Before I learned about SpaceX, I thought it was literally impossible to reuse rockets because they burned up in the atmosphere or something. After all, not a single space agency in the world was doing it. Surely the smart engineers working in those places had good reasons.
Watching SpaceX land rocket boosters on target again and again is kind of magical, like something out of Kerbal Space Program. I wonder how long it's gonna take to get a fully reusable rocket. They're working on the Starship and it's supposed to be fully reusable.
To be fair to the older models, I think the main issue is: only "recently" (let's say, some 30yrs ago), we have had computers and sensors fast enough to manage an automated reentry of rocket stages (and ship drones as well)
And SpaceX had to iterate a lot to get this working as well
The older model also didn't incentivize the things needed to support a reusable rocketry program. Companies were encouraged to have bloat and making satellites was still extremely expensive so there wasn't anywhere near enough demand to justify producing and reusing a handful of boosters.
Ariane space was still arguing that reuse wasn't viable for them due to their low flight rate only until a handful of years ago. As far as I'm aware, ULA's CEO still insists that anything more than their proposed engine segment reuse is going to require too high of a flight rate to pay for itself (IIRC he said they'd need 10 flights per booster).
This is also part of why SpaceX's model works. Even when disposable F9 is much cheaper than previous rockets in its class.
Cost-plus is part of it, but also just the launch vehicle acquisition process in general.
Early on, SpaceX's biggest hurdle was getting payload contracts. IIRC they had to sue the government several times to overturn bad acquisition policies. Government acquisition processes weren't suited to supporting small new entrances into the launch market and there was somewhat of a belief that you weren't a credible launch provider unless you were as large as Boeing or Lockmart.
In comparison nowadays it's relatively normal for lower cost programs or ones which can accept some loss to try flying on less proven rockets. New bold approaches to designing, building and flying rockets aren't immediately dismissed as just young naive engineers who haven't had the reality of spaceflight beat into them.
My understanding is that early contracts were awarded to ULA because SpaceX wasn't certified for military payloads. That, combined with your statement, implies the certification process levies bad requirements. Do we know what these bad requirements were?
As Musk had pointed out at the time, the requirements were essentially just large amounts of paperwork (since if NASA considers F9 reliable enough for cargo and even crew, it's kind of ridiculous to suggest it can't handle a GPS satellite), such that it was unreasonable for USAF to buy multiple years worth of launches without competitive bidding despite obviously having another option almost through certification. At the time they had made that award, SpaceX was already mostly through the certification paperwork with zero changes needed to the rocket.
Besides that, although not a lawsuit, Lori Garver's book "Escaping Gravity" goes into a lot of depth regarding the behind the scenes political struggle regarding commercial cargo and commercial crew and how Congress and some parts of NASA leaned on the idea that they couldn't tolerate any failure as a means of trying to disqualify anyone but ULA,Boeing etc from those programs.
I was wrong about them having to sue "several times" though.
A small nit: the DoD and NASA are two different customers who don't have to align on their risk thresholds. Just because NASA is okay with a particular risk, doesn't mean that the DoD should be. What I've seen happen in some cases is one side essentially says "That may be okay for them, but we don't trust their level of oversight and want it done our way." I would push back a bit on the "mostly certified" as being sufficient; you're either certified or you're not.
I've seen the paperwork argument cut both ways. It can definitely create a bureaucratic mess. On the other hand, one of the things NASA tries to drive is "risk informed decision making". Part of this process involves writing waivers when a requirement can't/won't be met to document what the new risk is (and they have to document that they find that risk acceptable). A lot of times, this process was dismissed at "just large amounts of paperwork" but in dismissing it, they made it very hard for decision makers to fully understand what the residual risk was. It was interesting to me that paperwork was used as an excuse to not commit to the process, but when it was streamlined, it still wasn't used. To me, that feels more like avoiding accountability than avoiding needless paperwork. This misunderstanding of risk is at the heart of most of NASA's spaceflight mishaps.
I wasn't aware of Garver's book, but thanks for bringing it up. Sounds like a good read.
The only thing that burned up in the atmosphere for the Space Shuttle program was the fuel tank. All the other components, including the booster rockets, were recovered and refurbished for reuse.
The innovation here isn't about reusability at all, it's about the economics of reuse. Maybe landing a rocket back on the pad rather than recovering them from the ocean makes it cheaper to reuse, has there been a study around that?
And I believe that the Falcon 9 with the fastest turnaround was B1062 with 21 days between flights. The previous record (also from a Falcon 9) was two separate boosters who launched and relaunched 27 days apart.
Yeah, SpaceX’s reuse is much, much simpler than the shuttle booster refurbishment. At the extreme, Starship is intended to have no refurbishment between flights - refuel and fly again, so that the cost of each flight will be mostly the fuel cost, on the order of single digit millions. Falcon 9 I believe is trickier since its kerosene fuel generates a lot of soot compared to the methane engines of Starship, and can be fouled. If you’ve ever used a camp stove with kerosene vs methane, one leaves your stove covered in black soot.
This is based on pretty old info, which probably comes from an offhand statement by Musk, so big grain of salt, but I think they’re hoping to get it to be on par with a jetliner, which if I understand correctly is just a maintenance schedule based on # of flight hours. Not sure if that’s going to be feasible given the difference in the violence of the flight profile.
That would be interesting to hear how they come up with those maintenance schedules. Even with the uptick in flights, it doesn't seem like they'd have the dearth of reliability data to make that determination with reasonable accuracy. But maybe they have a lot of internal reliability data we're not privy to.
The reliability data will be collected by inspections. Also from memory, their fleet leaders tended to do Starlink launches. Loosing one of those might suck, but can still be sold as due to pushing boundaries and something that wont happen to customers. They need backup launch equipment, ofc.
The gold standard for reliability studies involves collecting in situ failures though. You can infer a lot from inspections on non-failed components, but it's not as useful as having actual failure data. When I'm referencing their in-house data, I'm referring to failures (whether in situ or as a bench experiment)
Landing the rocket back on the pad requires significantly more fuel for most missions. The reason most Falcon 9s do landing at sea instead of RTLS is due to this. They have to boost up and out over the ocean to the correct insertion coordinates, and then they run out of fuel (for the mission). When they run out of mission fuel, they just use gravity and fall straight downwards to land on a drone ship.
In short, performing RTLS requires additional fuel. This is fuel that can no longer be used to boost the payload into the correct orbital insertion. So all RTLS missions tend to be lower performance missions that don't need the full power of the Falcon 9 / Heavy since they have fuel to spare for RTLS.
Note that for the Space Shuttle, the external tank was large proportion of the fuel-carrying capacity of the vehicle, i.e. "most of the rocket". And for the solid rocket boosters they were not fully recovered as they ejected the rocket engine bell just before impact with the water which sank to the bottom of the ocean on each launch.
Additionally, solid rocket boosters are some of the hardest things to re-use as the process of re-filling them is more like re-manufacturing them than actual true re-use. The sections (they had 4 sections) need to disassembled and refilled with the epoxy that is the solid rocket fuel. They were additionally made of steel rather than aluminum so were not substantially valuable components to recover versus the high end aluminum alloys in the external tanks.
> The innovation here isn't about reusability at all, it's about the economics of reuse.
It's about both.
SpaceX showed that the 1st stage of a rocket can be reliably reused in production. Which is really important from an architecture perspective because 2/3 stage rockets are just more efficient than 1/1.5 stage rockets (mostly due to the propellant mass fraction term of the rocket equation).
To give some examples, the payload mass fraction of some reusable rockets is:
You need economy of scale for reuse to makes sense. Space launch market still isn’t that big, and even for spacex half of their launches are from their internal project, starlink. Jury is still out if this is actually economical, as we don’t have insights into their financials and they keep on raising money (so we don’t know if some portion of that goes towards flacon subsidies).
There is no doubt that a reusable rocket will be more economical. You need cheap rockets first to get economies of scale. It's gonna be big in the future.
Except that the space shuttle solid rocket boosters were recovered and reused.
From Wikipediea: "Out of 270 SRBs launched over the Shuttle program, all but four were recovered – those from STS-4 (due to a parachute malfunction) and STS-51-L (destroyed by range safety during the Challenger disaster).[6] Over 5,000 parts were refurbished for reuse after each flight. The final set of SRBs that launched STS-135 included parts that had flown on 59 previous missions, including STS-1.[7] Recovery also allowed post-flight examination of the boosters,[8] identification of anomalies, and incremental design improvements"
Also the rockets on the shuttle itself were landed and reused.
Calling the shuttle reusable is such a massive stretch. It’s like the shuttle of Theseus they basically had to rebuild it every time and it cost something like 1.5 billion dollars. Basically could’ve built a brand new rocket every time.
Recovery of the first stage of SpaceX rockets is not even close to being the same thing as reuse of the shuttle. Surviving re-entry is a completely different set of specs than floating through atmosphere at a much much slower rate of descent.
The first stage of SpaceX is the most expensive part with nine engines. The second stage is just single engine. Plus SpaceX reuse fairings and the Dragon is also reusable. So re-usable parts are like 95% and expendable parts are vastly cheaper than Shuttle refurbishing.
I mean from a technological point of view sure, but that just shows how bad the design was(compared to what we know now). The falcon program shows that it’s obviously way more economical to just ditch the 2nd stage and reuse the booster. F9 launches cost sub 20 million and a shuttle launch costs on the order of a billion. Squabble over whatever details you want about performance or who had to do what R&D but F9 is clearly in a different league.
> Squabble over whatever details you want about performance or who had to do what R&D but F9 is clearly in a different league.
Falcon 9 First launch attempt 2007 [0]
Shuttle First launch 1981 [1]
I wonder if there's an advantage to having 26 years of watching someone else before designing yours? You also act like the rockets from the shuttle were not reused. They always (except for 2 instances) came back with the shuttle. The SRBs were also recovered, so it's not like these were wasted.
I’m not disputing SpaceX had the advantage of learning from peoples mistakes. And I’m not arguing that they didn’t physically get the engines and boosters back I’m just saying that calling the space shuttle and F9 both reusable may be semantically true but is really not giving an accurate representation.
SpaceX is also contributing to the increasing amounts of space junk. How long is the second stage just floating around after delivering payloads? Once the shuttle delivered payloads, it just left nothing but space.
No, it was a legit question. It depends on the payload though right? For StarLink deployments at LEO, I'm sure it re-enters soon-ish. But what about for higher altitude launches of other types of satellites? How long does the second stage float around?
If you take it out of context of your post, yes, it looks like a well posed question. When read in context it looks rhetorical and that's how I took it, too. Perhaps you didn't intend that. But that is how it came out.
Almost Immediately after the payload is deployed, the second stage does a deorbit burn. So maybe an hour? Each Shuttle launch ended with ditching the enormous external tank into like the Indian Ocean, far more dry mass in the atmosphere/ocean than a Falcon 9 upper stage. (Also, the ET was about $100-150 million apiece.)
A brief look would have you find that for most (but not all) launches the second stage de-orbits after one or two orbits and isn't even left in orbit for a day. (For some launches it's impossible to de-orbit the stage and no matter which provider you launch with will leave a second stage as junk in orbit.)
Also, this isn't just SpaceX, every launching country/company leaves second stages in orbit, some, unlike SpaceX, don't even try to dispose of their second stages and leave all second stages in orbit.
I'd recommend reading up more on the topic as I think you read something at some point and think that SpaceX is an especially bad actor in the space when in fact they are one of the most responsible actors, if not _the_ most responsible actor, in the space launch sector at the moment.
SpaceX is launching stuff at an incredible rate. That's the great thing, but it also means that as they continue to launch, the amount of 2nd stages left behind will only increase.
I didn't read anything one time in weird corner of the internet like seem to want to insinuate. There are plenty of sites where you can see the items being tracked. There is a large number of second stages in orbit that are not coming back any time soon. It might have been some coincidental bit of luck that on one particular perusal of one of these sites I just happened to click on 2 such items. Again, as they continue to increase the rate of launches, this will become an issue. And obviously, I don't believe SpaceX is the only such party doing this. That's just an unintelligent comment to have made.
> SpaceX is launching stuff at an incredible rate. That's the great thing, but it also means that as they continue to launch, the amount of 2nd stages left behind will only increase.
You seem to be saying that all entities globally should as a whole launch less objects into space, as you imply that putting more objects into space is a net negative. That's fine to think but I strongly disagree that we as a planet should launch less objects into space.
> There is a large number of second stages in orbit that are not coming back any time soon.
I agree, which is why we should push for regulations that require satellite operators to have plans to de-orbit their satellites at end of life and/or not launch to orbits where the launching stage cannot quickly de-orbit within a set number of years. However simply pushing for all countries to launch less things into space, as you seem to be saying, is not a good idea.
They also lugged a large amount of extra dry mass into space because of the mass of the orbiter. The part that was least important to re-use and most expensive to re-use is the part that was re-used the most.
maybe, but it was the coolest part to re-use. whatever inefficiencies the shuttle had that people squabble over, it was a very inspiring program. not sure how much inspiration to a small kid a SpaceX launch is, but i know exactly how much the shuttle was.
>not sure how much inspiration to a small kid a SpaceX launch is, but i know exactly how much the shuttle was.
From what I've seen, it sounds like many kids these days feel the same way about SpaceX launches now as you did about the shuttle (incredibly inspiring). I hope that's reassuring.
it is. i know that i flew my shuttle around my room much more than my apollo rockets. my fighter jets had a hard time keeping up with the shuttle in my room too. i once saw footage of the "chase" planes as the shuttle screamed past them on approach. just so much more "cool" to the shuttle than a capsule.
This video showing kids’ reaction to the Falcon Heavy inaugural launch is pretty fun. The dual-landing is absolutely insane. Straight up science fiction.
https://youtu.be/A0FZIwabctw
What do kids ever have to do with anything? They are the next generation that may or may not want to be involved in whatever they might be getting inspired by.
>The goal is to get to space affordably and reliably; the shuttle failed on both fronts SpaceX is succeeding at.
Shuttle had ~135 missions (number from memory) with 1 failure at launch. The second failure was at re-entry, so assumption is that the mission deployed whatever was being deployed (if that was part of the mission). How many missions has SpaceX lost payloads on? >1? If we do percentages, sure, but to say that 1 failure at launch is unreliable is just farcical.
Are you actually trying to say the mission when Columbia came apart on re entry and killed 7 people was successful? And you’re gonna say the challenger mission failure that killed people is the same as SpaceX losing a satellite and a dragon cargo capsule.
Rockets landing vertically is the most sci-fi thing ever and has been a staple of science fiction since the early 20th century. I think that's plenty more cool than landing it horizontally.
Anyway, coolness factor doesn't really matter for space. Economical things also tend to end up looking cool over time anyway simply because streamlined designs which are efficient tend to also look cool..
> inefficiencies the shuttle had that people squabble over, it was a very inspiring program
Exactly. The shuttle program was designed by bureaucrats to be "inspiring." Which explains why it was such a dismal failure economically and in turns of safety - killing 14 people which is far more than any other rocket, and costing insanely more than expendables.
The shuttle program set the U.S. space program back by decades. We are only now finally starting to recover from its dismal failure.
This number is total nonsense. It is derived from simply dividing the total cost of the program by the number of launches. This is useless. First, the R&D costs were sunk all the way back in the '70s. Second, the cost of personnel, operations, and facilities do not necessarily scale with the number of launches.
NASA put the cost of a launch at about $450 million in 2011, but that is pessimistic. That is simply the years cost divided by the number of launches in that year. The marginal cost, i.e. the cost of going from N launches to N+1, will always be lower. The Shuttle was designed to fly 24 missions per year and actually flew an average of under five missions per year. The lack of demand was a substantial reason for the high cost.
I find it tedious that every conversation on this topic has to start with the same tired old memes. No, the Shuttle did not cost $1.5 billion per launch. No, reusable launch vehicles are not new. No, not even reusable launch vehicles that land vertically under power are new. No, the price to LEO has not fallen by an order of magnitude. No, ULA is not dead in the water. No, neither is Ariane. Yes, the Shuttle program was badly mismanaged. Yes, the Falcon 9 is pretty cool.
I didn't know the SRB components were also re-used!
It's a useful clarification that the Space Shuttle program had a large amount of reuse in the design ... but it still cost over a billion dollars for each reuse, many multiples of e.g. fully expended soyuz. So until SpaceX proved that reuse could be done well, and economically, it wasn't obvious. Now it's super obvious and practically all space programs are frantically trying to catch up.
(The SS could carry fairly large ISS module payloads into orbit, while cheaper rockets/spacecraft couldn't ... but couldn't get to high orbits or beyond ... details details.)
Several weeks on-site vs years in a factory on the other side of the country. It’s a clear win. Marginal per launch cost is on the order of $15 million, including the price of a new upper stage. More full cost including GSE and amortization on the booster and fairing more like $25 million.
No outsider knows if Falcon reuse is economical. SpaceX keeps on raising money (they have a lot of going on, so that’s not surprising) and without insight into financials we don’t known if their launches are subsidized by VC money.
I think another factor many people neglect is that SpaceX benefits from economies of scale. SpaceX launches very large rockets and launches them very often. Large rockets and a regular cadence are both known to reduce the cost per ton to orbit. SpaceX has the ability to put a whole lot of tonnage in orbit and this immense capacity means they can do it cheaply.
However, this only works if there is enough demand to use all their capacity. At the moment this demand simply does not exist. SpaceX manufactures its own demand. The majority of its capacity goes to launching its own satellites. This lets it keep its cadence, but literally burns the company's own resources to do so.
Whether or not you think Starlink makes any sense in the long run, at the moment it is losing money hand over fist. It serves as an indirect subsidy for SpaceX's other customers.
You are right that no one knew that Space X's approach would work until they demonstrated it. However, it was clear that the shuttle's "reuse" approach was a dismal failure early on - even before it was built, the sheer mass shuttled to and from orbit every time was a giant red flag to anyone who understands the basic rocket equation. That also drastically limited its altitude and usefulness. In a world where practicality matters, it should have been abandoned early on in favor of a return to expendables, until we figured out something that actually worked.
Mostly the heavy steel booster cases were reused, iirc. I don’t believe the reuse process was ever cost-effective. As for the Shuttles themselves, so much inspection and refurb was required that it made reuse look unattractive.
SpaceX is getting the entire first stage back in mostly flight-worthy condition. Nobody outside the company truly knows how much work is done between flights but it appears to be vastly less than Shuttle.
Can you expand on this? I've looked for information on how much rework they require but have never come up with a good explanation. I'm curious because spaceflight requirements are typically very stringent and refurbishing back to those requirements isn't a trivial task. A lot of the cost for Shuttle was refurbishing to meet the quality requirements.
I could be wrong but I believe they use brand new first stages for any manned mission because the stakes are higher. And they allow customers to elect for a brand new first stage at added cost. So the standards for refurbishment may be not quite as strict for regular payload/resupply missions because worst case scenario you're blowing up tens of millions of dollars of equipment/supplies instead of 3 human beings.
This makes sense and falls inline with how NASA has always ratcheted up requirements as the risk increases. The highest risk level is human-rated flight, but they also gauge things like payload cost and "one-of-a-kind" nature of payloads.
I think someone more familiar with this than me could look at the cost of a Falcon 9 launch with a re-used first stage and compare it to a non-re-used first stage launch cost and make some assumptions/ballpark guess about how much work goes into refurb.
They weren't fully re-used as each shuttle booster ejected the engine nozzle right before impact with the water. Additionally, they had to be disassembled before re-filling with fuel. It was re-manufacturing versus re-use.
According to that list 2023 should be even more amazing. They have ~50 commercial or gov customer flights scheduled vs the ~30 they did this year.
I wonder how much they can increase the flight rate, if they can't they'll face the dilemma of not having enough capacity to also launch the same or more Starlinks.
What are the odds Starship begins launching Starlinks next year? If it doesn't they may want to open another pad, there's one available in Kourou.
I think an orbital launch attempt for Starship is pretty likely in 2023. They seemed to have planned to try deploying Starlink with it from the start but have changed those plans and seemingly don't plan to do so for the next 2-3 Starship prototypes (they sealed off the "PEZ" style dispensers).
Personally I don't think they'll be able to get more than 3-4 launches of it in, in part due to regulatory considerations. In which case maybe we'll see 1 or 2 Starlink launches on Starship depending on how well the first flight goes.
Idk man, I totally expected them to launch in 2022. They don't seem to have much confidence, or they would have done so. Especially since the "regulatory considerations" you mentioned include limits per year. Their production seems good enough that just disposing of them by test flight would be a win. So why no launch attempt, yet? Is risk not acceptable anymore since Artemis? Yeah, no bets from me for now, but will be interesting to see what happens.
My personal opinion is that somewhere along the line this year they had to sharply recalibrate their risk tolerance. I think they might have expected to be able to more easily extend their launch approval out of Starbase as needed (ie "regulatory considerations"). Alternatively maybe they're expecting a much sharper drop in investor capital in the next year or two.
As such they may have had to change their success criteria for the first launch such that while previously they would've considered simply not damaging launch infrastructure to be a success, now they might need to at least be able to make orbit.
An alternative explanation is simply that most of their work this year has been on "stage zero" - the launch infrastructure. It isn't as open to the sort of rapid iteration seen with the vehicles themselves but it needs to be sorted out so they can start building the various additional pads they want in Florida. So perhaps that's what's preventing them from launching for now.
Speaking of stage zero, I have difficulty wrapping my head around how they will test landing. The risk of damaging it seems like it would be quite high the first 20 times or so they try. Interrupting launches to repair damage seems like too large of a penalty.
Their plan appeared to be to test the flight profile at sea. So they would specify some precise GPS coordinates to perform a soft touchdown at off the coast of Hawaii and once they're confident with that they can test on the real thing.
Additionally, part of the reason for building so many pads up front (IIRC they have two in various stages of construction at Cape Canaveral) is probably so that one failure doesn't stop everything.
Besides that, the booster flies a trajectory such that it'll just crash into the water nearby if it isn't able to initiate its catch burn. On top of that, unlike F9, the Superheavy booster is meant to be able to hover meaning it has a few seconds to make any corrections that may be necessary. Catching the ships themselves will certainly be much trickier, although at least they have the somewhat uncanny looking velocity when skydiving at low altitudes.
I heard about how the first launch was planned to be expended near Hawaii, perhaps they'll attempt to hover before doing that. But even if they do that with the first few at some point they'll have to attempt a first catch, that will be quite a nail biter. It will be interesting to know how many they are willing to expend to get confident enough they won't destroy the launchpad.
But recall earlier this year one of the droneships was damaged by a landing of a returning F9. That was after over 100 landings.
Another part of this is how many successful landings are they going to do before people are allowed to be on board. How many would you want before you would feel comfortable? 5? 20? 100?
To achieve all this tolerances for Starship are going to have to be so much tighter than Falcon.
This seems generally right for a positive scenario.
One thing which they will need to complete before launching Starlinks V2 from Starship is launch a few from Falcon 9. They'll need some prototypes in orbit before they go into the kind of mass production they would want for Starship.
Hi, I help maintain the wiki page (different username than this one), and one thing to think about with regards to the "planned" number is that there's substantial numbers of satellites that are completely unknown until just before launch, either because they're a commercial customer who doesn't want to reveal business plans or whether it's a government customer wanting to launch a spy satellite with short notice given to possible other countries.
This is especially notable with SpaceX's own launches as SpaceX apparently shoves them into gaps in their schedule whenever there isn't a customer payload to launch. So suddenly long stretches of no launches can become suddenly filled with a bunch of Starlink launches.
Alternatively, many satellites are delayed (or canceled), often for multiple years, because of construction issues, regulatory issues, or funding issues (or sometimes bankruptcy). If you go back in Wikipedia history and compare what the list of launches for 2022 was at the end of 2021 versus what actually launched, there are substantial differences. See: https://en.wikipedia.org/w/index.php?title=List_of_Falcon_9_... (There were 38 non-Starlink launches planned for 2022 at the beginning of 2022, and only 27 launched.)
Those tables with the "planned" column explicitly only includes launches for which there is some kind of source stating that they'll be launching on Falcon 9.
> Alternatively, many satellites are delayed (or canceled), often for multiple years, because of construction issues, regulatory issues, or funding issues (or sometimes bankruptcy).
I think this was particularly exemplified by Falcon Heavy launches. I think there were supposed to be 5 in 2022. Only 1 actually made it across the line. We'll see how many happen in 2023.
This is a fantastic resource, thanks for posting it. Using it, I realized that in 2022, SpaceX had no landing failures. Their rockets nailed every landing they attempted, which is even more impressive when you realize they flew twice as many missions as in 2021.
Elon is self-aggrandizing, pompous, and indifferent toward the well-being of his employees - but does that mean that it's a bad thing when his company sends rockets into space? Thousands of people who aren't Elon contributed their effort and talent to this feat - Does Elon's behavior around twitter negate their contributions?
OK I see your points, still, a personal opinion (he disgusted me) is not an attack. The real world is not off topic. And using violent words like "hijack" to describe an online conversation seems to me inadequate.
Maybe not for you. However, other people do not see the world in black and white, and can get over their hatred for a single person, and appreciate the technical accomplishments of thousands of others.
> Being done with someone after too many shitshows implies no black or white worldviews
“Being done” with anything, with no chance for reëvaluation, redemption or shading, is by definition black-and-white. That’s fine. Maintaining full color or even grayscale is hard work, and if we’re honest, we all simplify with stereotypes when topics don’t merit full attention.
But we shouldn’t do so blindly. And if we’re going to act on such a simplified worldview, e.g. by expressing a view on it, we should be able to take time and attention to look at the facts more stoically.
I would love to read a minute by minute breakdown of what happens the 48hrs before a launch and 48 hrs after a launch. Like to the level of detail of the order of attaching straps to lift the booster off a drone ship and the individual commands to bring the flight computers online.
edit: actually, i'm goign to ask that everyday astronaut guy to figure that out :)
The Nasa Space Flight live streams go into a lot of nerdy detail on selected missions and tests. They seem to refer to spreadsheets that they have built up based on timing vents and lots of telescopic viewing.
Space activity is really picking up. There were 177 successful orbital launches this year, and 135 last year. The best year after that was 1967 with 120 launches... We are in interesting times!
Jeff Bezos' Blue Origin is a competitor in theory but in reality as far as I know they have not won a single contract from SpaceX. Not sure if they're even in the ring.
> Jeff Bezos' Blue Origin is a competitor in theory
Not really, even in theory. It's only even remote current participation in the orbital market is the very overdue BE-4 engines it is developing for ULA (which I assume is “the consortium” that the upthread poster excluded in asking about competition.) Currently, its a suborbital launch company with aspirations of sometime becoming something more.
RocketLab (Neutron), Relativity (Terran R), and Blue Origin (New Glenn) are all trying. But none of them are there yet.
Out of the bunch, RocketLab has the most operational experience. They're also the only company (of the above three) that isn't currently aiming for 2nd stage reuse.
For people reading, just for clarification, but all three of these rockets have shown no hardware publicly of any of the vehicles besides some engine testing. They're all years away from launching, let alone reaching any kind of scale.
The other poster responded, but RocketLab's Neutron rocket has not flown at all nor has any public plans yet to fly. Those are flights of the Electron rocket.
RocketLab (https://www.rocketlabusa.com/) services a slightly different segment (i.e. not as heavy) but does take things regularly to space (and in the case of Capstone, to the moon!)
RocketLab has not had much success in raising its launch rate. Meanwhile SpaceX undercut them on $/kg with their rideshare program which likely robbed them of a large number of potential customers, so RLs segment was not as separate as they would have liked. Indeed in their short history SXs rideshares have launched considerably more payloads and more weight than RL in their whole history. There are of course some payload's that want to go to a specific orbit not served by SXs rideshare.
Yes its David Vs Goliath, but that sums up all space companies not backed by billionaires or Government Pork. saying that I am rooting for both to succeed.
Odd that the other comments don't even mention ULA and Arianespace. SpaceX definitely dominates the competitive market but there is some more nuance than that.
In head to head competition ULA won a larger share of NSSL launches than SpaceX. Obviously a lot of people think this will result in taxpayers overspending and was not a really fair competition.
ULA and Arianespace have had more success against SpaceX for GEO than LEO missions, though I believe that has lessened over time. GEO is more competitive because SpaceX uses RP1 for its upper stage which is not as efficient as the others that use hydrogen.
IIRC ULA's accuracy is also a bit better, so for an expensive payload going high up it can be worth it to spend a bit more if it means your $1b payload is gonna last a few more years before running out of fuel. There's also the whole vertical vs horizontal integration thing
That's somewhat incorrect, and accuracy of the sort you're talking about only really matters for vehicles without any on-board propulsion, which is a rare thing on military satellites. The accuracy certainly doesn't save you "years" of propulsion.
> Obviously a lot of people think this will result in taxpayers overspending and was not a really fair competition.
Part of this is due to SpaceX bidding Starship for NSSL phase 1 (development) and getting rejected by the Airforce, while ULA, Blue Origin, and Orbital ATK (now Northrup Grumman) got development money for their respective rockets. This meant that SpaceX had to amortize the cost to develop vertical integration, a larger fairing, etc. over their NSSL phase 2 (operational) bid.
> ULA and Arianespace have had more success against SpaceX for GEO than LEO missions, though I believe that has lessened over time. GEO is more competitive because SpaceX uses RP1 for its upper stage which is not as efficient as the others that use hydrogen.
From what I understand, this is mostly ArianeGroup. ULA does some high energy missions, but they're mostly for NASA/DoD. I'm sure they're hoping to improve their commercial competitiveness with Vulcan, though.
Many GEO missions were bid so far in advance that they were booked on Ariane many years before.
There is also some reason to believe that some of the European GEO missions were given insentives in various ways to fly on Arianespace. And then finally Arianespace simply had to sell all the Ariane 5s that were left. The Ariane 5 had at base something close to competitive in a very niche market of dual GEO launches and had to go away.
ULA won a larger share of NSSL but have done almost no commercial contracts ever. The slightly larger share of NSSL probably has more to do with the existing relationship and the long history and understanding of airforce processes. However, ULA is already going to miss their target launch dates for NSSL.
So the reality is that Arianespace and ULA are closer to being state entities then anything else.
In terms of actually BUILDING rockets and engines they are not even playing in remotely the same league as SpaceX. Ariane 6 is a slightly upgraded Ariane 5 and cost a huge amount more then the whole Falcon program including reusability and the Heavy (including Marlin development).
And Ariane 6 given its further delays is probably adding another couple 100 million on top of what is already committed. A follow up to the Ariane 6 will likely not show up until more then a decade.
So while these are big players, I don't think they are actually endangers competitors, more like big niches.
> GEO is more competitive because SpaceX uses RP1 for its upper stage which is not as efficient as the others that use hydrogen.
This is actually not true. SpaceX second stage is perfectly capable of doing what it needs and cheaper to produce.
Yeah but those 120 launches were of way smaller vehicles then those now. While now we launch some small vehicles. Falcon 9 is huge historically, and so are many of the rockets today.
So really the true measure should be tons to orbit, not number of launches.
And by that measure I would assume SpaceX alone out-flies 1967.
I had a similar experience a month or two ago at dusk. There was a broad line of twinkles moving across the sky. It took me a few moments to realize it wasn't a meteor breaking up but a new set of Starlinks doing their orbits together before moving into separate tracks.
Central Floridian here. Considering blocking notifications from my "space launch alert" app. Feels like every day between spacex and the other companies. It's amazing.
What's the use case for the app? Is it to remind the user to go outside and look up to see something interesting? Or does it serve as a warning system so that one is not spooked by mysterious loud noises?
Space seems to be mostly about satellites, not about "real" space. The Moon, Mars, yes they represent interesting multi-billion dollar contract opportunities paid by a few big governments; but I don't see any other market opportunity besides that.
My dad has been a space fanatic for his whole life, but because of where he lives he never had a chance to see a rocket launch. Last summer I took him to see one of the SpaceX launches. I’m really glad I did. Personally, it was one of the coolest things I’ve ever seen.
140 comments
[ 2.8 ms ] story [ 211 ms ] threadWatching SpaceX land rocket boosters on target again and again is kind of magical, like something out of Kerbal Space Program. I wonder how long it's gonna take to get a fully reusable rocket. They're working on the Starship and it's supposed to be fully reusable.
And SpaceX had to iterate a lot to get this working as well
It is rocket science
Ariane space was still arguing that reuse wasn't viable for them due to their low flight rate only until a handful of years ago. As far as I'm aware, ULA's CEO still insists that anything more than their proposed engine segment reuse is going to require too high of a flight rate to pay for itself (IIRC he said they'd need 10 flights per booster).
This is also part of why SpaceX's model works. Even when disposable F9 is much cheaper than previous rockets in its class.
I assume you mean the cost-plus contracting model? Or is there a different incentive mechanism you're referring to?
Early on, SpaceX's biggest hurdle was getting payload contracts. IIRC they had to sue the government several times to overturn bad acquisition policies. Government acquisition processes weren't suited to supporting small new entrances into the launch market and there was somewhat of a belief that you weren't a credible launch provider unless you were as large as Boeing or Lockmart.
In comparison nowadays it's relatively normal for lower cost programs or ones which can accept some loss to try flying on less proven rockets. New bold approaches to designing, building and flying rockets aren't immediately dismissed as just young naive engineers who haven't had the reality of spaceflight beat into them.
My understanding is that early contracts were awarded to ULA because SpaceX wasn't certified for military payloads. That, combined with your statement, implies the certification process levies bad requirements. Do we know what these bad requirements were?
Besides that, although not a lawsuit, Lori Garver's book "Escaping Gravity" goes into a lot of depth regarding the behind the scenes political struggle regarding commercial cargo and commercial crew and how Congress and some parts of NASA leaned on the idea that they couldn't tolerate any failure as a means of trying to disqualify anyone but ULA,Boeing etc from those programs.
I was wrong about them having to sue "several times" though.
I've seen the paperwork argument cut both ways. It can definitely create a bureaucratic mess. On the other hand, one of the things NASA tries to drive is "risk informed decision making". Part of this process involves writing waivers when a requirement can't/won't be met to document what the new risk is (and they have to document that they find that risk acceptable). A lot of times, this process was dismissed at "just large amounts of paperwork" but in dismissing it, they made it very hard for decision makers to fully understand what the residual risk was. It was interesting to me that paperwork was used as an excuse to not commit to the process, but when it was streamlined, it still wasn't used. To me, that feels more like avoiding accountability than avoiding needless paperwork. This misunderstanding of risk is at the heart of most of NASA's spaceflight mishaps.
I wasn't aware of Garver's book, but thanks for bringing it up. Sounds like a good read.
The innovation here isn't about reusability at all, it's about the economics of reuse. Maybe landing a rocket back on the pad rather than recovering them from the ocean makes it cheaper to reuse, has there been a study around that?
Surely this entails a host of quality checks between flights? What is that cost estimate at?
In short, performing RTLS requires additional fuel. This is fuel that can no longer be used to boost the payload into the correct orbital insertion. So all RTLS missions tend to be lower performance missions that don't need the full power of the Falcon 9 / Heavy since they have fuel to spare for RTLS.
Additionally, solid rocket boosters are some of the hardest things to re-use as the process of re-filling them is more like re-manufacturing them than actual true re-use. The sections (they had 4 sections) need to disassembled and refilled with the epoxy that is the solid rocket fuel. They were additionally made of steel rather than aluminum so were not substantially valuable components to recover versus the high end aluminum alloys in the external tanks.
It's about both.
SpaceX showed that the 1st stage of a rocket can be reliably reused in production. Which is really important from an architecture perspective because 2/3 stage rockets are just more efficient than 1/1.5 stage rockets (mostly due to the propellant mass fraction term of the rocket equation).
To give some examples, the payload mass fraction of some reusable rockets is:
* Shuttle(partial reuse): 1%
* Falcon 9(partial reuse): 4%
* Starship(full reuse, in development): 2%
From Wikipediea: "Out of 270 SRBs launched over the Shuttle program, all but four were recovered – those from STS-4 (due to a parachute malfunction) and STS-51-L (destroyed by range safety during the Challenger disaster).[6] Over 5,000 parts were refurbished for reuse after each flight. The final set of SRBs that launched STS-135 included parts that had flown on 59 previous missions, including STS-1.[7] Recovery also allowed post-flight examination of the boosters,[8] identification of anomalies, and incremental design improvements"
Also the rockets on the shuttle itself were landed and reused.
Falcon 9 First launch attempt 2007 [0]
Shuttle First launch 1981 [1]
I wonder if there's an advantage to having 26 years of watching someone else before designing yours? You also act like the rockets from the shuttle were not reused. They always (except for 2 instances) came back with the shuttle. The SRBs were also recovered, so it's not like these were wasted.
[0] https://en.wikipedia.org/wiki/List_of_Falcon_9_first-stage_b...
[1] https://en.wikipedia.org/wiki/Space_Shuttle
How much more of a interrogative statement does it need to be made to be considered asking a non-rhetorical question?
Also, this isn't just SpaceX, every launching country/company leaves second stages in orbit, some, unlike SpaceX, don't even try to dispose of their second stages and leave all second stages in orbit.
I'd recommend reading up more on the topic as I think you read something at some point and think that SpaceX is an especially bad actor in the space when in fact they are one of the most responsible actors, if not _the_ most responsible actor, in the space launch sector at the moment.
I didn't read anything one time in weird corner of the internet like seem to want to insinuate. There are plenty of sites where you can see the items being tracked. There is a large number of second stages in orbit that are not coming back any time soon. It might have been some coincidental bit of luck that on one particular perusal of one of these sites I just happened to click on 2 such items. Again, as they continue to increase the rate of launches, this will become an issue. And obviously, I don't believe SpaceX is the only such party doing this. That's just an unintelligent comment to have made.
You seem to be saying that all entities globally should as a whole launch less objects into space, as you imply that putting more objects into space is a net negative. That's fine to think but I strongly disagree that we as a planet should launch less objects into space.
> There is a large number of second stages in orbit that are not coming back any time soon.
I agree, which is why we should push for regulations that require satellite operators to have plans to de-orbit their satellites at end of life and/or not launch to orbits where the launching stage cannot quickly de-orbit within a set number of years. However simply pushing for all countries to launch less things into space, as you seem to be saying, is not a good idea.
From what I've seen, it sounds like many kids these days feel the same way about SpaceX launches now as you did about the shuttle (incredibly inspiring). I hope that's reassuring.
What do kids ever have to do with anything? They are the next generation that may or may not want to be involved in whatever they might be getting inspired by.
>The goal is to get to space affordably and reliably; the shuttle failed on both fronts SpaceX is succeeding at.
Shuttle had ~135 missions (number from memory) with 1 failure at launch. The second failure was at re-entry, so assumption is that the mission deployed whatever was being deployed (if that was part of the mission). How many missions has SpaceX lost payloads on? >1? If we do percentages, sure, but to say that 1 failure at launch is unreliable is just farcical.
That’s one of the worst takes I’ve ever heard
Anyway, coolness factor doesn't really matter for space. Economical things also tend to end up looking cool over time anyway simply because streamlined designs which are efficient tend to also look cool..
Exactly. The shuttle program was designed by bureaucrats to be "inspiring." Which explains why it was such a dismal failure economically and in turns of safety - killing 14 people which is far more than any other rocket, and costing insanely more than expendables.
The shuttle program set the U.S. space program back by decades. We are only now finally starting to recover from its dismal failure.
It was designed in an environment they couldn't iterate and test, and tried to go from very little knowlage directly to a massively complex system.
And no only that, they were very limited in terms of funding, so anything that required to much development had to be kicked.
NASA put the cost of a launch at about $450 million in 2011, but that is pessimistic. That is simply the years cost divided by the number of launches in that year. The marginal cost, i.e. the cost of going from N launches to N+1, will always be lower. The Shuttle was designed to fly 24 missions per year and actually flew an average of under five missions per year. The lack of demand was a substantial reason for the high cost.
I find it tedious that every conversation on this topic has to start with the same tired old memes. No, the Shuttle did not cost $1.5 billion per launch. No, reusable launch vehicles are not new. No, not even reusable launch vehicles that land vertically under power are new. No, the price to LEO has not fallen by an order of magnitude. No, ULA is not dead in the water. No, neither is Ariane. Yes, the Shuttle program was badly mismanaged. Yes, the Falcon 9 is pretty cool.
It's a useful clarification that the Space Shuttle program had a large amount of reuse in the design ... but it still cost over a billion dollars for each reuse, many multiples of e.g. fully expended soyuz. So until SpaceX proved that reuse could be done well, and economically, it wasn't obvious. Now it's super obvious and practically all space programs are frantically trying to catch up.
(The SS could carry fairly large ISS module payloads into orbit, while cheaper rockets/spacecraft couldn't ... but couldn't get to high orbits or beyond ... details details.)
It is not clear if SpaceX is there per flight. They still need to spent several weeks on refurbishing of the landed booster.
You link seems to be specifically about Soyuz MS, a capsule. No idea how much SpaceX spends on refurbishing dragon.
However, this only works if there is enough demand to use all their capacity. At the moment this demand simply does not exist. SpaceX manufactures its own demand. The majority of its capacity goes to launching its own satellites. This lets it keep its cadence, but literally burns the company's own resources to do so.
Whether or not you think Starlink makes any sense in the long run, at the moment it is losing money hand over fist. It serves as an indirect subsidy for SpaceX's other customers.
SpaceX is getting the entire first stage back in mostly flight-worthy condition. Nobody outside the company truly knows how much work is done between flights but it appears to be vastly less than Shuttle.
Can you expand on this? I've looked for information on how much rework they require but have never come up with a good explanation. I'm curious because spaceflight requirements are typically very stringent and refurbishing back to those requirements isn't a trivial task. A lot of the cost for Shuttle was refurbishing to meet the quality requirements.
B1061 launched both Crew-1 & Crew-2, B1067 both Crew-3 & Crew-4 (with a commercial launch in-between). Crew-5 was new and Crew-6 seems to be planned as well. https://en.wikipedia.org/wiki/List_of_Falcon_9_first-stage_b...
https://en.wikipedia.org/wiki/List_of_Falcon_9_and_Falcon_He...
https://en.wikipedia.org/wiki/List_of_Falcon_9_first-stage_b...
I wonder how much they can increase the flight rate, if they can't they'll face the dilemma of not having enough capacity to also launch the same or more Starlinks.
What are the odds Starship begins launching Starlinks next year? If it doesn't they may want to open another pad, there's one available in Kourou.
Personally I don't think they'll be able to get more than 3-4 launches of it in, in part due to regulatory considerations. In which case maybe we'll see 1 or 2 Starlink launches on Starship depending on how well the first flight goes.
As such they may have had to change their success criteria for the first launch such that while previously they would've considered simply not damaging launch infrastructure to be a success, now they might need to at least be able to make orbit.
An alternative explanation is simply that most of their work this year has been on "stage zero" - the launch infrastructure. It isn't as open to the sort of rapid iteration seen with the vehicles themselves but it needs to be sorted out so they can start building the various additional pads they want in Florida. So perhaps that's what's preventing them from launching for now.
Additionally, part of the reason for building so many pads up front (IIRC they have two in various stages of construction at Cape Canaveral) is probably so that one failure doesn't stop everything.
Besides that, the booster flies a trajectory such that it'll just crash into the water nearby if it isn't able to initiate its catch burn. On top of that, unlike F9, the Superheavy booster is meant to be able to hover meaning it has a few seconds to make any corrections that may be necessary. Catching the ships themselves will certainly be much trickier, although at least they have the somewhat uncanny looking velocity when skydiving at low altitudes.
But recall earlier this year one of the droneships was damaged by a landing of a returning F9. That was after over 100 landings.
Another part of this is how many successful landings are they going to do before people are allowed to be on board. How many would you want before you would feel comfortable? 5? 20? 100?
To achieve all this tolerances for Starship are going to have to be so much tighter than Falcon.
One thing which they will need to complete before launching Starlinks V2 from Starship is launch a few from Falcon 9. They'll need some prototypes in orbit before they go into the kind of mass production they would want for Starship.
This is especially notable with SpaceX's own launches as SpaceX apparently shoves them into gaps in their schedule whenever there isn't a customer payload to launch. So suddenly long stretches of no launches can become suddenly filled with a bunch of Starlink launches.
Alternatively, many satellites are delayed (or canceled), often for multiple years, because of construction issues, regulatory issues, or funding issues (or sometimes bankruptcy). If you go back in Wikipedia history and compare what the list of launches for 2022 was at the end of 2021 versus what actually launched, there are substantial differences. See: https://en.wikipedia.org/w/index.php?title=List_of_Falcon_9_... (There were 38 non-Starlink launches planned for 2022 at the beginning of 2022, and only 27 launched.)
Those tables with the "planned" column explicitly only includes launches for which there is some kind of source stating that they'll be launching on Falcon 9.
I think this was particularly exemplified by Falcon Heavy launches. I think there were supposed to be 5 in 2022. Only 1 actually made it across the line. We'll see how many happen in 2023.
https://docs.google.com/spreadsheets/d/e/2PACX-1vQ4zCmVceLNd...
60% growth year on year still going strong.
Raw sheet, if someone wants to clone: https://docs.google.com/spreadsheets/d/11wy9u3hKQjgdmC1N3Clg...
But, yes, the negative emotional charge in that short a post is getting pretty close to flamewar.
“Being done” with anything, with no chance for reëvaluation, redemption or shading, is by definition black-and-white. That’s fine. Maintaining full color or even grayscale is hard work, and if we’re honest, we all simplify with stereotypes when topics don’t merit full attention.
But we shouldn’t do so blindly. And if we’re going to act on such a simplified worldview, e.g. by expressing a view on it, we should be able to take time and attention to look at the facts more stoically.
edit: actually, i'm goign to ask that everyday astronaut guy to figure that out :)
https://en.wikipedia.org/wiki/Timeline_of_spaceflight#Orbita...
Space activity is really picking up. There were 177 successful orbital launches this year, and 135 last year. The best year after that was 1967 with 120 launches... We are in interesting times!
This is a golden age of (new)spaceflight and I'm thrilled to be living it.
Not really, even in theory. It's only even remote current participation in the orbital market is the very overdue BE-4 engines it is developing for ULA (which I assume is “the consortium” that the upthread poster excluded in asking about competition.) Currently, its a suborbital launch company with aspirations of sometime becoming something more.
Could also be referring to "the National Team".
RocketLab (Neutron), Relativity (Terran R), and Blue Origin (New Glenn) are all trying. But none of them are there yet.
Out of the bunch, RocketLab has the most operational experience. They're also the only company (of the above three) that isn't currently aiming for 2nd stage reuse.
It'll be interesting to see how things shake out.
https://www.rocketlabusa.com/missions/upcoming-missions/
In head to head competition ULA won a larger share of NSSL launches than SpaceX. Obviously a lot of people think this will result in taxpayers overspending and was not a really fair competition.
ULA and Arianespace have had more success against SpaceX for GEO than LEO missions, though I believe that has lessened over time. GEO is more competitive because SpaceX uses RP1 for its upper stage which is not as efficient as the others that use hydrogen.
Part of this is due to SpaceX bidding Starship for NSSL phase 1 (development) and getting rejected by the Airforce, while ULA, Blue Origin, and Orbital ATK (now Northrup Grumman) got development money for their respective rockets. This meant that SpaceX had to amortize the cost to develop vertical integration, a larger fairing, etc. over their NSSL phase 2 (operational) bid.
> ULA and Arianespace have had more success against SpaceX for GEO than LEO missions, though I believe that has lessened over time. GEO is more competitive because SpaceX uses RP1 for its upper stage which is not as efficient as the others that use hydrogen.
From what I understand, this is mostly ArianeGroup. ULA does some high energy missions, but they're mostly for NASA/DoD. I'm sure they're hoping to improve their commercial competitiveness with Vulcan, though.
There is also some reason to believe that some of the European GEO missions were given insentives in various ways to fly on Arianespace. And then finally Arianespace simply had to sell all the Ariane 5s that were left. The Ariane 5 had at base something close to competitive in a very niche market of dual GEO launches and had to go away.
ULA won a larger share of NSSL but have done almost no commercial contracts ever. The slightly larger share of NSSL probably has more to do with the existing relationship and the long history and understanding of airforce processes. However, ULA is already going to miss their target launch dates for NSSL.
So the reality is that Arianespace and ULA are closer to being state entities then anything else.
In terms of actually BUILDING rockets and engines they are not even playing in remotely the same league as SpaceX. Ariane 6 is a slightly upgraded Ariane 5 and cost a huge amount more then the whole Falcon program including reusability and the Heavy (including Marlin development).
And Ariane 6 given its further delays is probably adding another couple 100 million on top of what is already committed. A follow up to the Ariane 6 will likely not show up until more then a decade.
So while these are big players, I don't think they are actually endangers competitors, more like big niches.
> GEO is more competitive because SpaceX uses RP1 for its upper stage which is not as efficient as the others that use hydrogen.
This is actually not true. SpaceX second stage is perfectly capable of doing what it needs and cheaper to produce.
So really the true measure should be tons to orbit, not number of launches.
And by that measure I would assume SpaceX alone out-flies 1967.
I was able to see, with my naked eye, as the rocket fired off some sort of burn.
It looked like an airplane with headlights pointing backwards.
It wasn't until the next day that I confirmed what I had seen when I read about it on my local Reddit sub.
It was really cool to have actually seen a the-future-is-now event like that.