[lutze]

I don't understand the need for the intermediary stages, couldn't the main rocket just deliver the payload and then land?

As it is in that video, it's having to carry engines and fuel for 3 separate stages, for no apparent reason.

[InclinedPlane]

It would be more convenient and cost effective to do so, if we could. But we lack the technology.

The rocket equation is a harsh mistress, it demands exponential amounts of fuel the faster you want to accelerate a given stage (specifically, as a ratio to the exhaust velocity of the rocket). Given that orbital velocity is quite high (about 8.5 km/s) relative to the exhaust velocity of the best chemical propellants (about 3 km/s for LOX/Kerosene) this results in impractical mass fractions to contend with (17:1 to get to orbit, and that's with no payload). But you can cheat. If you use staging and drop away the dead weight of empty fuel tanks and no longer needed engines from lower stages then you can make an end run around the rocket equation. You make a rocket that can accelerate a payload up to a certain velocity, then you make an even bigger rocket which can deliver the entire other rocket as a payload to a different velocity, and so on, until the sum total of all the velocities is the necessary total speed you require to get to orbit.

We're actually fairly close to being able to make single-stage-to-orbit (or SSTO) launchers workable, but it's a difficult problem. We can just about make a single stage with a high enough mass fraction to do it, but then there is almost no payload remaining. And the only way to make a vehicle with such a tiny payload cost effective would be to make it reusable, but enabling reusability would add additional weight which would destroy any payload whatsoever and probably prevent it from even reaching orbit, catch-22. Potentially we could use advanced engines, rocket fuels, and lightweight materials (like carbon fiber) to build a reusable SSTO which would have a reasonable payload, but such designs are hugely untested and very risky. So for now the best hope for reusability seems to be to incrementally advance the design of existing multi-stage rockets.

[elteto]

> The rocket equation is a harsh mistress

Love this, totally making it mine ;) Also, there is the problem of efficiency loss due to over/underexpansion as you switch from atmospheric to vacuum conditions, and I have never heard of a rocket engine with variable nozzle geometry to compensate for it.

[schiffern]

That was the promise of aerospike engines, but they have their own problems.

https://en.wikipedia.org/wiki/Aerospike_engine

[gus_massa]

I'd really love to see a spreadsheet with all this calculations. (Or better, an online html5 virtual rocket calculator tm.) With some default data (usual payload weight, fuel price, orbit speed, ...) and you can choose how many stages and haw much fuel each stage has and the spreadsheet calculate the total price of the mission.

You have a lot of information, but I understand that putting all of this in a xls would be a lot of work.

[numbsafari]

What you need is a transporter that can teleport you from a space-bound craft to a surface location and back without all of this rocket launching business to deal with.

[TheSOB88]

That would be quite convenient... A little too convenient?

[Florin_Andrei]

> Potentially we could use advanced engines

Nuclear rockets would be so awesome if they weren't so nasty.

[rst]

The video actually shows two stages and the Dragon capsule, which is payload. As to the use of staging, there are several reasons:

1) During the late stages of the boost, the one second stage engine is pushing only its single engine, fuel and tankage. You're no longer dragging around the nine engines of the first stage and their tanks. That weight saving means you get a lot more delta-V for each unit of expended fuel.

2) The engines themselves are also different. Rocket nozzles designed for optimal performance at sea level aren't optimal for high-altitude or vacuum conditions; those optimal for vacuum won't work at sea level (their exit pressure is so low that the exhaust has trouble pushing air out of the way). And compromise designs aren't optimal in either environment.

3) In the proposed SpaceX reuse architecture, they don't need to protect that long, thin first-stage tank from re-entry at orbital velocity. It's not clear how they could.

[lutze]

Good answer, thanks.

I've been playing too much Kerbal Space Program, I assumed the first stage had already reached orbital velocity.

[mikeash]

KSP is great fun and extremely educational, but it makes things way easier than they are in real life, presumably for the sake of fun.

One way this is done is by making Kerbin a lot smaller than Earth. Specifically, over ten times smaller. This reduces the speed needed for low Kerbin orbit to around 2000m/s, while low Earth orbit needs around 8000m/s.

A factor of four in speed probably already sounds bad, but it's much worse than one might naively expect. The amount of fuel needed to accelerate a given payload to a particular speed grows exponentially with the target speed. A typical rocket engine exhaust velocity, both in KSP and real life, might be 4000m/s. When the target velocity is half the exhaust velocity, as is the roughly the case with Kerbin, then you need about 40% of your rocket's mass to be fuel. In other words, you can orbit about 60% of your total mass. When the target velocity is double the exhaust velocity, as is roughly the case with Earth, you need 86% of the initial mass to be fuel, so you can only orbit about 14% of the rocket.

In other words, if you're putting 1000kg into orbit, then in KSP you need about 700kg of fuel, while on Earth you need over 6000kg of fuel. For a single stage rocket, that 1000kg that you're putting into orbit includes fuel tanks and engines. You need much bigger engines and fuel tanks to lift 7000kg than you do to lift 1700kg, so a lot of that 1000kg you get into orbit is going to be empty fuel tanks and spent engines, not actually useful stuff.

Staging lets you work around this problem by letting you reduce the amount of useless junk you put into orbit. By dropping large amounts of empty fuel tank and spent engine early on, you no longer have to lift it all the way up, and you have more orbited mass left over for actually useful widgets.

A single stage that goes from ground to orbit isn't too hard in KSP. On Earth, it's right at the limits of technology. Nobody has operated a rocket as a single stage to orbit, but a couple of pieces of larger rockets are theoretically capable of reaching orbit from the ground on their own if flown by themselves. It's possible, but the amount of useful payload that such a thing can put into orbit is so small that it's not cost effective.

[gamegoblin]

"A multistage (or multi-stage) rocket is a rocket that uses two or more stages, each of which contains its own engines and propellant. A tandem or serial stage is mounted on top of another stage; a parallel stage is attached alongside another stage. The result is effectively two or more rockets stacked on top of or attached next to each other. Taken together these are sometimes called a launch vehicle. Two stage rockets are quite common, but rockets with as many as five separate stages have been successfully launched. By jettisoning stages when they run out of propellant, the mass of the remaining rocket is decreased. This staging allows the thrust of the remaining stages to more easily accelerate the rocket to its final speed and height."

https://en.wikipedia.org/wiki/Multistage_rocket

[lutze]

Right, excellent.

Except the first stage doesn't run out of propellant does it, because it has to LAND again. Hence my puzzlement.

I know why multi-stage rockets are used... I'm asking why this rocket needs them, since they solve a problem it doesn't have by design.

[jessriedel]

Rockets don't have stages because they run out of propellant, they have stages so that you don't spend 99% percent of your fuel pushing engines and tanks into orbit which you don't need once you get there.

[jlgreco]

> "they have stages so that you don't spend 99% percent of your fuel pushing engines and tanks into orbit"

In fact you can kinda make a rocket that can reach orbit that doesn't have "stages" but rather jettisons engines themselves. Early Atlas rockets did this: they had one set of fuel tanks but two engines. About two minutes into the flight they would jettison one of the engines.

[nawitus]

Are you saying it should be single-stage-to-orbit? That's currently technologically nearly impossible, even if the rocket doesn't need to land back to Earth.

[nanofortnight]

Not as technologically impossible as you might think: http://www.reactionengines.co.uk/space_skylon.html

[tomelders]

I really don't understand why Skylon isn't getting more coverage.

[schiffern]

>they solve a problem it doesn't have by design.

This is where you're incorrect. Not having to accelerate the empty first stage to orbital speed (and back!) is a huge fuel savings.

[elteto]

A single stage rocket is desirable from a technological point of view since it sheds a lot of complexity: you don't have to deal with multiple engines, separation mechanisms, structural considerations, etc. However, multi-stage rockets can deliver more real payload to orbit, since they maximize the amount of fuel that is spent on taking the actual payload up there.

Single-stage rockets need to carry a lot of dead mass (empty tanks and engines) all the time, and that is a lot of wasted fuel. A multi-stage rocket can dispose the big first stage engines once it's cleared out most of the Earth's gravitational pull and use smaller engines to continue.

Another important consideration is that rocket engines don't run optimally during the whole burn. The first stages are optimized for atmospheric conditions, whereas later stages are optimized for vacuum conditions. Therefore having one big engine propel you up all the way incurs in an even grater loss of fuel due to the inefficiency at high altitudes. You could probably have a rocket engine capable of having a variable geometry to compensate for this but AFAIK is almost impossible to do it.

See the following for more details:

[1] http://en.wikipedia.org/wiki/Staging_(rocketry)#Advantages [2] http://en.wikipedia.org/wiki/Rocket_engine_nozzle

[schiffern]

>once it's cleared out most of the Earth's gravitational pull

There's about 90% as much gravity at the altitude of the ISS.

It's the atmosphere (and associated gravity drag), not the 1/r^2 scaling, that does it.

[elteto]

True! Really common misconception and yet I fell for it, thanks.