SpaceX Super Rocket?

[metaphor]

We can't just scale up smaller boosters to larger ones because of the square-cube law. Smaller boosters will always be more efficient than larger ones.

Aren't larger rockets more efficient because of the square-cube law? If volume is proportional to mass, mass increases as length^3, while surface area increases only as length^2. Since the structure of a ship is proportional to its surface area (a fuel tank holding 1000 tons of fuel requires only 100 times more surface area than a fuel tank holding 1 ton of fuel), and aerodynamic losses are also proportional to surface area (so a big rocket has a higher terminal velocity than a small rocket that's exactly scaled), larger rockets are more efficient.

I don't understand how you're using the square-cube law to support the idea that smaller boosters are more efficient than larger ones.

[Nibb31]

Metaphor is right. It's more efficient to launch a single SLS with a 130 ton payload than 10 Falcon 9s with a 13 ton payload.

However, there are practical limits to building bigger rockets. You need bigger factories to build the bigger stages, you need a bigger VAB to assemble the rocket, and you need logistics to transport all those bigger parts around. So as you get bigger, there is a point where you start losing economical efficiency.

For the US, anything bigger than SLS would require extension work on the VAB and at Michoud, new crawlers, larger capacity propellant facilities, and bigger barges to ferry the stages from Michoud to the Cape. Basically, you would need to redesign and rebuild the entire Cape Canaveral infrastructure and several factories around the country, which would cost billions more than just designing the SLS.

For Russia, there are no waterways to ferry rocket stages to the launch sites, so they are limited to railroad capacity for large stages. This is why the N1 was basically built on-site in sheds at Baikonour in very rudimentary conditions, which was the main cause of the failure of the program.

Designing rockets isn't just about assembling tanks and engines. It's also about railroads, ships, cranes, roads, industrial facilities, chemical handling and some very very large buildings.

Edited February 17, 2014 by Nibb31

[Scotius]

Which essentially boils down to the amount of money you can throw at the problem until it goes away. Too bad there is no way to use nuclear energy to power up crafts launching from the surface.

[Albert VDS]

There's one thing people are forgetting to consider when talking about sending the ISS up with one huge rocket: how it's going to be deployed.

Most of it was designed around the Space Shuttle. Al tough it had a large cargo bay, it still limited the total design of a module.

If you had one super rocket, you would "just" make it deployable or as one jigsaw puzzle.

[J.Random]

Aren't larger rockets more efficient because of the square-cube law?

Tsiolkovski's equation. To get the same dV for larger final mass, your initial mass should rise exponentially. Two 10-ton launches will use less fuel than one 20-ton launch. Also, they will need less powerful engines and probably even lesser amount of parts and assemblies, which means fewer points of failure.

On the Square-Cube law, vessel's mass and volume will rise as ^3, but connections between stages and space to mount engines will increase as ^2. At some point rocket will either crush under its own weight or you won't be able to mount enough engines to lift it even with side-mounted boosters.

Edited February 17, 2014 by J.Random

[Red Iron Crown]

I'm sorry, but with respect, do you honestly believe Shuttle was an economic way of putting a 500 tone station into LEO piece by piece ?

I'm pretty sure we could have used the $100 billion it cost to have developed a genuine super heavy lift booster and a non modular station and had an awful lot of spare change left over

Using a man rated launch system to simply put hardware into LEO is economically crazy

We need to compare apples to apples here.

Yes, STS was inefficient and expensive. But a version that could lift a 500t station in one launch would be even more expensive and inefficient.

If the hypothetical 500t disposable lifter existed, it would be less efficient than 5 100t disposable lifters or 10 50t disposable lifters.

I agree with you about using man-rated systems for bulk lifting, the only advantage I see is there was crew there to oversee the linking of modules and testing. I suppose it would've been better to use Soyuz to get that crew there separate from the module.

[Red Iron Crown]

Aren't larger rockets more efficient because of the square-cube law? If volume is proportional to mass, mass increases as length^3, while surface area increases only as length^2. Since the structure of a ship is proportional to its surface area (a fuel tank holding 1000 tons of fuel requires only 100 times more surface area than a fuel tank holding 1 ton of fuel), and aerodynamic losses are also proportional to surface area (so a big rocket has a higher terminal velocity than a small rocket that's exactly scaled), larger rockets are more efficient.

I don't understand how you're using the square-cube law to support the idea that smaller boosters are more efficient than larger ones.

That 1000 ton fuel tank would crumple like it was made of tissue paper. The structure that was adequate for the one ton tank would be entirely too weak for a 1000 ton one, because of the square-cube law. We would need to rework the structure to support the far greater mass, and we would be lucky if it scaled linearly with mass. Eventually we reach the limits of our materials, setting an upper limit on lifter size. It's a bit counter intuitive, sort of a reversal of economies of scale.

A good example is quadcopters. You can make a small electric copter that is efficient, controllable and has a decent range. Why can't we just scale it up to man-sized? Shouldn't a larger copter be more efficient?

[maccollo]

Tsiolkovski's equation. To get the same dV for larger final mass, your initial mass should rise exponentially.

No.

ÃŽâ€V = Ve * ln (m0/m1)

The equation doesn't change if your mass unit is ton, or kg, or grams, or solar masses.

To get the same deltaV for a larger final mass the initial increases proportionally to the final mass.

Same mass fraction, same delta V.

It's increasing the deltaV without increasing the ISP that makes the initial mass run amok.

Edited February 17, 2014 by maccollo

[J.Random]

No.

ÃŽâ€V = Ve * ln (m0/m1)

The equation doesn't change if your mass unit is ton, or kg, or grams, or solar masses.

To get the same deltaV for a larger final mass the initial increases proportionally to the final mass.

Same mass fraction, same delta V.

Crap. My fault.

Still, the problem of vessel's structural integrity and reliability remains.

[Simon Ross]

Crap. My fault.

Still, the problem of vessel's structural integrity and reliability remains.

So I'm kinda interested

What do you believe is the upper limit in terms of tonnage that can be raised to LEO ?

[SargeRho]

10000 tons, thereabout. The problem is if the "E" part in LEO will still be habitable afterwards. Orion drives aren't exactly clean.

[maccollo]

So I'm kinda interested

What do you believe is the upper limit in terms of tonnage that can be raised to LEO ?

Well the Saturn V supposedly had a payload to LEO of 120 tons.

This is over 4% of the initial mass, which is remarkable when compared to other launch vehicles, most of which only have payload fractions of 2-3% at most. In this 4% the mass of the third stage tank and engine is not included.It's just the fuel remaining in the third stage, and the TLI payload.

You can check these numbers yourself here.

http://history.nasa.gov/SP-4029/Apollo_18-19_Ground_Ignition_Weights.htm

http://history.nasa.gov/SP-4029/Apollo_18-23b_Launch_Vehicle_Propellant_Use.htm

So given the fact that the Saturn 5 has like the highest payload fraction ever I don't think that it is at the limit.

[SFJackBauer]

Metaphor is right. It's more efficient to launch a single SLS with a 130 ton payload than 10 Falcon 9s with a 13 ton payload.

However, there are practical limits to building bigger rockets. You need bigger factories to build the bigger stages, you need a bigger VAB to assemble the rocket, and you need logistics to transport all those bigger parts around. So as you get bigger, there is a point where you start losing economical efficiency.

For the US, anything bigger than SLS would require extension work on the VAB and at Michoud, new crawlers, larger capacity propellant facilities, and bigger barges to ferry the stages from Michoud to the Cape. Basically, you would need to redesign and rebuild the entire Cape Canaveral infrastructure and several factories around the country, which would cost billions more than just designing the SLS.

For Russia, there are no waterways to ferry rocket stages to the launch sites, so they are limited to railroad capacity for large stages. This is why the N1 was basically built on-site in sheds at Baikonour in very rudimentary conditions, which was the main cause of the failure of the program.

Designing rockets isn't just about assembling tanks and engines. It's also about railroads, ships, cranes, roads, industrial facilities, chemical handling and some very very large buildings.

Right on several accounts. But, while in your post you treated only the lifter part of the spacecraft, I'm also interested about the most important part in a rocket - the payload.

My question is a simple one - what is the most efficient way to design, build and deploy a 500 ton structure which is comprised of several layers of infrastructure - electrical, environment control, life system support, pressurization, electronics, attitude control, energy generation (solar panels), science instrumentation, habitation and the list goes on. Is it:

1) Build it as a monolith, a single bloc, and deploy it in one go;

2) Build it as a several modules, test them in the ground before launching, test them after assembling in orbit before declaring worth of being used, re-assemble in orbit if the need to expand, to fix defects, or to upgrade obsolete hardware arises.

Because you can't just say 1 without pondering the engineering consequences of doing so.

[Deta]

Using a man rated launch system to simply put hardware into LEO is economically crazy

Is it not all just MASS? hardware or flesh.

[SargeRho]

No. Man-rated launch systems are significantly more expensive than non-man rated launchers.

[crazyewok]

10000 tons, thereabout. The problem is if the "E" part in LEO will still be habitable afterwards. Orion drives aren't exactly clean.

Meh

It was calculated at causing 12 deaths per launch, 12 in 7 billion? We deal with worse odds for smaller again. And that was before there reworked the nukes think they got it down to just less than one per launch.

Eitherway dont launch over a city and you be fine,

[Roastduck]

So, crazyewok, knowing that every time it launched, 12 people would certainly die, would you be willing to be the one giving the launch command?

And you say "Don't launch over a city" when in reality it should be "don't launch where a city could be in range of the fallout" which pretty much eliminates anywhere near a powerful air current like the jet stream.

[metaphor]

That 1000 ton fuel tank would crumple like it was made of tissue paper. The structure that was adequate for the one ton tank would be entirely too weak for a 1000 ton one, because of the square-cube law. We would need to rework the structure to support the far greater mass, and we would be lucky if it scaled linearly with mass. Eventually we reach the limits of our materials, setting an upper limit on lifter size. It's a bit counter intuitive, sort of a reversal of economies of scale.

A good example is quadcopters. You can make a small electric copter that is efficient, controllable and has a decent range. Why can't we just scale it up to man-sized? Shouldn't a larger copter be more efficient?

We're nowhere near the structural limit of materials in terms of rocket size. We could easily make something that gets 500 tons into orbit with currently available materials.

Quadcopters are not a good example, the advantage of being smaller has no relation to how much mass they can carry. Larger helicopters are in fact more mass efficient than smaller ones.

But, while in your post you treated only the lifter part of the spacecraft, I'm also interested about the most important part in a rocket - the payload.

It is always a lot cheaper to assemble things on the ground than in orbit. Orbital assembly is expensive. Most of the cost of the ISS was not the modules themselves, but the lifters and the operations needed to assemble them.

[Red Iron Crown]

We're nowhere near the structural limit of materials in terms of rocket size. We could easily make something that gets 500 tons into orbit with currently available materials.

So where are these heavy lifters? Why has no space agency built anything as large as the 50-year-old Saturn V if it's so much more efficient?

Quadcopters are not a good example, the advantage of being smaller has no relation to how much mass they can carry. Larger helicopters are in fact more mass efficient than smaller ones.

Citation needed. An RC helicopter made from off the shelf components stomps all over full scale helicopters using custom designed, bleeding edge components, on a performance per unit mass scale. The world is full of examples of structures that work when small but don't scale up because of square-cube. It's the reason giant ants are impossible. It's the reason ornithopters aren't possible for anything other than trivially small sizes. It's the reason there are no large hummingbirds. Why do you think rockets are somehow immune to its effects?

It is always a lot cheaper to assemble things on the ground than in orbit. Orbital assembly is expensive. Most of the cost of the ISS was not the modules themselves, but the lifters and the operations needed to assemble them.

The assembly was done on the ground for ISS modules, in space was just docking and connections between modules for the most part. If it had been launched in one monolithic piece, as much or more in-orbit work would have been needed to configure it from something that was feasible to enclose in a fairing into a usable permanent configuration. Modular station construction is considered an advancement over monolithic stations for reasons of flexibility and removing the need for extremely heavy lifters. All future station plans that I'm aware of use the modular approach.

[Simon Ross]

So where are these heavy lifters? Why has no space agency built anything as large as the 50-year-old Saturn V if it's so much more efficient?

The answer is that there is no mission requirements in the pipeline that require a 500 ton booster. That does not invalidate the fact that we could build them if required.

Citation needed. An RC helicopter made from off the shelf components stomps all over full scale helicopters using custom designed, bleeding edge components, on a performance per unit mass scale. The world is full of examples of structures that work when small but don't scale up because of square-cube. It's the reason giant ants are impossible. It's the reason ornithopters aren't possible for anything other than trivially small sizes. It's the reason there are no large hummingbirds. Why do you think rockets are somehow immune to its effects?

Trying to compare a toy to an aviation certified helicopter really isn't a valid argument. It's about the same level as saying an RC model car stomps all over a real life people carrier

The assembly was done on the ground for ISS modules, in space was just docking and connections between modules for the most part. If it had been launched in one monolithic piece, as much or more in-orbit work would have been needed to configure it from something that was feasible to enclose in a fairing into a usable permanent configuration. Modular station construction is considered an advancement over monolithic stations for reasons of flexibility and removing the need for extremely heavy lifters. All future station plans that I'm aware of use the modular approach.

This really is chicken and egg stuff. Trust me, if we had a genuine EHL capability we certainly wouldn't be designing modular space stations, they would be much more along the lines of the dry stage concept used for Skylab.

With ISS we didn't have any EHL capability, therefore the only way to get it into orbit was as modules.

Edited February 18, 2014 by Simon Ross

[asmi]

The problem with building ISS in one launch is that all it's components would need to able to withstand whatever is above them plus launch loads (read - ~5 times it's own weight plus whatever is mounted on top of it). ISS is not able to support itself if it would be assembled on the ground. So in this case you'd end up overbuilding components mechanically just for the sake of launch, and that extra mass would stay there forever.

Another reason is that there is a limit on how wide can your fairings be, right now the biggest one is 5 meters, so it makes even more sense to assemble station on orbit from relatively small pieces.

Lastly, reliability. If one launch of super-heavy fails, it's devastating for the program since the price of the payload is huge, while loss of one component can be overcome more easily, and smaller LVs tend to be more reliable because they can have more test-launches within set budget, while it's not uncommon for heavies to have operational maiden flight (Saturn-V, Energia).

[Red Iron Crown]

I think it's the opposite way around, it was built in modules so that EHL wouldn't be needed.

[Simon Ross]

The problem with building ISS in one launch is that all it's components would need to able to withstand whatever is above them plus launch loads (read - ~5 times it's own weight plus whatever is mounted on top of it). ISS is not able to support itself if it would be assembled on the ground. So in this case you'd end up overbuilding components mechanically just for the sake of launch, and that extra mass would stay there forever.

Another reason is that there is a limit on how wide can your fairings be, right now the biggest one is 5 meters, so it makes even more sense to assemble station on orbit from relatively small pieces.

Lastly, reliability. If one launch of super-heavy fails, it's devastating for the program since the price of the payload is huge, while loss of one component can be overcome more easily, and smaller LVs tend to be more reliable because they can have more test-launches within set budget, while it's not uncommon for heavies to have operational maiden flight (Saturn-V, Energia).

And again, a one launch monobloc station would look nothing like the ISS, it would be designed in a completely different way, but please go do your research on the Skylab station to see what was possible with a SINGLE Saturn V launch nearly 40 years ago ! Even today, it's pretty damn impressive

Interesting fact, after dozens of launches to build it, ISS has around 15,000 cubic feet of habitable volume, Skylab had 11,290 cubic feet of habitable volume from a single launch !

J

[SargeRho]

No, it was built in modules because EHL wasn't available.

[Red Iron Crown]

ISS has 29,600 cubic feet of pressurized volume to Skylab's 11,290. More impressively, it masses 450,000kg to Skylab's 77,000kg, which is the really relevant metric.