SpaceX has long sold spacecraft as a fully and quickly reusable rocket designed to deliver thousands of pounds of cargo to Mars and have frequent lifespans. However, large-scale reusability means a spacecraft that can withstand disasters and faults. So a single obstacle does not spell out the catastrophe of the end of the mission.
The 10th Test Flight on Tuesday evening showed SpaceX focused on fault tolerance. In a post-flight update, SpaceX said the tests highlighted “limits in vehicle capabilities.” Understanding these edges is important to the company’s plan to ultimately use Starship to launch Starlink satellites, commercial payloads, and ultimately astronauts.
SpaceX did more than reaching a new milestone when the massive spacecraft rocket was lifted on its 10th test flight on Tuesday evening. It intentionally introduced several faults to test the heat shielding, propulsion redundancy, and re-illumination of its Raptor engine.
Heat shielding is one of the toughest engineering challenges SpaceX faces. As Elon Musk admitted in X in May 2024, reusable orbital return heat shields are the “biggest remaining problem” for 100% rocket reusability.
The belly of the upper stage, also known as the spacecraft, is covered with thousands of hexagonal ceramic and metal tiles that make up the heat shield.
Flight 10 was to learn how much damage the ship could accept and survive when it passed through air heating. During the 10th test, the engineers intentionally removed tiles from several sections of the ship, experimenting with new types of actively cooled tiles to collect actual data and refine the design.
Space Shuttle Columbia provided an unwelcome lesson in 2003 regarding the vulnerability of thermal shields. Some of the insulation foam hit the heat tiles on the left wing of Colombia.
TechCrunch Events
San Francisco
|
October 27th-29th, 2025
22 years later, SpaceX is focusing on performance mapping, even in the worst-case scenarios. If post-flight data indicates that the ship remains within the expected temperature margin, it moves the company towards its ultimate goal of regenerating and reusing the stage upright.
We also tested propulsion redundancy. The configuration of the Super Heavy Booster’s Landing Burn looked like a rehearsal of an engine failure. The engineer intentionally disabled one of the three Center Raptor engines at the final stage of the burn, and instead used the backup engine. This successfully rehearsed the engine out event.
Finally, SpaceX reported a re-light in the space of the Raptor engine. This is explained in the second broadcast that SpaceX has achieved this. A reliable engine reboot is required for deep space missions, propellant transfers, and perhaps some payload deployment missions.
NASA’s Artemis programme relies on SpaceX, which develops heat shields that can withstand reentry and a vessel that can regain reconstruction in orbit to safely deliver astronauts to the surface of the moon. The agency awarded SpaceX to just over $4 billion for the version of the spaceship that can land on the moon. The first spacecraft’s lunar landing is currently scheduled for mid-2027.
NASA accepts that, according to the mission profile, it calibrates the risks differently, and that the risk of irregular service missions is high and the risk of crew transport is very low. The agency will set quantitative safety targets that need to be demonstrated through tests and flight data before placing the astronaut on a new rocket. These levels do not change for the spacecraft just because they are bigger rockets, but they do mean potential fault modes.
Looking together, these experiments show that SpaceX is testing with these standards in mind. The company will introduce more changes in the next version of Starship, called Block 3, including Hailluc Raptor Engine, Flap Upgrades, Avionics and Guidance, Navigation and Control System Updates.
The next step is to convert Flight 10 data into future hardware upgrades and get closer to the days that Musk imagined, “Starship will be released more than 24 times in 24 hours.”
