Pretty impressive. The engineering team at SpaceX is really something. Some thoughts;
The 'chopsticks catch' was amazing to watch. Seems like it adds a lot of risk and clearly the booster needs additional fire suppression systems :-) perhaps the tower could mount something that sprays the booster like the barges have for the F9 boosters.
The heatshield held out for a much longer time, the asymmetric heating on the flaps was interesting. I had guessed that all four flaps would have equivalent heating based on an approach that was basically that side of the rocket perpendicular to the flow but it seems like that isn't the case. Still it seems like they are close to having something workable here.
The detonation at the end was pretty spectacular too, but I suspect that structurally the tanks failed as the rocket hit the water vs anything that was an engineering failure. Engineering it to be strong enough to land on water would presumably compromise the cargo to orbit number.
The use of Starlink was really interesting. The ability to get live video for the entire re-entry is pretty game changing for engineering. I'd guess there are even more 'views' than they showed (there would be if I were running things :-)) but overall that capability is something that really helps evaluate the changes made.
I can easily imagine that flight 6 will be nominal end to end without any unintended damage. That would enable, perhaps, one of their 'massive' Starlink missions to test cargo delivery. It will also start to give us some better numbers on exactly how much cargo Starship can put in orbit in 'full reuse' mode which is essential if they want to create a fueling station on orbit for the Artemis program.
Again, hats off to engineering at SpaceX, y'all did good.
As to whether catching the booster adds risk - I'm not sure it does.
First, to the extent the booster is out of position in the x (sideways) direction, the chopsticks can move to accomodate error. But actually I think this dimension is the easiest of the three, as the booster has plenty to time to null any error in this dimension.
In the y direction (direction of travel towards the tower), the rails on the chopsticks can cope with the booster touching down along quite a long distance. But importantly, they appear to be smooth, so the pins can initially skid along them and then the booster can swing if it has not fully nulled any horizontal movement.
In contrast, if the booster used legs and has not yet fully nulled any horizontal motion at touchdown, there is a greater risk of breaking a leg or simply tipping over.
And in the z direction, it should be possible for the chopsticks can absorb more vertical motion than legs can absorb, because you can easily build in huge springs/dampers/etc into the ground equipment without concern about mass.
Catching also puts the booster in tension rather than compression - it's easier to be rigid in tension than compression.
Finally, if legs were used, the engines would have to get close to the ground during landing, so reflected shock from the ground could cause damage. I know Falcon 9 does this, but the area of the base of Starship is much greater, so there's effectively less room for the reflected energy to escape. Catching completely removes this risk.
On balance, I think they would have better chance of success for each mission by catching. The main downside would be if you fail to catch, you may need to build a new tower, whereas a flat pad would be cheaper and easier to repair.
I am really curious what the maximum wind speed allowed for a booster landing will be. Upon landing, it has a lot of windage and not nearly as much mass as during take-off.
I have experience with docking large boats and it does seem to be a bit similar. In the case of boats, wind is a big deal, and the booster has nothing "below the waterline" to slow down the effects of wind.
The most challenging axis in my opinion is the roll axis of Super Heavy, if there is a roll angle error, the pins could not sit properly on the chopsticks and the whole booster slides off.
I'm actually wondering about that. If I understand correctly, the arms can move up and down, and pivot around the tower. This allows them to correct for some error in the rocket trajectory and also (presumably) "soften" the final contact. Between the nozzles and the arms, it gives SpaceX a lot of degrees of freedom in the final seconds (you can see how the booster kind of "hovered" right at the end) and in certain respects might even offer more forgiveness than the hard ground.
Could it smash into the tower? For sure. Would that be more dangerous than smashing into the pad? I don't know.
It's a new technique with which we don't have a lot of experience.
It helps enormously that unlike Falcon-9 this rocket can dial down the thrust of its engines low enough to be able to actually hover or to move arbitrarily slowly in the final meters before touchdown.
It can arrive to the designated intermediate point with some already good accuracy, and then take some time to trim the remaining errors to the noise level more slowly, possibly with feedback from the ground sensors.
The chopsticks also include rails with shock absorbers, the action of which can be seen in the view from the tower during the landing [1], so the required accuracy is probably relatively modest, provided one plans the maneuver carefully.
Yes, we should probably assume that the SpaceX engineers have considered all of the risks HN readers are able to come up with in a few hours. And that they have evaluated alternatives like the added weight etc of having foldable legs on the booster.
The catchzilla solution is an example of their amazing ability to think out of the box. This solution, and things like the rapid evolution of the Raptor engine (see picture, story here: https://medium.com/@futurespaceworld/the-evolution-of-spacex...), dynamic engine configuration (33, 13, 3, zero, up again) and control is almost magically impressive. This is the stuff of Sci-Fi, brought to life.
To me, it's not risk reduction that they're after with that booster catching mechanism, but weight reduction. Those landing legs that we've seen before (and the mechanism related to them) are costly weight that is absolutely necessary only on the rocket itself, because that is expected to land on its own somewhere on a bare rock. For booster however, it makes sense to have as much of such launch and landing weights externalized, considering booster's reduced use-case of starting from a spaceport and very soon ending up back there too.
The way the trajectory is designed is that it has to scoot over to the tower at the last second, and it only does that if it's really really sure it can make it, otherwise it crashes off to the side.
My thought (admittedly not well developed) is that smashing into a landing pad of concrete can damage that pad but it can be quickly repaired without affecting the ability to launch future rockets. If you damage the launch tower significantly you're going to have to suspend launches from it until you fix it. So the "higher risk" is more critical assets offline in the event of a non-optimal return.
> The detonation at the end was pretty spectacular too, but I suspect that structurally the tanks failed as the rocket hit the water vs anything that was an engineering failure.
It's possible that SpaceX programed the AFTS to trigger some time after the rocket touched down in the water. Just to make sure that it completely submerges quickly.
> I can easily imagine that flight 6 will be nominal end to end without any unintended damage.
I think it depends on what you mean by "nominal". SpaceX ultimately wants to catch the 2nd stage as well. I suspect that they are a ways off of that, since it would have to approach over land. The FAA is going to need to have very high confidence that it will do exactly what it's designed to do before they're going to allow that.
> It's possible that SpaceX programed the AFTS to trigger some time after the rocket touched down in the water. Just to make sure that it completely submerges quickly.
The SpaceX host on the stream said that they were going to try and touch it down on the water at more of an angle than the previous flight to attempt to get it to survive the initial splash-down so they could get some more data and video footage.
Obviously this wasn't guaranteed to succeed, but it indicates that they weren't planning to immediately detonate the ship on touchdown.
As I recall it was common on the early Falcon 9 landing tests that splashed down in the ocean to also explode after tipping over and smacking the water. Once they're actually landing them on a pad, tower, or ship that should be much less of an issue.
> Just to make sure that it completely submerges quickly.
Why would they want to do that? (genuinly curious).
I reckon there would be a lot of useful data left if they could recover or even just inspect the remains. The remains are one big tank, so it would have floated.
correct me if I’m wrong but I don’t think the second stage is meant to be caught. It will have legs to land on Earth/Mars without any landing infrastructure.
I imagine that they will be their own first customer - putting Starlink satellites into orbit while they are gaining confidence in the reliability of the system for external customers.
They have to prove out the landing of the second stage, either with another catch or with landing gear, which they need for the lunar lander anyway.
I think that has always been the plan. the V2 starlink sats were designed to fly on Starship. When Starship wasn't coming along as planned they shoe horned the guts of the V2 on to a smaller sat that became the V2 Mini.
If Im not mistaken, they already delivered a payload in one of the previous flights, because they cut the transmission for a while and didn’t show the payload bay like one previous flight.
They showed 4 streams at once during some of the reentry. One view of each control surface. They may have had still more views but just that 4 was a first.
I'm wondering if they'll be using the vertical equivalent of arresting gear on aircraft carriers[0]. See when fighter jets land on aircraft carriers? There's a cable that decelerates them. That, but for vertical landing.
The way these chopsticks are set forces the booster into a dangerous, snake like maneuver (a SnakeX maneuver) at the last second from a vertical setting, to get into the chopsticks. This maneuver is due to the fact chopsticks are short and the booster has to land on one point in space. No degrees of freedom.
Now, imagine if the chopsticks were long. The booster wouldn't have to land at one specific point, but it could now land on a line. One degree of freedom.
Now replace the long chopsticks with cables, and then add another pair of cables perpendicular to them. So you have a pair of parallel cables perpendicular to a second pair of parallel cables. Now the booster doesn't have to land on a line, but can land anywhere in the grid that's covered by the cables. Two degrees of freedom.
Pushing this thought leads to having a sort of iris diaphragm, like the ones in optics, but an iris diaphragm of cables. The diaphragm is open when the booster is about to land, then closes in quickly. Imagine this[1], but it's cables cinching in.
Now, it's a diaphragm of cables, not a diaphragm of rigid beams, so I imagine the deceleration to be even smoother as the cables elongate, and an additional system of springs and dampers to counter the weight of the booster.
The booster is vertical and stays vertical. Granted, an unstable equilibrium, but it beats doing the SnakeX maneuver to get to the chopsticks, and that's another story.
Now, imagine the iris cable diaphragm can move up and down like equipment handling containers, and now you have three degrees of freedom. That's less control to worry about on the booster's side at the worst possible moment, landing, where you can't make adjustments anymore.
This not only means being more forgiving on mistakes related to position, but also on speed and angle. The diaphragm catches the booster at any angle, and given that it cinches way above the center of gravity, the booster goes back to a more stable vertical position.
For the fire, maybe you just need a big hole down there.
This is a solution for a problem that's already solved, that is, booster maneuverability and accuracy. But they've just demonstrated that the booster is accurate and controlled enough to land on the chopsticks, and they have over a decade of experience in making rockets land accurately in a specific zone.
One thing to note is that the "Snake" maneuver was designed to keep the tower safe during the test, and not forced on the rocket by the chopsticks.
The rocket was set to come down on the pad just in front on the tower and launch mount until the final 3-engine landing burn started. This kept the infrastructure safe until the last moment, but also required that lateral translation you referred to.
I think the primary reason would be that landing legs are heavy, and it wastes performance to carry them. If your landing mechanism is mostly on the ground, you get that performance back.
Secondary reasons might include that it's simpler to get the booster right back to the pad. Once things have settled into an operational cadence, it's likely feasible to lower and lock the booster onto the stool, stack a ship on top, refuel, and relaunch -- no more messing around with barges, transport, weather issues, etc.
Why send the landing mechanism to space when it isn't needed there? Whatever kit you put on a rocket has to be brutally miniaturized to limit how much you eat into the payload mass. Also has to be rugged enough to withstand tremendous vibrations and thermal stresses. That adds cost and more points of failure. You want to move as much of the complexity off the rocket as possible. Then doesn't matter if the catching mechanism on the launch tower is big and heavy.
The rocket equation implies that if you want to maximize the delta-v a rocket gets out of a certain amount of fuel, then you should get the dry mass as close to zero as possible. Eliminating landing legs helps a lot.
The reflected sound of the engines is enough to destroy the engines, ironically. That's also why the launch mount is so high. You'd need truly enormous legs, which wouldn't work for weight.
You save the mass of the landing systems, you get to have all that mass on the ground and not have to lift it into space. Dramatically improves the performance of the rocket.
1) legs are heavy 2) empty rockets are stronger in tension than compression 3) the booster is large enough to make (1) and (2) matter more than they did for Falcon 9.
>The 'chopsticks catch' was amazing to watch. Seems like it adds a lot of risk and clearly the booster needs additional fire suppression systems :-) perhaps the tower could mount something that sprays the booster like the barges have for the F9 boosters.
There's no reason to reinvent the airport firetruck.
Unless you're gonna make it fully automated, it's not gonna work here as it can't be within kilometers of the landing site during landing in case there's a catastrophic failure.
>The heatshield held out for a much longer time, the asymmetric heating on the flaps was interesting. I had guessed that all four flaps would have equivalent heating based on an approach that was basically that side of the rocket perpendicular to the flow but it seems like that isn't the case. Still it seems like they are close to having something workable here.
Heat shielding didn't look relevant at this section of the flight at all. The booster didn't have any shielding.
The booster actually does have heat shielding behind the engine bells, to protect against aerodynamic heating on the return. In some of today's footage you can see it glowing yellow.
However, Raptor 3 is supposed to obviate the need for this shielding.
Nothing could have prepared me for how that catch looked. I was sure the rocket was careening into the tower at the last second before it straightened out. The control algorithms must be incredible for the landing system to work within those small tolerances.
MIMO and nonlinear control theories are probably some of the hardest topics in all of engineering. SpaceX control system also has to compensate for the fuel moving inside their rockets so the control algorithms probably involve some kind of fast numerical fluid simulation.
Another interesting thing SpaceX is doing is to use consumer-grade chips in triple redundancy configurations instead of using $100,000+ radiation-hardened aerospace/defense grade chips.
> control system also has to compensate for the fuel moving inside their rockets
My stepfather worked as a programmer on the Apollo program, and the thing he always talked about as his biggest accomplishment was working on the "slosh problem" -- so yeah, props to the SpaceX team for managing that landing. And props to my stepdad for managing it on hardware that was... a billion times less capable? :-)
> Another interesting thing SpaceX is doing is to use consumer-grade chips in triple redundancy configurations instead of using $100,000+ radiation-hardened aerospace/defense grade chips.
This has been known in the high availability and safety systems industry for a while and a good book to learn these reliability engineering techniques is "Reliability Evaluation of Engineering Systems".
I always thought the liquid sloshing would be one of the hardest to simulate (considering how chaotic fluid mechanics is). Interestingly, I think this caused the 2nd Falcon launch to fail (the LOX sloshing).
People don't realize how powerful applied math (especially in the areas you've mentioned) has become. Same tools can be applied to people in the ad tech/social media.
Just as a note, Space Engineers has a mod that accounts for fuel in the tanks and also various orbital mods. If one feels inclined to try it for themselves ;)
I know MPC takes a LOT of compute power. It's not like a finely tuned PID loop or even a cascade of PID loops, computationally.
Does anyone know (or have educated speculation on) what kind of hardware is running these algorithms? Like, do they have a linux machine that's running the control loops? Are we talking megabytes, gigabytes of SRAM?
I would think no -- you would definitely need hard real time for something like this. But my only experience with real time systems is in tiny MCUs with kb of SRAM. That's definitely too small for a controller like this.
"instead of using $100,000+ radiation-hardened aerospace/defense grade chips"
Well, that makes perfect sense considering that both the spaceport infrastructure, and the booster need to do their calculation on the ground level instead of the highly radiated environment that is space. However, for the rockets themselves, which happen to reach that harsh environment, they may use more resilient and expensive hardware in the future, after passing over the current "let it splash in the Indian Ocean" development and testing phases.
>probably involve some kind of fast numerical fluid simulation
Sometimes even a simple approach can work. On Apollo they developed (at the time cutting-edge) passive RC filter networks, to avoid the control system "exciting" the rocket at frequencies of the slosh/bend/torsion modes.
I never thought of using fluid dynamics in the rocket stabilization algorithm—maybe it's something that could be useful to prevent many of the accidents involving liquid-transport trucks
I have been told by people who worked on them that you get radiation hardened aerospace/defence grade chips by backing off the clock speed about 20% to give signal stabilization slightly longer time. I can understand the population being confused about this but industry being confused seems to have more to do with regulatory capture and beaurocratic moats which SpaceX does seem to be bypassing.
> SpaceX control system also has to compensate for the fuel moving inside their rockets so the control algorithms probably involve some kind of fast numerical fluid simulation.
Surely this isn't necessary with a small enough sensor granularity or whatever the terminology is. You can have very dumb software if it reacts quickly enough to changes in perception.
That's amazing footage, and you're right about the perspective: from the official feed the distances seem compressed compared to what we see in this footage.
I know the control algorithms are the mind-blowing part here but,
does anyone have any literature about how the Rocket localized itself with respect to the chopstick arms? It must've been some combination of GPS and Radar pings to the arms?
And then the onboard IMU to make sure it hits it straight.
Great question! Could just be Real-Time Kinematic (RTK) GPS like someone mentioned. Essentially the landing arms know their position very precisely and they measure the tiny errors in GPS data, and send that correction data live to the rocket in real-time as it's landing. Once the rocket gets very very close it could also just be using vision systems to zero-in on exactly where the chopsticks are.
To speculate more, they could also be using something like ultra-wide band positioning. This relies on the same time-of-flight principle as GPS but instead of using satellites in orbit to provide the precise time information you rely on various nearby ground stations. Would only be useful right at the final approach, the last couple hundred meters, but it's another way they could get very very precise position information. (fun fact: Ultra Wide band positioning is also how iPhones can locate AirTags with centimeter accuracy)
From the control point of view, isn't this exactly the same as F9 landing on a pad, except the pad is virtual, floating in between the chopsticks and the ground? Or course one difference is that the approach needs to be from the correct direction.
The booster was falling at 4500 Km/h 30 seconds before the catch with 2-3% fuel left. How is that amount of fuel remotely enough to stop the downward momentum?
First off, the booster was going about 1250 km/h when it started its landing burn, it relied purely on drag to get it slowed down to that speed.
Going by the telemetry of the seconds before the landing burn and noting the speed vs time, it seems drag was around 40 m/s^2 when it was going at around 3000 km/h. Since drag depends on velocity squared though, it had reduced to just above 10 m/s^2 just before the engines lit at 1250 km/h, and so would quickly become negligible once the engines lit.
Going by Wikipedia, the Super Heavy[1] has 3400000 kg of fuel at launch, so 3% of that is about 102000 kg. For the landing burn, it used 13 Raptor v3 engines[2] to scrub speed. Each Raptor flows about 650 kg/s max, so 3% fuel is enough for about 12 seconds for the 13 engines.
The empty mass of the Super Heavy is about 275000 kg, so about 377000 kg before the landing burn with 3% fuel.
Using the sea-level vs vacuum performance of the Raptor v2 engines, one can estimate that each Raptor v3 produces about 2.45 NM of force at sea-level. So 13 of them would produce about 31.85 MN of force.
Using Newton's second law, F=ma, this gives an initial deceleration of about 84 m/s^2 and about 104 m/s^2 when empty. If we do a rough spreadsheet integration, we get that a burn of roughly 4 seconds is needed to scrub the speed assuming no other forces.
Now, comparing this with reality, the full 13 engines were lit for a little over 5 seconds.
In my simplified calculations I was assuming full throttle the whole way, which obviously isn't realistic, and I also assumed 3% fuel. So over all I think that's a pretty decent estimation.
It weighs about 10% of what it did at liftoff, but half of the engines fire to slow it down.
Also I don't think the telemetry on the feed is that accurate, so with all of the atmospheric braking, it was probably going a bit slower than the 1200km/h at engine reignition.
Note that the energy of 3% of the propellant (~100 GJ) could theoretically get the empty booster (100,000 kg) to a little over 5000 km/h if properly applied.
I thought the same, screamed out "ouch that doesn't look good!" right before the catch.
The last part of the live stream they showed footage from a different angle and there it didn't look too bad though! For sure controlled.
Scott Manley put out a tweet that they went down towards a non-tower position until they were at three engine controlled burn, and only then did the side shift.
A clip from some news program popped up on YouTube, just a two minutes clip of the catch, I was convinced that it was reversed. The fact that this is possible, that they made it work is nothing short of amazing.
When Musk first proposed this, I thought he was crazy. It seemed like something a school boy would draw up. Now I think this will become routine and forgettable after a few more successes. Is there a word for that - something out of fiction becoming mundane?
I thought it was plausible given the accuracy of F9 landings, though I still wonder how it will work at scale if one failure destroys the landing site. That could ruin the cost benefits.
Where his vision hit a lot of speed bumps is second stage reusability. Starship is a beautiful second stage to throw in the ocean. They’ll probably get it landing but the heat tiles will require a lot of refurbishment between flights. They’re going to have to figure something else out.
2 related expressions:
1. Creeping Determinism - the idea that even magical leaps forward were somehow inevitable and were anticipated.
2. Nihil Admirari - the idea that wisdom is anticipating every possible thing that could happen, therefore a wise person would never be surprised.
But I lump both these together as the *"Wiseass Movable Feast"*
This happened with LLMs in a big way. Basically humanity surpassed some kind of AI milestone and we zoomed past the turing test in a big way. But thanks to social media everyone is sort of rolling their eyes at it.
Yes, indeed. But I will add that the sheer size of the rocket helps in this regard. I think it is rather hard to appreciate the massive scale of the feat by watching videos.
If I had to bet, I would bet against it. Boston Dynamics for example, for the longest time, didn't use anything other than Model-Predictive Control. Only recently have they started using RL
Naive question: I obviously expect there to be flames from the engines, but there were flames on the lower sides for quite a while after the catch – is that expected?
It is common for Starship prototypes to have uncontrolled fires, but it is obviously not a good thing.
For example, prototype number 10 exploded 8 minutes after landing [1] because of a seemingly insignificant fire at the bottom.
After today's flight there was a long lasting fire in the engine section, with occasional flaming pieces of plumbing raining down from the rocket. Examining the aftermath should help SpaceX to understand what improvements need to be made to prevent this from happening.
If not an intended vent, probably some methane leak. Given that they have the first stage intact, they will know exactly what happened very soon. Yet another advantage of having the rocket returned instead of sinking it in the ocean.
It does look like venting, but on the Everyday Astronaut video feed it also looked like a COPV inside one of the strakes looked like it exploded as well.
It looked to me like the fire was on the fuel intake valves and if you watch carefully that area was scoured by the nozzle output when it was first slowing down so it probably blew through the shutoff valves or something
They have mission goals which were achieved (booster was caught). Goals that they didn't think they could do (Starship being within the buoys). While there was flames and those could be dangerous you judge the mission based on the planned outcomes but they will try to eliminate the anomalies to improve the next mission while still achieving the goals.
I wonder what will happen when they get to 99% reliability? They clean up and rebuild the Mechazilla every once a hundred catches, on that occasion that one fails?
I suspect there's already a whole "refurbishment" process for the crane even for non-reusable launches, and once it's working darn reliably, they can just have a bunch of them ready to go, and cycle once in a while.
The booster aims towards the shore until the landing burn starts, only then does it swing towards the tower. So, for the most part, failures should mean that the booster safely crashes into the water.
I'm just so happy to see this level of progress. This another big step for opening up space. To think that one day this will be considered normal. 150 Metric tons sent on a fully reusable rocket.
So, reusable is supposed to reduce the cost. But the space shuttle was reusable and it has been shutdown because it was too expensive. What is the differences between the two?
SpaceX builds vehicles. The Shuttle was “reusable” because they needed a term between the default for transportation capital expenditures (e.g. trains, planes, cars and ships) and the modified missiles that defined post-War spaceflight. “Reusable” in the Shuttle’s context meant months of specialist overhaul time and the cost of a Falcon 9 launch in SRB booster replacements alone [0].
At the end of the day, in 2010, “the incremental cost per flight of the Space Shuttle was $409 million, or $14,186 per kilogram” [1]. ($591mm and $20,512 in 2024 dollars, respectively [2].) SpaceX’s prices per kg are around $3,170 on Falcon 9 [3] and $1,520 for Falcon Heavy [4]. Starship should bring those costs below $1000.
The main difference is that this is built by a private corporation who can't afford to throw money away, while the Space Shuttle was build by the government, and moreover it had to fulfill a number of conflicting requirements, and commercial profit was not one of them.
But on a more technical level. I think the vertical landing is the main difference. Vertical landing was obviously known and done by NASA, this is how the lunar modules landed on the Moon. But doing it on Earth, with vehicles weighing hundreds of times more, I don't think the world had that technical readiness a few decades ago, when the space shuttle was designed.
And another major difference is the mass manufacturing idea. From the start SpaceX planned for getting to mass manufacture its rockets. The Falcon rockets are much cheaper than any other alternatives even if you remove the reusability.
Then it's the methane burning engines. This was pure old fashioned engineering progress. SpaceX's engines are miracles of rocket engineering. Aside from that, the fuel choice is extremely smart. Methane is better than all other fuels, except for hydrogen. Hydrogen was the fuel of the space shuttle, but it's very tricky to work with. It has very low volumetric density, so the tank of the space shuttle was absolutely humongous. Hydrogen needs to be stored at an absurdly low cryogenic temperature, so this adds to the complexity. And that tank was not reusable, so it adds to the cost.
Think of the word ‘reusable’ in this case as less a binary descriptor but more of a scale of reusability.
Yes, both systems are reusable, but there are key differences in the refurbishment of the systems that partly explains the cost difference. It took more labor, resources and time to refurbish the shuttle. Also consider rapid reusability was a stretch goal when it was being designed, but we have come a loooong way since, spacex in particular has had it as a driving competitive differentiator for years now.
Another big difference is that NASA post Cold War was a skilled jobs program, with an incentive to do distributed, high overhead work to appease their bosses (congress), while SpaceX has the opposite.
Shuttle itself was refurbishable, but not rapidly re-usable. It was also incredibly expensive to build and refurbish. A shuttle launch also utilized boosters that were not re-usable.
Starship is supposed to be (and clearly well on the way to being) fully rapidly re-usable. That means all stages (in this case two) are re-usable, and that the capital and time required to get either stage flight ready again after a flight should be minimal.
Said another way – it is cheaper for SpaceX to build an entirely new Starship + Booster than it was to refurbish a Shuttle between flights, by a factor of about 4x ($90M for a Starship+Booster / $400m for Shuttle refurb).
Percentage of reusability: boosters of shuttle cannot be reused, maintenance of shuttle itself is also very expensive (heat shields were pricey). whereas the starship stack has higher reuse percentage and allegedly cheaper to maintain.
You've got a lot of responses on the difference of reusability, but the shuttle was also more expensive because it had to carry a lot of capabilities with it every time. If you were launching a satellite, you were carrying along the crew compartment and a couple astronauts. If you were bringing a few astronauts to the space station, you brought a cargo bay. And in either circumstance, you brought big wings. Starship can be filled with all cargo. And if you're just changing crew on the ISS, you could... not use Starship and launch a Falcon 9 instead. One of the mission profiles required by the Air Force for the shuttle was that it be able to rendezvous with a satellite, put it in the cargo bay, and return to Earth, all under 2 orbits and along a path that avoided flying over the Soviet Union, which required a rather large turn in-atmosphere to make it back to landing on the west coast.
No one has answered with one of the biggest issues with the shuttle: each one was extremely custom. Every single heat shield tile was unique to a specific position and a specific shuttle. There were probably over a hundred million individual components in the Shuttle, and with many critical ones being custom, the time to refurbish it for a new launch was much longer.
This is in contrast to something like the falcon, which has a very standardized mfg process and components, which allows for really rapid iteration
The Space Shuttle was not fully reusable as the biggest single part, the orange tank, was destroyed every time. But more importantly, the orbiter and boosters needed 2+ months of refurbishment/rebuilding after every flight.
One of the design goals of Starship is for the booster and ship to relaunch with zero refurbishment. To literally land over the launchpad, refuel, and go back to orbit within hours without people even approaching them. The heat shield is the biggest risk to that goal IMO, and we saw today that it definitely sustained serious damage despite improvements. But if they ever get there then per-launch costs will be a tiny fraction of the Shuttle with 6x the payload.
Falcon 9 has already proven that partial reusability is economical. SpaceX has dominated the entire worldwide launch industry and their competitors are nation states with no need to make a profit.
The difference is that they have already proven to be the lowest cost and most reliable launcher due to reuse. This is them lapping the industry with second stage reusability.
Reusability increases costs if you don't reuse often enough.
The shuttle would have been much, much, cheaper per launch if it had flown more often. The expected costs for the shuttle included a range based on how often it flew which turned out to be reasonably accurate. They were much worse at predicting which end of the range they would be flying in. At the rate they ended up flying they had the extra costs of reusability without any of the benefits.
Starship is ludicrously expensive, but still much cheaper than even the best case for the Shuttle, and it has a guaranteed source of launches to help it benefit from resuability.
The Shuttle consisted of the shuttle (orbiter) itself, the external tank (not reusable), and the two boosters which could be reused after ocean recovery. The orbiter itself was slow and expensive to reuse since (among other things) all the heat shield tiles were inspected and 30-100 replaced between each launch. I don't know how much work was done to the engines between launches, but SpaceX's parts and cost reduction on the Raptor engine have to give it an advantage there.
StarShip consists of the Super Heavy booster that we saw "caught" today, and the StarShip (orbiter) itself. Having the booster return to launch site vs requiring ocean recovery should potentially increase cadence and reduce cost of reuse. StarShip is also meant to be reusable, although it remains to be seen how that will pan out. On the previous flight there was burn through from inadequate heat shielding - maybe we'll see an improvement with today's vehicle. I'd expect SpaceX to iteratively arrive at a quicker and more cost effective orbiter reuse procedure than NASA had with the shuttle, but how quick remains to be seen. Of course they are planning many of these to go on one-way trips to Mars rather than being reused.
SpaceX took advantage of tech from both the US and Russia, including the experience with the Space Shuttle. They have better computers, better metallurgy, advanced 3D printing and their own experience with the Falcon 9.
There is no guarantee that it will reduce the costs that much, but will all that experience, the chances are success are higher than with the Space Shuttle.
One big issue that isn't talked about much and that SpaceX takes very seriously is simply a lack of demand. There is only so much stuff you want to put in space. Satellites are expensive, and even with disposable rockets, the launch is only a smaller fraction of the cost. It is already a problem with the Falcon 9 as they have a bunch of rockets and not much to do with them. Starlink, orbital refueling, and crazy ideas like earth to earth transport are all ways to address this problem.
It was a problem for the Space Shuttle too, they couldn't achieve the economies of scale they planned it for. It was supposed to fly for routine maintenance missions but it didn't work out.
the space shuttle was "reusable". It had to be taken apart and meticulously cleaned and tested and basically had to be rebuilt after each flight, in a process called turnaround. SpaceX's rockets are much closer to what you'd consider reusable.
The majority of damage to shuttle's TPS apparently came from foam strikes from the external fuel tank. Superheavy's optimized profile certainly helps here, since there are no large cryogenic tanks hanging ominously over the TPS while being shaken violently by solid rocket boosters.
To answer this oversimplified question with a simple answer, the Space Shuttle couldn't be a more different vehicle than this one. It truly is a comparison between apples and oranges.
Let's start with the fact that it was designed in the 1970's. If you had a Cadillac DeVille from 1970 it would get 8-12 miles per gallon. Just the mere fact that the design is about 70 years old makes that vehicle too expensive to operate, and that's before we even start talking about other issues with the design (performance, safety, reliability, etc).
The main thing is... Space shuttle wasn't all that reusable. It had to be launched with a massive rocket and two massive boosters that just fall into the ocean.
Estimated costs of the fully loaded cost of the shuttle program ended up at $600 million to $1 billion per flight. The refurbishment costs per flight and man hours/staffing were astronomical.
It actually would have been a lot cheaper if they had gone for serialized, mass assembly line production of Saturn V class disposable rockets to launch piece of space stations, satellites and manned missions into low earth orbit.
The shuttle wasn't fully reusable, just for starters. The boosters were reusable with a lot of refurbishment work. The center tank was expended every time. It was very expensive. Only five shuttles were ever made, which means that no effort was put into automation of production of engines etc., everything was custom, and everything required great care to save the sunk costs.
Manufacture and maintenance contracts for the Shuttle were deliberately spread out across many companies and states, especially in key congressional districts. It was a jobs programme; waste was a feature not a bug.
Because people are reflexively averse to government spending unless there's a billionaire making profits on the way through thanks to 100 years of academic capture by Austro-libertarian economists.
Is that 150t of payload or total? What’s the cost in fuel alone (let’s ignore maintenance and operations costs for now)? I’m trying to get a feel for the relative scale compared to today’s commercial flight.
They previously threw around a number of around $1M per flight, as mostly fuel costs.
Also, while 150t is the target payload capacity, the current test vehicles are closer to 50t in payload capacity, there are revisions in the pipeline based on data from these test flights which will bring it up to 150t.
Methane is about 900-1500$ / ton. About 1000 tons is used for the launch in addition to 3600 tons of lox. That should be a bit cheaper than methane per ton. Ballpark, the propellant might cost around 2M$.
A modern airliner on a long flight might burn around 80 tons of kerosene. It's slightly cheaper than methane. Call it 75-80K$.
Falcon 9 is still the most advanced rocket flying real missions to date. The only thing close to it is Blue Origin, which didn't even have their maiden flight yet.
Is their timeline too optimistic? Yeah, but if the industry still catches up with F9 and they are close to having something a lot more advanced it really doesn't matter.
Even without reusability nothing comes close to Starship's cargo capacity. If you don't have to put a lot of engineering into getting things as small and light as possible you can put things in space a lot faster.
Robert Zubrin's Mars Direct would have got "flags and footprints" on Mars without needing something like Starship. He worked on a version of SLS in 1989 in Martin Marietta that would have cost $1B and been ready in the second half of the 90s and would be retiring soon after many successful flight to Mars (and to the Moon), but instead we got SLS and Orion so we need Starship to get anything done.
The Mars Direct book is worth reading not only for things like transport and mission design but also a wealth of other good ideas about how to “live off the land” when we get there too. Truly worth reading for anyone remotely interested in space exploration
Well I mean going to Mars isn't that hard per se, a fair few countries have done it... but carrying enough supplies and shielding to last people 2 years until the return trip and then actually getting back are way harder problems.
Seeing Starship get burnthrough on these suborbital launches really shows how hard it'll be to do Mars return entry with a fair few extra km/s.
Getting to Mars is hard but not that hard compared to getting a large payload to Mars surface and back to Earth. Some people may assume Mars return is around twice as hard but it's orders of magnitude harder. And that's before even contemplating doing so with a cargo of humans and the necessary tonnage for them to survive on Mars at least 26 months.
Going to Mars is relatively easy compared to setting up a self-sustaining human civilization on Mars, which will be thousands of times harder. Many people don't understand the scale and scope of the biosphere services the Earth provides to humans free of charge, though, and forget that little of that exists on Mars, other than gravity.
It's difficult to overstate how important the milestone of catching the booster is. Now we have a reusable rocket an order of magnitude larger than anything we've had before, and the cost of kg to orbit just nosedived.
Second stage reuse seems the far more challenging problem. Other companies should have reusable boosters soon but if significant amounts of Starship continue to ablate on the way down they could be faced with a disposable Starship competing with smaller and cheaper second stages that are well sized for typical payloads. We already knew boosters can be flown back to launch sites reliably with high accuracy. We don't know if it is possible to make rapidly reusable thermal protection systems that can operate on an orbital vehicle of Starship's size until it is demonstrated.
> Second stage reuse seems the far more challenging problem.
Sure, SpaceX has been doing first stage reuse for a long time now. But they have demonstrated landing the second stage successfully at sea twice now with the same sort of smoothness that they demonstrated once for the booster before they then caught the booster on the first real try.
A partial list of unbelievably hard things that SpaceX has so far made seem easy:
- building a rocket from scratch
- landing Falcon 9 boosters
- landing Falcon 9 boosters *reliably*
- 3x weekly launch cadence (Falcon)
- the bellyflop manoeuver
- mass manufacturing(!) a rocket engine
- catching the booster
- simulated landings of the ship
Catching the booster is really just like landing a Falcon 9 booster w/o legs, but clearly much harder.
Anyways, if they can do all those things then it's pretty clear that they can catch-land the ship.
There's still a huge list of crazy-difficult things that SpaceX say they want to do that are hard to believe are possible, except for the fact that SpaceX has already done so many unbelievably difficult things already.
I found Jeff’s Bezos interview with everyday astronaut really illuminating on this topic.
Supposedly they’re working on both a reusable and cost optimised non-reusable second stage at the same time. And they don’t really know yet which one will end up being cheaper.
You also see this kind of thinking with Rocket Labs neutron rocket. Where they focus on making the reusable booster do more, while making the second stage smaller, cheaper and simpler.
I think if it wasn’t for the rocket engine this wouldn’t be a question at all. The tank doesn’t have much value. It’s just a thin shell and probably a fraction of the cost of the fuel.
So I’m thinking, perhaps the optimal solution is something like this: the bottom part of the second stage with the engines separates, and a small engine and fuel tanks places the engines in a stable orbit. The tank itself is deorbited and burns up.
At some point later something like the Starship collects several second stage engines and deorbits them safely to be reused.
Or perhaps just the engines can be immediately deorbited with an inflatable heat shield and parachutes.
Didn’t they look at all kinds of ideas earlier like squirting some propellant or water out over the skin on the way down, and wasn’t steel chosen for its thermal robustness? Did they get into the problem and realize it’s a lot harder and abandon those things for tiles?
Maybe they will have to sacrifice more payload mass for active or passive shielding or more fuel for powered deceleration. That would yield a less impressive lifter but with full reusability.
> Other companies should have reusable boosters soon
You're way too optimistic. Starship will deliver commercial payloads, with SpaceX phasing out Falcon 9 outside of ISS launches, before anyone has a reusable Falcon 9 equivalent.
It pains me to say this, but SpaceX is in a class all its own.
I'm not sure how critical "catching" the booster is to reusability, but it does save weight by not needing legs for landing, and perhaps the booster suffers less stress this way?
Note that the booster is not really being "caught" although this is the word it seems we're stuck with. It's really more like landing on the arms, since it throttles to a hover at that point.
> I'm not sure how critical "catching" the booster is to reusability
Not necessarily for reusability, but it helps significantly for rapid reusability: it eliminates the need to transport the booster from the landing site to the launch site. Given that it's 9m x 70m and weighs 270 tonnes, that's not an easy process.
I think saving weight is definitely one of the main issues, see those proportionally large legs on Falcon-9, I guess it simply doesn't scale on bigger vehicles like Starship / Heavy booster. Also, by catching the booster on site, they can even save the transportation and do the refurbishing on site, so even shorter turnaround time I guess.
They're still having some significant burn through problems on the upper stage fins during reentry it looks like. Way better than last time but the top part of the fin was glowing far a while after the main reentry finished.
I think that catching a grain silo in mid air that fell in a semi-controlled way from effin low earth orbit is undeniably incredible.
SpaceX continues to blow me away with these unbelievably lofty ideas. I remember seeing their "grasshopper" flights years ago and they blew it up because they couldn't land it at the time, who knew within a few years they'd be doing this.
I think they still have "Grasshopper" (the custom test unit) sat outside at McGregor, it was F9R Dev1 (a modified F9 booster) that they had to destruct after one of the engines failed mid-hop.
They were originally going to do high-altitude tests with an "F9 dev2" (similar to some of the early starship test launches) - but they gave up and just did some testing with real landing Falcon 9s instead.
Strategically this is huge for the US and NATO. Being able to put orders of magnitude more payload in orbit at a fraction of the cost of the competition is a huge advantage in controlling space. Starlink and starshield are already years ahead of what china and russia has, starship is going to widen that gap even further.
Amen! If it turns out we can’t make humans a multiplanetary species in any reasonable timeframe, at least we can make the last few decades of livable earth climate a friendly atmosphere for US business interests.
The most expensive part is to support the risk and cost of research & development. Sure, that renders you the first place for a while, but expect others to move easier (and thus faster) after you marked the trail.
To get an inert object through the atmostphere and do real damage it has to be very large. That's a very inefficient way to use mass that you've boosted to orbit, even if it's relatively cheap to get there. So any weaponary they put up there will likely look fairly conventional. The speed of deployment and difficult of intercept would be the game-changer.
Readit News
The 'chopsticks catch' was amazing to watch. Seems like it adds a lot of risk and clearly the booster needs additional fire suppression systems :-) perhaps the tower could mount something that sprays the booster like the barges have for the F9 boosters.
The heatshield held out for a much longer time, the asymmetric heating on the flaps was interesting. I had guessed that all four flaps would have equivalent heating based on an approach that was basically that side of the rocket perpendicular to the flow but it seems like that isn't the case. Still it seems like they are close to having something workable here.
The detonation at the end was pretty spectacular too, but I suspect that structurally the tanks failed as the rocket hit the water vs anything that was an engineering failure. Engineering it to be strong enough to land on water would presumably compromise the cargo to orbit number.
The use of Starlink was really interesting. The ability to get live video for the entire re-entry is pretty game changing for engineering. I'd guess there are even more 'views' than they showed (there would be if I were running things :-)) but overall that capability is something that really helps evaluate the changes made.
I can easily imagine that flight 6 will be nominal end to end without any unintended damage. That would enable, perhaps, one of their 'massive' Starlink missions to test cargo delivery. It will also start to give us some better numbers on exactly how much cargo Starship can put in orbit in 'full reuse' mode which is essential if they want to create a fueling station on orbit for the Artemis program.
Again, hats off to engineering at SpaceX, y'all did good.
First, to the extent the booster is out of position in the x (sideways) direction, the chopsticks can move to accomodate error. But actually I think this dimension is the easiest of the three, as the booster has plenty to time to null any error in this dimension.
In the y direction (direction of travel towards the tower), the rails on the chopsticks can cope with the booster touching down along quite a long distance. But importantly, they appear to be smooth, so the pins can initially skid along them and then the booster can swing if it has not fully nulled any horizontal movement. In contrast, if the booster used legs and has not yet fully nulled any horizontal motion at touchdown, there is a greater risk of breaking a leg or simply tipping over.
And in the z direction, it should be possible for the chopsticks can absorb more vertical motion than legs can absorb, because you can easily build in huge springs/dampers/etc into the ground equipment without concern about mass.
Catching also puts the booster in tension rather than compression - it's easier to be rigid in tension than compression.
Finally, if legs were used, the engines would have to get close to the ground during landing, so reflected shock from the ground could cause damage. I know Falcon 9 does this, but the area of the base of Starship is much greater, so there's effectively less room for the reflected energy to escape. Catching completely removes this risk.
On balance, I think they would have better chance of success for each mission by catching. The main downside would be if you fail to catch, you may need to build a new tower, whereas a flat pad would be cheaper and easier to repair.
I have experience with docking large boats and it does seem to be a bit similar. In the case of boats, wind is a big deal, and the booster has nothing "below the waterline" to slow down the effects of wind.
I'm actually wondering about that. If I understand correctly, the arms can move up and down, and pivot around the tower. This allows them to correct for some error in the rocket trajectory and also (presumably) "soften" the final contact. Between the nozzles and the arms, it gives SpaceX a lot of degrees of freedom in the final seconds (you can see how the booster kind of "hovered" right at the end) and in certain respects might even offer more forgiveness than the hard ground.
Could it smash into the tower? For sure. Would that be more dangerous than smashing into the pad? I don't know.
It's a new technique with which we don't have a lot of experience.
It can arrive to the designated intermediate point with some already good accuracy, and then take some time to trim the remaining errors to the noise level more slowly, possibly with feedback from the ground sensors.
The chopsticks also include rails with shock absorbers, the action of which can be seen in the view from the tower during the landing [1], so the required accuracy is probably relatively modest, provided one plans the maneuver carefully.
[1] https://youtu.be/Ysx4t7ICO58?t=678
The catchzilla solution is an example of their amazing ability to think out of the box. This solution, and things like the rapid evolution of the Raptor engine (see picture, story here: https://medium.com/@futurespaceworld/the-evolution-of-spacex...), dynamic engine configuration (33, 13, 3, zero, up again) and control is almost magically impressive. This is the stuff of Sci-Fi, brought to life.
It's possible that SpaceX programed the AFTS to trigger some time after the rocket touched down in the water. Just to make sure that it completely submerges quickly.
> I can easily imagine that flight 6 will be nominal end to end without any unintended damage.
I think it depends on what you mean by "nominal". SpaceX ultimately wants to catch the 2nd stage as well. I suspect that they are a ways off of that, since it would have to approach over land. The FAA is going to need to have very high confidence that it will do exactly what it's designed to do before they're going to allow that.
The SpaceX host on the stream said that they were going to try and touch it down on the water at more of an angle than the previous flight to attempt to get it to survive the initial splash-down so they could get some more data and video footage.
Obviously this wasn't guaranteed to succeed, but it indicates that they weren't planning to immediately detonate the ship on touchdown.
As I recall it was common on the early Falcon 9 landing tests that splashed down in the ocean to also explode after tipping over and smacking the water. Once they're actually landing them on a pad, tower, or ship that should be much less of an issue.
Why would they want to do that? (genuinly curious).
I reckon there would be a lot of useful data left if they could recover or even just inspect the remains. The remains are one big tank, so it would have floated.
They have to prove out the landing of the second stage, either with another catch or with landing gear, which they need for the lunar lander anyway.
The way these chopsticks are set forces the booster into a dangerous, snake like maneuver (a SnakeX maneuver) at the last second from a vertical setting, to get into the chopsticks. This maneuver is due to the fact chopsticks are short and the booster has to land on one point in space. No degrees of freedom.
Now, imagine if the chopsticks were long. The booster wouldn't have to land at one specific point, but it could now land on a line. One degree of freedom.
Now replace the long chopsticks with cables, and then add another pair of cables perpendicular to them. So you have a pair of parallel cables perpendicular to a second pair of parallel cables. Now the booster doesn't have to land on a line, but can land anywhere in the grid that's covered by the cables. Two degrees of freedom.
Pushing this thought leads to having a sort of iris diaphragm, like the ones in optics, but an iris diaphragm of cables. The diaphragm is open when the booster is about to land, then closes in quickly. Imagine this[1], but it's cables cinching in.
Now, it's a diaphragm of cables, not a diaphragm of rigid beams, so I imagine the deceleration to be even smoother as the cables elongate, and an additional system of springs and dampers to counter the weight of the booster.
The booster is vertical and stays vertical. Granted, an unstable equilibrium, but it beats doing the SnakeX maneuver to get to the chopsticks, and that's another story.
Now, imagine the iris cable diaphragm can move up and down like equipment handling containers, and now you have three degrees of freedom. That's less control to worry about on the booster's side at the worst possible moment, landing, where you can't make adjustments anymore.
This not only means being more forgiving on mistakes related to position, but also on speed and angle. The diaphragm catches the booster at any angle, and given that it cinches way above the center of gravity, the booster goes back to a more stable vertical position.
For the fire, maybe you just need a big hole down there.
- [0]: https://en.wikipedia.org/wiki/Arresting_gear
- [1]: https://en.wikipedia.org/wiki/File:Iris_Diaphragm.gif
Moreover I think Elon discusses the mechazilla arms in one of Tim Dodd interviews. They dont want longer arms, ratger shoeter.
Think about physics involved for longer arms and how much more stress you will putt on the connection points.
https://www.youtube.com/watch?v=27TvGDpPLNw
It will be interesting to see how these two approaches fare vs. one another.
https://youtu.be/OYWBmu6H0ik?feature=shared&t=35
The rocket was set to come down on the pad just in front on the tower and launch mount until the final 3-engine landing burn started. This kept the infrastructure safe until the last moment, but also required that lateral translation you referred to.
I think the primary reason would be that landing legs are heavy, and it wastes performance to carry them. If your landing mechanism is mostly on the ground, you get that performance back.
Secondary reasons might include that it's simpler to get the booster right back to the pad. Once things have settled into an operational cadence, it's likely feasible to lower and lock the booster onto the stool, stack a ship on top, refuel, and relaunch -- no more messing around with barges, transport, weather issues, etc.
There's no reason to reinvent the airport firetruck.
There is. The booster is high above, and is larger than basically any aircraft. There's no flat concrete airfield around either.
I would suggest blowing CO2 or nitrogen through pipes positioned at the right height on the tower.
Heat shielding didn't look relevant at this section of the flight at all. The booster didn't have any shielding.
However, Raptor 3 is supposed to obviate the need for this shielding.
Another interesting thing SpaceX is doing is to use consumer-grade chips in triple redundancy configurations instead of using $100,000+ radiation-hardened aerospace/defense grade chips.
My stepfather worked as a programmer on the Apollo program, and the thing he always talked about as his biggest accomplishment was working on the "slosh problem" -- so yeah, props to the SpaceX team for managing that landing. And props to my stepdad for managing it on hardware that was... a billion times less capable? :-)
This has been known in the high availability and safety systems industry for a while and a good book to learn these reliability engineering techniques is "Reliability Evaluation of Engineering Systems".
The book is available on amazon: https://a.co/d/1nH824K
They used trios of regular consumer grade disks/servers etc as a cluster and it looked like a single node.
They had to replace a LOT of hardware but this was still cheaper than big iron industrial grades servers.
Does anyone know (or have educated speculation on) what kind of hardware is running these algorithms? Like, do they have a linux machine that's running the control loops? Are we talking megabytes, gigabytes of SRAM?
I would think no -- you would definitely need hard real time for something like this. But my only experience with real time systems is in tiny MCUs with kb of SRAM. That's definitely too small for a controller like this.
Really curious about the nuts and bolts of this.
Well, that makes perfect sense considering that both the spaceport infrastructure, and the booster need to do their calculation on the ground level instead of the highly radiated environment that is space. However, for the rockets themselves, which happen to reach that harsh environment, they may use more resilient and expensive hardware in the future, after passing over the current "let it splash in the Indian Ocean" development and testing phases.
https://ntrs.nasa.gov/api/citations/19700023342/downloads/19... (search for "slosh" or "shaping network")
http://cse.lab.imtlucca.it/~bemporad/publications/papers/ecc...
Surely this isn't necessary with a small enough sensor granularity or whatever the terminology is. You can have very dumb software if it reacts quickly enough to changes in perception.
I am curious but clueless about these problems, can you expand more?
https://x.com/shaunmmaguire/status/1845444890764644694
does anyone have any literature about how the Rocket localized itself with respect to the chopstick arms? It must've been some combination of GPS and Radar pings to the arms?
And then the onboard IMU to make sure it hits it straight.
To speculate more, they could also be using something like ultra-wide band positioning. This relies on the same time-of-flight principle as GPS but instead of using satellites in orbit to provide the precise time information you rely on various nearby ground stations. Would only be useful right at the final approach, the last couple hundred meters, but it's another way they could get very very precise position information. (fun fact: Ultra Wide band positioning is also how iPhones can locate AirTags with centimeter accuracy)
Going by the telemetry of the seconds before the landing burn and noting the speed vs time, it seems drag was around 40 m/s^2 when it was going at around 3000 km/h. Since drag depends on velocity squared though, it had reduced to just above 10 m/s^2 just before the engines lit at 1250 km/h, and so would quickly become negligible once the engines lit.
Going by Wikipedia, the Super Heavy[1] has 3400000 kg of fuel at launch, so 3% of that is about 102000 kg. For the landing burn, it used 13 Raptor v3 engines[2] to scrub speed. Each Raptor flows about 650 kg/s max, so 3% fuel is enough for about 12 seconds for the 13 engines.
The empty mass of the Super Heavy is about 275000 kg, so about 377000 kg before the landing burn with 3% fuel.
Using the sea-level vs vacuum performance of the Raptor v2 engines, one can estimate that each Raptor v3 produces about 2.45 NM of force at sea-level. So 13 of them would produce about 31.85 MN of force.
Using Newton's second law, F=ma, this gives an initial deceleration of about 84 m/s^2 and about 104 m/s^2 when empty. If we do a rough spreadsheet integration, we get that a burn of roughly 4 seconds is needed to scrub the speed assuming no other forces.
Now, comparing this with reality, the full 13 engines were lit for a little over 5 seconds.
In my simplified calculations I was assuming full throttle the whole way, which obviously isn't realistic, and I also assumed 3% fuel. So over all I think that's a pretty decent estimation.
[1]: https://en.wikipedia.org/wiki/SpaceX_Super_Heavy#Engines
[2]: https://en.wikipedia.org/wiki/SpaceX_Raptor
Fuel is the vast majority of the vehicle weight at launch, kind of like an empty vs full can of soda.
Also I don't think the telemetry on the feed is that accurate, so with all of the atmospheric braking, it was probably going a bit slower than the 1200km/h at engine reignition.
Had flashbacks to playing Jupiter Landing on the C64!
https://en.m.wikipedia.org/wiki/Jupiter_Lander
Edit: SpaceX should create a simple 'catch' the rocket game. Play in browser style. Just for kicks and marketing.
They did: https://starshipthegame.spacex.com/
there is, it was discussed in the FAA thread.
https://news.ycombinator.com/item?id=41821220
lunar.unnecessarymodification.com
The last part of the live stream they showed footage from a different angle and there it didn't look too bad though! For sure controlled.
Scott Manley put out a tweet that they went down towards a non-tower position until they were at three engine controlled burn, and only then did the side shift.
Absolutely impressive accuracy and precision.
Where his vision hit a lot of speed bumps is second stage reusability. Starship is a beautiful second stage to throw in the ocean. They’ll probably get it landing but the heat tiles will require a lot of refurbishment between flights. They’re going to have to figure something else out.
But I lump both these together as the *"Wiseass Movable Feast"*
"Is there a word for that - something out of fiction becoming mundane?"
Musking?
Check Steve Brunton youtube channel, he is one of the leaders in this area: https://youtu.be/gb_C9LcjDSI?si=xUjqUZ9-0MIFohX6
> Mechazilla has caught the Super Heavy booster!
https://x.com/spacex/status/1845442658397049011
Historic viewing :)
https://www.youtube.com/watch?v=YC87WmFN_As&t=3h25m14s
For example, prototype number 10 exploded 8 minutes after landing [1] because of a seemingly insignificant fire at the bottom.
After today's flight there was a long lasting fire in the engine section, with occasional flaming pieces of plumbing raining down from the rocket. Examining the aftermath should help SpaceX to understand what improvements need to be made to prevent this from happening.
[1] https://youtu.be/XOQkk3ojNfM?t=38346
The vent out the side before touchdown didn’t look right, though. Something blew, but non-critically.
[1] https://x.com/Cosmo_556/status/1845554958604657051
Thats like a 747 to space.
SpaceX builds vehicles. The Shuttle was “reusable” because they needed a term between the default for transportation capital expenditures (e.g. trains, planes, cars and ships) and the modified missiles that defined post-War spaceflight. “Reusable” in the Shuttle’s context meant months of specialist overhaul time and the cost of a Falcon 9 launch in SRB booster replacements alone [0].
At the end of the day, in 2010, “the incremental cost per flight of the Space Shuttle was $409 million, or $14,186 per kilogram” [1]. ($591mm and $20,512 in 2024 dollars, respectively [2].) SpaceX’s prices per kg are around $3,170 on Falcon 9 [3] and $1,520 for Falcon Heavy [4]. Starship should bring those costs below $1000.
[0] https://forum.nasaspaceflight.com/index.php?topic=51959.0
[1] https://en.m.wikipedia.org/wiki/Space_Shuttle_program
[2] https://www.usinflationcalculator.com/
[3] https://www.spacex.com/media/Capabilities&Services.pdf LEO
[4] https://en.m.wikipedia.org/wiki/Falcon_Heavy LEO, theoretical
But on a more technical level. I think the vertical landing is the main difference. Vertical landing was obviously known and done by NASA, this is how the lunar modules landed on the Moon. But doing it on Earth, with vehicles weighing hundreds of times more, I don't think the world had that technical readiness a few decades ago, when the space shuttle was designed.
And another major difference is the mass manufacturing idea. From the start SpaceX planned for getting to mass manufacture its rockets. The Falcon rockets are much cheaper than any other alternatives even if you remove the reusability.
Then it's the methane burning engines. This was pure old fashioned engineering progress. SpaceX's engines are miracles of rocket engineering. Aside from that, the fuel choice is extremely smart. Methane is better than all other fuels, except for hydrogen. Hydrogen was the fuel of the space shuttle, but it's very tricky to work with. It has very low volumetric density, so the tank of the space shuttle was absolutely humongous. Hydrogen needs to be stored at an absurdly low cryogenic temperature, so this adds to the complexity. And that tank was not reusable, so it adds to the cost.
Yes, both systems are reusable, but there are key differences in the refurbishment of the systems that partly explains the cost difference. It took more labor, resources and time to refurbish the shuttle. Also consider rapid reusability was a stretch goal when it was being designed, but we have come a loooong way since, spacex in particular has had it as a driving competitive differentiator for years now.
Another big difference is that NASA post Cold War was a skilled jobs program, with an incentive to do distributed, high overhead work to appease their bosses (congress), while SpaceX has the opposite.
Starship is supposed to be (and clearly well on the way to being) fully rapidly re-usable. That means all stages (in this case two) are re-usable, and that the capital and time required to get either stage flight ready again after a flight should be minimal.
Said another way – it is cheaper for SpaceX to build an entirely new Starship + Booster than it was to refurbish a Shuttle between flights, by a factor of about 4x ($90M for a Starship+Booster / $400m for Shuttle refurb).
Material: stainless steel is much cheaper
Percentage of reusability: boosters of shuttle cannot be reused, maintenance of shuttle itself is also very expensive (heat shields were pricey). whereas the starship stack has higher reuse percentage and allegedly cheaper to maintain.
This is in contrast to something like the falcon, which has a very standardized mfg process and components, which allows for really rapid iteration
a. Cheaper, more durable material (stainless steel).
b. Cheaper, easier to manufacture engines.
c. Easier to use fuel (methane is much "tamer" than hydrogen).
d. Standardized heat shield with much smaller requirements for manual work.
One of the design goals of Starship is for the booster and ship to relaunch with zero refurbishment. To literally land over the launchpad, refuel, and go back to orbit within hours without people even approaching them. The heat shield is the biggest risk to that goal IMO, and we saw today that it definitely sustained serious damage despite improvements. But if they ever get there then per-launch costs will be a tiny fraction of the Shuttle with 6x the payload.
The difference is that they have already proven to be the lowest cost and most reliable launcher due to reuse. This is them lapping the industry with second stage reusability.
The shuttle would have been much, much, cheaper per launch if it had flown more often. The expected costs for the shuttle included a range based on how often it flew which turned out to be reasonably accurate. They were much worse at predicting which end of the range they would be flying in. At the rate they ended up flying they had the extra costs of reusability without any of the benefits.
Starship is ludicrously expensive, but still much cheaper than even the best case for the Shuttle, and it has a guaranteed source of launches to help it benefit from resuability.
StarShip consists of the Super Heavy booster that we saw "caught" today, and the StarShip (orbiter) itself. Having the booster return to launch site vs requiring ocean recovery should potentially increase cadence and reduce cost of reuse. StarShip is also meant to be reusable, although it remains to be seen how that will pan out. On the previous flight there was burn through from inadequate heat shielding - maybe we'll see an improvement with today's vehicle. I'd expect SpaceX to iteratively arrive at a quicker and more cost effective orbiter reuse procedure than NASA had with the shuttle, but how quick remains to be seen. Of course they are planning many of these to go on one-way trips to Mars rather than being reused.
More than 40 years and many lessons learned.
SpaceX took advantage of tech from both the US and Russia, including the experience with the Space Shuttle. They have better computers, better metallurgy, advanced 3D printing and their own experience with the Falcon 9.
There is no guarantee that it will reduce the costs that much, but will all that experience, the chances are success are higher than with the Space Shuttle.
One big issue that isn't talked about much and that SpaceX takes very seriously is simply a lack of demand. There is only so much stuff you want to put in space. Satellites are expensive, and even with disposable rockets, the launch is only a smaller fraction of the cost. It is already a problem with the Falcon 9 as they have a bunch of rockets and not much to do with them. Starlink, orbital refueling, and crazy ideas like earth to earth transport are all ways to address this problem.
It was a problem for the Space Shuttle too, they couldn't achieve the economies of scale they planned it for. It was supposed to fly for routine maintenance missions but it didn't work out.
Let's start with the fact that it was designed in the 1970's. If you had a Cadillac DeVille from 1970 it would get 8-12 miles per gallon. Just the mere fact that the design is about 70 years old makes that vehicle too expensive to operate, and that's before we even start talking about other issues with the design (performance, safety, reliability, etc).
It actually would have been a lot cheaper if they had gone for serialized, mass assembly line production of Saturn V class disposable rockets to launch piece of space stations, satellites and manned missions into low earth orbit.
Same thing for SLS.
Also, while 150t is the target payload capacity, the current test vehicles are closer to 50t in payload capacity, there are revisions in the pipeline based on data from these test flights which will bring it up to 150t.
A modern airliner on a long flight might burn around 80 tons of kerosene. It's slightly cheaper than methane. Call it 75-80K$.
No idea about the other costs.
https://en.m.wikipedia.org/wiki/ʻOumuamua
Is their timeline too optimistic? Yeah, but if the industry still catches up with F9 and they are close to having something a lot more advanced it really doesn't matter.
Even without reusability nothing comes close to Starship's cargo capacity. If you don't have to put a lot of engineering into getting things as small and light as possible you can put things in space a lot faster.
She's not an enthusiast; she's got an impression from SciFi that going to mars shouldn't be that hard.
Seeing Starship get burnthrough on these suborbital launches really shows how hard it'll be to do Mars return entry with a fair few extra km/s.
Other countries have sent satellites to orbit around Mars, which is not the same thing.
she got me a while back with ink solvents. Monks may have been more illuminated making those pretty manuscripts than we appreciated.
Sure, SpaceX has been doing first stage reuse for a long time now. But they have demonstrated landing the second stage successfully at sea twice now with the same sort of smoothness that they demonstrated once for the booster before they then caught the booster on the first real try.
A partial list of unbelievably hard things that SpaceX has so far made seem easy:
Catching the booster is really just like landing a Falcon 9 booster w/o legs, but clearly much harder.Anyways, if they can do all those things then it's pretty clear that they can catch-land the ship.
There's still a huge list of crazy-difficult things that SpaceX say they want to do that are hard to believe are possible, except for the fact that SpaceX has already done so many unbelievably difficult things already.
Supposedly they’re working on both a reusable and cost optimised non-reusable second stage at the same time. And they don’t really know yet which one will end up being cheaper.
You also see this kind of thinking with Rocket Labs neutron rocket. Where they focus on making the reusable booster do more, while making the second stage smaller, cheaper and simpler.
I think if it wasn’t for the rocket engine this wouldn’t be a question at all. The tank doesn’t have much value. It’s just a thin shell and probably a fraction of the cost of the fuel.
So I’m thinking, perhaps the optimal solution is something like this: the bottom part of the second stage with the engines separates, and a small engine and fuel tanks places the engines in a stable orbit. The tank itself is deorbited and burns up.
At some point later something like the Starship collects several second stage engines and deorbits them safely to be reused.
Or perhaps just the engines can be immediately deorbited with an inflatable heat shield and parachutes.
Maybe they will have to sacrifice more payload mass for active or passive shielding or more fuel for powered deceleration. That would yield a less impressive lifter but with full reusability.
You're way too optimistic. Starship will deliver commercial payloads, with SpaceX phasing out Falcon 9 outside of ISS launches, before anyone has a reusable Falcon 9 equivalent.
It pains me to say this, but SpaceX is in a class all its own.
Note that the booster is not really being "caught" although this is the word it seems we're stuck with. It's really more like landing on the arms, since it throttles to a hover at that point.
Not necessarily for reusability, but it helps significantly for rapid reusability: it eliminates the need to transport the booster from the landing site to the launch site. Given that it's 9m x 70m and weighs 270 tonnes, that's not an easy process.
That was the exact reason they went this way.
SpaceX continues to blow me away with these unbelievably lofty ideas. I remember seeing their "grasshopper" flights years ago and they blew it up because they couldn't land it at the time, who knew within a few years they'd be doing this.
They were originally going to do high-altitude tests with an "F9 dev2" (similar to some of the early starship test launches) - but they gave up and just did some testing with real landing Falcon 9s instead.