“I’m not surprised that they are having these issues,” says Jonathan McDowell, an astronomer and astrophysicist at the Harvard-Smithsonian Centre for Astrophysics in Cambridge, Massachusetts. That doesn’t mean those issues will perpetuate forever, however. “They will probably eventually get Starship working, but it is going to take a lot longer than they are thinking, because it is such a step forward,” he says.
There are multiple reasons, but they come down to two main areas. One is Starship’s massive size – scaling the systems up to such a level, then dealing with the physics that result, can be challenging. Starship is 10 metres taller than the Saturn V, the previous record-holder for tallest rocket, and more than twice as tall as the Space Shuttle stack.
At lift-off, mass is about 5,000 tonnes, nearly double the Saturn V's 2,900 tonnes and more than twice the Space Shuttle's stack. Both of those cause engineering issues that need to be consistently resolved. “Scaling these things up is never trivial, even if you're using exactly the same technology,” says McDowell.
Along with the enormous scale is the fact that Starship layers new technologies on systems seen in prior generations of spacecraft. One particularly thorny new contribution is the methane and liquid oxygen system that the vehicle uses for propulsion. One reason that we may have seen so many fiery failures of Starship in recent months is difficulty managing those ‘methalox’ (methane and liquid oxygen) molecules as a fuel source for the Raptor engines, compared to the kerosene or hydrogen and liquid oxygen fuel sources that other rockets before it have often used.
“Methane is a different size molecule from either liquid hydrogen or kerosene,” says McDowell. “And so it's going to get through different sized, tiny holes. It doesn't surprise me that they've had leak issues and plumbing issues.”
The vastly different fuel source is a headache for SpaceX engineers, reckons McDowell. “There's plumbing physics and combustion physics [to solve],” he explains. One is the physics of how things leak and how to prevent that happening, and the other is managing fires or other issues when those gases ignite within the rocket’s engines.
The design of Starship is also something that engineers have had to grapple with. It has been widely reported that the design has had issues with vibrations – particularly ‘harmonic response’, where vibrations at certain frequencies resonate with the rocket's structure, amplifying stresses and causing hardware failures. That can cascade down in a series of errors, proving fatal to the flight’s ability to take off successfully.
In January 2025 at Starship’s seventh test flight, for instance, the vessel faced vibrations several times stronger than it had faced during ground testing, which triggered the harmonic response. That stressed the rocket’s propulsion system, causing propellant leaks in the part of Starship known as the ‘attic’, an unpressurised section between the bottom of the liquid oxygen tank and the aft heat shield. Those leaks overwhelmed the venting system, resulting in sustained fires that led to the destruction of the vehicle.
Unwanted vibrations are not unique to Starship. A similar problem called ‘pogo oscillations’ plagued rockets such as the Saturn V. But because of Starship’s enormous scale, any vibrations can compound in intensity, and therefore result in more damage.
To try to counter this, SpaceX has said it will add more vents and a new nitrogen purge system to the attic to reduce the risk of combustion there, while in the long term they plan to move to a newer Raptor 3 engine, while simultaneously reducing the number of leak-prone joints across the whole vehicle.
The recent higher rate of failure is not necessarily an indication that things are going wrong, says McDowell. “This was always going to be a hard development,” he says. The fact that the cause of failure is different with each launch should be seen as a positive. “They’re not repeating failures again and again,” he says. “It’s like debugging code: you get rid of a bug, and then you get rid of another bug, and so on. Except it's a lot more expensive and spectacular – but I understand the process, as a software guy.”
Provided they continue to find the funds for the rockets, McDowell feels confident that SpaceX will eventually solve the problems they are encountering. “It feels like you’re not making progress because you keep blowing up,” he says. “But in fact, as long as it doesn’t go wrong the same way twice, you are making progress.”
Want the best engineering stories delivered straight to your inbox? The Professional Engineering newsletter gives you vital updates on the most cutting-edge engineering and exciting new job opportunities. To sign up, click here.
Content published by Professional Engineering does not necessarily represent the views of the Institution of Mechanical Engineers.