Starship V3 delayed, engine drama

SpaceX has slipped the first flight of Starship V3 into May 2026. Elon Musk said on April 3 that the “next flight of Starship and first flight of V3 ship & booster is 4 to 6 weeks away,” which moved the public target from April into early or mid-May after weeks of expectations that the new vehicle might fly sooner (teslarati.com, techgenyz.com). That sounds like a small schedule slip. It is not. V3 is supposed to be the version that turns Starship from a giant prototype into something closer to a working transport system. That is why the engine matters so much. SpaceX says Starship is meant to carry up to 150 metric tons fully reusable, and the company’s own updates say it has already logged more than 40,000 seconds of run time on the next-generation Raptor 3 engine (spacex.com, spacex.com). V3 is built around that engine and around a larger redesign of both ship and booster. The promise is more thrust, simpler plumbing, faster reuse, and eventually the launch rate SpaceX needs for Starlink, NASA’s lunar plans, and Musk’s Mars pitch. The problem is that engines do not care about promises. This week, one of them caught fire. Gizmodo reported that a Raptor test at SpaceX’s McGregor, Texas, site on April 7 ended in a large fire during V3-related engine work, with video showing a dramatic blast at the stand (gizmodo.com). McGregor is where SpaceX pushes engines hard before they ever reach the pad. A failure there is better than a failure in flight. It is still a failure in the exact hardware family that is now pacing the whole program. The trouble is not confined to McGregor. At Starbase in Boca Chica, where SpaceX assembles and tests the full vehicle, a separate rapid unscheduled disassembly was captured on camera on April 6, according to Basenor’s report on the incident (basenor.com). SpaceX has not publicly explained in detail what broke. That gap matters. A pad anomaly can mean anything from a sacrificed test article to damage that ripples through the launch flow. When a company is trying to debut a new rocket generation, even a small mystery becomes a schedule problem. The backdrop here is a program that has already been living on test data and breakage. SpaceX’s last full Starship integrated flight test, Flight 11, flew on October 13, 2025. The company said that mission reached planned velocity, deployed eight Starlink simulators, and completed an in-space Raptor relight before reentry and splashdown (spacex.com). Since then, SpaceX has been doing what it always does between flights: swapping hardware, changing designs, and trying to learn faster than the vehicle fails. In March, the company said Ship 39 had completed cryoproof operations as “the first campaign with a next-generation Starship V3,” which showed that the new architecture had at least reached serious ground-test territory (space.com). That still leaves the oldest Starship question untouched. Can SpaceX turn constant redesign into a reliable rocket before the redesign itself becomes the bottleneck? The FAA’s Boca Chica Starship license remains current through 2028, so regulation is not the obvious blocker here (faa.gov, faa.gov). The blocker looks more mechanical than bureaucratic. SpaceX is trying to field a new ship, a new booster, and a more mature Raptor at the same time. On April 7, one of those engines burned on a Texas test stand. On April 6, something else at Starbase came apart hard enough to earn the program’s favorite euphemism.