United Launch Alliance's (ULA) Vulcan rocket failed to reach orbit on November 16, 2026 (Japan time), on its first flight in about nine months, and 40 production satellites for Amazon's "Amazon Leo" satellite communications network were lost. The vehicle lifted off from Space Launch Complex 41 (SLC-41) at Cape Canaveral Space Force Station in Florida at 2:40 a.m. on November 16 Japan time (12:40 p.m. Eastern Standard Time on November 15). Vulcan had placed its payloads into the planned orbit on all four of its previous flights; the fifth was its first failure. Amazon's official mission list also confirms the loss of the 40 satellites.

The mission, known as "LV-01," was the first launch under Amazon's contract for 38 Vulcan flights. It was also the first flight of a configuration with a shortened upper stage, designed to carry many satellites to low Earth orbit at once. In a statement, ULA said an anomaly occurred about 10 minutes after the start of the flight and that the satellites could not be placed into orbit. For Amazon, which lost 40 satellites, the challenges ahead include its future launch schedule and the deadline set by the U.S. Federal Communications Commission (FCC) to deploy and operate its first-generation constellation by July 30, 2029.

AD

About 10 minutes into the flight, the plan had the upper stage in its first burn

ULA posted its statement on X about 50 minutes after liftoff, at 3:30:32 a.m. on November 16 Japan time (1:30:32 p.m. Eastern Standard Time on November 15). The company said the anomaly occurred about 10 minutes into the flight. According to ULA's flight plan, that moment falls in the window when the Centaur V upper stage is performing its first engine burn to reach a parking orbit.

Under the plan, the six solid rocket boosters mounted on the sides of the vehicle would separate first, four and then two. The main engines of the liquid-fueled first stage would then shut down, the stage would be jettisoned, and the Centaur V upper stage would take over the acceleration.

Planned time in flight plan Event Source
1 min 36.6 sec, 1 min 38.1 sec Solid rocket booster separation ULA flight plan
4 min 57.1 sec First-stage main engine cutoff ULA flight plan
5 min 03.1 sec First stage and upper stage separate ULA flight plan
5 min 13.1 sec Upper stage first ignition ULA flight plan
About 10 min into flight Anomaly occurs in flight ULA statement
16 min 58.4 sec Upper stage first burn cutoff ULA flight plan
1 hr 5 min 49.5 sec Upper stage second ignition ULA flight plan
1 hr 9 min 41.5 sec to 1 hr 35 min 21.5 sec Satellite deployment ULA flight plan

The planned times in the table are based on the flight profile chart released by ULA. In the chart, liftoff is set 0.9 seconds after the reference time, so these should be distinguished from the times at which each step actually occurred in flight.

The plan called for the upper stage to enter an elliptical parking orbit through a first burn of about 12 minutes. It would then shut down the engine and coast for about 50 minutes, reigniting near apogee, the point farthest from Earth, to approach a circular orbit. The orbit for the first satellite deployment had a perigee of about 423 km, an apogee of about 442 km, and an inclination of 51.9 degrees. The satellites were to be released in 16 batches: one at a time in the first four, then three at a time in the remaining 12.

As for the actual flight, Ars Technica analyzed the live broadcast and reported that the first-stage main engine cutoff came at least 30 seconds later than planned. Immediately after cutoff, the broadcast's animation based on flight data showed the vehicle beginning to spin, and after stage separation there was an announcement that the upper-stage engine had ignited, the outlet said. Several sources who spoke to the magazine said the problem occurred on the first stage. However, the specific component responsible has not been identified.

Because the upper stage inherits the position and velocity it has at separation from the first stage, if the first stage fails to deliver the planned acceleration, the amount of acceleration needed to reach the target orbit changes as well. If the vehicle's attitude is off, it must also be corrected so that the stage accelerates in the proper direction. In this flight, a problem on the first stage may have affected the upper stage's flight, but ULA has not yet identified where the anomaly occurred or its cause.

Why the upper stage was shortened for low Earth orbit

The "VC6L" configuration, used for the first time on LV-01, is based on the Vulcan built for high-energy orbits and optimized to carry many satellites to low Earth orbit. The Centaur V upper stage has been shortened to 10.7 m in length and carries less propellant. By lightening the upper stage to match the propellant needed for low Earth orbit, the mass allocated to satellites was increased. The fairing that encloses the 40 satellites and protects them from airflow during ascent is a large type 21.3 m long.

Liftoff thrust comes from six GEM 63XL solid rocket boosters on the sides of the vehicle and two BE-4 engines on the first stage. The solid boosters provide acceleration at liftoff and early in the ascent, while the first stage's BE-4 engines burn methane-based fuel and liquid oxygen to keep accelerating after booster separation. The Centaur V upper stage carries two RL10C-1-1A engines fueled by liquid hydrogen and liquid oxygen. This configuration is described in ULA's mission materials and a pre-flight technical explainer.

Vulcan's previous flight was the U.S. Space Force mission "USSF-87" on February 12, 2026. On that flight, one of the four solid rocket boosters attached suffered a significant performance anomaly early in flight. Vulcan nonetheless continued flying and placed its payload directly into geosynchronous orbit as planned. ULA acknowledged the anomaly in a post-flight announcement, and afterward worked with booster supplier Northrop Grumman, the U.S. Space Force and others on materials and component testing and full-scale ground firing tests of the GEM 63XL.

ULA had explained that, regarding the February anomaly, it investigated the booster nozzle, identified three candidate causes, and took corrective measures in both design and manufacturing processes. LV-01 carried boosters incorporating these measures. To understand the nozzle's behavior and the effect of the corrective measures in more detail, the company planned to continue additional testing and analysis even after returning to flight.

AD

Loss of 40 satellites and the impact on deployment pace

According to the mission list published by Amazon, 396 production satellites had been launched in 14 flights before LV-01. This figure is a tally of the number carried on each flight, by rocket, covering Mission No. 1 to 14 on the official list. It excludes the two prototype satellites, LV-01 and upcoming planned flights.

Rocket Number of launches Satellites per launch Total satellites carried
Atlas V 8 27 on 4 flights, 29 on 4 flights 224
Falcon 9 3 24 each 72
Ariane 6 3 32 on 2 flights, 36 on 1 flight 100
Total 14 396

This tally reflects the number carried at launch and differs from the number of satellites actually deployed in orbit and operating. Of the 14 flights so far, the 224 satellites carried by ULA's Atlas V accounted for more than half of the total. LV-01 was to carry 40 satellites, well above the Atlas V's per-flight count. To raise the deployment pace with this configuration, Vulcan, which can carry 40 satellites, needs to be launched continuously.

The new ground facilities built for Amazon are also meant to raise launch cadence. ULA set up a dedicated vehicle assembly building so that launch preparations can proceed in parallel, separate from existing facilities. Assembled vehicles are moved on a dedicated mobile launch platform, but launches still use the same SLC-41 as before. Even if ground facility upgrades make preparation more efficient, resuming satellite deliveries requires Vulcan to return to flight.

Amazon's next launch, on another rocket, is "LE-04" on Europe's Ariane 64. It plans to launch 36 satellites on or after November 23 and deploy them at an altitude of about 465 km. Amazon says that after finishing communications with the satellites and checking their condition, it will raise them to the operating altitude of 630 km. Having secured multiple rockets means satellite deployment can continue on another vehicle while one type cannot fly. However, how much this failure affects the deployment plan will depend on how many flights can be secured on other rockets and when each launches.

There is also precedent for switching the launch vehicle to a different type. On March 20, while investigating the February booster anomaly, the U.S. Space Force announced that it was moving the launch of the GPS satellite "GPS III-8" from Vulcan to SpaceX's Falcon 9, and assigning "USSF-70" to Vulcan instead. If Amazon changes its launch plan in response to this failure, both the timing of Vulcan's return to flight and available launch slots on other rockets will factor into the decision.

The 2029 deadline remains after the FCC's exception

On June 5, 2026, the U.S. Federal Communications Commission (FCC) granted an exception allowing Amazon to continue deploying its first-generation Amazon Leo constellation even if it fails to meet the milestone. However, the final deadline of July 30, 2029, to deploy and operate 3,232 satellites remains in place. Amazon had originally been required to deploy and operate 1,616 satellites, half of the authorized total, by July 30, 2026. According to the FCC decision, "DA 26-553", the exception waives termination of the authorization for missing the milestone, while attaching conditions related to frequency coordination.

Item Conditions applying to Amazon Leo's first generation
July 30, 2026 milestone 1,616 satellites. If missed, authorization for the remaining satellites is not terminated, and the requirement to meet the milestone as a precondition for reaching the final target is also waived
Frequency coordination Satellites not deployed and operating by July 30, 2026 temporarily lose the priority treatment they had from the 2020 Ku/Ka-band and 2021 V-band authorization rounds, and are treated with the same priority as systems authorized in later rounds
July 30, 2029 final deadline 3,232 satellites deployed and operating. If missed, the number of authorized satellites is reduced to the number operating on the deadline date

The FCC reviews satellite systems filed around the same time in categories called "authorization rounds," and frequency coordination is meant to prevent harmful radio interference between different systems. The temporary lowering of priority under this exception will in principle continue until March 30, 2028. However, if Amazon certifies to the FCC that it has built all the satellites needed to meet the milestone and has secured enough launches to deploy them, the end date can be moved up to October 30, 2027. If it achieves deployment and operation of 1,616 satellites before the deadline, its original priority is restored at that point.

The FCC also waived the procedure of demonstrating, before beginning operations, that Amazon will not cause harmful interference to other systems authorized in the original two rounds. However, even while its priority is temporarily lowered, Amazon must coordinate frequencies with the relevant operators or operate under conditions that do not cause harmful interference.

How much the loss of these 40 satellites affects the final 2029 deadline will depend on when ULA completes its cause investigation and corrective action and returns Vulcan to flight, and on the launch opportunities Amazon can secure on other rockets. What the FCC's final target requires is not the number of launches but the number of satellites deployed in orbit and operating by the deadline.