I added the date clarification to the original title.
The article discusses SpaceX launching the same Falcon 9 rocket, serial number B1058, for the 18th time. Its first use was to send Crew Dragon Demo-2 to the ISS in May 2020; it was the first manned launch from US soil since the shuttle's retirement in 2011.
The 18th launch broke a tie with another SpaceX booster with 17 launches. SpaceX currently limits itself to 20 uses per booster for Starlink launches but may raise the limit. Its customers, according to the article, always get boosters with fewer uses than that.
I hope they continue to push some boosters until failures start happening. There isn't really another way to improve the longevity of a booster without learning the long term failure modes.
Yes, that's the likely scenario: Keep reusing boosters for Starlink launches until they fail. If it turns out that they can be reused 30, 40, 50, 100 times, all the better.
Are they still building Falcon 9s? I thought they had directed all manufacturing capacity to Starship. If so, they need this fleet to last as long as possible to continue to generate revenue for Starship development until it is flight proven, at which point the Falcon fleet becomes disposable (Heavy missions only?).
Yes. They still build new boosters, not often, a few a year, and new second stages are cranked out every few days. F9 is still in production and will be for years to come. It shares no manufacturing with Starship.
The second stages are still single-use only, and they're still occasionally building a first stage, if only to be sure they don't lose the capacity. (They're less public about engine replacements, but those are still probably happening too, on "flight proven" boosters in inventory.)
I wonder how Rocket of/Ship of Theseus this is. I wonder how many different rocket engines have been on it. I wonder how much of those engines have been replaced.
This is a huge accomplishment. That we can wonder about more detailer aspects of what needs to be replaced is a marvel.
The concern is not so much about number of parts replaced ($) but about turnaround time because cadence is ($$$). If building a new booster from scratch with its tanks, its engines, the avionics, etc takes n days and refurbishing a booster takes n-1 days, refurbishment's the way to go.
More simply put, they could not have hired and built factories enough to keep up with their manifest. I think at peak production they were churning out a bit better than 1 Merlin per day and that's far lower today. So, let's say they could double that peak rate with brute force hiring and factory expansion. That could get them 800 Merlins produced per year. But that's only 80 launches a year at best. They're gonna do very close to 100 this year and well over 100 next year. Further, they couldn't magically double Merlin production as I imagined above. Since boosters are being re-used so often, however, they've actually cut production to something like 1 Merlin every 3 days which gets them a few new boosters each year to replace expended ones, and all the second stages they need.
It's not. In some dialects H would be silent, and you can certainly write "an historic" because of that in informal writing, but it's discouraged in style guides that sites like this would use.
This is incorrect information. It’s actually quite controversial and most learned people point to “an.” Either way, one shouldn’t talk factually as if the issue is settled. Particularly if you’re citing style guides, like you are in this response, rather than informal usage.
When style guides recommend you to use "a historic" in formal writing, why should I not talk factually about what they recommend. You were the first to complain about the usage in this article.
Ironically, that's how it'd be transliterated in greek - where h really is a strange exception that gets a breath mark rather than its own proper letter in the alphabet.
Formally, the word historic begins with a consonant sound and so the form a historic is preferred in formal writing. However, many people prefer the form an historic in informal writing and speech for personal reasons.
This makes sense! It’s like lede vs lead. It’s really lede, but common usage has made lead acceptable. At least this is a better explanation and intuition than the poster below who simply asserts one thing based on the first letter without beginning to understand any context.
I disagree (NZ English), but am curious. Would you say "an history book", or "an historical society", or "an historian"? My dad is definitely a historian to me.
Historic in standard dictionaries (Oxford, Webster) has the h pronounced (IPA: /hiˈstôrik/ ) so we must use A instead of An.
It is different for some other words start with h, for example hour (IPA: /ˈou(ə)r/) that the h is silent and it starts with the vowel sound, so we should use an hour.
It's kinda wild that just a few years ago, NASA preferred to launch humans on brand-new rockets, rather than on ones that have flown multiple times and shown their capabilities.
It'll be interesting to see how the bathtub curve will look like for reused rockets, once we have more data on them.
Are individual rockets and subs considered safer after each trip? I thought it was the opposite since they endure so much stress, and one has to do expensive inspections and scans to ensure they're still flight/dive worthy.
I think that's why there is a bathtub curve. New rockets are more risky because they could have manufacturing flaws and old ones may be worn to the point they fail. The middle is the safest because they have been thoroughly tested but not worn. We seek to find how many times is the limit.
It’s important to note that this is technically not true.
It is an assumption that is true for other stuff and -maybe- could be true for rockets but 1) the last hundred Falcon 9 launches worked flawlessly regardless if the booster was new or not and 2) every other rocket company launches only new rockets.
Would you rather fly on an airplain's maiden flight or its second flight. For me, the wear is far less a concern than a quality control failure at manufacture. I want flight proven but I also don't want the flight leader in the fleet. There's a sweet spot between, "they're still working out bugs from the factory" and "the core, non-replaceable systems are starting to wear out." My guess is that's around 20 or so flights for an F9 booster, maybe a bit more.
My guess is we'll find F9 boosters good for between 20 and 25 launches before refurb becomes expensive enough that it's better to weld up some new tanks and build another batch of Merlins. There will be a pretty big range I think. We already see they've got favorites that are far outpacing the rest of the fleet presumably because they're easier to refurb and refly. So, I think some will retire around 15 flights, and some around 30, with the low 20s being the top of the bell curve.
> We already see they've got favorites that are far outpacing the rest of the fleet
My guess is that they are keeping some in the low #s so they are available for external customers, and running others through the gauntlet for Starlink launches. Like all things equal they'd rather have 1 rocket with 4 launches and 1 with 20, than both with 12.
We know for sure that even later in the Block 5 generation the re-usability got better. So that isn't controversial.
I would assume their 'favorites' are being pushed in order to get engineering information on those flights, not because they are superior to the others of the same build.
Watching the dual boosters come in for a synchronized landing back at KSC with the center core almost landing on the ASDS (which though that was a failure on the day, SpaceX can do reliably now) was the moment where I realized it.
Apparently the rocket is now very sooty. Think about it: SpaceX is the only company launching rockets covered in soot from previous launches. Because they can.
I think RocketLab came closest to being able to attempt a reflight of a booster, but in the end they too have shelved reuse of Electron as they build Neutron from the ground up for F9 style reusability.
Blue Origin have a (significantly less capable) reusable suborbital rocket, and Rocket Lab have reflown an engine.
It's worth noting that the Falcon 9 only reuses the first stage and payload fairings, with the second stage expended. I believe SpaceX's Starship is the only fully reusable rocket in realistic development
what good is a suborbital rocket? is the expectation that the payload will then have its own booster functionality to finish the boost to orbital altitudes?
I think the idea was that they could demonstrate and develop vertical landing technology while selling the vehicle (tourism and some freefall science experiments). I think they've done two or three tourist flights using it, and maybe a couple more science flights. They lost the booster a year or so ago and have been making sure they fix the problems since.
In practice, SpaceX's approach of starting with an orbital rocket and adding reuse worked a lot better, given they started later.
Vulcan is supposedly launching next month, but they're not planning on reusing the whole first stage (ULA claims that with their design the economics favor just reusing the engines). BO is working on a reusable rocket but I think it's a long ways off from launching (and no guarantee it will).
>Has anyone been able to even launch the same rocket twice?
DC-X did in the 90s, the first prototype launched 3 times., DC-X was canceled before they could actually make it to orbit.
I asked one of the launch managers last year about the soot, and apparently they used to clean them after each use, but it was taking several days and costing quite a lot. So they just stopped doing it.
Reminds me of why the Space Shuttle's external tank was a rusty red color: after the first couple of launches, they stopped painting it white, because of cost and weight.
How do you know how many new engines it has had? I thought thus far we didn't have concrete data on how often SpaceX are swapping engines during refurb, which also means we don't have reliable info on how much it actually costs to refurb for each flight.
57 comments
[ 3.4 ms ] story [ 101 ms ] threadThe article discusses SpaceX launching the same Falcon 9 rocket, serial number B1058, for the 18th time. Its first use was to send Crew Dragon Demo-2 to the ISS in May 2020; it was the first manned launch from US soil since the shuttle's retirement in 2011.
The 18th launch broke a tie with another SpaceX booster with 17 launches. SpaceX currently limits itself to 20 uses per booster for Starlink launches but may raise the limit. Its customers, according to the article, always get boosters with fewer uses than that.
Edit: Appreciate the replies providing context!
The only problem with that is the media going wild with headlines of "Musk's SpaceX rocket blows up, catastrophic failure", etc. etc.
While those in the know will know, the general public will eat it up.
This is a huge accomplishment. That we can wonder about more detailer aspects of what needs to be replaced is a marvel.
More simply put, they could not have hired and built factories enough to keep up with their manifest. I think at peak production they were churning out a bit better than 1 Merlin per day and that's far lower today. So, let's say they could double that peak rate with brute force hiring and factory expansion. That could get them 800 Merlins produced per year. But that's only 80 launches a year at best. They're gonna do very close to 100 this year and well over 100 next year. Further, they couldn't magically double Merlin production as I imagined above. Since boosters are being re-used so often, however, they've actually cut production to something like 1 Merlin every 3 days which gets them a few new boosters each year to replace expended ones, and all the second stages they need.
Me, I voice my 'h's, so "a historic" is correct in my locale.
It is different for some other words start with h, for example hour (IPA: /ˈou(ə)r/) that the h is silent and it starts with the vowel sound, so we should use an hour.
It'll be interesting to see how the bathtub curve will look like for reused rockets, once we have more data on them.
It’s important to note that this is technically not true.
It is an assumption that is true for other stuff and -maybe- could be true for rockets but 1) the last hundred Falcon 9 launches worked flawlessly regardless if the booster was new or not and 2) every other rocket company launches only new rockets.
My guess is that they are keeping some in the low #s so they are available for external customers, and running others through the gauntlet for Starlink launches. Like all things equal they'd rather have 1 rocket with 4 launches and 1 with 20, than both with 12.
I would assume their 'favorites' are being pushed in order to get engineering information on those flights, not because they are superior to the others of the same build.
I think in a few decades' time Falcon 9's remarkableness will pale in comparison to later rockets.
Watching the dual boosters come in for a synchronized landing back at KSC with the center core almost landing on the ASDS (which though that was a failure on the day, SpaceX can do reliably now) was the moment where I realized it.
https://i.stack.imgur.com/1Gsl2.jpg
Engines are also cleaned. In fact, one landing failure happened because left over cleaning alcohol in the engine itself.
https://twitter.com/elonmusk/status/1252985622219960327
Remarkable.
Has anyone been able to even launch the same rocket twice ? Or does space X effectively hold a 19X advantage over them now ?
It's worth noting that the Falcon 9 only reuses the first stage and payload fairings, with the second stage expended. I believe SpaceX's Starship is the only fully reusable rocket in realistic development
In practice, SpaceX's approach of starting with an orbital rocket and adding reuse worked a lot better, given they started later.
This is something I forget about. Blue Origin in 2000, SpaceX in 2002 (according to WikiP). It really makes me think less of BO's "accomplishments".
New Glenn is so silly that no one else wanted to make one with so little capability and so much cost. It’s a toy, basically.
>Has anyone been able to even launch the same rocket twice?
DC-X did in the 90s, the first prototype launched 3 times., DC-X was canceled before they could actually make it to orbit.
I think the saying goes: “Look after your booster…”
“… this particular booster has launched 846 satellites, …”
Now do Starship….