It is the cheapest AND most advanced rocket design ever conceived. You can thank Musk for saving taxpayers' money; if he didn't come, you would have to remain dumping massive tons of money (IIRC 10x more per year than the whole Starship program has cost SpaceX - SpaceX, not taxpayers, they contributed only a very minor part of that budget) to Boeing and other very expensive parties. BTW I am sure you know that SpaceX was funded with private money at the beginning and does a lot of private flights.
BTW 2, I am sure that you know that advancing spaceflight has widespread positive effects on humanity as a whole; the people using Starlink won't be billionaires, but ordinary dudes.
In a lot of ways the Space Shuttle was more advanced. But more advanced is not always better. Sometimes the simple solution is the best.
Complaining that a rocket company gets government funding is dumb. All of the major players get government funding to one degree or another. SpaceX gets less than most.
Does anyone with real expertise in aerospace have thoughts about Musk's choice to use steel instead of carbon fiber for star ship? Does his arguments make sense or is it likely just a cost saving measure?
It's like the open source software of materials, not only is it cheap but it's fast because all the info you need is out there in industry.
Steel is just so much more workable every step of the way. You can order your special steel from close to anywhere, your fancy epoxy might have two sources. Any engineer can design for steel and be really good at it (they probably even took classes on it in college) so you can have more cheaper engineers working parallel. Your engineers that know their way around CF are gonna be much more rare and expensive. The tool and die side of things is well known. The little tricks that are tribal knowledge for CF have long been compiled and published into textbooks for steel. If you fat finger a relief radius the die maker is gonna ask you to clarify whereas with some modern material that doesn't have that body of documentation yet they'll just assume you know what you're doing. The material handling side of things is well known, you'll never have to trash a part of delay a deadline because someone somewhere didn't store one of your constituent products in the right environment. When you've finally built your rocket anyone can test it. The techniques for inspecting steel are well known and widely practiced. For exotic materials you either have to track down experts or build that tribal knowledge internally (expensive and slow).
These little conveniences compound throughout your entire "technology stack" so to speak and a rocket company has a really tall stack.
It's hard to overstate just how much easier and faster you can work in a known-known like steel vs something that's new, hot and sexy that everyone wants to treat like a trade secret (as is the case with CF).
Source: Former defense sector employee, granted on the software side but you still learn a lot of things through proximity.
My understanding is lots of steel grades are trade secrets. Though how to actually make it isn't necessarily a secret, others might not know which alloy is specifically employed for a particular purpose
It's a trade secret in the way that a pizza franchise's sauce is a trade secret. The specific combination of component parts and why that specific combination was chosen is a secret but what the components do and how they interact and affect the final product is well known.
I'm not sure why this is down-voted. Do people not believe that people with expertise in food can tell you what the amount of salt or sugar does to a sauce or that people in the steel industry knows what the amount of chrome does to a steel?
I would also say that I suspect it is much easier to ensure and test quality of steel. Ensuring every square inch of compound material like carbon fiber meets all requirements is much more difficult.
Steel can be made pretty uniform in large quantities, shapes and sizes. On the other hand carbon fiber can easily have local faults ingredient imbalances, etc. that will be difficult to detect but might significantly affect strength of the material.
In case of a huge rocket it does not matter how small part of it fails, even a single substandard square inch of it might cause entire rocket to fail catastrophically.
I also understand that steel fails in relatively predictable ways compared to compound material. Carbon fiber has different properties in different directions which depend on how it was layered and that is probably going to cause any calculations to have potentially very large margin of error. As building the rocket is very much about calculating how much can be shaved off, unpredictability in material performance might be a huge issue.
There are a variety of non-destructive test methods for carbon fiber. CF is definitely prone to defects, but it is now common enough, and sufficiently well-understood to be quite reliable.
Outside of Re-entry concerns, a large reason for the move away from CF was that tooling to create CF tubes at the size of Starship didn't exist, and would have to be created from scratch at large expense.
SpaceX had CF tooling (most notably including a mandrel), and had built at least one tank. They recently scrapped the tooling, after settling on stainless steel.
One of the old Atlas rockets was stainless steel. The point is optimizing the entire rocket instead of a small section.
Starship will spend most of its time being in high heating conditions (like reentry) or very cold because the rocket fuel is loaded into the rocket at very low temperatures. It turns out that while the strength to weight ratio of carbon fiber is better than this alloy of steel at room temperature, it is impacted negatively by both hot and cold. In contrast, the steel alloy they are using actually has higher strength at low temperature and performs much better than carbon fiber at higher temperatures.
So the whole rocket can be lighter and simpler. Carbon fiber rockets that are going to reenter require a lot of heat shielding that this ship won't need.
And rockets and their engines are designed together. So it would be very difficult to make a design change like moving from aluminum to steel on Falcon without essentially designing a new rocket. Instead they will focus all their energy on Starship and eventually replace the Falcon series.
SpaceX has said it is possible that the build cost of a Starship/Superheavy stack could be lower than a Falcon 9. The variable costs of a launch would also be cheaper while delivering more payload.
This is the rocket that could make a lot of science fiction come true. That is what the excitement is about and we should all hope they get it to work.
> This is the rocket that could make a lot of science fiction come true
The irony is that those 1930s scifi stories with silver rockets landing vertically on distant planets, which looked so ridiculous in the 80s with the Shuttle being the future, look like that's exactly what the future will be.
Seeing the 2 boosters landing side by side after the first Heavy launch was something out of a scifi film.
I can clearly remember the day of the Falcon Heavy launch. I almost cried when the fairing came off and starman in his tesla was in open space, it was so incredible. I think a lot of people clearly remember that day, probably more so than the 2018 superbowl which was played 2 days earlier.
I was at kennedy space center at the Apollo / Saturn V center and was pretty much in hysterical happy tears when the side boosters came in for the landing. It was magical.
I'm not into food, but that video - and SpaceX know how to do publicity - made me very hopeful that I will get to see people landing on Mars in my lifetime.
Aren't we at the point that you could have lots of bots ruining sites like this by posting things like that? It seems doable to do even today. I really enjoy technical/cultural/watercooler conversations with a like group of technical scientific people. hope it doesn't go away.
I remember seeing it around or just after midnight when I was visiting a customer in Taipei. I had fallen asleep two hours earlier, but set a timer to wake up just for this historic event.
Totally worth it. That was one the best executived live streams of all time:
That really was an incredible thing to watch. I still tear up a bit when I watch the "Falcon Heavy and Starman" video that they posted a few days later...
I've noticed that- I watch a lot of MST3K and it's always a 50s Sci-fi flick that show rockets landing vertically like SpaceX. They also really like to explain gravity and orbital mechanics at length in every movie because this stuff was all so new and cool!
The implication of the sentence is to ask if there is a practical advantage or only a cost advantage (or both). I think you could give the writer more credit.
Only if weight is the least concern. Most engineering is gravity bound and can replace thin expensive things with thick heavy ones for less cost. If weight or volume become sufficiently important then practicality moves away from cost: you could make a rocket out of bricks and bricks are cheap but it's not the best solution.
I don't know about steel but carbon fiber has so far never succeeded as a material for cryogenic rockets or fuel tanks.
The proposed space shuttle successor was designed around carbon fiber fuel tanks / hull and engineering problems with them eventually killed the project.
The material struggles with low temperatures of cryogenic fuels (brittle), high temperatures of re-entry (just burns), oxidation in oxygen tanks. Also they apparently tend to sponatiously combust with rocket fuels.
But the scale is much, much smaller than Starship.
If I understand correctly after you pass a certain tank radius you basically step into “nobody’s ever done this with CF before”, which is not a good place to be for SpaceX. Starship was on that spectrum.
Elon Musk cares about one over-arching thing: making humanity an interplanetary species. I believe that everything he does is in service to that goal.
In order to make humanity interplanetary, lift costs must be greatly and sustainably reduced.
In order for lift costs to be greatly and sustainably reduced, everything in the lift vehicle except the propellant must be fully reusable. Also, the vehicle needs to be inexpensive, quick to manufacture and reasonably performant.
In order for the orbital stage of a lift vehicle to be re-used, it must re-enter the earth's atmosphere and effectively shed a large fraction of the chemical energy that was used to accelerate it to 7.8km/s.
In short, the 2nd stage must be able to repeatably handle intense re-entry heat.
For performance (and other) reasons, it must also be able to handle the very low temperatures of liquid propellants.
Given all of these constraints, specific types of steel are literally the only choice.
1. Low material cost
2. Low manufacture cost
3. Excellent high temperature performance
4. Excellent low temperature performance
5. Sufficient normal temperature performance
In fact, Musk has said that his initial focus on carbon fiber was a big error.
Re: Falcon: as others have stated, Falcon 9 is effectively a fixed design. It's now a 'cash cow', and there's no reason to substantially change it. In fact, changing it would risk a lot of profitable contracts that are effectively funding Starship/Superheavy development.
Then why wasn't Falcon 9 steel to start out with? Just more familiarity of aluminum (which I assume is its main structural element) within the rocket biz? Or does Falcon 9 just not deal with that re-entry requirement (it's only first stage, not the much faster second stage) and therefore aluminum is the right answer because it is cheaper/lighter?
The reason is that this is a completely different use case.
The second stage of a F9 is not meant to survive re-entry, so there is no need to do anything interesting in terms of heat shielding. Starhopper is meant to survive re-entry, so the calculation had to be made to have it survive the heat.
It was deemed "better" to rely on steel rather than an ablative heatshield and aluminum because inspection requirements are pretty high.
As for the 1st stage also being steel instead of aluminum like the F9. The 1st stage booster will be much larger than the F9 and so the strength of steel is an advantage since it avoids the issue of the structure not being able to stay upright unpressurized.
I never claimed it was the _only_ reason, but it is certainly one of them. Even if you wanted to make F9 out of steel (for whatever reason) you still couldn't do it with Merlins in an economically feasible way.
This is a simple fraction:
thrust-to-weight ratio = thrust / weight.
If you increase the denominator but keep your thrust the same then you lose payload capacity.
This is incorrect. A T/W ratio by itself is meaningless since it ignores mass.
The propellant mass fraction is what actually matters which takes into account the mass of the vehicle.
The fact that the Raptor is more powerful than the Merlin is barely a consideration in the "Build the Starship out of steel discussion". They would have loved to not have to bite into their mass budget, but other engineering concerns triumphed.
Payload capacity would not be diminished. Steel is denser but also stronger than aluminum, so you make the walls thinner and there's no mass penalty. Thin walls, though, are difficult to work with, though that's been done before with Atlas and Centaur.
To be mass-efficient, steel for a rocket the size of Falcon 9 would have to be so thin as to crumple easily—a "balloon tank". This makes manufacturing and handling difficult, as the structure must be either fully supported or pressurized at all times. (Starship is large enough that the steel skin ends up a workable 4mm thick)
This has precedent in early Atlas rockets and the Centaur second stage still used to this day, but that's not what SpaceX had experience with on Falcon 1 (where it would have been even more unreasonably thin) and they probably just leveraged their knowledge to create as derivative a vehicle as possible when making F9.
Like really, there is a compressor next to the rocket that keeps constant pressure inside against the slight leaks. Pretty impressive for 40-50 old rocket left outside exposed to the elements to be still this airtight. :)
> Then why wasn't Falcon 9 steel to start out with?
The Falcon 9 has between no and minimal high heat requirements.
The 2nd stage does not attempt to survive re-entry.
The 1st stage enters at (if memory serves) around 1.5km/s compared to nearly 8km/s (or beyond for interplanetary returns) that Starship will need to handle.
Also the Falcon 9 didn't have the lofty re-usable and manufacturing goals that Starship has.
I'm not anything like an expert on this matter, so I can't say with certainty, but I think aluminum is a better choice given these slightly different constraints.
It's also possible that steel would have been a better choice for Falcon 9 even give the more limited requirements. Understanding is always an incomplete quantity.
According to Elon Musk it's not about saving costs but about speeding up development. Welding steel needs basically no tooling or setup time. It's hard to imagine their current methodology of rapid prototyping with carbon fiber. Other advantages are just a bonus.
The arguments for it make a lot of sense. One of them is that it's a cost saving measure. All other things being equal, that's a good thing, not a bad one.
The Falcon family is wildly successful, but it's small potatoes next to SpaceX's vision for the economics of space launch in the near future, so it seems unlikely that they'd go back to fix what isn't broken.
Elon is all about tight, quick iterations on design and testing: “if it takes long is wrong” is an actual quote from him regarding Starship.
With carbon fiber you get a lot of theoretical nice-to-haves, but you can’t iterate quickly with it. You need specialized tooling for manufacturing plus there is a lot of actual R&D that still needs to be done to manufacture with CF at Starship scale.
SpaceX sees all this effort as a diversion from the goal of actually building a new rocket. Compare with steel where you can just hire a bunch of welders and build the damn thing out in the open.
Falcon will never switch to steel. It makes no sense from a performance POV (Merlins are much less powerful than Raptors so your payload capacity hit is too high) and Falcon is a “dead” platform: only minor incremental updates from now on. Any change to it will trigger recertification with NASA, the USAF and a bunch of clients, which is a no-go. Falcon is SpaceX’s workhorse and money making machine, and it will continue on that role until Starship retires it.
> Compare with steel where you can just hire a bunch of welders and build the damn thing out in the open.
I don't really know anything about welding nor about the forces acting on a rocket, but I remember reading that the effort in welding the Saturn V rocket was a feat in itself. It took a long time to train the workforce to perform the welds and extensive inspection was needed.
Does anyone know if this applies to this rocket too? Has welding progressed (or simply kept the skill alive)?
Edit: To anyone interested. The excellent book, Stages to Saturn, covered this (and everything else) in quite some detail.
You have to remember that Starship is not going orbital any time soon, this is all for testing and development. I can guarantee you that the final version of Starship will not be manufactured so haphazardly as these ones.
And they’ve already seen several failures with their prototypes so even with steel they have learning to do.
I'm sure it's unlikely now because of the failures you mentioned but Elon said they are aspirationally targeting an orbital flight with SN5 or SN6 before the end of the year. Elon had earlier said he thought it was possible to get orbital flights in 6 months but that was in September and was certainly also aspirational.
It looks like SN5 will be doing the 20km test flight (whereas they probably planned on doing that with SN4 originally but then SN3 was destroyed in the pressure test failure).
AFAIK Falcon 9 is using automated friction stir welding, which was only invented in 1991 and first used on rockets in 1999 by ULA. So there is definitely considerable progress in welding that has been done since Apollo.
As for Starship, IIRC friction stir welding of stainless steel has not really been done, definitely not at scale so SpaceX will have to use more "traditional" welding methods, which have still got a lot of improvements and new techniques as well since the Saturn 5 times, as mentioned in other comments.
What do you mean "just" a cost saving measure? It's 3% the cost. So they can build 30 rockets for the price of one. Pretty important detail, when your goal is to reduce launch cost be two orders of magnitude.
Elon's long term plan for Falcon is to switch customers to Starship as soon as it's available, and they're comfortable switching. (Which might be a while for the Air Force and NASA.) In the meantime, any major changes to Falcon would be a serious distraction, which he'd probably forgo for that reason alone, regardless of the technical merits.
I'm just a hobbiest but I have tried to look into this. We really need to talk about the first and second stage differently. Of course Elon is a great one for communiality in the way he build things, common fuel, common engines and so on. He does this agian with Starship/Super Heavy, Metholox, Raptor Engine, Steel tanks.
Lets talk about the most difficult part, the second stage, ie Starship. The big problem you have to solve, is to build a vehicle that can go from incredibly cold structure to a very hot one. Because you start very cold, and when coming back, it gets really hot.
SpaceX loves deep cryo, meaning not just going to the point of making that fuel/oxidiser liquid, but really pushing it down as far as possible. This has a number of advantages but also many disadvantages. SpaceX took a long time to perfect the deep cryo technology with Falcon 9 (that was of course Kerolox).
Because the fuel heating up it forced their operations to be very tight, and in the beginning cause many launch delays. It also cause the famous AMOS 6 failure, as a pretty interesting mostly unknown effect was taking place dring these temparatures.
The upside is that you increase desnisty and thus vehicle performance, and it helps keep your turbo pumps cool as well.
SpaceX very much wanted to continue this for Starship architecture, as they think its a huge win you are leaving on the table if you don't do it.
What this means is that Starship is incredibly cold on the pad, really as cold as they can get it. Any structural material, needs to be able to handle this. This by itself is very challanging with carbon fiber even with the best carbon fiber technology perfomrance degrades the deeper you go. It forces you to really build a carbon fiber structure like we have never seen before.
While some steels are terrible brittle under that condition, 301 stainless (and of course the new alloy SpaceX is developing) are the opposite. Their strengh actually increases, and basically the colder the better. So with incredibly cheap steel that you can make pots out of, you can basically match the performance of super expensive carbon fiber.
On the hot side, you have the reentry. Starship is design not just to come in from LEO like the Shuttle, but actually come from interplantary and then use the air to break. So this is a much hotter profile then Shuttle. And as most people know, Shuttle heat shield caused endless problems.
With both carbon fiber and aluminum you really don't want to let them get them to hot (500C is pushing the issue). Any such design basically requires packaging most of the Starship in a very thick Heatshield. Steel however is very tolerant of heat, meaning that even under fairly high tempratures (800C ish), once it cools down, your structure is still as good as before.
When designing and insulating heat shield (rather then an non-reusable ablative one) the question is really how thick, and thus how heavy does it have to be. The amount of heat that can be handled by the structure behind the heat shield, directly impacts how thick the shield needs to be.
So the combination of being almost as good as peak carbon fiber in deep cryo, and being clearly a much superior choice for reentery makes it a good choice. Once you combine that with the operational apspects, meaning both the per-unit material cost and much cheaper manufacturing its a no-brainer at that point. This is why I believe Musk called it 'The best desision in the whole design', it just made everything easier.
Another bonus is that you have access to a far larger group of engineers with experinace, plus an even larger gorup of workers that can work with steel reliably compared with carbon fiber.
Now for the First Stage, there is a good argument to be made that steal probebly can't quite reach the peak performance level of a carbon fiber design. Maybe not of an aluminum design. In this case I think its really more about material and manufacturing price and of communality with the Starship.
65 comments
[ 4.0 ms ] story [ 126 ms ] threadA reminder SpaceX lives off taxes to enrich the lives of Billionaires.
Congrats on the working design. Sounds expensive.
BTW 2, I am sure that you know that advancing spaceflight has widespread positive effects on humanity as a whole; the people using Starlink won't be billionaires, but ordinary dudes.
Complaining that a rocket company gets government funding is dumb. All of the major players get government funding to one degree or another. SpaceX gets less than most.
You also probably use a hundred objects every day directly funded by taxpayer money given to private corporations.
Would future Falcon rockets also switch to steel?
Additionally the ease of manufacture is not just a manufacturing cost saver, it's a development cost and time saver.
Falcon is a frozen design and won't change, beyond a proposed longer fairing.
Steel is just so much more workable every step of the way. You can order your special steel from close to anywhere, your fancy epoxy might have two sources. Any engineer can design for steel and be really good at it (they probably even took classes on it in college) so you can have more cheaper engineers working parallel. Your engineers that know their way around CF are gonna be much more rare and expensive. The tool and die side of things is well known. The little tricks that are tribal knowledge for CF have long been compiled and published into textbooks for steel. If you fat finger a relief radius the die maker is gonna ask you to clarify whereas with some modern material that doesn't have that body of documentation yet they'll just assume you know what you're doing. The material handling side of things is well known, you'll never have to trash a part of delay a deadline because someone somewhere didn't store one of your constituent products in the right environment. When you've finally built your rocket anyone can test it. The techniques for inspecting steel are well known and widely practiced. For exotic materials you either have to track down experts or build that tribal knowledge internally (expensive and slow).
These little conveniences compound throughout your entire "technology stack" so to speak and a rocket company has a really tall stack.
It's hard to overstate just how much easier and faster you can work in a known-known like steel vs something that's new, hot and sexy that everyone wants to treat like a trade secret (as is the case with CF).
Source: Former defense sector employee, granted on the software side but you still learn a lot of things through proximity.
not an expert, but I heard that spacex's steel alloy composition is a trade secret.
https://en.wikipedia.org/wiki/SAE_steel_grades
Same is true for aluminum https://en.wikipedia.org/wiki/Aluminium_alloy#Alloy_designat...
I would also say that I suspect it is much easier to ensure and test quality of steel. Ensuring every square inch of compound material like carbon fiber meets all requirements is much more difficult.
Steel can be made pretty uniform in large quantities, shapes and sizes. On the other hand carbon fiber can easily have local faults ingredient imbalances, etc. that will be difficult to detect but might significantly affect strength of the material.
In case of a huge rocket it does not matter how small part of it fails, even a single substandard square inch of it might cause entire rocket to fail catastrophically.
I also understand that steel fails in relatively predictable ways compared to compound material. Carbon fiber has different properties in different directions which depend on how it was layered and that is probably going to cause any calculations to have potentially very large margin of error. As building the rocket is very much about calculating how much can be shaved off, unpredictability in material performance might be a huge issue.
Starship will spend most of its time being in high heating conditions (like reentry) or very cold because the rocket fuel is loaded into the rocket at very low temperatures. It turns out that while the strength to weight ratio of carbon fiber is better than this alloy of steel at room temperature, it is impacted negatively by both hot and cold. In contrast, the steel alloy they are using actually has higher strength at low temperature and performs much better than carbon fiber at higher temperatures.
So the whole rocket can be lighter and simpler. Carbon fiber rockets that are going to reenter require a lot of heat shielding that this ship won't need.
And rockets and their engines are designed together. So it would be very difficult to make a design change like moving from aluminum to steel on Falcon without essentially designing a new rocket. Instead they will focus all their energy on Starship and eventually replace the Falcon series.
SpaceX has said it is possible that the build cost of a Starship/Superheavy stack could be lower than a Falcon 9. The variable costs of a launch would also be cheaper while delivering more payload.
This is the rocket that could make a lot of science fiction come true. That is what the excitement is about and we should all hope they get it to work.
The irony is that those 1930s scifi stories with silver rockets landing vertically on distant planets, which looked so ridiculous in the 80s with the Shuttle being the future, look like that's exactly what the future will be.
Seeing the 2 boosters landing side by side after the first Heavy launch was something out of a scifi film.
Aren't we at the point that you could have lots of bots ruining sites like this by posting things like that? It seems doable to do even today. I really enjoy technical/cultural/watercooler conversations with a like group of technical scientific people. hope it doesn't go away.
Totally worth it. That was one the best executived live streams of all time:
https://www.youtube.com/watch?v=wbSwFU6tY1c
How is that a dichotomy?
The proposed space shuttle successor was designed around carbon fiber fuel tanks / hull and engineering problems with them eventually killed the project.
The material struggles with low temperatures of cryogenic fuels (brittle), high temperatures of re-entry (just burns), oxidation in oxygen tanks. Also they apparently tend to sponatiously combust with rocket fuels.
Some reference: https://www.compositesworld.com/articles/an-update-on-compos...
https://en.wikipedia.org/wiki/Lockheed_Martin_X-33
If I understand correctly after you pass a certain tank radius you basically step into “nobody’s ever done this with CF before”, which is not a good place to be for SpaceX. Starship was on that spectrum.
Elon Musk cares about one over-arching thing: making humanity an interplanetary species. I believe that everything he does is in service to that goal.
In order to make humanity interplanetary, lift costs must be greatly and sustainably reduced.
In order for lift costs to be greatly and sustainably reduced, everything in the lift vehicle except the propellant must be fully reusable. Also, the vehicle needs to be inexpensive, quick to manufacture and reasonably performant.
In order for the orbital stage of a lift vehicle to be re-used, it must re-enter the earth's atmosphere and effectively shed a large fraction of the chemical energy that was used to accelerate it to 7.8km/s.
In short, the 2nd stage must be able to repeatably handle intense re-entry heat.
For performance (and other) reasons, it must also be able to handle the very low temperatures of liquid propellants.
Given all of these constraints, specific types of steel are literally the only choice.
1. Low material cost
2. Low manufacture cost
3. Excellent high temperature performance
4. Excellent low temperature performance
5. Sufficient normal temperature performance
In fact, Musk has said that his initial focus on carbon fiber was a big error.
Re: Falcon: as others have stated, Falcon 9 is effectively a fixed design. It's now a 'cash cow', and there's no reason to substantially change it. In fact, changing it would risk a lot of profitable contracts that are effectively funding Starship/Superheavy development.
With Starship there is no urge to make money with it yet, they can afford to be somewhat inefficient.
The reason is that this is a completely different use case.
The second stage of a F9 is not meant to survive re-entry, so there is no need to do anything interesting in terms of heat shielding. Starhopper is meant to survive re-entry, so the calculation had to be made to have it survive the heat.
It was deemed "better" to rely on steel rather than an ablative heatshield and aluminum because inspection requirements are pretty high.
As for the 1st stage also being steel instead of aluminum like the F9. The 1st stage booster will be much larger than the F9 and so the strength of steel is an advantage since it avoids the issue of the structure not being able to stay upright unpressurized.
This is a simple fraction:
If you increase the denominator but keep your thrust the same then you lose payload capacity.The propellant mass fraction is what actually matters which takes into account the mass of the vehicle.
The fact that the Raptor is more powerful than the Merlin is barely a consideration in the "Build the Starship out of steel discussion". They would have loved to not have to bite into their mass budget, but other engineering concerns triumphed.
This has precedent in early Atlas rockets and the Centaur second stage still used to this day, but that's not what SpaceX had experience with on Falcon 1 (where it would have been even more unreasonably thin) and they probably just leveraged their knowledge to create as derivative a vehicle as possible when making F9.
And some Atlas missiles on display are still being pressurized to this day: http://heroicrelics.org/ussrc/atlas/dsc18272.jpg.html
Like really, there is a compressor next to the rocket that keeps constant pressure inside against the slight leaks. Pretty impressive for 40-50 old rocket left outside exposed to the elements to be still this airtight. :)
The Falcon 9 has between no and minimal high heat requirements.
The 2nd stage does not attempt to survive re-entry.
The 1st stage enters at (if memory serves) around 1.5km/s compared to nearly 8km/s (or beyond for interplanetary returns) that Starship will need to handle.
Also the Falcon 9 didn't have the lofty re-usable and manufacturing goals that Starship has.
I'm not anything like an expert on this matter, so I can't say with certainty, but I think aluminum is a better choice given these slightly different constraints.
It's also possible that steel would have been a better choice for Falcon 9 even give the more limited requirements. Understanding is always an incomplete quantity.
The Falcon family is wildly successful, but it's small potatoes next to SpaceX's vision for the economics of space launch in the near future, so it seems unlikely that they'd go back to fix what isn't broken.
With carbon fiber you get a lot of theoretical nice-to-haves, but you can’t iterate quickly with it. You need specialized tooling for manufacturing plus there is a lot of actual R&D that still needs to be done to manufacture with CF at Starship scale.
SpaceX sees all this effort as a diversion from the goal of actually building a new rocket. Compare with steel where you can just hire a bunch of welders and build the damn thing out in the open.
Falcon will never switch to steel. It makes no sense from a performance POV (Merlins are much less powerful than Raptors so your payload capacity hit is too high) and Falcon is a “dead” platform: only minor incremental updates from now on. Any change to it will trigger recertification with NASA, the USAF and a bunch of clients, which is a no-go. Falcon is SpaceX’s workhorse and money making machine, and it will continue on that role until Starship retires it.
I don't really know anything about welding nor about the forces acting on a rocket, but I remember reading that the effort in welding the Saturn V rocket was a feat in itself. It took a long time to train the workforce to perform the welds and extensive inspection was needed.
Does anyone know if this applies to this rocket too? Has welding progressed (or simply kept the skill alive)?
Edit: To anyone interested. The excellent book, Stages to Saturn, covered this (and everything else) in quite some detail.
And they’ve already seen several failures with their prototypes so even with steel they have learning to do.
It looks like SN5 will be doing the 20km test flight (whereas they probably planned on doing that with SN4 originally but then SN3 was destroyed in the pressure test failure).
As for Starship, IIRC friction stir welding of stainless steel has not really been done, definitely not at scale so SpaceX will have to use more "traditional" welding methods, which have still got a lot of improvements and new techniques as well since the Saturn 5 times, as mentioned in other comments.
And no, F9 won't change.
I don't know why you think that cost saving measures are different from measures that "make sense".
Lets talk about the most difficult part, the second stage, ie Starship. The big problem you have to solve, is to build a vehicle that can go from incredibly cold structure to a very hot one. Because you start very cold, and when coming back, it gets really hot.
SpaceX loves deep cryo, meaning not just going to the point of making that fuel/oxidiser liquid, but really pushing it down as far as possible. This has a number of advantages but also many disadvantages. SpaceX took a long time to perfect the deep cryo technology with Falcon 9 (that was of course Kerolox).
Because the fuel heating up it forced their operations to be very tight, and in the beginning cause many launch delays. It also cause the famous AMOS 6 failure, as a pretty interesting mostly unknown effect was taking place dring these temparatures.
The upside is that you increase desnisty and thus vehicle performance, and it helps keep your turbo pumps cool as well.
SpaceX very much wanted to continue this for Starship architecture, as they think its a huge win you are leaving on the table if you don't do it.
What this means is that Starship is incredibly cold on the pad, really as cold as they can get it. Any structural material, needs to be able to handle this. This by itself is very challanging with carbon fiber even with the best carbon fiber technology perfomrance degrades the deeper you go. It forces you to really build a carbon fiber structure like we have never seen before.
While some steels are terrible brittle under that condition, 301 stainless (and of course the new alloy SpaceX is developing) are the opposite. Their strengh actually increases, and basically the colder the better. So with incredibly cheap steel that you can make pots out of, you can basically match the performance of super expensive carbon fiber.
On the hot side, you have the reentry. Starship is design not just to come in from LEO like the Shuttle, but actually come from interplantary and then use the air to break. So this is a much hotter profile then Shuttle. And as most people know, Shuttle heat shield caused endless problems.
With both carbon fiber and aluminum you really don't want to let them get them to hot (500C is pushing the issue). Any such design basically requires packaging most of the Starship in a very thick Heatshield. Steel however is very tolerant of heat, meaning that even under fairly high tempratures (800C ish), once it cools down, your structure is still as good as before.
When designing and insulating heat shield (rather then an non-reusable ablative one) the question is really how thick, and thus how heavy does it have to be. The amount of heat that can be handled by the structure behind the heat shield, directly impacts how thick the shield needs to be.
So the combination of being almost as good as peak carbon fiber in deep cryo, and being clearly a much superior choice for reentery makes it a good choice. Once you combine that with the operational apspects, meaning both the per-unit material cost and much cheaper manufacturing its a no-brainer at that point. This is why I believe Musk called it 'The best desision in the whole design', it just made everything easier.
Another bonus is that you have access to a far larger group of engineers with experinace, plus an even larger gorup of workers that can work with steel reliably compared with carbon fiber.
Now for the First Stage, there is a good argument to be made that steal probebly can't quite reach the peak performance level of a carbon fiber design. Maybe not of an aluminum design. In this case I think its really more about material and manufacturing price and of communality with the Starship.
Metals can be used in supersonic and cryogenic aerospace applications.
Composites and carbon-fiber cannot, regardless of cost.