Disclaimer: I've been watching the Starship program since it started and I want to see it succeed. I'm fairly certain that Super Heavy and Starship weigh more than this today, dry. And I have just...
Exemplary
Disclaimer: I've been watching the Starship program since it started and I want to see it succeed.
The Super Heavy Booster weighs 160 tonnes dry, doesn’t make it to space, and its peak speed is only roughly 4,600 mph. Meanwhile, Starship has a dry mass of around 150 tonnes, makes it to space, and reaches an orbital speed of at least 17,500 mph. This means that during landing, Starship has over 13.57 times the kinetic energy of the Booster! And that doesn’t account for the fact that Starship carries significantly more propellant during landing than the Booster.
I'm fairly certain that Super Heavy and Starship weigh more than this today, dry. And I have just one more nitpick here: calling the entirety of EDL (entry, descent, landing), "landing", is imprecise.
There's no way Starship carries more propellant than the Super Heavy booster during their respective landings. The Super Heavy booster is igniting 13 Raptors for a few seconds, and then scaling down to 3, for a total burn time of 20 seconds.
Meanwhile, the Starship second stage starts out with 3 Raptors, and switches down to 2, for a total burn time of 10 seconds.
Because the bellyflop manoeuvre is a no-go, Block 2 is designed to slow down more with its retro rockets (where the rockets are fired in the direction of travel to slow down). This should make landing more viable, as the craft should be more controllable. This would also enable the front fins to be shrunk and the heat shield to be thinned, saving weight. But this will also require more propellant, especially as the rockets have less thrust than planned. This is why Block 2 is larger and heavier than Block 1.
The only way that using Raptor as retro rockets is going to help out with heating is with hypersonic retropropulsion (the heatshield has heat load from 110km to 30km, roughly, where the current landing burn ignites at <1km altitude.), and I don't believe this is feasible. There simply is not the propellant load to do any kind of reentry burn.
If you want to know what retrorockets for an orbital spacecraft looks like, Stoke Space's Nova rocketis engineered from the ground up to support hypersonic retropropulsion! It has its heatshield surrounding the engine bay, with a ring of nozzles around it. I believe they also vent hydrogen into the bubble of plasma that the engines' exhaust traps? Very cool technology, I want them to succeed.
Will could be proposing some other retro rockets on Starship's heat shield, but Block 2 Starship doesn't have any rockets on its heatshield, which you can tell by just looking at it. Starship today has only Raptors and ullage-powered cold gas thrusters for propulsion, and cold gas thrusters can barely control Starship while it's in space. (See flights 3 and 9!)
But then, saying "the craft should be more controllable" implies starting the Raptors sooner for landing burn? You can't start it much sooner. Starship is subsonic for 1m45s, and the Raptors only have the propellant for maybe 10-20 seconds? And even if you have to run the Raptors for longer, that's still a bellyflop and a flip!
But Starship isn't going to switch away from Raptors any time soon.
This puts incredible pressure on SpaceX to save weight anywhere they possibly can, which is why the rockets keep failing.You can’t make them much lighter than they already are.
There is always incredible pressure on SpaceX to save weight anywhere. A ton removed from the second stage is a ton more payload to orbit (not counting the 0%-10% decrease in fuel required to land.) A ton removed from the first stage is anywhere from 1/5th to 1/10th ton of payload to orbit.
Indeed, we can see this with the planned Block 3 version of Starship. It is even longer than Block 2 to accommodate even more propellant to help increase its shrinking payload and ensure it can slow down enough to land. But somehow, it weighs significantly less than the smaller Block 2.
Where have these weight savings come from? They aren’t changing any major materials. They aren’t changing any structural designs. They aren’t redesigning the entire engine or fuel tank setup. The only way is if major systems are built with a smaller safety factor, making crucial systems vulnerable and weak.
But somehow, it weighs significantly less than the smaller Block 2.
Okay, I was going to write a rebuttal to this, but I didn't actually find a source for Block 3 being lighter? I expected it to be on one of SpaceX's presentations where they show different Starship versions, but those make no mention of dry mass. Admittedly, I didn't look very far. Maybe Elon has said this at some point; he says many things.
Also, they're using Raptor 3's on Block 3 Starship, which (theoretically) will not require external shielding, as SpaceX has stated:
You never achieve iterative design with a full-scale prototype. It is incredibly wasteful and can lead you down several problematic and dead-end solutions. I used to engineer high-speed boats — another weight- and safety-sensitive engineering field. We would always conduct scale model tests of every aspect of design, iteratively changing it as we went so that when we did build the full-scale version, we were solving the problems of scale, not design and scale simultaneously.
Some (petty!) quibbling on this point. My understanding of SpaceX's philosophy is this:
You can always reduce risk with more money.
No matter how much you reduce risk, there will be unknown unknowns.
Therefore, it's better to get every component to 1% risk, and plan for a few failure, than get 0.01% risk on every component.
SpaceX could have spent twice as much and three times longer and gotten to where they are today, with no debris sent into the Gulf or the Turks and Caicos or the Turks and Caicos (again!).
Flight 10 shows the result of this philosophy; after 3 brutal failures, after the first design of the Payload bay didn't close, everything worked.
(Tangent Alert: In my opinion, as soon as you're in the philosophy of "it cannot fail", improving any part of the process becomes vastly harder. Why redesign anything if the old one is flight proven and the new one could doom a launch?)
NASA has taken the other side of this tradeoff, where nothing can fail because they're launching people.
They're launching Artemis II with astronauts on board, with Artemis I's heatshield design which showed unexpected behavior. (Artemis I was the first in-flight test of Orion's heatshield since EFT-1 in 2014, and they changed the design in-between.) They're mitigating this with a new reentry trajectory, which has never been tested. And it will work! Heck, Artemis I would have been a success with astronauts on board!
Also, they're going to revise the heatshield to fix this problem! This will probably (unless they decide to do a flight test, which personally I'd consider unlikely) debut on Artemis III, again with astronauts on board.
Artemis II is also the first Orion flight to launch with working life support:
So for Artemis I, we had a pressure control system that maintained, you know, habitable temperatures within the cabin. But for Artemis II, we’re flying new hardware to this time to, you know, to provide CO2 removal and humidity removal for the, for the for the crew. [...]
Again, this will work! NASA will spent a lot of money and a lot of time to make things that won't fail, because NASA is risk-averse at every level and also they're launching people and willing to spend near-infinite money to not kill people.
It's worth noting that life support is one of the easier systems to test on the ground. Not so for reentry or aerodynamic control or propellant tanks.
This is a tradeoff! Feel free to debate it!
SpaceX could have easily done this. They already proved they could land a 1st stage/Booster with the Falcon 9, and Falcon 9’s Booster could launch a 1/10 scale Starship into orbit. Tests of such a scaled-down model would help SpaceX determine the best compromise for using the bellyflop manoeuvre and retro rockets to land. It would help them iteratively improve the design around such a compromise, especially as they will be far cheaper and quicker to redesign and build than the full-scale versions.
From where I'm standing, SpaceX has already found very nearly the best compromise between the bellyflop and using the landing engines. High-altitude flight tests with SN10 and SN15 proved the bellyflop and landing will work. Integrated flight tests 4, 5, 6, and now 10 have shown successful bellyflop and landing. They did this with extensive computer modeling and the high-altitude Starship hop campaign.
Not only that, but these tests would highlight any of the design’s shortcomings, such as the rocket engines not having enough thrust-to-weight ratio to enable a high enough payload. This allows engineers to do crucial, complete redesigns before the large-scale version is even built.
Today, the rocket engines are ~18t of a ~150t second stage. Reducing their mass to 0 [i.e. infinite thrust-to-weight] would add ~18t of payload capacity.
It's on the first stage that thrust matters more. On the first stage, each engine lifts a exit-area-sized column of propellant. Increasing the thrust here allows for a taller rocket, which increases the fraction of mass dedicated to tankage. The weight (as part of the thrust-to-weight) of a rocket engine is kind of not important?
Besides, I feel like the Raptor engine is in a pretty good place already. They're test-firing Raptor 3 on stands today, which has a theoretical thrust-to-weight of 160, while Raptor 2 is at 80.
Well, through some transparent corruption and cronyism, he could secure multi-billion-dollar contracts from NASA to build this mythical rocket. But, by going for full-scale testing, he could not only hide the inherent flaws of Starship long enough for the cash to be handed over to him but also put NASA in a position of the sunk cost fallacy. NASA has given SpaceX so much money, and their plans rely so heavily on Starship that they can’t walk away; they might as well keep shoving money at the beast)
As far as I can tell, NASA's Starship HLS contract is for upwards of $4 billion ($2.9 billion for Artemis III, and $1.1 billion for IV). These contracts are milestone-based, meaning that NASA and SpaceX agree on milestones, and SpaceX is only awarded money for demonstrating these milestones.
That is the real reason why Starship was doomed to fail from the beginning. It’s not trying to revolutionise the space industry; if it were, its concept, design, and testing plan would be totally different. Instead, the entire project is optimised to fleece as much money from the US taxpayer as possible, and as such, that is all it will ever do.
It is trying to revolutionise the space industry. Its concept, design, and testing plan are totally different from any rocket developed in the last 50 years.
It has already launched 10 times, which would be a staggering achievement for any program "optimised to fleece as much money from the US taxpayer as possible".
In summary, this post is vague, is barely refutable, and where it is refutable, wrong.
Since this article was written, there was a test flight (In August) that was mostly a success. The project might still fail in the end, but “doomed from the beginning” would be hard to prove, and...
Since this article was written, there was a test flight (In August) that was mostly a success.
The project might still fail in the end, but “doomed from the beginning” would be hard to prove, and I don’t see this article proving it. It’s a rant with little technical content.
Consider what we know of Elon Musk, and then consider that this was his idea. Elon Musk is a very rich con man who has never actually done engineering. He doesn't know what the fuck he's doing....
Consider what we know of Elon Musk, and then consider that this was his idea.
Elon Musk is a very rich con man who has never actually done engineering. He doesn't know what the fuck he's doing. Odds are the engineers in his employ do, but they're being led by a con man.
If a space company that's being led by rich con man who has never done any engineering can launch 29 customer payloads in a year, why aren't any other companies getting better CEOs and...
If a space company that's being led by rich con man who has never done any engineering can launch 29 customer payloads in a year, why aren't any other companies getting better CEOs and outcompeting them?
[edited to improve clarity and be less argumentative]
Yes. I’ve basically been screaming stuff like this since the project started back before musks complete fall from grace. The plan has always from day one been batshit insane and makes no logical...
Yes. I’ve basically been screaming stuff like this since the project started back before musks complete fall from grace.
The plan has always from day one been batshit insane and makes no logical or physical sense given what we know and what we’ve done.
It has always included WILD projections that have no basis in reality and NASA is, in my eyes, well aware they are wasting their time on this.
We already had this video https://m.youtube.com/watch?v=OoJsPvmFixU&pp=ygUMSSB0b2xkIG5hc2Eg which brings up, in very nice way, how totally fucked this all is form a core level. “Well maybe it’s actually X refuels” is NOT something you can have be a question mark at this stage of development.
At best, absolute best, from the start we were maybe getting starship in orbit, and that’s it. It looks like that isn’t even happening.
The moon was always unlikely, and you can tell how bullshit this all is because the claim was Mars and that was 10000000% dumb on its face for a million reasons and in my eyes showed just how many people like the “dream” of science more than the reality.
I think it’s important to distinguish which project you’re talking about. Starship could succeed as a launch platform (for carrying lots of cargo to orbit) even if other projects it was designed...
I think it’s important to distinguish which project you’re talking about. Starship could succeed as a launch platform (for carrying lots of cargo to orbit) even if other projects it was designed for fail.
It could also succeed technically as a launch platform while being a commercial failure (not paying for itself) if there’s not enough demand for all that cargo space.
I cannot understand how our design plan for rockets being. “Well maybe this total scam might have a use “ is remotely acceptable. I think there’s also remedies evidence that neither of those...
I cannot understand how our design plan for rockets being. “Well maybe this total scam might have a use “ is remotely acceptable.
I think there’s also remedies evidence that neither of those outcomes are likely, and if they were the intent there are better ways to approach it.
Disclaimer: I've been watching the Starship program since it started and I want to see it succeed.
I'm fairly certain that Super Heavy and Starship weigh more than this today, dry. And I have just one more nitpick here: calling the entirety of EDL (entry, descent, landing), "landing", is imprecise.
There's no way Starship carries more propellant than the Super Heavy booster during their respective landings. The Super Heavy booster is igniting 13 Raptors for a few seconds, and then scaling down to 3, for a total burn time of 20 seconds.
Meanwhile, the Starship second stage starts out with 3 Raptors, and switches down to 2, for a total burn time of 10 seconds.
The proposed trajectory is idea-shaped nonsense.
SpaceX has planned to shrink the front flaps since before Flight 3, from what I remember. Starship Flight 10's second stage landed in the same way as Block 1 ships did, with a bellyflop, landing burn, and flip.
The only way that using Raptor as retro rockets is going to help out with heating is with hypersonic retropropulsion (the heatshield has heat load from 110km to 30km, roughly, where the current landing burn ignites at <1km altitude.), and I don't believe this is feasible. There simply is not the propellant load to do any kind of reentry burn.
If you want to know what retrorockets for an orbital spacecraft looks like, Stoke Space's Nova rocket is engineered from the ground up to support hypersonic retropropulsion! It has its heatshield surrounding the engine bay, with a ring of nozzles around it. I believe they also vent hydrogen into the bubble of plasma that the engines' exhaust traps? Very cool technology, I want them to succeed.
Will could be proposing some other retro rockets on Starship's heat shield, but Block 2 Starship doesn't have any rockets on its heatshield, which you can tell by just looking at it. Starship today has only Raptors and ullage-powered cold gas thrusters for propulsion, and cold gas thrusters can barely control Starship while it's in space. (See flights 3 and 9!)
But then, saying "the craft should be more controllable" implies starting the Raptors sooner for landing burn? You can't start it much sooner. Starship is subsonic for 1m45s, and the Raptors only have the propellant for maybe 10-20 seconds? And even if you have to run the Raptors for longer, that's still a bellyflop and a flip!
But Starship isn't going to switch away from Raptors any time soon.
There is always incredible pressure on SpaceX to save weight anywhere. A ton removed from the second stage is a ton more payload to orbit (not counting the 0%-10% decrease in fuel required to land.) A ton removed from the first stage is anywhere from 1/5th to 1/10th ton of payload to orbit.
Okay, I was going to write a rebuttal to this, but I didn't actually find a source for Block 3 being lighter? I expected it to be on one of SpaceX's presentations where they show different Starship versions, but those make no mention of dry mass. Admittedly, I didn't look very far. Maybe Elon has said this at some point; he says many things.
Also, they're using Raptor 3's on Block 3 Starship, which (theoretically) will not require external shielding, as SpaceX has stated:
SpaceX — [...] Raptor 3 is designed for rapid reuse, eliminating the need for engine heatshields [...]
@SpaceX
Raptor 3 (sea level variant)
Thrust: 280tf
Specific impulse: 350s
Engine mass: 1525kg
Engine + vehicle-side commodities and hardware mass : 1720kg
9:29 AM · Aug 3, 2024
@SpaceX
Performance stats of previous versions:
Raptor 1 (sea level variant)
Thrust: 185tf
Specific impulse: 350s
Engine mass: 2080kg
Engine + vehicle-side commodities and hardware mass: 3630kg
Raptor 2 (sea level variant)
Thrust: 230tf
Specific impulse: 347s
Engine mass: 1630kg
Engine + vehicle-side commodities and hardware mass: 2875kg
Raptor 3 is designed for rapid reuse, eliminating the need for engine heatshields while continuing to increase performance and manufacturability
10:59 AM · Aug 3, 2024
Some (petty!) quibbling on this point. My understanding of SpaceX's philosophy is this:
You can always reduce risk with more money.
No matter how much you reduce risk, there will be unknown unknowns.
Therefore, it's better to get every component to 1% risk, and plan for a few failure, than get 0.01% risk on every component.
SpaceX could have spent twice as much and three times longer and gotten to where they are today, with no debris sent into the Gulf or the Turks and Caicos or the Turks and Caicos (again!).
Flight 10 shows the result of this philosophy; after 3 brutal failures, after the first design of the Payload bay didn't close, everything worked.
(Tangent Alert: In my opinion, as soon as you're in the philosophy of "it cannot fail", improving any part of the process becomes vastly harder. Why redesign anything if the old one is flight proven and the new one could doom a launch?)
NASA has taken the other side of this tradeoff, where nothing can fail because they're launching people.
They're launching Artemis II with astronauts on board, with Artemis I's heatshield design which showed unexpected behavior. (Artemis I was the first in-flight test of Orion's heatshield since EFT-1 in 2014, and they changed the design in-between.) They're mitigating this with a new reentry trajectory, which has never been tested. And it will work! Heck, Artemis I would have been a success with astronauts on board!
Also, they're going to revise the heatshield to fix this problem! This will probably (unless they decide to do a flight test, which personally I'd consider unlikely) debut on Artemis III, again with astronauts on board.
Artemis II is also the first Orion flight to launch with working life support:
Again, this will work! NASA will spent a lot of money and a lot of time to make things that won't fail, because NASA is risk-averse at every level and also they're launching people and willing to spend near-infinite money to not kill people.
It's worth noting that life support is one of the easier systems to test on the ground. Not so for reentry or aerodynamic control or propellant tanks.
This is a tradeoff! Feel free to debate it!
From where I'm standing, SpaceX has already found very nearly the best compromise between the bellyflop and using the landing engines. High-altitude flight tests with SN10 and SN15 proved the bellyflop and landing will work. Integrated flight tests 4, 5, 6, and now 10 have shown successful bellyflop and landing. They did this with extensive computer modeling and the high-altitude Starship hop campaign.
Today, the rocket engines are ~18t of a ~150t second stage. Reducing their mass to 0 [i.e. infinite thrust-to-weight] would add ~18t of payload capacity.
It's on the first stage that thrust matters more. On the first stage, each engine lifts a exit-area-sized column of propellant. Increasing the thrust here allows for a taller rocket, which increases the fraction of mass dedicated to tankage. The weight (as part of the thrust-to-weight) of a rocket engine is kind of not important?
Besides, I feel like the Raptor engine is in a pretty good place already. They're test-firing Raptor 3 on stands today, which has a theoretical thrust-to-weight of 160, while Raptor 2 is at 80.
As far as I can tell, NASA's Starship HLS contract is for upwards of $4 billion ($2.9 billion for Artemis III, and $1.1 billion for IV). These contracts are milestone-based, meaning that NASA and SpaceX agree on milestones, and SpaceX is only awarded money for demonstrating these milestones.
It is trying to revolutionise the space industry. Its concept, design, and testing plan are totally different from any rocket developed in the last 50 years.
It has already launched 10 times, which would be a staggering achievement for any program "optimised to fleece as much money from the US taxpayer as possible".
In summary, this post is vague, is barely refutable, and where it is refutable, wrong.
I haven't followed Space stuff very closely, is this plausibly true?
Since this article was written, there was a test flight (In August) that was mostly a success.
The project might still fail in the end, but “doomed from the beginning” would be hard to prove, and I don’t see this article proving it. It’s a rant with little technical content.
Consider what we know of Elon Musk, and then consider that this was his idea.
Elon Musk is a very rich con man who has never actually done engineering. He doesn't know what the fuck he's doing. Odds are the engineers in his employ do, but they're being led by a con man.
This doesn’t prove anything one way or the other about whether SpaceX will succeed.
If a space company that's being led by rich con man who has never done any engineering can launch 29 customer payloads in a year, why aren't any other companies getting better CEOs and outcompeting them?
[edited to improve clarity and be less argumentative]
Yes. I’ve basically been screaming stuff like this since the project started back before musks complete fall from grace.
The plan has always from day one been batshit insane and makes no logical or physical sense given what we know and what we’ve done.
It has always included WILD projections that have no basis in reality and NASA is, in my eyes, well aware they are wasting their time on this.
We already had this video https://m.youtube.com/watch?v=OoJsPvmFixU&pp=ygUMSSB0b2xkIG5hc2Eg which brings up, in very nice way, how totally fucked this all is form a core level. “Well maybe it’s actually X refuels” is NOT something you can have be a question mark at this stage of development.
At best, absolute best, from the start we were maybe getting starship in orbit, and that’s it. It looks like that isn’t even happening.
The moon was always unlikely, and you can tell how bullshit this all is because the claim was Mars and that was 10000000% dumb on its face for a million reasons and in my eyes showed just how many people like the “dream” of science more than the reality.
I think it’s important to distinguish which project you’re talking about. Starship could succeed as a launch platform (for carrying lots of cargo to orbit) even if other projects it was designed for fail.
It could also succeed technically as a launch platform while being a commercial failure (not paying for itself) if there’s not enough demand for all that cargo space.
I cannot understand how our design plan for rockets being. “Well maybe this total scam might have a use “ is remotely acceptable.
I think there’s also remedies evidence that neither of those outcomes are likely, and if they were the intent there are better ways to approach it.