Unlike most of Musk's other ventures, Starship keeps it together for Flight Test 10 SpaceX has finally managed a test flight of Starship without anything creating an impromptu firework display. Aside from some very minor issues, Flight Test 10 was a complete success for the company. This will be a great relief to its engineers and NASA managers, who depend on Elon Musk's mighty rocket for their lunar …
My expectations are none and have been since I compared the weight, fuel weight, and thrust of Saturn V and Starship. The payload target is now 50 tons, less than Falcon Heavy. Also, they've used $3bn of taxpayer money on this so far and should have been on the moon by the end of the first half of this year.
Where did you get that figure?
Space X website shows ~68t for Falcon Heavy and 100-150t for Starship
From what I can remember, they are aiming to get more with the Raptor 3.
"Where did you get that figure?"
Elon. The aspirational payload mass for Starship was 100t initially but calculated to be about 30t for v1 according to The Man Boy-Child. Right now the proven mass aside from some dummy payload satellites is grams (a banana) and that's on a sub-orbital parabolic trajectory, not orbit. To get to orbit, Starship needs to speed up another 2,000kph. Dissipating that increased energy and coming to a demonstrated controlled "landing" in the ocean is going to be a big hurdle. I'll bet Mexico will not allow an overflight to bring a test article back to Boca Chica before that happens. Enough politicians in the US can be convinced($$$) to let them land in Florida when the time comes, but even that is going to require the Space Force to allow them to operate Starship from the Cape in the first place. A big accident there could severely cripple space flight operations for lots of craft.
"What would be your expectations for a realistic but ambitious test programme of a novel, reusable rocket of this weight class?"
Novel? Get a copy of "Rocket Propulsion Elements" by Gary Sutton.
The chant has been for "reusable" but nobody is doing the sums to see if it's even a good idea in every situation. Launching a mega-constellation, yes. Other stuff, not really. The only application for something as large as Starship is dispensing litter into LEO and maybe sending cargo to the moon. Mars is not a thing. The companies that build satellite busses aren't building anything so large as to need Starship or in a form factor that would be squirted out a thin and wide door in the side. Falcon 9 Heavy last flew in Oct 2024. There's an expected flight in Dec 2025 and a few tentatively in 2026 including lunar lander missions for Astrobotic. Not a lot of flights and New Glenn will be coming online which might steal some heavier payloads from SpaceX.
Exactly.
I don't know who is in charge, but that person is obviously capable of keeping His Muskiness at arm's length (maybe suggesting another problem with X to keep him away) while doing real, actual Science (and work).
Well done to that wizard.
Wizard...
Gwynne Shotwell
This. Gwynne Shotwell is very capable, and that she keeps Elmer Muskrat away from the real operations is a blessing.
That said, the countless engineers making this stuff happen definitely deserve the credit they get for pulling this off (eventually, finally), because it means they're making progress towards heaving big stuff up (ok, they already do with Falcon Heavy, but this is new and different).
As much as I utterly detest that wankstain with the four-letter names, Shotwell and her merry band of engineers deserve that beer tonight.
Lets take a look at the competition. The Long March family of rockets have launched 591 times with no successful landing. Not all of them smahed into the sea... Some performed energetic lithobraking manoeuvers instead. Soyuz 2 has launched 154 times with a similar success rate to the Long Marches. Ariane 5 launched 117 times and almost all their boosters reached the sea - at rapid velocity. Atlas V has launched 102 times and not a single booster survived. Vulcan and Ariane 6 have each launched three times with no successful landings but it took SpaceX 7 attempts to land a Starship booster so give them time. New Glenn has launched once. I could die of old age before they reach 7. SLS will probably not get that far.
RocketLab launched 66 times and have recovered 7 boosters. Their first success was on the 16th attempt - ahead of SpaceX Falcon 9 which first landed on flight 20. The Space Shuttle is a bit of a mixture. The orbiter mostly came back intact but without the main tank. Some of the SRBs were recovered but recycled would be more accurate than re-used.
I take your point, but your comment is a bit disingenuous - Up until relatively recently, boosters have been throwaway components rather than carrying extra fuel or parachutes to enable recovery.
So the fair comparison is "how many rockets exploded before they had finished their task?".
Now recovery is a goal too. I assume SpaceX has the real numbers for recoverable vs throwaway costs and has decided that recoverables will be cheaper in the long run.
Different rockets had different goals so the only fair comparison is to judge each against its own goals. The most common was "How much can I screw the government for?". Atlas V and Delta IV Heavy were both really successful at achieving that goal. For Soyuz, the answer was "Bugger all", which made it a viable commercial launch vehicle. Ariane V's clever trick was to deliver two satellites at a time to GTO. Getting paid twice for the same launch made it a commercial success.
Falcon 9 was still intended to be a cash cow, but with the assumption: "If we focus on reducing price the market size will grow". This worked, and ate Ariane's and some of Soyuz's commercial lunch. Then SpaceX went for re-use. Before that started working ULA decided to retain their US government business by replacing Atlas & Delta with Vulcan. That business did not require re-use. Vulcan is very late and now that it is mostly here the market has changed. Ariane is in a similar position with Ariane 6: there were only about a dozen EU launches per year. That could be done with one re-usable booster per year. It was much easier to sell A6 as a jobs program by building 12 expendable boosters per year. SLS is by far the most successful rocket ever. The goal was to cost as much as possible. No-one has made a more expensive rocket.
Now the market has completely changed. Low prices have brought in new customers and all the big governments want their own Starlink (including the US military :-). Those low prices come from re-use and dividing fixed costs by a large number of launches. ULA and Ariane committed to designs that are not compatible with re-use. Roscosmos has no budget. RocketLab aimed for small launch. They are good at it but there is only a small number of small launches (to unusual orbits). Most of the small launch market takes a ride share on a medium rocket to the most popular destination: Sun Synchronous Orbit.
The real competitors for re-use are China's collection of new commercial space companies, RocketLab's Neutron, Stoke's Nova, Firefly's Eclipse, Rocket Factory Augsberg and Blue Origin. All of them have barely started. Too early to decide which are going to be successful.
The really different one is Starship. The goal for that is colonise Mars. That goal comes from the mouth of a very unreliable source. Independent evidence is required. 42000 Starlinks with a five year life span launched 60 at a time requires one Starship launching two or three times per week. Starfactory is massively oversized for Starlink. The factory is oversized for the entire launch industry. Even a dozen refueling launches for Artemis (every two years) is a small contract. The only thing that matches Starfactory's scale is hundreds of ships going to Mars fueled up by thousands of tanker flights.
The big thing going on is Starfactory and the Starship development launches are an entertaining side show created as a by-product. The current generation of ships and boosters have minimal value because SpaceX produce faster than they are licensed to launch. Most of the hardware is scrapped. Currently the only reason to land a booster is if it would be educational. There is a better one that will be ready to launch before a landed one can be re-furbished. One booster has launched twice - because finding out how to do refurbishment was educational. Finding the limits of what a Starship can do without exploding shows which parts of the design are weak and which are stronger than necessary. Such tests include a significant risk of crashing. This "landing" was aimed at the sea because that is cheaper to repair than rebuilding the launch infrastructure again.
Starship has not colonised Mars so it is currently a very long way away from being a success. Starfactory is ramping up to the cadence required for success.
"The only thing that matches Starfactory's scale is hundreds of ships going to Mars fueled up by thousands of tanker flights."
Who need hundreds of ships going to Mars? Elon, his techbro mates, and his political affiliates could all be crammed into a handful.
Well, to be fair to the NASA engineers who worked on SLS (a friend, dearly departed for around a decade now, worked on it too for quite a while and he was not impressed by the politics involved), the whole pig-snout-meet-trough thing started between Constitution and Independence Avenue in Washington, *not* at NASA. NASA had to get on with delivering stuff (via contractors with their snouts in said trough) or not get any money.
That's been the case for decades; and yes, you were absolutely 100% spot-on with ULA. Guess who is one half of ULA (no telling, Flocke)? Yeeeeeeah... that bunch of shysters. At least Lockheed had (via Skunk Works) a solid reputation for delivering stuff. The other lot, well... how delayed is their capsule now?
:-)
So the V2 bombing campaign was a complete failure
And rocket sleds are failures
And both virgin and blue origin's rockets, and of course all the thousands of sounding rockets which have been used over the years.
Some rockets are designed to put things in orbit - none of the flight profiles have yet included orbital injection as a planned event - quite deliberately so.
They are very, very, close to orbital velocity when they cut the engines - and that is exactly where they are aiming to be. It is not failure when you don't do something you're not aiming for.
"So the V2 bombing campaign was a complete failure"
Are you asserting that it was a success? Economically and militarily it was a disaster for Germany, although I suppose it was was a huge success for Nazi scientists who got a meal ticket to the US with no consequences. In a way poetic justice to watch fascism being recreated in America today.
They are very, very, close to orbital velocity when they cut the engines
empty, without payload ... where will it be with 100t of stuff on board ? The whole aerodynamic and mechanical balance will be changed when it's loaded with 100t on top ! I could bet 2 beers that this rocket is never going to put 100t in orbit. NEVER. I actually doubt that it will ever put anything in orbit, not even a banana.
I don't know of any example of a rocket that has been flight-tested empty. Heck, even the first Falcon was loaded with Musk's own Tesla roadster to make for weight. Testing a huge rocket empty and then loading it and expecting to behave as it was tested empty is unprecedented. As are the Raptor engines, the stainless steel structure ... something stinks fishy here
"empty, without payload ... where will it be with 100t of stuff on board?"
Hopefully in orbit, but that will be v3, not a v2.
We all know it's not how most rockets have been developed, but this is a deliberate choice.
As is the choice to continually assume that SpaceX are developing the rocket in the "conventional" manner, and that every flight that doesn't achieve orbit is therefore a failure...
"No, all rockets have the same goal : put something in Earth's orbit. Which Starship has failed to ever do in 10 attempts. Not even a banana"
No, some rockets are not meant to put things in Earth orbit. Instead, they take rovers to Mars or satellites to Saturn or into more heliocentric orbits.
Starship has yet to attempt getting into orbit. They need to prove out a lot of stuff such as once getting there, they can deorbit to land in a precise location. Just a burp of a raptor engine only shows they can relight the engine, which is good, but there's still a need to see how quickly propellants are going to boil off and if the can relight and engine, navigate to a defined location and land.
Starship is doing new things and things are being learned from failures. In fact, the recent launch had planned weak points. Few companies/governments have even gone to the trouble of landing their boosters. Some of the Falcon 9's have lasted for many launches, and even if not reused for financial efficiency, that aspect has purpose in developing experience. SpaceX JUST MIGHT have to rescue the current American space efforts again, yet with such huge resources, why can't other companies/NASA do it? Why so many screw-ups with human payloads? Is NASA really testing the equipment that they purchase? Is the vendor adequately testing the product? Are they flying people after only a few flight tests? I don't expect SpaceX to be perfect, but some of the other space vendors "starts with a B' should be doing much better.
NASA was too focused on reusability for human flights via the shuttle, and didn't care about reusability for launching stuff to orbit. They didn't do enough launches to make it worth the trouble of developing that capability.
And yes a lot of that had to do with politics. There have always been a lot of congressmen who wanted to see NASA's budget cut. Some called for them to be eliminated, claiming the private sector could take over - as if the private sector is ever going to do stuff like Hubble, Webb, Voyager or LISA. But that's science, and if you believe all the science you need is found in the Old Testament you don't want any of that stuff anyway.
>didn't care about reusability for launching stuff to orbit.
Launching stuff to orbit isn't NASA's job. That's what United Launch Alliance do, having a government agency do it would be communism. A monopoly cartel of all US space+defense companies providing services to the government is the only free-market efficient solution.
"Arguably they were used - decades later in the smaller scale X37B unmanned shuttle."
The massive size of the Shuttle wasn't needed for the ever smaller sensing platforms the spies, err, agents/agencies were able to build. There was no need for talkative astronauts to be along for the ride and "see things" the military mob didn't want them to know about.
Dream Chaser seems to be much closer to reality now.
"and didn't care about reusability for launching stuff to orbit. "
The Shuttle was horribly inefficient as a satellite launch vehicle. It was faster and cheaper to do it the old fashioned way and drop the used rocket in the drink afterwards. It was a truck and until ISS was built, it didn't have truck things to do. Servicing unique assets such as the Hubble Space Telescope was another application that worked well.
NASA does science and the private sector hardly does basic research anymore. By having the agency and giving them good funding, we get new ways to look at Earth, other planets and the stars. We gain knowledge about how human activities affect our ecosphere. The data can go on to help manage things like food production, water management and forestry. The bunch of failed attorneys in government don't seem to get it since there is no "spend this, get this return". Pure research always pays off, but not on a time table nor in ways that one expects. Bean Counters blow fuses with that sort of thing.
NASA did get a large amount of money from congress but were required to spend it on specific things. Top of the list was SLS. NASA's funding came with a warning: fund anything that competes with SLS and your budget will be slashed. Do not expect much from NASA until US voters can equate "jobs program" with "welfare for companies that make large election campaign contributions". In the mean time the Orion crew capsule can only be launched on SLS. SLS+Orion costs over $4B/launch. Artemis 2 must launch with crew for financial and political reasons. The heat shield problem identified by Artemis 1 had to be fixed without an extra uncrewed test launch. NASA put a lot of time and effort into that. I hope it will turn out to be sufficient.
Luckily SLS+Orion is overkill for sending crew to ISS. Commercial crew was a firm fixed price contract. Boeing and their supply chain are set up for cost plus. They are first rate at being able to correctly justify every possible cent of expenditure and every year of delay. For firm fixed price any money saved by trimming out some essential testing will be eaten by one of the suppliers. There has been little recent news on Starliner. NASA wants a diverse supply of crew launch vehicles. Boeing and congress do not want to pay to fix Starliner. Boeing and ARJ are in legal discussions over responsibility for thruster failures on the Starliner crew demo flight.
SpaceX avoided supply chain problems by producing in house anything that was not available at a competitive price. SpaceX also benefited from being in competition against a monopoly supplier that had spent years allowing high prices creep into high costs. SpaceX did the opposite. They spent years ploughing profits back into R&D to reduce costs. Falcon+Dragon have an excellent reliability record. The only crewed Starship contract with a "date" is the lunar lander for Artemis 3 (Starship HLS). Firstly ignore the date. The space suits and SLS will not be ready either so there is plenty of time. During that time you will see rapid iteration on the Starship design followed by a large number of Starlink launches. Last year I would have said Starship will not launch with crew until it is at least as safe as Starliner. With the US government's new focus on killing Americans I am not so sure anymore.
Getting to orbit is hard and expensive enough. Creating a complete supply chain because the available suppliers have no intention or ability to control costs makes the job more time consuming and expensive. Investment did not really become available in the required quantities until after SpaceX became successful. New entrants to the market will not be competing against SpaceX's current prices. They will have to go much lower because SpaceX has huge margins. To do that, new rockets must be developed with minimal R&D costs so they do not have too great an impact on the final price of launches. Stoke and RocketLab are showing excellent and extremely rapid progress on rockets with the potential to be cheaper to operate than Falcon 9. They might have working products next year. Wait at least two years beyond that before they have the experience to minimise refurbishment costs.
ESA's old funding model was money supplied by each nation was spent in that nation even if a more cost effective solution was available elsewhere. This was eventually identified as a barrier to progress. More recently significant funds have been distributed to companies showing progress with no national restrictions on spending. Some of the companies involved may have a potentially re-usable first stage in a couple of years.
China is funding a bunch of new commercial launch companies which are at different points along the line of progress towards re-usable rockets. The Chinese government understands the value of cheap launch and will spend the money required to get it. It has taken them a considerable amount of time to transition from solids and hypergolics to the cryogenic propellants required for cost effective re-use. At best they are on a level with Stoke and RocketLab. At worst they may be a little behind Europe - but with more reliable funding.
The only competitor that started earlier than SpaceX and had a bigger budget was Blue Origin. They started off as a minimally funded think tank exploring possible alternatives to rockets for getting to space. They patented landing a booster on a drone ship before SpaceX tried it. When size and funding increased it was not focused on what the market wanted. Blue won the contract to build the engines for ULA's Vulcan over a bid from ARJ. This hurt them two different ways. Increasing the thrust of BE-4 up to Vulcan's requirements added significant delays. Jeff Bezos then hired the management team from ARJ who turned Blue into an old space company: instead of milking the US government they turned their skill to milking Bezos. The drastic reduction in progress caused the most enthusiastic engineers to leave and found Stoke. Bob Smith has been replaced by Dave Limp and now Blue Origin is making slow rather than glacial progress.
So plenty of good and bad reasons why their isn't yet strong competition for SpaceX. Musk has put an exceptional amount of effort into changing that. Thanks to his efforts many people, businesses and governments are looking to invest in and contract with anyone but SpaceX.
Driving down the freeway at 55 MPH, the driver of the car you're riding in announces, "The left rear tyre just blew out, but the mission must continue apace. Everybody in the car needs to lean forward and to the right, to take the weight off the failed tyre."
And why would you design a mission where at the end, all your stuff explodes as it does in the end of the original Thunderbirds opening credits sequence?
You'd seem to prefer that there was no redundancy and the failure of one engine caused the whole mission to fail? Plenty of vehicles can handle a tyre blowout, some of them can even continue with their "mission" after such event - it doesn't work that way on cars because "cost"
I think the eventual commercial and government clients (not to mention any future astronauts sat on top) will be well pleased that the mission continued despite such an occurrence, and it demonstrates that actually as well as all the fancy fireworks, they are actually getting many things right
why would you design a mission where at the end, all your stuff explodesThe craft's capabilities need to be proven before it's tested on land, where it would make a much bigger and more expensive mess if something went wrong. They previously went through all of this while proving Falcon 9's Return To Launch Site capability.
And... well... my car doesn't have 33 wheels but if it did and I lost one I wouldn't be too concerned.
The test was all the not-blowing-up that it did before that, all the way to a soft touchdown on the ocean right on target. Which is the entirety of its future mission profile when coming down on catch arms, so the explosion after that was more of a bonus lightshow.
That they could bring the vehicle to a controlled landing at the designated location - when they do this for real (without exploding), the landing approach will be over populated areas of the US, so they need to demonstrate to the regulator that this can be done safely, even if the vehicle has had a rough ride on re-entry.
"That they could bring the vehicle to a controlled landing at the designated location"
The site in the Indian Ocean where they are discarding the used Starships is not that "controlled". It's a parabolic, sub-orbital trajectory so all they need to do is burn the engines long enough with the rocket going in that direction and it will impact that spot in the ocean. If the first orbital flights with recovering at Boca Chica come to pass, the descent will be over Mexico. Cape Canaveral has still not given permission for Starship operations so SpaceX's building there is taking a lot for granted. It would take traversing the whole of the contiguous 48 State of the US to RTB in Florida. It's also where the vast majority of US space operations are done so large scale incidents even within the boundaries of Cape Canaveral would be very bad and could cripple the ability of the US to launch some things for a long period of time. Some rockets only have facilities at one launch location. If that location were impacted (in any sense), the military may not be able to order up a quick bird to observe something that they want a look at. Not every payload can be launched on just any rocket.
Some of the earlier boosters reached the sea without exploding. The booster from IFT4 was recovered and dismantled. I don't know if that provided useful data for SpaceX but in theory it could have provided useful data to others.
for all we know the DoD might insist the vehicle is not left in a reusable state if it doesnt land on US soil,or can be recovered quickly by SpaceX, in case it were to be captured by foreign adversaries who might put what they learn from the technology to use in bad ways.
regardless, the size of these vehicles that are designed to stand vertically will never react well to falling on their sides from any height with a mixture of flammable gases waiting for an ignition source,its practically a design feature
>The damage wasn't caused by the explosion
I disagree; I don't see that you can assert that based on the SpaceX footage, and the footage definitely suggests the explosion is related. The explosion was directly beside the rear left flap (see below for engine positions). Ten seconds after the explosion, the damage to the rear left flap was first visible. Here's a timeline, using video timestamps from the Associated Press re-broadcast of SpaceX's coverage - Video link: https://www.youtube.com/watch?v=kh0dSsroJ2o
T+45:18 - rear left flap looks completely OK
T+46:59 - "energetic event" between engine positions E5 and E6, i.e. exactly where the rear left flap is
T+47:10 - view from front left flap along left side of Ship, can see damage behind rear left flap now. Material clearly moving in plasma stream behind flap.
Determining 'directly beside': The "left flap" is taken from looking at Ship from above the non-heat-shielded side. From Scott Manley's quick IFT-10 video, E5 is the engine position at the middle of the heat-shield side. You can see this from the plasma flow visible at T+52:38. That puts the explosion directly beside the left flap.
Note - the rear right flap was damaged earlier, I first noticed that visible at T+39:55. This is after the engine relight at T+37:55, and also just after a very brief "standby for Starcams" interlude, but well before the explosion.
Many times they mentioned they were stress-testing the frame to detect failure points, to find borderlines and places for improvement. They also removed heat tiles in key locations to simulate failures. There's your burn-through explanation.
Though they did suffer burn through on previous flights when stressing wasn't an objective.
I think the important thing with this flight was that the front flaps didn't appear to suffer any damage or plasma leaks round the hinge regions, showing significant progress.
You might be surprised to know that every currently man-rated reentry vehicle has "bits" flying off all the way through reentry - that's what an ablative heat shield is (and of course ditching all the other bits they don't intend to recover).
I do wonder whether the forward flaps will need to be enlarged to increase their authority if the SS is returning with, say, life support systems still in the payload bay - that would be a significant change to the centre of mass, which would probably require a much more forward balance of flap deployment. I haven't looked at the comparative flap angles through reentry, the forward flaps might be generally at only a small angle of attack already, implying that they have plenty of authority left.
https://youtu.be/f8hGjwaMKO8?t=9992
It did not ignite this time, OK, but I suspect that propellant is not expected to be free-floating within the structure.
The impulse from the gas leaks challenging the attitude control system after MECO was even be discussed on live stream.
So I'm not completely sure this should be called an overwhelming success.
More a repeat of IFT-7 with a bit more luck and less active ignition sources.
