Monday, May 10, 2021

SpaceX Breaks the 10-Launch "Barrier"

Sunday Morning at 2:42 EDT, SpaceX launched their 27th Starlink mission atop booster B1051, completing the 10th mission of one Falcon 9 booster, a number which has been an aspirational goal for the fleet.  The flight appeared completely nominal in all regards, with B1051 and the upper stage staying on the center line of their planned path, followed by deployment of the 60 Starlink satellites. And this:

SpaceX's live webcast, screen capture, from a looping animation on Twitter posted by SpaceX corporate.  The landed booster looking down from the interstage area at the recovery drone ship's deck on the left, and a view of the booster through the smoke cloud left by the engines. 

The number of ten flights was probably just pulled out of the air (PFA), as a number of launches without major refurbishment.  SpaceX Vice President of Mission Assurance Hans Koenigsman as said essentially the same.

It goes without saying that this is another world record because nobody else is launching recoverable, liquid-fueled, orbital boosters.  In an interesting aside, Eric Ralph at Teslarati lets us in on some industry chatter. 
For the entirety of SpaceX’s operational life, its only two real competitors have – and continue to be – US conglomerate United Launch Alliance (ULA) and European conglomerate Arianespace. Almost like clockwork, both extremely conservative groups – comprised of numerous traditional, entrenched aerospace and military contractors – have gone through a similar cycle of belittlement and dismissal, denial, goalpost-moving, disbelief, and resignation as SpaceX announced plans for reusability, began real-world attempts, and gradually worked out the kinks.

As it became clear that SpaceX would succeed in its efforts to vertically launch and land Falcon 9 boosters and ULA and Arianespace had to move their goalposts from “it’ll never work,” both generally settled on largely arbitrary claims that even if SpaceX could land rockets, reuse would never be economical. ULA went even further than Arianespace with an explicit claim – derived from armchair analysis built on opaque, unspecified assumptions – that SpaceX’s approach to Falcon reuse would “require ten [booster] uses to be profitable.” [PDF]

There's a story that ULA had been analyzing an alternative approach to reuse since 2015.  It seems like an expensive, complex arrangement and it would only save the engine.  It was to be called, “SMART (Sensible Modular, Autonomous Return Technology) Reuse” and was intended for its next-generation Vulcan rocket, still working toward its first launch probably in '22.  Unlike SpaceX returning the entire first stage to a semi-autonomous drone ship, ULA would develop an extremely complex engine section that would detach from Vulcan in mid-air, deploy an experimental inflatable heat shield, and be grabbed out of the sky with a helicopter.  ULA’s original schedule for SMART reuse would would debut no sooner than the mid '20s.

A Spanish-language site featured this montage of B1051's first nine flights.   


  1. I do not believe that Bezo's BE-4 engine will be ready to fly in 2022 or 2023, maybe even 2024. ULA's reliance on Mr. Confidence Game in order to try to catch up on Musk is going to bite them in the arse.

  2. And... Go go SpaceX! Be interesting to see if they keep launching the booster without any major refurbishment to see what happens. But that would be too risky, so be interesting to see how many launches they get with decent refurbishment.

    Really, will be interesting to find out how many times a Merlin can be used, how many times all the other pieces parts can be used with and without major rework.

    I am so glad that someone is actually giving us the reusability that we were promised with the Shuttle and it's unbirthed derivatives (insert long screed over lost opportunities and blatant lies by those in power...)

    Lots of conspiracy rumors that the democratic craphola republic of China has supposedly stolen and copied both Falcon and Starship technology. Which I don't but do believe sort of, because that is what the ChiComs do.

    1. Another thing about this 10th launch is that the turnaround time from the flight in March was a pretty routine 56 days. That implies that there was nothing unexpected after nine flights; they just did their regular preparations, scheduled a launch, and it was fine.

      I was wondering if they were going to give this one a more thorough tear down and checkout but haven't seen anything saying that.

      The other side of this is that the only booster that failed to be recovered, a few months ago, had a problem they'd never seen before. You know they're checking for it now, so it seems the only barrier is the "unknown unknowns" - the things they don't know that they don't know. That number will get smaller with every new thing they learn.

    2. Graybeard, do you expect they'll keep launching this booster until it fails, for the sake of finding the "magic number"? I have to assume that even with routine maintenance, the booster will eventually wear out and fail.

    3. I honestly don't know. I tend to think there's no such thing as a "magic number." That is, 1051 might fail at one number of launches while another booster fails at a different number. At some point, something will wear out and fail, but I think it could take a major dissection with tons of time and effort to find out what the failure might be.

      There's an application in statistics that uses a Poisson distribution, shaped differently from a Gaussian or Normal distribution, that tells you the probability of first failure. It gets used in industry to check things like the percentage of parts that are too big/too small/out of spec in some way. You inspect parts until you find the first one bad and get answer like, "there's a 90% chance the lot of parts is 5% defective."

      There are nine Merlin engines on the Falcon 9, and all it takes is for one to fail. The Poisson distribution says if the real probability of failure was 1%, there's some small percent chance you see one fail on the first flight and a much bigger probability you'd see one fail by the tenth flight. Remember, though, that "Space level" hardware means it's the best hardware we know how to make.

      As you can see by looking at that final image with the different flights, SpaceX has always done the testing of high usage boosters on their own dime, so that if the rocket fails, it's not for a paying customer. I think that means they'll do their usual inspections and keep using it on Starlink missions with some risk of losing a load of Starlink satellites. I've read they're making those satellites faster than they can launch them.

    4. Actually, with nine Merlin engines one or more can and have failed and the booster still goes on to complete the mission. How it will affect booster recovery, I don't know. Have there been cases of this documented yet ?

    5. As long as one of the Merlins doesn't really rapidly disassemble (explode), then the mission should be able to be completed.

      I think as long as the central engine doesn't fail, the booster is recoverable. Though the longer burn from 8 or 7 engines after a shutdown of 1 or 2 failed Merlins may push the booster past the recoverable window.

      One of the reasons 9 engines was touted as better than 3 or 2 or 1 is for that very reason, for mission survivability over platform survivability. I seem to remember they had early issues with one engine out in a mission, which got the payload to space. But that was, I think, before the reusable/landable stuff started happening.

    6. There was a mission within the last year, I think it was the first booster that made five flights, where one engine shut down and while it did burn longer and make orbit, the booster couldn't land.

      Found it. Booster 1048. Summary on Teslarati

  3. I'm starting to feel like the "Old Uncle" on my Mom's side. He saw the Wright brothers fly when he was little, and was just awestruck when Apollo 11 landed on the Moon.

  4. This comment has been removed by a blog administrator.