It turns out orbital launchers scale rather well with size.
The rocket is already 90+% propellant (fuel & oxydizer), with the few percent left being the engines, structure and payload. So if you want the rocket to have a meaningful payload, it needs to be big.
For example the Falcon 9 weights 350 tons and can place 22 tons to low earth orbit (LEO). So about 6% of initial launcher weight ends up in orbit.
If you want to place the 100+ tons estimated for a reasonable Mars trip, your launcher will be either huge or you will need to do a lot of launches and assemble your ship in orbit with all the overhed that requires.
Starship also aims to be fully reusable, which will eat to your payload yet again but provides a huge benefit of not throwing away any part of your rocket, just burning a lot of propellant, which is dirt cheap compared to rocket hardware (estimates say about 5% of conventional rocket launch is fuel, the less is hardware that you normally simply crash into the ocean after use).
So in short - big payload -> big rocket. Reusable rocket with big payload -> humungous rocket that is really really cheap to operate.
Efficiencies of scale. Too small and you can't pack enough food for the trip or proper redundancies, still too small and you can't build the base fast enough, don't bring enough stuff on the trip and you need to make more trips..which are time limited due to the rendezvous. You end up inclined to 'as big as possible'
Basically to drive down the cost of payload mass as much as possible. First you have to build a lot of rocket to get anything up into space. If it only launches 1kg of payload, that's going to be a very expensive kilogram. As you continue to scale the rocket up, assuming you can design for a decent payload ratio, the amount of cargo capacity increases and the cost per kilogram drops.
People need to live in it. A 1 Kg probe can take some pictures, the tiny Apollo Lunar Module was 16,400Kg and only designed to support 2 people for 75 hours. Meanwhile Mars (3.72 m/s²) has significantly more gravity than the Moon (1.62 m/s²), making a trip back from the surface even more difficult.
Launch vehicles tend to scale payload capacity at a greater rate than dry mass. So a small launch vehicle like RocketLab's Electron might have a wet mass of 12,000kg at launch and an LEO payload capacity of 225kg (225/12000 ~ 1.8% payload), the SpaceX Falcon9 has a wet mass of 550,000kg and an LEO payload capacity of 22,800kg (22800/550000 ~ 4.1% payload).
This just tends to mean that, very generally and with exceptions, larger vehicles can get mass into space with less fuel.
A Falcon 9 has a LEO payload capacity of 22,800kg in expendable mode, something SpaceX does not like to do. The largest payload mass they've carried to LEO was Starlink-2 at approximately 15,600kg (60 satellites at 227 kg each).
The rocket equation [0] disagrees: "In what has been called 'the tyranny of the rocket equation', there is a limit to the amount of payload that the rocket can carry, as higher amounts of propellant increment the overall weight, and thus also increase the fuel consumption"
The pure, theoretical rocket equation still scales linearly, the tyranny lies in the hard limit of how much mass you can theoretically accelerate out of a gravity well per unit off fuel.
But bigger rockets can get closer to that theoretical optimum because some dead weight components remain fixed-size and because some masses scale with surface instead of volume. Rockets can scale better than linearly with size, but they will never exceed the rocket equation.
I'm having a mental blockage on this, as you can always launch 10 rockets at the same time and get 10 times the payload. Having those 10 rockets strapped together should not change the payload of them flying independently.
What am I missing here -- I'm sure I'm overlooking something.
If you strap 10 rockets together, you don't need to keep all 10 copies of some components. You don't need 10 computers, or ten emergency abort systems, or ten communications systems, etc. One large tank will have less mass/volume than 10 small ones, etc.
Interestingly there are couple rockets that are close to "strapping couple rockets together" for various reasons:
The R7 rocket
- mostly to avoid engine ignition in flight & due to having a lot of nozzles you need to fit on a rocket
- this was then kept for Soyuz even after an additional stage has been added that is started in flight
- Soyuz 2V uses a more efficient engine & uses only the central stage
The proton Rocket
- the central tank has the maximum diameter you can ship via rail from the factory to Baykonur
- by mounting smaller tanks around it that are shipped separately you can avoid building an overly long rocket, that could be fragile and unstable
Saturn 1/1B
- basically a stop gap using existing Jupiter and Redstone rocket tankage tooling
- strap 8 Redstone tanks around 1 Jupiter tank and you get the S-I first stage
Delta IV Heavy/Falcon Heavy
- you have a rocket that can launch by itself with smaller payload or by strapping 3 first stages together can launch a bigger payload
- better economies of scale as you can doe more with a single rocket design instead of maintaining 2 separate one (big and small)
- in Falcon Heavy case you can also save all the first stages from more demanding trajectories where you would otherwise have to expend the regular F9 first stage
OTRAG
- make dirt cheap and as simple as possible standardized "rocket tubes"
- strap a lot of them together
- fire and jettison in the right order to achieve orbit (check your staging! ;-) )
While this does work, the Falcon Heavy has convinced Elon not to go into that direction again. Its actually not that simple to strap these together and just making a bigger tank is actually easier.
Poorly for the current market using a FH makes sense, but to break to another level, building one big rocket is better. Both the US Saturn 5 and the Soviet N1 were big single tube vehicles.
Correct. You can (super¹)linearly increase launch mass by adding fuel. The rocket equation hits when you're trying to go farther, rather than more massive.
¹ Superlinear because tank mass scales with a lower exponent than tank volume.
You aren't missing anything, the comment you are responding to is confusing the per unit of fuel payload limit to be the total limit of the rocket. You can add more engines/ use bigger ones which would allow for more fuel and thus more payload( in theory at least).
That's correct so far. Now think further: a single, larger rocket has the advantage of leaving away the tank walls that would be in the inside now. Less mass needed for the tank walls (the outer walls need to become stronger though), more available for payload, extra fuel and oxidizer or extra stuff like things needed for reusability.
Why does it need to be super-large? To simulate gravity?