Monday, September 28, 2026

Flight 14 shortened but still seemed to prove everything out

This morning's Starship Flight 14 had its launch delayed and a few glitches along the way that made them shorten the mission, but it sure seemed to prove out the big things they had listed for the mission. 

Almost two months to the day after the July 24th flight 13 which met or exceeded its goals, Ship 41 lifted off on top of Booster 21 from Starbase, Texas, at 8:49 am EDT (1249 UTC). While counting off milestones going by, SpaceX coverage noted that one of the 33 Raptor 2 engines had shut down fairly early in the flight, but the difference is small and the remaining 31 kept pushing the vehicle to orbit. A second Raptor 3, in the same ring of engines, shut down closer to Main Engines Cutoff, before stage separation. That's two Raptor sea-level engines out of 33.

After the stage separation, the ship looked better with all six of its engines working fine. Meanwhile the booster continued its return to a hovering stop off the beaches it had launched from. Much better than flight 13. The booster did appear to explode while standing vertically in the Gulf, but there is speculation that was the Flight Termination System commanding the detonation. 

Meanwhile the ship continued toward orbit, done by burning its six engines for around another five minutes. One of the vacuum raptor engines (larger engine bells) also shut down; the only noticeable effect of that was the system ran the engines a few seconds longer to achieve orbit. 

This led to the only endurance test for folks watching this one. Based on the performance of losing two of the booster's 33 raptors and one of the ships 3 vacuum raptors, SpaceX spent several minutes trying decide if they were going to do the orbital insertion burn or if they were just going to stay suborbital. It took on the order of 12 minutes before they decided it was a minimal risk and did the short (honestly, I think it was around two seconds) burn of just one of the three sea-level raptors. Then the team announced "nominal orbital insertion." 

SpaceX's 14th Starship test flight began with a Monday morning liftoff from Starbase, Texas. The city of South Padre Island is visible in the background, on the far (north) side of the Brownsville Ship Channel. Credit: SpaceX

Once in orbit, attention shifted rapidly to deploying the 26 Starlink 3 satellites that were the main goal of this mission. We watched them deploy the new version Starlinks, at a rate of approximately one every 65 seconds, using the same basic approach as they've used before, a system called a Pez dispenser. Honestly, calling it that is something that has always puzzled me. I mean, it doesn't look like any Pez dispenser I've ever seen. 

It seems looking back that it was very soon after the last Starlinks were released the NASA Spaceflight coverage started covering that SpaceX had decided the mission had proven enough and they would bring it down for a wet landing in the Pacific a little north of Hawaii. SpaceX's post-mission summary said, 

Out of an abundance of caution given the engine issue experienced during ascent, the flight control team decided to limit the duration spent on orbit and proceeded with deorbiting Starship to a splashdown location in the northern Pacific Ocean. This area was one of two pre-coordinated landing zones identified before launch that was cleared of sea and air traffic.

Starship then relit a single Raptor sea-level engine and performed its first ever deorbit burn. That maneuver successfully brought the vehicle out of orbit and sent Starship through atmospheric reentry and descent to the intended splashdown point.

Obviously, I don't have access to any inside information on SpaceX goals and how they set them, but it seemed to me that this went very well, just maybe not as well as flight 13. That might just be that both met or exceeded all goals but 13 achieved a few more. We have 93 days left in the year, and with 14 launching two months after 13, call that around 60 days. That makes it sound like we're more likely to get only one Starship launch than two launches before the end of the year. 

They have lots to do, but the pace of improvement has been picking up lately. This is where the problem of over-optimizing products comes in. They lost three sea-level raptor 3 engines? Has one part in those incredible engines been made a little too light, a little too thin? Only the people who spent the last decade (or however long) optimizing those engines would have a clue where to start looking. 

NASA Spaceflight has done a good summary video on the mission. It's about a 10 minute video and worth the time.



11 comments:

  1. It was very impressive. Two orbits was good, lots of data accumulated, all major goals reached.

    And the plasma flowing over the lower flaps was just beautiful.

    Can't wait to see what they discover and what's next.

    ReplyDelete
    Replies
    1. The reentry looked smoother than the way they were doing the (slower) suborbital reentries. Being able to watch that in real time is still pretty amazing. I have to expect that they instrumented those two reused tiles from another flight just to be sure they got more data. Not as good as having them in front of you to run any kind of test on, but better than just "the hull didn't get melted through in that spot."

      I tell myself that they must have done everything they wanted, but then I think that was two orbits instead of six and I can't imagine they didn't have much more planned than just sitting and counting time. Can you imagine these guys having hours available to test things and just sitting there?

      Delete
    2. SiG, the greatest fear at that point was that they might not be able to relight for the deorbit burn. Losing one of the engines on Ship rightfully spooked them. All more orbits would have done would have been to give more time for something to settle wrong, or freeze up, or ? so why hang around. So they used the first opportunity to come down. Since it worked the first time, they didn't need the two backups.

      Yes, I would like to know what the problem is with the engine, also. One doesn't want to have to rely on redundancy to get through an ordinary mission.

      Delete
    3. "SiG, the greatest fear at that point was that they might not be able to relight for the deorbit burn. Losing one of the engines on Ship rightfully spooked them."

      I got that. The thing that stood out to me was that the one that failed on the Ship was one of the vacuum raptors while the two that failed on the booster were sea-level raptors. While out of my engineering world, it seems the "moving parts" at the top have to be pretty much the same but the interfaces to the (definitely different) exhaust bells are probably different, too. Does that matter?

      In brute numbers, the sea-level raptors had a failure rate of 2/33 while the vacuum raptors had a rate of 1/3. Is the failure rate really 3/33? Is it really fair to lump them together?

      Nothing like idle speculation.

      Delete
    4. Idle speculation is the best, because it can go anywhere!

      Delete
    5. And my last number should have been 3/36. So 1/12 not 1/11. 0.83% vs 0.91%.

      Delete
    6. Elon said, a while ago, that they'll be transitioning soon to Raptor V4. Which should correct (hopefully) the continued small issues they have with V3.

      Delete
  2. Just think of their Starlink deployment mechanism as a shoe, like they use in casinos.

    ReplyDelete
  3. The booster *successfully* decelerating to a landing is a big deal after the flight 13 failure. The stagnation heating on the engines and transom structure is insane. And as a retired aerospace engineer it was very cool to see a textbook detached bow shock as it came down through the clouds (a flight state the terminology inventors hadn't considered?).

    ReplyDelete
    Replies
    1. I never tire of watching those monstrously huge boosters decelerate and land. There has been exactly one catch of a booster in the giant chopsticks and I wouldn't be surprised if I've watched the video a hundred times.

      Delete
  4. As a Certified Space Nerd I can confidently state that 90%+ of this Flight's goal was reached. They *almost* have the ice (in the Booster's fuel tanks) problem solved, they probably have the exact same problem in the Ship since the RVac in the Ship shut down after the autogeneous tank pressurization could have had the same effect. Tiles looked pretty well stuck to the Ship, although an after-flight inspection would've been A Good Thing. See: Ship 40, IFT 13.
    The deployment of the Ver 3 Starlink birds was flawless!
    They were unsure of the elongated cold soak screwing up the fuel feed(s) and a Sea-level Raptor failing to start back up. BTW, the only real difference between the RSea and the RVacs is the bell. And the different Isp in vacuum. Control program, and over 95% of hardware, is identical.

    SpaceX is still being cautious - they had 3 different landing sites planned Just In Case, and like I said, they didn't want to push their luck and have Starship stuck in orbit.

    Rumor has it they are planning TWO launches in October. We'll see, stuff will probably crop up - but you can bet your bippy they have the capability *right now* to accomplish this!

    ReplyDelete