Can SpaceX Solve Its ‘Exploding Starships’ Problem?
After a string of fiery failures, SpaceX’s biggest rocket faces another test flight with sky-high stakes for U.S. space ambitions
SpaceX’s Starship soars through the sky during the rocket’s ninth test flight on May 27, 2025. Both the “Super Heavy” booster and the Starship upper stage suffered spectacular failures before the flight’s completion, marking the rocket’s third unsuccessful test in a row.
“Kaboom” is not the sound you want a rocket ship to make, as a rule. Yet that’s the problem facing the private aerospace company SpaceX and its leader, Elon Musk. Instead of going to space, their newest rocket ship keeps going kaboom.
The last three flights of Starship, a two-stage, 400-foot tall behemoth, ended in fiery disaster—what Musk has sometimes jokingly called a “rapid unplanned disassembly.” In January and again in March the launch vehicle’s Super Heavy booster stage made it back to a massive, pincer-equipped gantry, but Starship’s upper stage didn’t. In May the booster exploded just before splashdown, and Starship broke up spectacularly in the atmosphere, raining debris that commercial aircraft had to dodge. As a bonus, in June the upper stage detonated on the launchpad while Starship was getting fueled for a test firing of its engines. The tally for 2025 thus far is: explosions, four; SpaceX, zero.
Today Starship’s Super Heavy booster and upper stage are on the launchpad yet again. The 10th test flight is scheduled for liftoff on Sunday, circa 7:30 P.M. EDT, from SpaceX’s Starbase launch site in South Texas. If all goes to plan, the booster will use its 33 rocket engines to push the whole shebang to the edge of space, then drop off, somersault, execute a “boost-back burn” and descend to a soft splashdown in the Gulf of Mexico. Meanwhile Starship’s upper stage should be firing its rockets to reach orbit, where it will deploy some cargo before flying itself back down through the atmosphere to its own splashdown about an hour and 15 minutes after launch. “Excitement guaranteed,” a SpaceX announcement promises.
If you’re enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
But there’s excitement and there’s excitement. Look, going to space is hard. It’s even harder to do in the way SpaceX is attempting. “It’s one of the biggest rockets ever. It’s, for sure, the biggest rocket that has attempted reuse,” says Jonathan McDowell, an astrophysicist at the Center for Astrophysics | Harvard & Smithsonian, who tracks space launches in his spare time. “Developing a vehicle this big and launching it repeatedly ain’t easy.”
Starship isn’t just an oligarch’s folly. It’s a launch system meant to revolutionize spaceflight by flying cargo and crews to orbit at a cost that’s almost too cheap to meter. It’s supposed to take NASA astronauts back to the moon and human settlers to Mars. And it represents the kind of gleaming, hardware-forward future that Silicon Valley’s techno-optimists are always promising. Starship is the linchpin of a lot of plans and schemes.
Over the past few months, SpaceX has acknowledged which pieces of the ship broke with each flight but hasn’t gone into any great detail about why. The company didn’t return requests for comment from Scientific American. But SpaceX’s very online rocket-spotting fans—and the half-dozen aerospace engineers I talked to—have been willing to speculate what the problems might be. Mostly, they believe the company has some brilliant people who stand every chance of solving them. But they also wonder what will happen if SpaceX can’t figure out what’s wrong—or, even worse, if some fundamental engineering issue means the idea of a reusable, reliable, workhorse spaceship stays confined to science fiction.
A Starship Super Heavy booster returns to the launchpad during a test flight from SpaceX’s Starbase facility in South Texas on January 16, 2025. The Starship upper stage exploded and was lost during the flight.
Though SpaceX characterizes this differently, Starship had essentially the same types of mishap in all three of the most recent flights—leaks, fires and explosions in the fuel system. On flight seven, there was a flash and then a fire in the unpressurized “attic” below the bottom of Starship’s liquid oxygen tank. On flight eight, that happened near one of the rocket engines. On flight nine, fuel leaked into the nose cone.
That fuel, and the plumbing to move it around, might be the problem. It’s a mix of liquid methane and liquid oxygen—a volatile cryogenic cocktail that’s still, by rocket science standards, experimental. To stay liquid, methane has to be below –259 degrees Fahrenheit (–162 degrees Celsius), and oxygen has to be even colder—below –297 degrees F (–183 degrees C). That means a lot of mechanical effort to keep it cold, to move it around on the ground and on the vehicle and to accommodate it as it shifts from liquid to gas and gets lit on fire. Going back and forth from supercold to hot is called thermal cycling; without careful design and maintenance, almost anything under those conditions will break.
In a Muskian science-fiction future, that’s all worth it. Cryogenic fuel is a pain in the asteroid, but it has more oomph per pound as go juice—what engineers call “specific impulse.” And fuels like methane offer the tantalizing possibility that they could be harvested “in situ” on another world—that they could be synthesized from carbon dioxide and frozen water in Martian regolith or, say, slurped up from the roiling methane seas of Titan. That makes “living off the land” in space seem feasible, even though nobody really knows how to do it yet. “Methane is a new rocket propellant for space launch, so we’re still learning how to do the methane plumbing. The fact that they’ve had leaks, the fact that they’ve had overheating, doesn’t really surprise me,” McDowell says. “It’s a different-sized molecule, bigger than liquid hydrogen but smaller than kerosene, so it leaks differently in different circumstances. Its chemistry is different.”
But the cryogenic chemistry here might be less relevant than cold mathematics. Anything going to space has to carry its own fuel, but that fuel itself has mass. “That’s the tyranny of the rocket equation,” says Hassan Saad Ifti, an aerospace engineer at Texas A&M University, referring to the calculation that vexes every would-be space jockey. “You need to carry more to deliver what you want, but more fuel means more fuel for the fuel.”
Starship’s two stages have to structurally support nearly 11 million pounds of fuel; the upper stage is meant to carry as much as 330,000 pounds of payload. So the vessel itself has to be as light as possible—yet still withstand the buffeting forces of launch and reentry. So far, it has not. “They’ve designed their structure light enough to perform when the rocket ignites and wants to fly, but maybe—and this is speculation—when they’re loading the fuel, that’s causing cracking,” Dec says. “When a structure is cooled, it shrinks. If it’s rigid and can’t move, that’s going to cause a stress, and it’s going to break.”
A couple other pieces of evidence fit this theory. One reason the booster may have survived flights that the upper stage did not is that the booster doesn’t go all the way to space, and it comes back to the ground at only about 4,600 miles per hour. Starship’s upper stage goes all the way to orbit and reaches 17,500 mph. That’s a lot of kinetic energy to get rid of on reentry—usually as heat. “This is the physical constraint,” Ifti says. “We can’t get away from it. We have to manage this energy being generated through heating.”
An early version of Starship tried to bleed off that kinetic energy with a kind of aerodynamic belly flop that ended in a catastrophic loss of control. Now the vehicle uses its control surfaces and rockets to slow its descent and relies on heat-resistant tiles (which, of course, add weight). One persistent critic of SpaceX, Will Lockett, has argued that Starship simply must use more propellant than its builders expected for its return flights, adding even more weight. “This puts incredible pressure on SpaceX to save weight anywhere they possibly can,” . “SpaceX is having to make the rockets too light, resulting in them being fragile, meaning that just the vibrations from operation with a fraction of its expected payload would be enough to destroy the rocket.”
Kaboom.
Maybe this build-test-destroy-rebuild cycle is what you’d expect from a cutting-edge company like SpaceX, which owes much of its astonishing success to iterating like a software start-up. The version of Super Heavy that’s set to launch on Sunday has some major design changes, increasing the size and strength of the winglets called “grid fins” but reducing their number from four to three and aiming for a more controlled, higher angle-of-attack descent. Starship’s upper stage will also test several new kinds of tiles to protect against the ferocious heat of reentry. This is what coders call “agile.”
In practice, though, this Silicon Valley–style approach forces SpaceX to play a very expensive game of Whac-A-Mole. “The way I read what Elon’s trying to do, wow, is it complicated. And when you deal with a very complicated device, there’s multiple modes of failure,” says Joseph Powers, an aerospace engineer at the University of Notre Dame and editor in chief of the Journal of Propulsion and Power. “With a rocket, that almost always results in detonation.”
Each failure is supposed to be an opportunity to learn to avoid disaster the next time. “They’re facing challenges, but I don’t see any showstoppers,” McDowell says. “I don’t want to minimize the problems they’re having. It is embarrassing for SpaceX, and they do have to fix these things, but they are making progress.”
So there’s an easy solution: reduce the weight of the payload Starship can carry and charge more per pound. But even if SpaceX and its customers can absorb the higher price, not all of Starship’s planned missions can necessarily wait for a more reliable spacecraft. NASA’s Artemis III is supposed to use Starship to land astronauts on the moon’s south pole in 2027. That’s practically tomorrow, in aerospace time. Plus, even if you didn’t already think that ionizing radiation and toxic regolith make Musk’s dreams for Mars settlement about as likely as finding canals there, a reduction in Starship’s cargo capacity and rapid reusability would seem to doom the plan. One model for making the trip in three months instead of the usual six or nine requires four cargo Starships and two crew Starships and assumes a total of 45 launches—a mere fraction of the 1,000 Starship launches per year SpaceX foresees.
If you enjoyed this article, I’d like to ask for your support. Scientific American has served as an advocate for science and industry for 180 years, and right now may be the most critical moment in that two-century history.
I’ve been a Scientific American subscriber since I was 12 years old, and it helped shape the way I look at the world. SciAm always educates and delights me, and inspires a sense of awe for our vast, beautiful universe. I hope it does that for you, too.
If you , you help ensure that our coverage is centered on meaningful research and discovery; that we have the resources to report on the decisions that threaten labs across the U.S.; and that we support both budding and working scientists at a time when the value of science itself too often goes unrecognized.
In return, you get essential news, captivating podcasts, brilliant infographics, , must-watch videos, challenging games, and the science world’s best writing and reporting.
There has never been a more important time for us to stand up and show why science matters. I hope you’ll support us in that mission.
Thank you,
David M. Ewalt, Editor in Chief, Scientific American
Source: www.scientificamerican.com