top | item 8537390

The Tyranny of the Rocket Equation (2012)

214 points| bane | 11 years ago |nasa.gov | reply

157 comments

order
[+] Animats|11 years ago|reply
Right. This has been known for a long time. It's why rockets aren't much better than they were 40 years ago. Chemical fuels are as good as they can get. Space travel with chemical fuels is just barely feasible.

In the 1960s, it was assumed that nuclear power would be necessary for space flight. Everybody involved knew the rocket equation. The original plan for Apollo included a nuclear upper stage. The engine (NERVA) was built and tested. A Nuclear Assembly Building at Canaveral was planned. But, because the goal was so narrow ("man, moon, decade"), the solution chosen was a disposable rocket the size of a 50-story building to send an RV-sized payload to the moon.

The crash of a nuclear rocket would produce a radioactive mess. Not Chernobyl or Fukishima sized, but at least small-town sized. Launching from an isolated island would help.

Various schemes have been tried or proposed to beat the rocket equation. Launching from a balloon was tried early. Launching from an aircraft is still used by Virgin Aerospace. It helps a little.

A space vehicle that's an air-breather while it's in the atmosphere and transitions to rocket mode once out has been proposed many times, but making something that's both a rocket and an airplane is hard and adds a lot of weight. As an airplane, it has to go hypersonic to get up enough speed that it's worth doing this. Building a hypersonic aircraft is very hard; so far, only a few small demo craft have done it. The National Space Plane (hypersonic single-stage-to-orbit) was proposed in the 1980s. Ben Rich, head of the Lockheed Skunk Works and the designer of the SR-71's propulsion system, declined to let Lockheed bid on it. (His comment: "We used titanium (on the SR-71). You know anything stronger?") Remember, it has to be strong at a few thousand degrees.

The same problems apply to launch track systems. Going hypersonic near the ground is possible; the Holloman AFB test track, 50,000 feet of very straight railroad track, has been used to reach Mach 8.6. The required acceleration is about 14g. Far too much for humans.

The "space elevator" requires not only unreasonable strong materials but the ability to put so much mass in space that you wouldn't need a space elevator if you had that kind of launch capacity.

[+] DennisP|11 years ago|reply
NASA did a study on space elevators about a decade ago, and found that it would require carbon nanotubes several centimeters long, bound together by a realistically strong epoxy. Launching their design would require seven space-shuttle flights to deploy a minimal elevator, which you use to pull up additional construction material.

They addressed a lot of other practical issues too. Here's their final report (pdf), it's an interesting read.

http://www.niac.usra.edu/files/studies/final_report/521Edwar...

[+] mikeash|11 years ago|reply
Everyone Knows™ that putting stuff into space is expensive. Then Everyone Assumes™ that it's because of all the fuel. But no, fuel is cheap, hardware is what's expensive. If you look at the costs involved in putting something into orbit, the cost of fuel is a trivial detail, on the order of 1% of the total costs. Compare that to an airliner, where fuel is around 1/3rd of the total costs, or a car, where fuel can easily be over 50% of the total costs.

I happened to be reading up on rocket efficiency and I was surprised to learn that rockets are fairly efficient for launching stuff into orbit. Wikipedia uses the example of the Space Shuttle, where 16% of the energy in the propellants ends up in the kinetic and potential energy of the orbiter. That's pretty good!

The problem is that you use a rocket once and then throw it away. Imagine if your car was one-time-use. How often would you drive somewhere? How often would anyone drive anywhere? It wouldn't matter how efficient they are and it wouldn't matter if they didn't even require fuel at all.

Now, the rocket equation still comes into play here, because it means you need a lot of rocket for a little bit of payload. But the main problem is the one-time-use thing. A nuclear disposable rocket wouldn't improve things much. A reusable nuclear rocket would be great, but then so would a reusable chemical rocket.

This is the genius of SpaceX. For decades, rocket designers have looked at the rocket equation and tried their hardest to save fuel. SpaceX looked at the economics of rocketry and realized that fuel costs more or less don't matter, and instead concentrated on building their machines cheaply, and on making them reusable. We'll see how it works out, but if they succeed in making reusable rockets then they'll cut the cost of launches by an order of magnitude or more.

[+] weavejester|11 years ago|reply
What about momentum exchange tethers, AKA skyhooks? The smaller ones require significantly less mass than a full space elevator, but mean that rockets need carry less fuel, because they're taking from the momentum of the tether. The lost momentum of the tether can then be replenished over time, using either efficient engines like ion drives, or if the tether is electrodynamic, using the Earth's magnetic field.
[+] EliRivers|11 years ago|reply
"A space vehicle that's an air-breather while it's in the atmosphere and transitions to rocket mode once out has been proposed many times, but making something that's both a rocket and an airplane is hard and adds a lot of weight."

For those interested, these people are working on it. http://www.reactionengines.co.uk/ (They're the Skylon makers mentioned in another comment here)

[+] ericd|11 years ago|reply
I wouldn't dismiss the space elevator out of hand. It requires carbon nanotubes of a few meters length to achieve the required strength, and you wouldn't need to lift it pre-built - you could build it with a guideline and cable laying cars traveling up and down, adding to the cable, much like they do with suspension bridges. Long term, it seems like far and away the best approach if we can solve the materials science problem.

EDIT: I found this book to be a pretty great primer on the subject: http://www.amazon.com/Space-Elevator-Earth-Space-Transportat...

[+] bcoates|11 years ago|reply
Whatever happened to transmitted power designs, like using a ground-based laser to lift a payload? If you can leave the powerplant on the ground and only send the power, you no longer have to lift the fuel, just reaction mass.
[+] sehugg|11 years ago|reply
NERVA was fine if you didn't start it up until you were in LEO and took care to avoid any trajectories that might involve Earth reentry. But even for manned flight outside the atmosphere, shielding was an issue and servicing it was pretty much a suicide mission. Not insurmountable problems, but more than we needed for our modest goals at the time which didn't involve going beyond the Moon.

http://www.wired.com/2012/09/nuclear-flight-system-definitio...

[+] snowwrestler|11 years ago|reply
I wonder about a fusion rocket drive, like Larry Niven wrote about (not the Bussard, the reaction drives).

There are two big problem in fusion energy research: plasma leaks, and high-energy neutrons. It seems to me that a fusion rocket answers both questions: just throw it all out the back. And hydrogen is abundant and cheap.

[+] knz42|11 years ago|reply
About your last point: if you have a space elevator you can have a mass going down to balance for the mass going up, which reduces the energy needed merely to working against the friction on the cable and the connection point at the top. That's cheap.
[+] FranOntanaya|11 years ago|reply
> Space travel with chemical fuels is just barely feasible.

To certain extent, that only applies while assuming rocket launches are hard to iterate. Otherwise it would be a matter of launching enough of them.

[+] lordnacho|11 years ago|reply
"If our planet was 50% larger in diameter, we would not be able to venture into space, at least using rockets for transport."

I thought this was quite though-provoking from a Drake Equation standpoint. From what I've gleaned, the earth-like planets we know of seem to be a big bigger than Earth. Perhaps if there's civilization out there, they are hampered by the misfortune of being on a planet that's practically impossible to escape from. If they find it hard to put up a Hubble Telescope, perhaps they'll just not be as likely to bother.

My other remark is to the engineering. There's a quip that you have to be an engineer to make something that only just satisfies the requirements. Plenty of people build houses without much in the way of calculation. Even cars can be built by enthusiasts without degrees. This is what makes the space stuff such amazing engineering.

[+] Sharlin|11 years ago|reply
> From what I've gleaned, the earth-like planets we know of seem to be a big bigger than Earth.

This is entirely a selection effect. Big planets are easier to find, that's why we have mostly found big planets.

[+] chton|11 years ago|reply
We're still very early in the development of spaceflight. If you compare with the car industry, we're at the early industrialized phase, where large companies carefully start to develop the necessary technology. Now, 100 years later, a lot is standardized and the knowledge is so common that everybody has the basics, and people can build cars in their back yards. It's the same development that happened with airplanes.

At its core, a rocket isn't more complicated than a car. The challenges are just different. My dearest hope is that eventually rocket components will be as commoditized as car parts, so people can build and maintain their own spacecraft. I want to see rockets held together with duck tape and spit, because that's the point where spaceflight is available to everybody and gravity stops becoming a hurdle.

A quick disclaimer: I have the greatest respect for rocket engineers. They are taking the first steps, the most difficult ones, and I don't believe for a second that they their work is easy. I just believe that eventually, they'll become obsolete to the majority of spacetravel :)

[+] kabdib|11 years ago|reply
You could still do it with nuclear propulsion. Orion, if you absolutely have to go into space today and don't mind riding a string of atomic explosions.
[+] throwawayaway|11 years ago|reply
degrees are overrated. does john carmack have a degree?
[+] edraferi|11 years ago|reply
The common soda can, a marvel of mass production, is 94% soda and 6% can by mass. Compare that to the external tank for the Space Shuttle at 96% propellant and thus, 4% structure. The external tank, big enough inside to hold a barn dance, contains cryogenic fluids at 20 degrees above absolute zero (0 Kelvin), pressurized to 60 pounds per square inch, (for a tank this size, such pressure represents a huge amount of stored energy) and can withstand 3gs while pumping out propellant at 1.5 metric tons per second. The level of engineering knowledge behind such a device in our time is every bit as amazing and cutting-edge as the construction of the pyramids was for their time.

That's awesome, I had no idea.

[+] tempestn|11 years ago|reply
> In the 1970’s, an experimental nuclear thermal rocket engine gave an energy equivalent of 8.3 km/s. This engine used a nuclear reactor as the source of energy and hydrogen as the propellant.

That was intriguing, but didn't go into detail on why a nuclear thermal rocket hasn't been tried since. The obvious explanation is that there could be serious consequences if such a rocket exploded, spreading radioactive material, etc. And rockets tend to explode sometimes. However, it sounds like while that is certainly a concern, it is not as great of one as it might seem:

http://en.wikipedia.org/wiki/Nuclear_thermal_rocket#Risks

And indeed there is still work ongoing on such designs.

[+] anovikov|11 years ago|reply
Nuclear thermal rockets also provide a very poor mass fraction of fuel because their propellant has very low density (liquid hydrogen) and the reactors are heavy. So their advantage is much small than it seems. And low thrust makes things even worse, they can be used only for upper stages/space tugs. In the end they were never used mainly because cost outweighed benefits. Imagine, nuclear fuel contains about millionx the energy per mass of chemical fuel, and yet exhaust velocity is only 2x higher, and effective combined ISP of the stage (velocity change vs mass fraction assuming dead weight of stage being zero, and with gravity/ballistic losses subtracted) is only about 50% higher than state of the art chemical systems, at vastly higher cost and risks involved. If i was Elon Musk i won't launch a nuclear rocket unless i had a really good liability insurance, and i was an insurer i'd say nah unless federal government backs me, and if i was uncle Sam i'd say nah, too. It simply doesn't worth it.

Also, problems with space access are mostly market size-related problems. Cheap access to space is possible, only requires a vast market to pay back the investments, which is simply not there.

[+] daeken|11 years ago|reply
> And rockets tend to explode sometimes.

Chemical rockets -- the only ones we've ever actually used -- explode because that's what they're intended to do. The only difference between a successful rocket firing and a catastrophic rocket failure is the speed at which the explosion happens. A nuclear rocket engine has basically no risk of explosion; tearing itself apart at speed maybe, if the aerodynamics aren't done properly or there's a structural weakness. But that's about it.

[+] yoha|11 years ago|reply
Nuclear engines are not practical for lift-off. They can, however, be used for inter-planetary travel for a big enough payload.
[+] RBerenguel|11 years ago|reply
I wonder if the treaty about nuclear weapons in higher atmosphere/space has something to do with it. A reactor is far from being a bomb, even if it explodes mid-air the explosion would essentially be hydrogen+oxygen (with maybe a high amount of radioactive plume, but not a proper nuclear detonation).
[+] jwilliams|11 years ago|reply
People that have played different versions of Civilization will remember the Triremes.

They couldn't end a term in the ocean, only the coast. However, you could take a punt and travel over the ocean - with only the hope there was land on the other side.

Horribly expensive, but discovering new land or another civilization early could be transformative for the same.

[+] ars|11 years ago|reply
Lest you despair of ever making space flight routine, a rocket is not the only way to get into orbit.

Virtually all of the needed velocity is tangent to the surface, not away from it. So you can accelerate the vehicle along the ground at least part of the way, and only then turn heavenward and burn fuel to get into orbit. With this boost you significantly reduce the amount of fuel needed.

There are many way of doing this. You can have magnetic propulsion (very futuristic and powerful), you can have a simple motor on the vehicle, powered by contact with rails on the ground (but motors have a limit of how fast they can work). You could have fuel "guns" fired into the back of the vehicle as it passes them, which would then capture the container and burn them as a normal rocket would. This has the benefit of not requiring a large rail, just fuel stations that the rocket would pass over.

These are just some basic ideas, there are many more.

[+] yoha|11 years ago|reply
Starting from a ground propulsion is a priori a bad idea due to friction forces in the atmosphere. An almost plausible design though, would be to have a first stage taking advantage of the atmosphere (i.e. using jet engines with oxygen from the environment).

To build one's intuition about space exploration, Kerbal Space Program is an excellent start.

[+] fpgaminer|11 years ago|reply
I wonder if you could build a particle accelerator of sorts. A toroidal tube in which the vehicle is placed, and magnetically accelerated to launch velocity. Would just need some way to let the vehicle out; some kind of track switch. Such a setup would not require a large area to bring the spacecraft up to speed, and could be vacuum sealed to negate air resistance.
[+] has2k1|11 years ago|reply
Given that earth is a ball, aren't you always pointed heavenward if you are between 0-180 degrees. So you might as well start of by pointing directly heavenward, i.e 90 degrees. Plus, turning requires acceleration you cannot just capitalise on the velocity you have built up otherwise you would violate the 1st law of thermodynamics.
[+] ChuckMcM|11 years ago|reply
I really enjoyed this. It lays out in some very accessible ways, the challenges of getting into space. The recoverability of the Falcon9 will cut its costs dramatically as it reduces the cost of the launch by several tens of millions of $. I'll mention on-orbit refueling as well since you don't need recoverability per-se if you can refuel in orbit. Then your Mars lander / Crew Module can launch with enough fuel to get to Low Earth orbit, refuel, and then head out to Mars.

Air launched (Skylon, Pegasus, Et alia) are also interesting, laser boosting (using lasers to add energy during the initial launch) would also help. As Elon points out though, a multi-gigawatt laser for boost to orbit is impractical both from a construction standpoint and a diplomatic stability standpoint.

If the quantity of water on the moon is accurate, then it should be possible to create a 'refinery' on the Moon which could more easily get material into Earth orbit. We'll see though if we can get a group established there.

I had hoped to visit the Moon at some point (I was assured by NASA in my youth that would be able to :-)) but I don't expect that to come to pass unless something amazing changes.

[+] nickff|11 years ago|reply
If you do the math, air-launch does not save much in fuel costs, while it adds a great deal of complexity to the launch system, and you do not save much in terms of reduced energy requirements. Most of the rocket fuel and oxidizer are used to increase velocity; comparatively little is used to gain (the first 30k') altitude or lost to atmospheric drag. The main benefit to air-launch is reduced range safety costs, and ease of scheduling, as the launch vehicle can be taken out over an unoccupied stretch of ocean, and range rental/lease costs are basically eliminated.
[+] AndrewDucker|11 years ago|reply
Why not assemble some rockets in space?

Get the pieces up there as efficiently as possible, and then assemble a more efficient rocket that's not designed to leave the gravity well.

[+] danielweber|11 years ago|reply
If you skip the middle step, that's called staging.

It's really expensive to try and keep a permanent base in space.

[+] clumsysmurf|11 years ago|reply
If anyone is interested, these researchers from Russia have proposed a way to accelerate a spaceship while in flight – firing a ground-based laser up its backside

http://www.osa.org/en-us/about_osa/newsroom/news_releases/20...

[+] neurobro|11 years ago|reply
Well that seems less preposterous than my own idea of using lasers to accelerate a ring of air, generating a plasma circuit that would pull the craft up magnetically. (Version 2 would have a series of plasma rings that the craft would fly through like a coil gun.)
[+] Pxtl|11 years ago|reply
The soda can analogy is horrifyingly informative. I wonder what kind of efficiencies could be obtained with semi-science-fiction fuels like the Orion nuclear-explosion propulsion system (which is different from a nuclear rocket) or even straight-up antimatter.

en.wikipedia.org/wiki/Project_Orion_a(nuclear_propulsion)

[+] Lambdanaut|11 years ago|reply
If you've got a poor mind for Math but want to understand rocket science on a more intuitive level, check out Kerbal Space Program. It's the most educational game I've ever played that still retains being fun.

I usually avoid video games because I feel like I'm wasting time, but I make an exception for KSP.

[+] agopinath|11 years ago|reply
Fascinating read. Are there more blog-type posts like this from NASA? I'd be interested in reading more but I can't seem to find them from the main page and can't find links to these posts...
[+] ytturbed|11 years ago|reply
>Currently, all our human rated rocket engines use chemical reactions (combustion of a fuel and oxidizer) to produce the energy.

Yes, however, for completeness: an explanation of why we must limit designs to chemical rockets ought to include an explanation of why the dozen or so fusion projects underway around the world will all fail, i.e. let's inject some rational optimism. Note that the Apollo programme began before its tech was ready.

[+] pjc50|11 years ago|reply
The fusion projects have been running for a long time with little success. Apollo was built on scaling tech that already worked (1940s rocketry could reach space although not achieve orbit)
[+] shalmanese|11 years ago|reply
Fusion devices aren't rockets and are thus, not subject to the Tyranny of the rocket equation.
[+] im2w1l|11 years ago|reply
What about firing the rocket from Jules Verne's space gun[0]? Now, I don't mean an actual gun what with the high g-forces and so, but only some machinery that gives the rocket high initial speed. Maybe a super sonic evacuated tube maglev that ends with a ramp? Put it on Mt. Everest for less air resistance.

[0] http://en.wikipedia.org/wiki/Space_gun

[+] mrfusion|11 years ago|reply
So this is something I've always wondered about.

Can anyone explain the equation for getting to an earth sun Lagrange point?

Since were already on earth orbiting the sun it would seem we already have the correct orbital velocity to hang out at a Lagrange point. So we could really approach these points at any speed no? Is there potential to use less fuel than you'd need to reach an earth orbit?

[+] mcguire|11 years ago|reply
"If a vehicle is less than 10% propellant, [c]hanges to its structure are readily done without engineering analysis; you simple weld on another hunk of steel to reinforce the frame according to what your intuition might say."

This is why you don't let cabinet makers build ships.

[+] 001sky|11 years ago|reply
"If our planet was 50% larger in diameter, we would not be able to venture into space, at least using rockets for transport."

Don't tell hollywood.