At 4:48 p.m. CST on November 13, 2026 (5:48 p.m. EST, 22:48 UTC), SpaceX's Starship lifted off on its 15th flight from Orbital Launch Pad 2 at Starbase, Texas. The main goal was to catch both the first-stage Super Heavy Booster 22 and the second-stage Ship 42 with the launch tower's mechanical arms, proving full recovery of an orbital-class rocket on a single flight for the first time. About 6.5 minutes after liftoff, however, Booster 22, which was supposed to return to the launch site, was diverted toward the Gulf of Mexico and exploded before reaching the water. Ship 42, meanwhile, reached low Earth orbit and released 26 next-generation Starlink V3 satellites. The booster's loss pushes back the simultaneous catch, but the next turning point is whether Ship 42, now on an orbital coast of about 22.5 hours, can make its first return to the launch tower.
Starship Flight 15's first full-recovery attempt: mixed results
In conventional orbital rocket development, the second stage has typically been expendable, or only the first stage has been returned to a landing zone at sea or on land and then transported back. SpaceX's stated goal is different: to return both stages directly to the launch tower at the launch site and catch them in midair with the tower's mechanical arms, known as "chopsticks." Because the vehicles carry no heavy landing legs and the ground equipment grabs them instead, the turnaround to reuse can be shortened considerably. As SpaceX made clear in its mission overview, catching both stages on the same launch would have been the first full recovery of an orbital launch system.
The vehicles flown on Flight 15 were Super Heavy Booster 22 (hereafter Booster 22) and the upper stage Ship 42. They were built as the fourth flight pair of SpaceX's latest Version 3 (Block 3) hardware. Version 3 updates the vehicle structure and propulsion systems, and its operational requirements are a step beyond those of the early test vehicles. Earlier flight tests pursued recovery experiments with early-generation boosters and orbital-insertion tests separately. Flight 15 was the first attempt to combine simultaneous tower recovery of both stages with the orbital delivery of operational satellites in one mission, using the latest Version 3 vehicles.
On launch day, two different outcomes were written within just a few minutes. After leaving Orbital Launch Pad 2 (hereafter Pad 2), Booster 22 pushed the upper stage toward space. It then veered off its return course to Pad 2, where launch and recovery were meant to be completed at the same site, was guided out over the Gulf of Mexico, and was lost in an explosion. By contrast, after separation Ship 42 continued accelerating toward orbit, reached its target altitude, and completed its satellite deployment mission on schedule. The original goal of recovering the whole system on one flight was lost with the first stage, but because the upper stage's on-orbit operations survived, the mission ended up neither a complete failure nor a complete success.
Comparing the original plan for each stage with what had been confirmed by the time of orbital insertion shows a contrast: a recovery barrier tied to the first stage's hardware update, and a second stage that has reached a more mature level of orbital operations.
| Stage and vehicle | Vehicle version and track record | Main objective on Flight 15 | Confirmed result and current status |
|---|---|---|---|
| Super Heavy Booster 22 (first stage) | Fourth Version 3 (Block 3) flight pair. Successful Version 3 booster recoveries: 0 (Version 1: 3 boosters caught, 2 of them reused) | Return to Pad 2 at Starbase about 6.5 minutes after launch and be caught by the tower's mechanical arms | The catch was called off for reasons not disclosed, and the booster was diverted toward the Gulf of Mexico. It exploded just before reaching the water at about T+6.5 minutes |
| Ship 42 (second stage, upper stage) | Fourth Version 3 (Block 3) flight pair. All 14 previous Ships ended their flights at sea; successful recoveries: 0 | Be placed in low Earth orbit, deploy 26 Starlink V3 satellites, then attempt the first tower catch after an orbital coast of about 22.5 hours | Reached orbit at an altitude of about 275 km and completed deployment of all 26 satellites. Now coasting for up to 14 orbits, with return scheduled for the afternoon of November 14 |
As the table shows, the loss of Booster 22 cost Flight 15 the record of full recovery on a single flight, but the practical side of operating a space launcher, orbital insertion of the upper stage and payload release, went as planned. To gauge where the Version 3 vehicles stand, it is therefore necessary to separate clearly the investigation into the first stage already lost at sea from the outcome of the second stage's return, which is still to come.
Why only the Version 3 booster was kept from returning to the tower
Booster 22's launch began with the simultaneous firing of its new engines. The 33 Raptor 3 engines at the base of the vehicle ignited together, and the booster rose vertically from Pad 2 with more than 18 million pounds (about 8,165 tonnes-force) of thrust. When so many engines are bundled together, slight differences in the combustion pressure or thrust of individual engines directly affect the attitude and vibration of the whole vehicle, so extremely sophisticated control is required throughout the climb. From liftoff to separation, Booster 22 fulfilled its role as the propulsion stage that carries the upper stage to its planned speed and altitude without trouble.
About 2.5 minutes after launch, the flight reached its biggest turning point, the separation of the first and second stages. The method used was hot staging. In conventional multistage rockets, the first-stage engines are shut down completely and the stage is jettisoned, and the second stage is ignited only after a safe distance has been secured. With Starship, by contrast, almost simultaneously with Booster 22's main engine cutoff (MECO), the still-attached upper stage Ship 42 ignites its own Raptor engines and pushes the booster away with the pressure of its exhaust as it separates. Because thrust is not interrupted at the moment of separation, the stage avoids the problem of propellant floating up in the tanks in weightlessness and interrupting the supply, and can accelerate the upper stage more efficiently. On the other hand, the hot exhaust of the second stage blasts directly onto the top of the first stage, so after separation the first stage must endure heat loads and attitude disturbances while setting itself up for its return orientation.
Having completed separation, Booster 22 began turning its course back toward Pad 2 at Starbase, where it was launched. In the procedure anticipated by the flight plan, the booster would glide toward the launch site while reentering the atmosphere, slow down by relighting its engines, and settle into the tower's mechanical arms about 6.5 minutes after launch. The whole sequence, in which a giant booster that lifted off from the pad returns minutes later to the narrow space of the same tower, hangs motionless in the air, and is caught by the arms, works only when the vehicle's guidance and control and the ground-side readiness mesh with no margin for error.
As the scheduled landing time approached, however, SpaceX decided not to attempt the tower catch. The company switched Booster 22's guidance target from the tower to the open waters of the Gulf of Mexico, diverting it to a splashdown at sea away from ground facilities. At about 6.5 minutes after launch (T+6.5 minutes), Booster 22, now heading out over the sea, exploded in midair just before reaching the surface. After a violent explosion the vehicle momentarily disappeared from view, but immediately afterward some of the wreckage was seen rising again from the surface as propellant remaining inside the vehicle continued to burn. Damage to the launch tower and ground infrastructure was avoided, but the vehicle itself was unrecoverable.
One fact that must be kept strictly separate in assessing the Flight 15 results is that, as of this report, the specific cause of Booster 22 abandoning the catch and exploding has not been identified at all. In the launch broadcast and initial reports, SpaceX did not state why it called off the tower catch. Whether there was an anomaly in the relight procedure of the 33 Raptor 3 engines, whether the guidance software judged that safety criteria for approaching the tower were not met, or whether damage occurred to the vehicle structure or thermal protection, and in which area (machinery, software or structure) the cause lies, has not been disclosed. What is confirmed at this stage is only that the company aborted the return to the tower for safety reasons and diverted to the sea, and the observation that Booster 22 exploded just before splashdown.
Still, set against earlier vehicle generations, it is clear that Version 3 boosters face a barrier of their own in recovery. SpaceX has so far succeeded three times in catching the older Version 1 Super Heavy booster with the tower's mechanical arms, and has actually flown two of them again. The concept of catching a first stage at the tower had thus already been technically shown to be feasible with Version 1. Yet the redesigned Version 3 Super Heavy, including Booster 22's failure, has never succeeded in returning to the tower. With propulsion and vehicle design having moved from Version 1 to Version 3, establishing the reliability to bring a large booster safely back to the tower remains an unconquered challenge for the next-generation vehicle.
Two burns and a 42-minute deployment: where orbital operations stand
Even after Booster 22 disappeared into the waters of the Gulf of Mexico, the upper stage Ship 42 kept accelerating toward space. Although attention tends to gather on the dramatic scene of the first stage's destruction, what ultimately determines a rocket's value is whether it can reliably deliver its cargo to the target orbit. Ship 42's flight on Flight 15 confirms that Starship is moving beyond its early test flights toward an orbital vehicle that can handle practical satellite delivery.
To enter a stable low Earth orbit, Ship 42 used a two-burn orbital insertion method. After separating from the first stage by hot staging, Ship 42 kept its Raptor engines running and finished its first ascent burn a little after the eight-minute mark. Immediately after that first burn ended, however, the vehicle remained on a trajectory that would pull it back into the atmosphere under Earth's gravity, and it could not keep circling the planet stably. So at 26 minutes 10 seconds after launch (T+26:10), Ship 42 performed a second engine burn of about 20 seconds. This orbital insertion burn adjusted its speed and orbit in space, placing the vehicle precisely into low Earth orbit at an altitude of about 275 km (about 170 miles).
Reaching a full orbit around Earth after two burns carries great significance in the history of the Starship program. Before the previous Flight 14, flown on September 28, 2026, Starship's flight tests had been suborbital flights, which reenter the atmosphere without completing a single orbit. Suborbital flights end at sea roughly an hour after launch, and have the advantage of allowing reentry testing while minimizing the risks of an extended stay in orbit and engine relight. By contrast, putting the vehicle into a full orbit through two burns has been done only twice in the program's history, on Flight 14 and now on this flight. Flight 15 was also meant to confirm that the orbital insertion procedure can be executed repeatably rather than ending as a one-off success.
After confirming insertion into low Earth orbit, Ship 42 moved straight to deploying its commercial payload. Satellite release began about 35 minutes after launch and continued for 42 minutes, sending 26 next-generation Version 3 (V3) Starlink satellites into space as planned. Releasing as many as 26 satellites in sequence from the cargo bay of a large spacecraft demands a precise sequence that, in weightlessness, prevents contact between satellites and with the vehicle, and times each push so that every satellite enters its own independent orbit. Deploying all 26 within the planned 42 minutes is evidence that Ship 42's attitude control and payload release mechanism worked as intended on orbit.
The insertion of the 26 satellites also brought a practical benefit for SpaceX's own satellite communications network. This was the second time the company has delivered 26 Starlink V3 satellites to orbit with Starship, following Flight 14. According to SpaceX, the 26 V3 satellites deployed this time will add 1 terabit of data capacity to the Starlink constellation as a whole. The pain of Booster 22's failed return is considerable, but on the core mission of a space launcher, delivering customer and in-house satellites to the specified altitude at the right time, Flight 15 delivered as planned.
Ship 42's 22.5 hours: a first catch over land, or a retreat to the ocean
Having sent the 26 Starlink V3 satellites into space, Ship 42 did not return to the atmosphere right away but entered a coast phase of about 22.5 hours, circling Earth up to 14 times. This long stay in orbit is another technical test linking the first and second halves of Flight 15. While exposed to sharp temperature swings as day and night alternate roughly every 90 minutes, the ship must maintain the pressure and remaining quantity of its cryogenically stored reentry propellant and keep its attitude control and power systems healthy for nearly a full day; otherwise it cannot proceed to the deorbit burn for its return.
If all vehicle systems remain healthy, Ship 42 is scheduled to attempt its return to Starbase around 3:20 p.m. CST on Saturday, November 14 (4:20 p.m. EST, 21:20 UTC), 22.5 hours after launch. This return and recovery, however, is only a conditional planned event, not an accomplished result. If the ship completes the deorbit burn after its 22.5 hours in orbit and is ultimately caught by the mechanical arms of the tower at Starbase, it would be the first time in the Starship program that an upper-stage vehicle is recovered on the ground.
Still, precedent suggests that recovering an upper stage is extremely difficult. All 14 Ships launched so far, including Flight 14, ended their flights at sea, and not a single vehicle has returned to the ground intact. Compared with a first-stage booster that comes back down just minutes after launch, an upper-stage vehicle reentering from low-Earth-orbit speed must endure a far longer and harsher process of aerodynamic heating and deceleration. And because Booster 22, a first stage of the same Version 3 configuration, failed to complete its return to the tower just before, it remains uncertain whether Ship 42 can satisfy all the conditions for a tower catch.
Ship 42's return also comes with geographic and safety constraints that did not exist for the first-stage recovery. Unlike a booster that flies out east over the Gulf of Mexico and turns back to the launch pad from the sea side, Ship 42, which returns to the atmosphere from west to east after orbiting Earth, must take a path over land in Mexico and southern Texas to land at Starbase. Because inhabited areas lie beneath this return track, permission to fly over land cannot be given if even slight uncertainty remains about attitude control or heat-shield performance during reentry.
To manage that risk, SpaceX built multiple abort routes into the plan from the start. If telemetry during the orbital coast raises any doubt about system health or landing safety, the company intends either to cancel the return to Starbase and end the mission early, or to switch guidance to a splashdown at sea. Specific fallback options secured are waters off Hawaii, off Chile, and the Gulf of Mexico, so that in the event of an anomaly the vehicle can be disposed of in an area where it poses no danger to land.
How close Starship Flight 15 has brought the program to practical, fully reusable rockets will be settled by two remaining questions. First, whether the investigation into Booster 22's called-off catch and explosion, whose cause had not been disclosed at the time of reporting, can identify the technical factors that kept a Version 3 booster from returning. Second, at the return decision scheduled for around 3:20 p.m. CST on November 14, whether Ship 42, after about 22.5 hours in orbit, remains healthy enough to pass over Mexico and southern Texas and carry out the first upper-stage tower catch, or instead opts for an ocean splashdown such as off Hawaii. The loss of Booster 22 pushes full recovery on a single flight to later missions, but if Ship 42 achieves the first tower catch after its stay in orbit, one of the biggest remaining technical barriers, recovering an upper stage from orbit, would be cleared.