On his blog, Warren Zoell asked the question "What went wrong?" on his excellent modeling blog. What went wrong with the space program, that is; how did we go from the seeming optimism of the late 1960s and early 1970s to this current sorry state of affairs? You can read his post here:
.
I've been so busy at work I haven't had time to work on a model or write anything about the demise of the space dream in general, but I've at least thought about it.
We all know the arguments, one way or the other.
I'm not going going to recapitulate the arguments, partly because I'm never going to change anyone's mind, and partly because I'm so tired, frustrated and aggravated with work issues I'm not sure I'm capable of generating a coherent argument in favor of anything (the only thing I'm sure of right now is this:
It starts with the basic technology of the Apollo Program, specifically the Saturn V booster. You then add a nuclear upper stage to the Saturn V.
This is what a flight-ready nuclear engine would have looked like had the NERVA (Nuclear Engine for Rocket Vehicle Applications) program been allowed to produce one. It basically amounts to a modest nuclear reactor roughly the size of a large curbside garbage can. You pump liquid hydrogen into the left end (and, because you're pumping liquid hydrogen, you can exploit odd exothermal effects to bootstrap the pumps so you don't need anything extra to run the pumps). The liquid hydrogen is heated to about 2,000 degrees C in the reactor core in the middle, and then you allow it to squirt out the back.
But why? What advantages does this offer over an existing chemical rocket engine?
The efficiency of a rocket engine, and thus its final ability to do useful work, depends upon an arcane measurement called specific impulse. Without going into too much detail, suffice it to say that the specific impulse rating of a rocket engine is sort of like the miles-per-gallon rating of an automobile. People get all worked up over total thrust, which is sort of like a car's horsepower rating, and total thrust is a factor when you're actually trying to leave Earth. But assuming you're already in space (that is, assuming that the Saturn V booster has already gotten you into orbit), thrust is meaningless. All that matters, once you're in space, are two numbers: the rocket's mass ratio and the engine's specific impulse. That's all. (Mass ratio, by the way, is a measurement of how much of the spacecraft's total weight is fuel.)
The point is that once you get in space, all the other measurements fall away and all you're working with is mass ratio and specific impulse. Nothing else matters, at least from the performance point of view.
The higher the specific impulse, the better. A given engine's specific impulse is determined basically by one thing: how fast the junk squirts out the back. And as it turns out, how fast the junk squirts out the back depends on two things: how light the junk is in the first place, and how hot it is when it squirts out the back. Lighter junk squirts out faster than heavier junk, and hotter junk squirts out faster than cooler junk.
So let's say we have some rocket engine, and we want to send some imaginary payload to some groovy place like Mars. We've done all we can to make the payload as light as possible, but it turns out that we still can't get there from here - to make the mission work, we either have to increase the mass ratio to the point where the booster is the size of a small moon and sends a payload the size of a lunch box, or we have to increase the specific impulse of the engine.
How do we increase the specific impulse? One is to use propellants with a lighter atomic weight - that is, make the junk lighter. Another is to make the engine run hotter by burning more energetic fuels.
Chemical rocket engines work by combining two different chemicals, igniting them, and allowing the heat of combustion to force the burned junk out the back. The choice of fuels determines how hot the fire will be, and how light the burned junk will be. Sometimes there are good engineering reasons for using less than optimal propellants - the Saturn V first stage burned kerosene and liquid oxygen, a combination that isn't great for specific impulse but has certain advantages in terms of cost and ease of handling. But once you're in space, you want to use the best combination you can so you can get the best specific impulse, and that means oxygen and hydrogen.
Burning liquid oxygen and liquid hydrogen gives you hottest fire, and it gives you the lightest burned junk squirting out the back. And it turns out that this combination gives you a specific impulse of about 460 seconds (don't worry about what the seconds means). And there's nothing you can do to make it better. There's no way to engineer more specific impulse out of the engine - no magnets on the fuel lines, no chips, no vortex generators in the air filter. You can add stuff like fluorine to the liquid oxygen to make it a little hotter, but the engineering challenges of a fluorine oxidizer are so formidable they aren't worth pursuing.
The point is that the best chemical propellants will give you a specific impulse of about 460, because they only burn so hot and the burned junk is only so light (the combustion product of oxygen and hydrogen is water, two hydrogen atoms and an oxygen atom, and there's just no way to make that any lighter than it already is).
And it turns out that a specific impulse of 460 seconds is pretty limiting. It becomes hard to do things with such a low specific impulse. Missions take much longer (remember how long it took Cassini to Saturn?) and payloads have to become very small (remember how dinky Mars Pathfinder was?) for anything to get done, and the combination of long mission times and small payloads makes manned landings on Mars (the sine qua non of any space program, as far as I'm concerned) exceedingly impractical.
What can you do? Nothing. There's no chemical propellants that give a better specific impulse, and you can't squash men down until they have the size, weight, and life support requirements of fruit flies.
But you can abandon chemical rocket engines entirely and use a nuclear engine. Nuclear engines can run arbitrarily hot - ideally, as close to the melting point of the engine's components as you can get without it actually melting. And gaseous-core reactors are potentially possible, which have no arbitrary temperature limit at all. They're already in the form of a gas, so the melting point of the components doesn't matter. And since you don't have to actually burn anything to produce the junk that squirts out the back, you can pick the lightest junk possible - hydrogen.
This gives you an engine that runs much hotter than a chemical engine, and squirts out the lightest possible junk. All you need is the will (and the money) to build it. And with such an engine doing the pushing, mission times drop dramatically, and payloads increase dramatically. Want to get to Mars in sixty days? Can do. Want to land men on Mars? Can do. Want to send a probe to Pluto in a year? Can do. Want to build a kind of space tugboat that can efficiently shuttle cargo back and forth between a space station and the Moon? Can do.
The advantages of this sort of nuclear-thermal propulsion have been known since the days of Tsiolkovsky, and all of the major pioneers (Goddard, von Braun and Oberth, among others) understood the advantages of them too. The only limit on the specific impulse of the nuclear rocket is the temperature of the reactor core, and there's no arbitrary upper limit on that; it's simply a matter of how good your engineering is.
The first tests were performed by the Americans after World War Two. The first tests were pretty unpleasant affairs - the reactors tended to disintegrate and it was common for them to spit glowing chunks of uranium fuel a thousand feet into the air, which nobody wants. But development went smoothly and quite quickly, and by the mid-1960s the NERVA program had produced engines that ran reliably for hours at a time, remained under good control while being throttled, could be started up and shut down relatively quickly, and most important of all tended to stay together in one piece. And they produced specific impulses that started at about 850 (that is, almost twice as efficient as the best conceivable chemical rocket engines) and proceeded upward from there, some of them nudging 1,000 as the high-temperature engineering of the fuel assemblies proceeded. The Soviets, who came into the game later, were said to have built an engine that ran with a specific impulse of 1,200 in the 1980s.
The success of the American NERVA program is not well understood, even in the aerospace field. I've talked to engineers in the field, and though my results aren't scientific by any means, the common perception is that the nuclear-thermal engines either didn't work at all, only barely worked, or tended to blow up in glowing mushroom clouds. But they didn't. They worked. They worked spectacularly well.
There were accidents, of course. During one engine test, they ran out of liquid hydrogen and the reactor melted down almost instantaneously. In other tests, the fuel elements broke and glowing bits of the core shot out the nozzle. And all the tests tended to make the test stand radioactive through the process of neutron activation on hitherto-nonradioactive materials. And there was always the problem of the exhaust plume blowing atoms of uranium and other radioactive fission products into the atmosphere, which probably wasn't smart, Ann Coulter's belief in radiation hormesis notwithstanding. And in one test, they deliberately allowed a reactor to catastrophically overheat and blow up, just to see how bad a launch pad accident would be (the answer: not as bad as the doomsayers predicted, but still bad enough - it's still a nuclear reactor, and nuclear reactors aren't the same as a vinegar-and-baking soda rocket at all).
But my central point is that if you combine the power and relative safety of the Saturn V booster with the efficiency of the nuclear rocket engine, many things suddenly become possible. Not just possible in the engineering sense, but even practical. Without getting all bogged down in the mathematics, teaming a nuclear upper stage with a booster like the Saturn V doubles the payload that can be dispatched to any point in the Solar System, and halves the time it takes to get there. And that's the worst case. By clustering nuclear engines, or by using better nuclear engines, the payload just keeps going up and the mission time just keeps going down.
Consider the Orion shuttle from the movie 2001. I always assumed it used nuclear-thermal engines in orbit. You wouldn't want to light up its nuclear engines on the ground, because you don't want to make the spaceport itself radioactive. And you wouldn't want to run the nuclear engines in fairly dense air, because the air itself tends to backscatter neutrons and gamma rays from the reactor core. But once you're above 100,000 feet, backscatter radiation becomes negligible (neutrons and gamma rays from the cores fly out in straight lines and no longer reflect back into the passengers). So all you have to do is get Orion to about 90,000 feet or so, and we already know how to do that using turbofan engines. And the use of the nuclear engines would completely transform the economics of the Orion shuttle.
A useful rule of thumb is that it'll cost you about $2,000 per pound to get yourself into orbit using an expendable booster like the Delta or Ariane. The Saturn V was cheaper, about $900 per pound, partially because those were 1970 dollars and partially because the Saturn V lifted so goddamned much the economics of scale came into play. The American Space Shuttle was about $7,000 per pound, which is pretty pricey. But suppose I fly from here to London on the Concorde (which no longer flies, but we'll pretend it does). The total bill is about $10 per pound, and it's even less for subsonic airliners like the 777 (the last time I flew to England, the bill came out to about $2.50 per pound, not including my luggage).
I read somewhere that a shuttle employing jet engines to take off and nuclear engines to reach orbit would cost the user about $14 per pound. Let's say that's wildly optimistic and the real cost is double that. Say, $28 per pound. And someone has to pay for all that orbital infrastructure. Let's say it costs $50 per pound, once you add everything up. An Orion shuttle, using mixed turbofan and nuclear propulsion, would still be 40 times cheaper than a Delta, and about 140 times cheaper than an American Space Shuttle.
In other words, a nuclear shuttle would operate on airliner economics, not rocket economics, cheap enough that I'd probably do it just to say I'd done it.
(Here we are combining, of course, the efficiency of nuclear rocket engines with the efficiency of jet engines. Modern jet engines are so efficient they're almost embarrassing to rocket engineers. Specific impulse is kind of meaningless for jet engines since they ingest air from the environment as reaction mass and oxidizer, but their specific impulses are in the vicinity of 15,000, which gives you an idea of how good they are. Jet engines aren't perfect, though. They suffer from thermal problems at high speed, and at high altitudes they just flat run out of air. But if you let jet engines do what they do best, and then at high altitude and high speed hand over to nuclear engines and let them do what they do best, well, you're in bidness, as they say.)
So why not? Why aren't we using Saturn Vs (or derivatives thereof) to boost heavy stuff into orbit and using nuclear rockets to push the heavy stuff around? Why aren't there shuttles that combine the efficiencies of jet engines and nuclear rockets? Why aren't there footprints on Mars? Why isn't there a moon base, or multiple moon bases, or an actual goldurn city on the moon? Why aren't there radio telescopes on the far side of the moon, and helium-3 mines on the moon, and people extracting water from permafrost on Mars?
Part of it is ignorance. If even engineers in the aerospace biz think that nuclear rocket tests in the 1960s were a fiasco, what must the layperson think of them? Part of it is apathy. Even in the 1970s most people didn't give a rat's ass for space exploration any more, and they probably only got interested in it in a "let's beat the Commies" sort of way in the first place. Part of it is that nobody seems to care much about anything at all any more, other than jailbreaking their iPhones and finding cool apps and watching Lady Gaga videos on the train. Part of it is a failure of education. I don't think schools do a good job of preparing people to think about space exploration, and as a result we as a people don't realize that in the 1960s we had in our grasp two keys to doing serious work in space, the Saturn V and the nuclear rocket.
Part of it is an anti-nuclear backlash. The word "nuclear" came to mean "evil" by definition, and I can sort of understand that. You say the world "nuclear" to the average person on the street, and they're liable to think "Three Mile Island" and "thermonuclear war", not "a specific impulse of 1,000" and "vastly improved capability in space". Nuclear rockets aren't nuclear weapons, but they're still nuclear reactors, and they aren't toys and they have certain rather serious risks, and I'd be the last person to say that the risks of operating things like nuclear-thermal rocket engines and SNAP reactors in space should be ignored (even I, a proponent of nuclear propulsion, view photographs of the NERVA engine tests with misgivings, and my toes curl when I see the plumes of hot hydrogen and radioactive junk spewing out of the NERVA engine test rigs and into the open atmosphere. By way of vast digression, our family lived for many years in Flagstaff, Arizona, which was more or less downwind of the nuclear test facilities in Nevada. When my dad died, the US government paid us a not inconsiderable sum of money because it was plausible that his cancer may - I emphasize may - have been caused by fallout from nuclear tests in Nevada. My cancer, however, isn't on that government list, as there is no reliable statistical link between Hodgkin's Lymphoma and nuclear tests).
The Vietnam War played into it too. That war cost a fortune, and it's even worse now, what with unending wars in Afghanistan and Iraq. The Vietnam War's voracious appetite for funding was arguably the final straw that caused the Nixon Administration to shut down the Saturn production line and terminate the NERVA nuclear rocket program.
People of good conscience make the argument that space exploration isn't worth the cost, given the scale of problems we already have on Earth. Wouldn't the $1.5 billion we spent on the NERVA program have been better used finding a cure for cancer, or improving public transportation, or taking care of people with serious mental illnesses? I can understand those arguments. I don't necessarily agree with them, but I can at least understand them. My main counterarguments to them are A) it's difficult to say what sorts of world-changing spinoffs a large space program might have produced, and B) money taken from NASA was never redirected into programs designed to fight cancer or end childhood hunger anyway; for all practical purposes it just vanished, and a lot of it just vanished into the ever-hungry maw of the Vietnam War. If you want to defund something in order to pay for things like fighting cancer or educating children or ending mental illness, wouldn't it have made more sense to defund the Vietnam War, whose yearly budget vastly, vastly, exceeded NASA's paltry $5 billion?
But as I said, I can at least understand arguments against the space program on the grounds of fiscal conservatism, or social progress, or cost-versus-benefit calculations. The arguments I have difficulty wrapping my mind around are the ones based on ignorance and anti-intellectualism, which are unfortunately pretty common in America too.
That's a subject for a different day.
My point is that as a kid in the 1960s, it was easy for me to imagine a future that included the following:
* Saturn V heavy-lift boosters remaining in production and carrying big payloads into low Earth orbit at reasonable cost and with minimal risk to the planet.
* NERVA-style nuclear-thermal rockets moving those payloads around the Solar System on a scale and with a speed and efficiency that no chemical rocket could ever hope to match.
* A jet-nuclear shuttle like Orion providing cheap and reliable access to space for ordinary people, on an economic scale more akin to airline operations than rocket operations.
* A whole infrastructure springing up in space, not necessarily because some bureaucrat in Washington said so, but because they made economic or scientific sense. Space stations, mines, bases, exploration camps, telescopes, particle accelerators, factories...
* All of this leading in the end to the ultimate in cheap and reliable access to space, a space elevator as described by Arthur C. Clarke. It wouldn't happen in my lifetime, but it would happen, because once you had the nuclear rockets to do the work with, the idea of fetching hither a suitably large rock to serve as an anchor for the far end of the space elevator stops being impossible and merely becomes difficult.
Goodness. That got long and tedious, didn't it?
