Friday, October 28, 2011

What Might Have Been, Part III

Forty years ago, when they were drawing up the plans for the Space Transportation System, the original plans called for a flight rate of about fifty times per year. About the most we ever managed on a consistent basis was six. Something doesn't quite add up, here. What went wrong?

Well, one thing that went wrong is that there was never enough traffic to justify a fifty-per-year sortie rate. And another thing that went wrong is that it takes about three to four months to turn an orbiter around for re-flight. Early turn-around estimates were wildly optimistic. Now, we could have achieved a fifty-per-year sortie rate. But we would have needed more orbiters. With each orbiter flying at most four times per year, you need at least fifteen orbiters to keep the flight rate up.

The additional expense of those orbiters probably isn't as much as you're thinking. A large part of a Shuttle's price tag came from the fact that we had to amortize the entire RDT&E budget over five units. Six, if you count Enterprise. Similarly, part of the reason that a Bugatti Veyron cost $2 million and a Toyota Camry costs $20 thousand is that only 200 Bugatti Veryons were ever built, and there are about 5 million Camrys out there. Once you build the factory and tooling, the marginal cost of each additional unit isn't astronomical; and if you build enough of them, you get better at it, and the efficiency begins to show in the unit cost.

Which still begs the question: you don't need such a high sortie rate, unless you're moving a lot of cargo upstairs. Which is what went wrong with my teaser from back in May. Without such cargo volume, why pursue the matter any further?

For a couple of reasons. First, it keeps my mind occupied when I'm on the treadmill. And second, counterfactual scenarios sometimes provide a glimpse into why things in the real world turned out the way they did. So, without further ado, we're going to board the bus for Crazytown. Don't worry, we've got return tickets.

First, we go back to the year 1969. The lynchpin of the Soviet answer to Project Apollo was Sergei Korolev's giant N-1 rocket. It was about as big, about as powerful, could lift about as much stuff into space ... and it had 30 engines in its first stage. As I've mentioned before, Korolev had spent the last ten or fifteen years in a pissing match with Chelomei and also with Glushko, who was the engine expert. Korolev had to use less powerful engines, which meant that he had to use a lot of them. The first flight of the N-1 was in February of 1969, and by "flight" I mean "explosion". Getting thirty engines to play nicely together is not exactly an easy feat.

Between 1969 and 1972, three more test flights took place. The N-1 program was not officially cancelled until 1974. The Soviet Union never did land a man on the moon, but it wasn't for lack of effort. At cancellation, two N-1 rockets were still ready for test flights.

Which brings up a very interesting question, and the springboard for our counterfactual exercise: Why, five years after they'd already lost the Moon Race, were they still working on a Moon rocket?

The most likely answer is simply inertia. Soviet programs tended not to be cancelled until someone with authority looked at it and said, "Why are we still doing this?" And sometimes not even then. Voskhod 3, for instance was never officially cancelled. The spacecraft stayed in a shed, kept ready, even as Soyuz 1 was being prepared for flight.

The more entertaining answer is that the Soviets were planning a propaganda coup, by the establishment of a permanent Lunar base. There were some plans drawn up to this effect, which is another reason why the plug wasn't pulled right away when Apollo 11 was successful. Part of the reason that the project was cancelled in 1974 is that none of the tests had been successful. But the truth is, each one got a little bit closer. The fifth test flight might well have done the trick, had there ever been one.

Now, in this scenario, it's 1976. Two successful test flights prove the design, and more rockets are built. While America celebrates its Bicentennial, giant Soviet rockets are delivering payloads to a rapidly-growing Soviet base on the Moon.

What I'm trying to sell here is a scenario where Reagan, as part of his defense build -up, buys a whole bunch of Shuttles, and plays catch-up in a Moonbase race. The problem with this scenario is that it requires everyone to go crazy, in the same way, all at once.

And, at the end of the day, I just can't buy it. No part of this is plausible.

The Soviet Union cancelled the N-1 in 1974 because at that point, it was a white elephant with no useful purpose. Even if it worked, it wasn't going to do anything especially useful for them. They had decided to focus on a long-duration spaceflight program, and score their propaganda points that way. It worked, after a fashion. To this day, all of the duration records are held by Russians, except only longest flight by a woman. The point is, they had found a way to make their case at an acceptable cost in time and materials.

And for us, as I've said several times, we've proven by trial and error that the American public is willing to spend about 0.5% to 1.0% of the Federal budget on NASA, to include all of its aeronautical research programs. There was never a political case to be made for a giant program involving a moonbase in the 1980s or 1990s. Which meant that the "design" sortie rate for the Shuttle was a moot point. Part of the reason it only flew four to six times a year is that there was only enough traffic to keep it busy four to six times a year. And even so, look at the other side of the record books: the people with six or seven missions to their credit? Only two Russians on that list. The Shuttle put more human beings into orbit than any other spacecraft. That's not an achievement to sneeze at.

As we turn the page on this fine project, and as we look back at the other things we might have done in its stead, I have to say that we probably did about as well as we could have. We lost fourteen fine people. But we gained an immeasurable amount of knowledge. Only time will tell if that was a good trade. All I know is, the people who made that sacrifice believed so.

I hope -- and I also believe -- they were right.

Friday, October 21, 2011

Video Del Fuego, Part XLVIII

When Space-X recently announced their proposed Falcon Heavy rocket, one of the selling points they touted was extreme engine-out reliability. This comes from the fact that, at liftoff, the three core stages have 27 Merlin rocket engines between them. This got me to thinking. Specifically, it got me to thinking of an earlier rocket, that also had a fair number of first-stage engines.

The Soviet entry into the Moon Race, their equivalent of the Saturn V, was the N-1. The N-1 was a giant beast of a rocket, with thirty engines. I had always called that first stage a plumbers' nightmare, because of the complicated piping that I associated with so many engines firing simultaneously. It occurs that, perhaps, I was being unfair to Mr. Korolev.

And another curious thing: the last test flight of the Soviet Moon rocket was in 1974. Five years after the first American landing, and nearly two years after the last. If they'd given up on going to the Moon, then what in the world were they doing still trying to perfect a rocket for doing just that?

Therein, perhaps, lay a curious tale. In history as it actually happened, all four N-1 test flights ended more or less like this:



But what if they hadn't? What if they'd gotten all the kinks worked out? From everything I've read, they came close. One more test might have done the trick.

Maybe not next time, but soon, I'll take up the question of what might have happened next.

Friday, October 14, 2011

Election 2012 Preview: Handicapping the Primaries, Part III

Well, as I said before, prognostication is probably something I shouldn't do. My record is rather less than perfect. Still, it's fun, so I'm going to have another shot at it. Quite a lot has happened, and a few people have dropped out since the last time we looked at the race. It's still a long way to the conventions, but we've got a pretty good idea of who won't be in the running.

Democratic Party: Again, this entry is for completeness' sake only. Ralph Nader's pot-stirring notwithstanding, incumbent Presidents who still want the job always win re-nomination. Incumbency is a powerful advantage. You'd have to be a fool to throw that aside. Granted, the Democrats almost did in 1980, but it's not 1980. For the same reason, we won't see any movement on the VP side of the ticket, either. The last President who was re-elected after switching running mates was Franklin D. Roosevelt in 1944 ... and let's be honest here, Roosevelt could have had a burlesque dancer for a running mate, and he still would have won. That was a special case. These days, such a shift would be a huge display of weakness. So, no change here, it's Obama/Biden once again for Team Blue, unless one or both of them gets run over by a combine harvester in the meantime.

Republican Party: Oh dear, where do I begin? Let's just do this by the numbers (according to Intrade, current as of Friday afternoon):

Mitt Romney, 67%: He's managing a difficult dance extraordinarily well. As I said earlier, one of the biggest problems that he faces is that Obamacare is Romneycare with the serial numbers filed off, and everyone knows this. He had to find a way to credibly run against something he basically invented ... and so far, he's actually doing exactly that. An amazing feat, really. The basic thing to understand about the Romney campaign at this point is that he's not playing to win so much as he's playing to not lose. His selling points are competence and business acumen. Those are just about his only cards, and so far, he's playing them very well. He doesn't lead many polls, though, because his other problem may well be his Achilles' heel. He's a Mormon, stumping before a heavily evangelical Christian electorate. That's going to cost him. Will it cost him enough to turf him from the campaign again? Maybe, but only maybe. He still out-polls all other potential Republican contenders against President Obama. Electability may win out over religious prejudice.

Rick Perry, 11%: I've listened to Rick Perry several times in GOP Gubernatorial Debates here in Texas, and I have to ask: Who is this man, and what has he done with the real Rick Perry? Oh, I never expected him to be a debating all-star. But I never expected him to crater this horribly. Nevertheless, he'll make a strong run this coming Spring, once the campaign heads down South. This is going to be a two-way race, between Mitt Romney and "not Mitt Romney", whoever the hell that ends up being. Rick Perry still has a fighting chance to be "not Mitt Romney". But he's going to have to fight off several contenders for that slot, including Mitt Romney. (This is going to be a really weird year.)

Herman Cain, 9%: Fitting, since his signature tax plan is called "9-9-9". John Huntsman was right about that, by the way; it does sound like a pizza price. (And a pretty good one, at that.) Herman Cain leads the polls at the moment in the contentious race for "not Mitt Romney", but he's going to peak and stall pretty soon. He's ... how do I say this without sounding crass? Let's just say that his campaign will do a Titanic once the campaign goes down South. You can guess the reason.

And no one else is above 5%. This includes Huntsman and Gingrich (2%), Michelle Bachmann (1%), and Sarah Palin (under 1%). Sarah Palin has bailed out of the running for the Republican nomination, but this may not be the last we've seen of her this year. The interesting thing about the way Intrade works is that her chances will never drop to zero, even though she's dropped out. There are a number of poor slobs who bought Palin shares when they were at 7%, and didn't have to good sense to sell while the selling was good. They're stuck.

I still think Bachmann would be a good, if risky buy. She's liable to do quite well in Iowa, maybe even pulling a win. In that case, I'd expect a spike, and if you bought in at 1%, you could do quite well. But, as I say, it's risky. Bachmann may well have peaked already. She's been on a slump since debate season started. Still, if she were to climb back towards double digits, a buck or two could get you a night at the movies. (I think. I'm still not sure any real money changes hands, here.)

Palin is an interesting case. I see a scenario where she could be back in the running. Let's say that Romney does win the nomination. Let's further say that the Tea Party finds that unacceptable, and revolts. Who would be the standard-bearer for a third-party challenge? I don't see Rick Perry doing that. I don't see John Huntsman doing that. Both of them could reload for the 2016 GOP nomination, unless they burn their bridges with a third-party run. But Palin, who's already burnt her fair share of bridges, would certainly be game.

In any case, the battle for the Republican nomination will be a two-way scrum between Romney and Perry, barring a major breakout by another candidate. And the field is essentially fixed, there will be no new candidacies at this point. Either one could win. Romney's long-term position looks pretty good, and he's frankly the strongest candidate the Republicans could get out of this field. If they're smart, he's the one they'll run.

If they're not smart, or if there's a Tea Party revolt and a third-party run, we might well see the curious sight of a President, holding a bag full of 9% unemployment and lousy economy, winning re-election anyway because the opposition is either divided or too horrifying to contemplate.

I'd say stranger things have happened, except that I'm not sure that's true. As I said, this is going to be a really weird election year.

Saturday, October 01, 2011

Moneyball

My wife and I recently went to see the film Moneyball. I'm not a huge fan of baseball, but I had heard about the concept that the Oakland A's used to build a very good team on a shoestring budget. It was a very good film. I say that not because it sparked a latent interest in baseball, but because it got me to thinking about a few other things. To wit: what, precisely, will the United States do with its manned spaceflight program now that the Shuttles are going off to various museums and exhibits around the country?

I know, last time we talked about this, I promised a peek into a scenario where they might actually have reached the Shuttle's original advertised sortie rate of one flight per week. I got to thinking about it, though, and that requires a detour into Crazytown that I'm not quite ready for yet. I'll get around to that sooner rather than later, just not today.

So, we're stuck with the basic question of what to do next. And then, on the way home from the theater, it struck me. The problem all along is that we've been trying to field a New York Yankees program on an Oakland Athletics budget. And that's worked out about as well as anyone ought to expect it to. The main problem is that, ever since 1968, the long-range plans have all assumed massive budget increases that just won't happen.

It's time for a new paradigm. Taking a cue from Moneyball, I'm going to identify a couple of over-valued and under-valued "players" whose status needs to be re-evaluated. In no particular order:

(1) Heavy Lift: The question has to be asked -- do we really need a dedicated heavy lift booster? Do we need to re-create a Saturn V class rocket? I used to think so, but I've changed my mind. Dedicated heavy lift rockets are problematic at best from an economic point of view. They only make sense if you have a lot of heavy payloads going up on a fairly consistent schedule. If you only use them once or twice a year, the unit cost becomes hideous. You build and use at most ten or twenty in a decade, which means that you have to spread the cost of the factory and tooling over at most ten or twenty flights. This alone massively inflates the cost of a project that uses heavy lift. Which, in turn, makes the up-front "sticker shock" so harsh that the project never climbs up out of the planning stage. The Constellation program is only the most recent example of this. So, in our new paradigm, screw heavy lift. We're going to figure out a way to get by without it. And, with a few key enabling technologies, we can do just that.

(2) Closed-Loop Life Support: Part of the rationale for heavy lift is that interplanetary manned spacecraft need to be huge. If you have to carry all of your consumables (food, air, water) as cargo, you have to have about 30 kilograms of supplies for each crew, each day. That's just about a ton per crew per month. A minimum-energy trajectory to Mars takes nine months, and it can take as long as a year for the return launch window to open. So, at thirty months duration, we're talking 30 tons of food, water, and oxygen for each crew member. That's 120 tons for a crew of four. The largest part of that figure is water. If you can figure out how to recycle the water, you can cut that figure down drastically. Freeze-dried food and oxygen come to a little more than 1 kilogram per day. You would need a week or two of reserve water supply, but you could cut 120 tons down to less than ten. Savings like this cascade through the entire system.

(3) High-Efficiency Propulsion: Another part of the rationale for heavy lift is the fact that interplanetary spacecraft need so much fuel, not just for Earth escape, but for returning to Earth later on. Obviously, if you have more fuel-efficient engines, you don't have to haul as much fuel along with you. Taken together, these last two items make the spacecraft design much lighter. And since weight lifted into orbit is a big part of your cost, this makes the program as a whole more affordable.

(4) In-Space Refueling: That's all well and good, of course, but if you still have to lift the fully-fueled spacecraft into orbit all at once, you still need a heavy lift booster. That's where our last enabling technology comes in. If you develop the techniques for transferring fuel in orbit, you don't have to lift the whole thing all at once. You can lift the crew cabin first, then the fuel tanks, then the engines, then the fuel. And, once the ship returns from its first flight, it can be refurbished, refueled, and used again.

(5) Reusable Launch Vehicles: That's all nice, but we still have the problem of being able to get into low Earth orbit economically. Re-use is fairly important, if you want to get costs down. Operational simplicity is also important, of course, but you really want to be able to use expensively-machined components like engines more than once before dunking them into the ocean. Fortunately, we've run into a spot of luck on this score:



It's not completely reusable. But, if it works, enough components can be reused to bring operating costs down substantially. Also, the Falcon 9 core is the key ingredient for:



(Honest, I don't work for these guys. I just like their work.)

Bearing all of this in mind, where do I think we should go from here? Mainly, I think it's not really NASA's job to build a new rocket. Rockets, we've got. Good ones. What we need is for them to work on (2) through (4) above. If you've been paying attention, this is the "flexible path" option outlined in last year's Augustine Committee report. These goals ought to be achievable on a fairly modest budget. With those three things in hand, they can leverage private industry's work, and carry off an awesome program of exploration for a very reasonable price. And isn't that what we're paying them for?