Space & Technology

SpaceX Starship Flight 13 Aborts at the Last Second — What Went Wrong

On July 16, 2026, the world's most powerful rocket sat on its launch pad in South Texas, all 33 Raptor engines beginning their ignition sequence — then, at T-0, the computer called a halt. The scrub delayed Starship's first attempt to deploy satellites and marked another setback for a vehicle that has yet to complete a fully successful orbital mission.

• SpaceX's Starship Flight 13 was aborted at the last second on July 16, 2026, when several of the 33 Raptor engines on the Super Heavy booster failed to ignite, triggering an automatic shutdown. The vehicle remained intact on the pad and propellant was safely offloaded.
• This was Starship's first flight attempt since SpaceX's blockbuster June 2026 IPO, which raised an estimated $86 billion. The stock closed below its IPO price for the first time following the scrub, reflecting the high market stakes attached to Starship's progress.
• Flight 13 was originally planned to deploy 20 Starlink V3 satellites — the first operational payload ever carried by Starship — and test a single-engine Raptor relight in space, a crucial step toward orbital reuse.

At approximately 5:45 p.m. Central Time on July 16, the 33 Raptor engines of the Super Heavy booster began their pre-launch ignition sequence at SpaceX's Starbase facility in Boca Chica, Texas. Unlike a traditional countdown, where all engines must confirm stable combustion before the hold-down clamps release, several engines failed to reach proper operating condition. An automatic abort command fired before any vehicle movement occurred, and the launch was called off for the day.

The scrub came just weeks after SpaceX's highly publicized initial public offering in June 2026, which valued the company at over $250 billion and raised roughly $86 billion from public investors. The stock market reacted immediately: shares of SpaceX fell about 3% in aftermarket trading, closing below the $135 IPO price for the first time. Analysts noted that the market had priced in a successful flight, and the abort reset expectations for Starship's operational timeline.

Flight 13 was not just another test. It was the first Starship mission intended to carry a real payload: 20 next-generation Starlink V3 satellites designed to provide high-speed internet to rural and remote regions. The V3 satellites are larger and more capable than previous Starlink versions, and their deployment from Starship's payload bay would have been a major validation of the vehicle's commercial utility. The mission also aimed to perform a Raptor engine relight in orbit — a capability essential for future missions that require orbital maneuvering or deorbit burns.

SpaceX has not yet announced a new launch date. Engineers will need to investigate why the engines failed to ignite, a process that mirrors the troubleshooting after Starship's previous test in May 2026, which ended in the vehicle's loss during reentry. The company has described the prior failure as the result of "interconnected causes" and introduced multiple fixes to the Raptor engine design. Flight 13's abort suggests that some of those fixes may require further refinement.

For context, Starship is by far the largest rocket ever built — standing 121 meters tall and generating more thrust than the Saturn V that took humans to the Moon. The complexity of synchronizing 33 Raptor engines during startup is unmatched in rocketry, and engine ignition failures have been a recurring challenge across the program's test campaign. Each scrub, while frustrating, provides engineers with valuable data on edge cases that only real hardware can reveal.

The broader picture is that Starship's development is following a pattern familiar from Falcon 9's early years: iterative testing, frequent failures, gradual reliability improvements. Falcon 9's first successful landing took five attempts; today it has landed more than 300 times. Starship's path to routine reusability is likely to be longer, but the trajectory remains consistent with SpaceX's engineering philosophy of test, fail, fix, repeat.