Starship Reaches Orbit With 26 Starlink V3s Amid Engine Faults

A milestone with an asterisk

SpaceX's Starship reached Earth orbit for the first time on September 28, 2026. On its fourteenth integrated flight test, it deployed 26 operational Starlink V3 satellites 1. The vehicle flew on the new Super Heavy V3 booster from Starbase in Boca Chica, Texas 5. This was the first time Starship carried useful cargo to a stable orbit rather than a suborbital trajectory 24.

The headline result is real. So is the fine print. Engine problems on both stages turned a planned showcase into what Ars Technica called a dramatic morning 2. The Register reported that the flight was planned to last about ten hours and six orbits. Instead, SpaceX brought the ship home after three 3.

What went wrong on the ship

During Ship 41's ascent, one of its Raptor engines shut down early. The surviving engines burned longer to make up the lost performance 12. Accounts differ on which engine it was. Wikipedia's flight log says it was one of the three Raptor Vacuum engines 1. Ars Technica says only that one of the ship's six Raptors failed 2. Either way, the ship lost thrust at the most critical point of the flight.

SpaceX first said the spacecraft would not commit to orbit. It reversed that call after reviewing the vehicle's condition 1. That reversal was the key decision of the mission. It shows flight controllers were willing to go ahead with a degraded vehicle once the margins checked out. It also shows that, at first, those margins were not clearly there.

The booster's quieter problems

Super Heavy had its own troubles, and the outlets describe them in different terms. Ars Technica counted two Raptor failures among the booster's 33 engines: one during launch and one during the return maneuvers 2. The Register gave a more detailed breakdown:

• All 33 Raptor 3 engines ignited.
• One engine shut down during ascent.
• For boostback, 26 of 28 planned engines reignited.
• For the high-thrust part of the landing burn, 11 of 13 planned engines reignited before the booster stepped down to five and then three 3.

By The Register's count, the booster missed several planned relights, so "two failures" may understate the problem. Ars Technica noted that the booster issues mattered less for reaching orbit. They still matter for future launches 2. That point is central to SpaceX's business case. A booster that cannot reliably relight its engines cannot reliably land, and without landings there is no rapid reuse.

Why the Starlink payload matters

The July flight, flight 13, also carried Starlink V3 satellites, 20 of them, but it was suborbital 12. Those satellites burned up as they fell back into the atmosphere 2. Flight 13 ended with a booster destroyed in a hard splashdown. The ship did splash down softly in the Indian Ocean 1. Flight 14 is the first time V3 hardware has actually joined the constellation.

This matters because the V3 satellites can only launch on Starship. They are too large for the Falcon 9's payload fairing 2. Starlink's next performance step therefore depends on Starship's progress. The Mobile Internet Resource Center reported that SpaceX plans to offer gigabit speeds to some Starlink customers later this year 4. The same outlet said SpaceX has begun winding down Falcon 9, turning away customers, competitors and even NASA to focus on Starship 4. That raises the stakes on every Starship flight.

Competitive context

The timing helps SpaceX. Rivals Blue Origin in launch and Amazon Leo in satellite broadband are still recovering from major setbacks 4. CNN framed the megarocket as on the verge of paradigm-shifting success 5. The Mobile Internet Resource Center also cautioned that Starship remains far from Falcon 9-style rapid reusability. Its argument is that once Starship can deliver cargo to orbit, every development flight can carry Starlink satellites and do useful work 4.

The reading

The most accurate reading is that flight 14 proved capability, not reliability. Starship showed it can reach orbit and deliver payload even after losing an engine. That kind of engine-out tolerance is valuable. But the ship needed a mid-flight reassessment to commit to orbit, the mission was shortened by half, and the booster missed several planned relights. On this evidence, the Raptor fleet is not yet dependable.

SpaceX is retiring Falcon 9 and tying Starlink's upgrade path to Starship. That leaves the company dependent on a vehicle that is still working through engine anomalies. The company has turned that kind of risk into progress before. Still, the coming flights will matter more for whether the engine failures go away than for any new firsts. Until they do, each Starlink V3 batch is a test flight that happens to carry revenue.