 litres, storing something like 250 kg of compressed gas. There have also been references to a 10,000 psi vessel rating, which isn't necessarily the actual operating pressure. Either way, this is a fairly absurd amount of pressure to have in a road car, and I say that with admiration.

Now make the air very, very cold. Liquid air is roughly 870 kg per cubic metre around its boiling point. So 250 kg occupies about 290 litres as a liquid. The back-of-the-envelope arithmetic is delightfully simple: that's less than half the stated combined vessel volume.

Of course, those vessels would need to be specially designed for cryogenic use. You cannot take an ordinary warm-gas pressure tank and casually introduce liquid at around minus 190 degrees Celsius (minus 310 degrees Fahrenheit). But this is a hypothetical car with SpaceX-designed thrusters, so I feel we're already allowing ourselves a little engineering ambition.

The bit where we pour it in

Suppose a purpose-built station could transfer cryogenic liquid at 30–50 litres per minute. That isn't a claim about a real Tesla refilling system; it's just the flow rate for our imaginary setup. At 50 litres per minute, 290 litres takes less than six minutes. At 30 litres per minute, just under ten.

Add some time for connections, chilling the lines and making sure nothing explodes, and the actual transfer could potentially be done in minutes.

There is something wonderfully daft about a car that can supposedly accelerate with rockets being refilled using what looks, in principle, like an industrial-scale version of pouring a very cold drink.

Fine. But now you have a car full of extremely cold liquid.

This is where the physics sends an invoice.

The liquid needs energy to vaporize and warm up, and the pressure needs to be managed properly. As a loose order-of-magnitude calculation, bringing 250 kg of a nitrogen-like cryogenic fluid towards room temperature involves around 100 megajoules, roughly 28 kWh, of thermal energy. That's a useful mental model, not an exact constant-volume tank calculation.

If we imagine a station capable of delivering heat effectively and safely, a ridiculously simplified energy-divided-by-power calculation looks like this:

• 50 kW: about 33 minutes of idealized heating.
• 100 kW: about 17 minutes.
• 150 kW: about 11 minutes.
• 200 kW: about 8 minutes.

I can already hear an actual cryogenic engineer shouting that you cannot just put 200 kW into a pressurized carbon-composite vessel and call it a day. Quite right. Heat transfer, tank gradients, pressure evolution, insulation, controls and relief systems are all rather more complicated. I am not suggesting Elon install four industrial hairdryers in the boot.

But the point of the daydream is that the energy could, in principle, come from an external refilling station rather than a compressor on the car. If you can engineer the hardware and thermal management, the 15–25 minute neighborhood starts to look like a question worth exploring, not a number plucked entirely from the ether.

And perhaps you don't even have to wait until it's completely warm

Here's the part I find most intriguing. The thrusters don't necessarily need every last kilogram of propellant sitting at room temperature before they can do anything. If the feed system reaches a useful pressure and mass flow earlier in the warm-up, partial thruster capability might be possible.

Think less of an instant switch from ‘off’ to ‘ready to impersonate a Falcon 9’ and more of a progressively increasing capability as the system reaches its intended operating state.

To be clear, I cannot tell you at what minute or pressure that happens. It depends on the real fluid, fill fraction, temperature, tank design and nozzle system. Cryogenic liquid does not magically turn into 10,000-psi gas just because we want a dramatic conclusion. But the distinction between some usable thrust and a full warm charge makes the refill-time question more interesting.

The catches, because even my imaginary rocket car has them

You would need a pressure vessel that genuinely tolerates cryogenic temperatures and repeated cycles, specialized connectors, carefully controlled warming, and a whole lot of safety engineering. Liquid air also brings oxygen-enrichment and changing-mixture hazards; liquid nitrogen might actually be the more sensible candidate if the thruster architecture allows it. None of this is something to attempt on a normal tank, or at home, unless your garage is staffed entirely by qualified rocket engineers.

Then there's cost. Cryogenic liquid is an industrial commodity, but the total price of a refill station includes storage, delivery, boil-off, equipment and extraordinarily low utilization if the average Roadster owner fires the thrusters once to demonstrate them to a terrified passenger and then spends six months telling the story.

The economics could still be manageable for an extravagant halo car. But I'm not pretending we've produced a reliable dollar-per-refill quote from a few litres, kilograms and kilowatts.

Would Tesla actually do it?

Maybe not. A clever high-pressure gas station with buffer storage and proper cooling might be much simpler, lighter and cheaper. Tesla's actual engineers may already have evaluated cryogenic refilling, rejected it for ten excellent reasons, and moved on to something better.

And frankly, I hope they do find the best solution. This is not a dunk on Tesla or Elon. It's precisely the opposite. I love the ambition of building a Roadster so hilariously capable that a discussion about refilling its rocket thrusters is even possible.

Nor do I imagine millions of drivers will be stopping at their local liquid-air forecourt on the school run. How frequently is anyone realistically going to use this feature? Most owners might deploy it a handful of times, or perhaps simply enjoy knowing the car could do something borderline absurd. That's half the charm.

My thoroughly unqualified verdict

Could rapid liquid transfer plus an engineered warm-up system help Tesla reach its targeted 30-minute refill time? I think it's a plausible avenue to explore. Could it reliably produce a fully usable refill in 15 minutes? I have absolutely not proved that. But the rough numbers make it fun enough to ask the people who actually know what they're doing.

I would love to know whether the SpaceX team has modeled it. I assume they have. They are, after all, rocket scientists, whereas my professional expertise is making SaaS applications and wondering why the button padding looks wrong on mobile.

Still, I rather like the idea that the solution to refilling the world's most ridiculous electric sports car might be: make the air very cold, put it in quickly, and persuade it to warm up on schedule.

Simple, really. Apart from the entire field of cryogenic engineering.