The April 2013 issue of Scientific American has the latest ("The Low-cost Ticket to Space" by S. Alan Stern) in a long series of articles predicting that the era of cheap and easy spaceflight is just about to break out. These stories usually have a "manned" bias. Explicitly or implicitly they assume that it is important for people to have direct access to space. I found the story just sad. It saddens me to see how little has changed in the past 50 years.
Fifty years ago the Apollo moon landings had just wrapped up and it was early days for the Space Shuttle. Some things have changed. The Shuttle has been retired. Some things have not. We haven't sent a man (or woman) to the moon since. We are yet again full of wide eyed plans to revolutionize access to space. Fifty years ago it was the Shuttle. Today it is private space launches. That's a change but not that big of one. And, depressingly, the sales pitch is still the same. Then the Shuttle was going to make it cheap and reliable to get to space. Why? Reusable components. Now private launches are going to make it cheap and reliable to get to space. Why? Reusable components. Sheesh.
Then the Space Shuttle was supposed to be the gateway to manned exploration of the moon and beyond. Guess what? The same is true for private space launches. It is nice that the Scientific American article does have another pitch. It says it is important for scientists to have access to space. Cheap and reliable (both pitched and predicted in the article) launches by private companies would actually be helpful to scientists if they ever come into being. And I truly hope it happens. But remember the International Space Station. Well, it's still there and it's still gobbling great chunks of NASA's budget. The justification for the ISS was all the great science it would enable. It never happened.
The premier technical journal of the science community that is published in the U.S. is called Science. It is published by the AAAS (American Academy for the Advancement of Science). I have subscribed to it for years. Several times a year it publishes a cluster of articles based on some space based scientific program. Recently, for instance, they published a cluster based on scientific returns from the MESSENGER probe orbiting Mercury. Scientists have found out some very interesting things about Mercury that completely justified the Science coverage.
And it's not just Mercury. They have also published clusters based on various probes to Mars, Jupiter, Saturn, and other parts of the Solar System. Science has published several clusters based on findings of the Hubble Space Telescope (no surprise). But they have also published clusters based on other "Astronomical Observatories". Several missions featured in Science packages are unfamiliar to most of the public. (Quick - what's the "Dawn" spacecraft up to?) My point is that Science does these packages frequently (several times per year) and on a wide variety of space missions. But they have yet to do a package based on science produced on either the Space Shuttle or the ISS. (I recall no Scientific American articles based on scientific research hosted on the Shuttle or ISS either.) "Supporting more scientific research" is an argument that has never had any traction when it comes to increasing support for space related activities. But I have digressed.
The main argument in the Scientific American article is one about cost. "Private companies will make space flight really cheap" paraphrases the article's main point nicely As I mentioned before, this argument parallels the previous argument that "the Space Shuttle will make space flight really cheap" that was propounded in the '70s while the Shuttle was under development. In the Shuttle's case it never really worked out. During its midlife the Shuttle had a cargo capacity of about sixty thousand pounds. And a launch cost (not counting Shuttle development costs of many billion dollars) about six hundred million dollars. This works out to a cost of ten thousand dollars per pound. That's really expensive and explains why space flight has always been a low volume operation. And remember the Shuttle was supposed to deliver low cost access to space.
Cost figures are hard to come by in coverage of the space program aimed at the general public. The Scientific American article is no exception. It is full of gee whiz and "trust me, it'll be really cheap" but it's short on actual numbers. A rare exception is the information that Virgin Galactic intends to (but hasn't actually) charge $200,000 to take 200 pounds on a suborbital mission. That's only a thousand dollars a pound, a 90% savings over the shuttle, right? But the Shuttle delivered its pound to low earth orbit and Virgin Galactic is only delivering its pound to "suborbital". Figure at least a doubling of cost to achieve LEO (low earth orbit - e.g. to the Space Station). In fact, Space X, one of the companies featured in the story, claims their costs figure out to be just under $2,000/lb. However, a Space X mission is costing NASA $133 million to deliver 29,000 lbs. of cargo to the ISS. This figures out to about $4,500/lb., a modest improvement over the older Shuttle costs. (Late stage Shuttle flights after all the safety fixes and allowing for general inflation penciled in at over a billion dollars per flight to deliver only about 50,000 lbs. of payload, about $20,000/lb.). The Shuttle carried all that extra "man rated" weight and overhead plus all that purported "government inefficiency" overhead, so this is not as impressive an improvement as it might otherwise seem. So what's the story?
It turns out the basic science of rocketry was figured out by Robert H. Goddard in the 1930s. He developed the first rockets that were not toys. And he almost single handedly developed the science of rocketry. The Nazis were almost the only ones who paid any attention. They used his work as the starting point for the design of their V-2, the first military rocket of any practical value. The V-2 opened the eyes of the rest of the world to the potential of rocketry. This led to the first successful satellite launch in 1957 and the first successful manned space flight a couple of years later. By the end of the '60s rocket science had reached a level of maturity that permitted the first manned moon landing. Unfortunately, since then not much has changed. The only big improvement has been the development of light, cheap, reliable computer technology. Coupled with great advances in instrumentation, helped greatly by the same advances in computer technology, it has since become possible to create very sophisticated robotic vehicles, some of which are quite small. But the basic "rocket" part of "rocket science" has changed very little.
Goddard figured out in the '30s that hydrogen/oxygen was the best rocket fuel in terms of efficiency. No one has come up with anything better since. By the '60s solid fuel rockets had been developed. These were less efficient but the fuel was a lot less expensive. This has led to a number of hybrid designs consisting of a solid fuel lower stage or solid fuel strap-on boosters combined with liquid fuel upper stages. (BTW, Goddard also pioneered the multistage rocket in the '30s.) This is a small advance in technology driven by cost considerations. The third type of rocket motor is one where the hydrogen is replaced by something cheaper and easier to handle like kerosene. This is intermediate in efficiency and cost between a solid fuel rocket motor and a hydrogen/oxygen rocket motor. Goddard pioneered kerosene/oxygen rocket motors in the '30s.
By the '60s most rocket designs were based on a 90/10 ratio. The rocket consisted of 90% fuel (by weight) and 10% everything else (structure, tanks, motors, guidance and control etc., and payload). As far as I can tell this ratio still holds. You can figure that a typical rocket on the launch pad is 90% fuel. And that "everything else", well there is a 90/10 rule here too. It turns out that the payload usually represents about 10%. Combining these two factors leads us to the conclusion that it takes about 99 lbs. of fuel to put one lb. of payload into LEO. So there is an absolute limit to how much you can reduce the cost of a rocket launch. You have to pay for all the fuel, which gets all burned up by the launch.
This leads me to ignore 99.9% of what I read about whatever the new idea is about how launching rockets into space is all of a sudden going to be real cheap. A hydrogen/oxygen rocket motor burns two lbs. of hydrogen and 8 lbs. of oxygen to produce ten lbs. of water. Those ratios are just chemistry. I showed above that it takes about 99 lbs. of fuel to provide the boost to put one lb. of payload into LEO. So tell me, how much does it cost to deliver 11 lbs. of liquid hydrogen and 88 lbs. of liquid oxygen to the fuel tanks of a rocket on the launch pad? That's the absolute minimum cost it is going to take to get that single pound into LEO.
A quick Internet search finds liquid hydrogen in large quantities running about $2.50/lb. and liquid oxygen (again in large quantities) running about $0.10/lb. I don't know how good these numbers are but I am going to use them for the rest of this article. They would translate to a cost of about $36 for our 99 lbs. of hydrogen/oxygen fuel (mostly for the hydrogen). Now the rest of it (structure and other components, design, launch services, etc.) is going to cost something. But figure it in multiples of the cost of our 99 lbs. of fuel. The current Space X contract of $4,500/lb. works out to 145 times fuel costs. It looks like there is room for improvement. Space X itself apparently thinks that it can improve the situation as it works down the learning curve. If it can deliver an improvement to $2,000/lb. this would work out to 55 times fuel.
The article denigrates the Atlas V, a vehicle developed and operated in the standard NASA / Military Industrial Complex environment. The article pegs Atlas V launch costs as being in the $150 million to $350 million range. Wikipedia lists the lift capacity of an Atlas V to LEO as 64,000 lbs. (a little more than the capacity of the pre safety fix Shuttle). This pegs the cost at $2350 - $5500 per lb. This translates to 65 to 150 times fuel cost. That puts it in the same range as Space X. So where's the big cost savings of Space X (and other "new and more efficient" vendors) over NASA? It looks like unmanned rockets save significant costs over a manned rocket like the Space Shuttle. But that's hardly a surprise. And it also looks like the new vendors like Space X have yet to come up with some great new way to get launch costs down to the point where they are way below an Atlas.
So, if we ignore history, ancient and new, what is the lowest feasible cost multiplier using fuel cost as our base and LEO as the target of our benchmark? An old article in Science in support of a now discontinued NASA program to develop a cheap efficient booster (it was discontinued not because it failed but because it never garnered any political support) thought that a target of $300/lb. was feasible. That was a few years ago so let's round the figure up to $360/lb. to allow for inflation (this probably underestimates inflation but it makes the math simple). This would work out to a cost of 10 times fuel. I have certainly seen no one suggest that a more aggressive target is feasible. Anyone who can deliver a cost of 10 (or even 20 or 30) times fuel cost is someone who has found a way to build a better mousetrap. But especially if the cost is 100 or more times the fuel cost "pay no attention to the man behind the curtain". What's going on is pure flimflam (or perhaps just good marketing).
Now let me focus for a few minutes on why I have been fixated on LEO. It turns out that if you don't have to be in a hurry (e.g. because you have people aboard) LEO gets you most of the way to anywhere you want to go in the solar system. We have already noted that it costs about half as much to get to "suborbital" (usually a "ballistic" flight with a peak altitude of about 100 miles but not enough oomph to get you into orbit at 100 miles) costs about half as much as LEO. So, if your rocket can put one pound into LEO it can put two pounds into suborbital. Similarly, if your rocket can put 1 lb. into LEO it can put about 0.7 lbs. into GEO (Geostationary orbit where all the communications satellites live). I don't know what the ratio is but I think the same rocket could put about a half a pound into the vicinity of the moon. Note: It takes extra energy to translate "vicinity of the moon" into lunar orbit).
"Vicinity of the moon" will also get you to the surface of the moon. But you would be going at a pretty good clip when you got there. If you want to survive the experience, you need extra energy to slow down. It goes without saying (that's why I am saying it) that it would take more energy to get you off the moon and back home too. I don't know how much energy it takes to get you to the vicinity of other planets but I think it's mostly pointing in a different direction rather than applying more energy. So let's say you could send about 0.4 lb. wherever you wanted in the rest of the solar system. Again, going into orbit or landing takes more energy. And the trip can take years. The "New Horizons" Pluto flyby mission was launched in 2006 and will finally fly by Pluto in 2015, about a decade later.
To summarize, when you are reading one of these gee whiz stories about how we are about to crack space wide open ask yourself "how much progress are we making on the cost front?" The Shuttle was billed as a giant cost saver. It wasn't. We are currently saying bad things about the Atlas V, Delta IV, and other traditionally developed and operated unmanned launch vehicles. But the new "cheaper and better" alternatives like the Space X Falcon don't look like a big improvement. Maybe these new options will get a lot cheaper as time goes by. We'll see.
Finally, if you want a web site that is chock full of hard core "this is how it really is" information then go here: http://www.projectrho.com/public_html/rocket/index.php. The site is maintained by Winchell Chung who does art work for various SciFi projects and comic books. The whole site is a lot of fun. But the link sends you to a lot of technical material. Subjects include the real world (e.g. how much energy does it take to get from here to there) to various fantasy/scifi subjects (e.g. if it was possible to build a nuclear rocket how would it work and how well would it work).
No comments:
Post a Comment