The SLS Frenzy

So apparently, the SLF fanbois (and fangirls) going crazy over a giant welder on Twitter.

Anyway, I was rereading this essay I wrote half a decade ago. It was depressing. Here’s how little of some of it I’d have to change to keep it relevant to today.

Rethinking the Vision

To get past the misperceived lessons of the past four decades and to develop a “safe, innovative, affordable, and sustainable” plan for manned spaceflight, we must begin by stating plainly why we should go into space, for the why gives shape to the how.

The United States should become a spacefaring nation, and the leader of a spacefaring civilization.

That means that access to space should be almost as routine (if not quite as affordable) as access to the oceans, and with similar laws and regulations. It means thousands, or millions, of people in space — and not just handpicked government employees, but private citizens spending their own money for their own purposes. It means that we should have the capability to detect an asteroid or comet heading for Earth and to deflect it in a timely manner. Similarly it means we should be able to mine asteroids or comets for their resources, for use in space or on Earth, potentially opening up new wealth for the planet. It means that we should explore the solar system the way we did the West: not by sending off small teams of government explorers — Lewis and Clark were the extreme exception, not the rule — but by having lots of people wandering around and peering over the next rill in search of adventure or profit.

We should have massively parallel exploration — and not just exploration, but development, as it has worked on every previous frontier. We need to expand the economic sphere into the solar system, as John Marburger, George W. Bush’s science adviser, used to say in his speeches. We need to think in terms of wealth creation, not just job creation. That would be “affordable and sustainable,” almost by definition.

You may say I’m a dreamer, but I’m not the only one: Apollo left many orphans. But it’s not a dream shared by NASA, successive presidents, or members of Congress, at least to judge by their plans over the past four decades. We have had a monolithic government space agency for over half a century at a cumulative cost of roughly half a trillion dollars (in current-year dollars). If we are going to continue to spend that order of magnitude of money — as, for political reasons, it seems we are going to do indefinitely — we should at least have something more to show for it than just a couple hundred brief trips to orbit for elite civil servants at an average cost over that period of about a couple billion dollars per flight. NASA needn’t do all the work of making space affordable and sustainable, but it ought to do something. To put it another way, it isn’t NASA’s job to put humans on Mars; it’s NASA’s job to make it possible for the National Geographic Society, or an offshoot of the Latter-Day Saints, or an adventure tourism company, to put humans on Mars.

In concrete economic terms, that implies that whatever infrastructure we establish for reaching space should have low, not high, marginal costs of operation. Low marginal costs mean that as demand for a service grows, the price can drop rapidly. For example, a large restaurant with a full staff (a high fixed cost) but only a couple of diners would have to charge thousands of dollars each for a meal. But the marginal cost of feeding the next diner is only the cost of the food, and as the restaurant fills, the average cost can drop to where the price of a meal becomes affordable. (In this analogy, our current spaceflight practices are akin to burning down all the restaurant’s furniture after every meal and buying it all anew before the next one; marginal costs are quite high in that scenario.)

High marginal costs will forever constrain the level of activity that’s possible. That was true of Apollo, it is true of what NASA currently plansned with Constellation, and it is true of any Constellation-like architecture (such as DIRECTSLS/Orion): every flight will require throwing away tens if not hundreds of millions of dollars worth of hardware. If we were, say, to discover something on the Moon really worth going after, our ability to ramp up activity with ConstellationSLS/Orion would be severely limited by our budget. Low marginal costs provide scalability, which is essential for any technology that is going to open up large new markets. NASA’s and Congress’s plans completely lack any understanding of this crucial principle.

The principle of low marginal (and average) costs was why the shuttle was created, except that it ended up combining the worst of all cost worlds: the shuttle hasd high fixed costs (for the standing army needed to service it), high average costs (resulting from the low flight rates), and high marginal costs (due to the hardware thrown away with each flight). When you hear that a space shuttle flight cost hundreds of millions of dollars, that figure iwas an average cost — the annual cost of the overall shuttle program divided by the number of flights that year (dividing the total cost of the shuttle program since its inception by the total number of flights would result in a yet higher number). The actual marginal cost (the cost of flying one more mission, given that you are already flying) iwas much lower, at most $150 million — still ridiculously high, but comparable to other launch vehicles with much less capability.

How does ConstellationSLS/Orion stack up? Let’s ignore the crew module Orion and just look at Ares ISLS, the crew launch vehicle. A recent Aerospace Corporation study estimated that the total cost of the Ares I program — all the costs of the program from inception to grave — would be $19 billion for fourteen flights. This figure was obviously based on the initial development cost estimate of $14 billion; using the current estimated development costs of $35 billion, the total cost for Ares I would actually be about $40 billion. Even under the most charitable interpretation of the numbers — not including all the development costs and generously assuming four flights per year and fixed costs of just $1 billion per year — each Ares I flight would still cost roughly the same as each shuttle flight, although with much less capacity. [Insert numerous analyses indicating that SLS will cost several billion per flight.] And that figure doesn’t include the Orion capsule, let alone the Ares V heavy-lift vehicle with all the expensive lunar mission hardware aboard. Each lunar mission, in this architecture, will cost several billion dollars.

Mike Griffin was wont to compare Constellation to the U.S. Interstate Highway System. But the interstate was a national investment that resulted in a system with very low marginal costs and affordable for all. Anyone with a car could get on it, drive at high speeds, and just gas up when they got to a station. Its network of roads was also a boon to national security (which was in fact its initial justification). By contrast, Constellation (and any similar architecture) doesn’t just fail to support national security (thereby ignoring one of the Aldridge commission recommendations). It is also a huge money sink that will result in a system with high marginal costs, low flight rates, and only for use by government employees.

Gas Stations in Space

The only reliable way to lower marginal costs is to pursue full reusability — that is, to make the entire spacecraft, including the launch vehicle, reusable. To return to our analogy, the restaurant is the vehicle fleet, facilities, and staff to service it. The food is the propellant. The first flight is hugely expensive. After that, costs will drop rapidly. The ultimate floor on the marginal cost of any form of transportation is the cost of the energy required to get from one point to the other. Getting to orbit is not that different, energetically, from flying across the Pacific, and there’s no reason that we shouldn’t be able to lower the marginal cost of getting to orbit to within an order of magnitude of the marginal cost of air transport, given sufficient demand.

The critical requirement of a reusable space system is refuelability. Consider a thought experiment from an earlier frontier. Imagine that, on the settlers’ hard trek to the western United States, there had been no vegetation along the way for the wagon-pulling horses or oxen to eat. To get across the country, each Conestoga would have to carry enough hay to feed the animals (not to mention supplies for the pioneers for months). The wagon would have been so large that the animals wouldn’t have been able to pull it. The longest distance that could be traveled would be dictated by the largest size of wagon that they could pull when it was full, and the initial speed would be very slow, picking up as the wagon grew lighter. Once the final destination was attained, the wagon and the animals would be useless without more fuel, so presumably the wagon parts would be used to build a cabin or saloon. In reality, of course, such a system would never have been affordable; had the settlers not been able to avail themselves of food and water along the way, the West would never have been settled.

Now apply that logic to space. The vast majority of the payload for heavy-lift launch vehicles is the propellant needed to send a relatively miniscule spacecraft to the Moon (or Mars or whatever destination) and back. Recall the Apollo missions’ gargantuan Saturn V rocket; the tiny capsule atop it was all that came back. And much of the propellant used by Saturn V was needed just to deliver into space the propellant that will be used for the trip back, since there were no gas stations on the Moon. The Apollo missions’ marginal costs were astonishingly high — but acceptable in the context of a race, since we did not have the time to set up the infrastructure, the needed service stations for fuel and food, along the way.

The lack of refuelability shapes every aspect of the Constellationany SLS-based architecture. Why is it that an SLS-based architectureConstellation, like Apollo before it, will discard each lunar lander? Because, at current launch costs, it is very expensive to deliver the propellants needed to reuse the lander — more expensive than the cost of the hardware. It is cheaper to simply throw it away and send another one. But if it were possible to refuel on the lunar surface, and in lunar or other high Earth orbits, a lander and other transportation elements could be reused many times for trips between those nodes. Reusing space elements like the lunar lander is feasible with current technology — but it would require the presence of “gas stations” at which such spacecraft could refuel, which in turn would require an infrastructure of transporting and manufacturing propellants at much lower cost.

Resources and Refueling

Broadly speaking, then, if we want to make human spaceflight affordable and sustainable, we must develop an infrastructure that makes it possible to refuel in space. A person might reasonably object that refueling makes no difference: we will still have to pay to get the fuel into space (or to the Moon or wherever), whether it is sent in the fuel tanks or is sent in some other container. What, in the end, do we save by sending fuel ahead of time? There are three responses to this.

First, if we had an infrastructure that allowed fueling in orbit, we could get by with much smaller launch systems, because we wouldn’t have to carry as much propellant merely to deliver propellant. This would save us the high development and operating costs of a heavy-lift launch vehicle. In addition, fueled vehicles (like those presumably planned for ConstellationSLS missions) must be built to handle the stresses of launch with their tanks full; if they could instead fill up after launching, their structure could be much lighter because accelerations in space are much more benign. This would further reduce launch costs as well as in-space propellant requirements since the vehicle would be less massive. (This “dry-launch” concept was one of the options presented in the Boeing CE&R study I consulted on.)

Second, in a space infrastructure that permits refueling, the means of getting propellant to a depot will not necessarily be the same as the means of getting other hardware there. After all, gas stations are resupplied by tanker trucks, not automobiles. This results in economies of scale, and it can also dramatically reduce transportation costs — because while manned spacecraft will generally be in a hurry to get places, propellant tankers can move more slowly so long as there is a steady supply of other tankers. Thus, while manned spacecraft must use high-thrust propulsion with low fuel economy, propellant tankers can use much more efficient propulsion systems, such as ion thrusters (an existing technology already used on communications satellites), dramatically reducing the cost of propellant delivery. As long as there is demand, a series of tankers in continuous motion would ensure that depots are always near full at low cost, just as there is a steady stream of slow oil tankers every few miles in the oceans between Japan and the Persian Gulf.

Third, part of the purpose of the original Vision for Space Exploration was to use the Moon as a steppingstone to other destinations. There is an abundance of oxygen that can be cracked from the silicates of the lunar regolith, and oxygen is a major component of rocket propellant. Yes, developing the capability to do so will cost money, but it would surely cost less than the tens of billions of dollars that NASA’s heavy-lift vehicle will cost. (If there is abundant ice on the Moon as well — a question that has not yet been settled — on-site fuel production will be much easier since there will also be a source of hydrogen, which would otherwise have to be transported from Earth.)

Not only might lunar resources be used to fill up the tanks of lunar landers, but the Moon might conceivably become a regular source of propellant, or at least the oxidizer component of propellant, for the entire fuel infrastructure. With a production infrastructure in place, propellant made on the Moon could become cheaper in space than propellant made on Earth and shipped to space, since it would not have to be freed from Earth’s gravity well. Later, propellant might be made from resources found on asteroids or comets (which, because they hold water, could provide the resources needed for both fuel and oxidizer), further reducing the demand for propellant made on and launched from Earth.

In short, a space-refueling infrastructure would vastly reduce the cost of propellant (the vast bulk of the mass required for extraterrestrial exploration), it would allow full reusability of all transportation elements (at first between Earth and the Moon, and eventually out into the solar system), and it would result in low marginal transportation costs and great scalability.

There is another key advantage to this approach, one that will be felt right away. Propellant is cheap — liquid oxygen costs about the same as milk — and almost infinitely divisible, so it can go into orbit on less reliable (and presumably less expensive) vehicles of all sizes, with cost being the deciding factor. This would create a market for reusable space transports that may not yet be trusted for carrying passenger or expensive cargo, but could deliver the low-cost payload of propellant at a low financial risk. This opens up business opportunities for anyone who wants to provide access to orbit to sell propellant into the fuel infrastructure. Propellant could be as fungible in orbit as oil is on Earth. This would satisfy two key requirements of the Aldridge commission that NASA has heretofore ignored — supporting commercial providers and incorporating international partnerships, without becoming too dependent on any single business or country. The proliferation of profit-seeking private enterprises heading into space could result in a robust diversity and redundancy of launch capability, so that if any particular launch system is temporarily grounded (as the space shuttle has been twice), overall access to space will not be devastated (just as temporarily grounding a particular type of aircraft does not shut down the airline industry).

In this space-refueling infrastructure, propellant would be cheaper, flight hardware wouldn’t have to be as heavy, and alternative launch vehicles would flourish. Every year that we starve the kind of research and technology that would make this possible and instead spend our money on mega-launchers like the Ares V is another year that we delay developing a truly sustainable space transportation infrastructure — and becoming a truly spacefaring people.

Safety and Space

The space-refueling infrastructure I’ve proposed is silent on the question of the optimal way of reaching low Earth orbit. This is a decision best left to the market. But NASA’s (or to be more accurate, Congress’s)ConstellationSLS/Orion approach is too costly and too much of a fragile monoculture to provide affordable and reliable access to space.

There are already several existing private launch providers, with more starting up all the time, and they are seeking a variety of innovative ways to put humans and cargo into orbit. United Launch Alliance (ULA, a joint venture of Boeing and Lockheed Martin) has two proven vehicles — Atlas V and Delta IV — that could be made safe for transporting humans. SpaceX has a rocket sitting on the launch pad in Florida, expected to have its first flight this yearanother successful flight with a turnaround of less than two weeks from the last one, in a few days; it was designed from the beginning to be capable of delivering SpaceX’s crew capsule, Dragon, to low Earth orbit. Other companies are developing reusable suborbital vehicles designed for high flight rates and low marginal costs; this experience will allow them eventually to advance to orbit, particularly if they can make money by delivering propellant as part of an extensive flight test program prior to operations with more valuable payloads. Bigelow Aerospace is building and launching inflatable orbital habitats that could, over time, be used as boarding houses for workers at orbital assembly facilities and propellant depots.

One issue that frequently arises in discussions about private spaceflight is the matter of safety. Will it be safe to trust our precious astronauts to private launchers?

Let us be clear: Perfect safety does not exist on this side of the grave. “Safe” and “unsafe” are not binary conditions. All we can do is to make things reasonably safe — keeping in mind such factors as expense. If our attitude toward the space frontier is that safety is paramount, that we must never lose an astronaut, then that frontier will remain closed. If our ancestors who opened the West, or who came from Europe, had had such an attitude, we would still be over there. It has never been “safe” to open a frontier, and space is the harshest frontier man has ever faced, but fortunately, we have sufficiently advanced technology to allow us to do it anyway, and probably with much less loss of life than any previous one.

NASA has developed a system for determining whether spacecraft or launch vehicles are safe enough for transporting its astronauts; spacecraft so deemed are said to be “human-rated.” In practice, this is a useless term, wielded arbitrarily and inconsistently over the years, and it serves to frustrate and confound serious analysis by its simplistic implication that there is a bright line dividing the safe from the unsafe. No NASA vehicle — including the space shuttle — has met the agency’s own human-rating standards since the 1960s.

Instead of using it as a true indicator of whether spacecraft are worthy of carrying passengers, NASA seems to use its human-rating system to protect agency jobs. Before he became administrator, Mike Griffin said in 2003 that the concept of human-rating was outdated and “no longer very relevant.” But, as engineer and space blogger Jonathan Goff has pointed out, Griffin changed his tune when he took over NASA, saying that “a bunch of changes” would have to be made to ULA’s Atlas and Delta rockets before they could be human-rated. Why? “In part,” Griffin said, because sticking with shuttle-derived rocketry “gives us the best work force transition issues.”

Given its arbitrary and politicized application, private companies should not waste their time formally having their spacecraft human-rated. Whether SpaceX and its Dragon capsule, or ULA and its launchers, private companies should make their spacecraft as safe as they can afford to without losing their customers. Private individuals willingly climb Mount Everest every year (and many have died trying); within reason, private individuals should be left to their own recognizance when it comes to flying on a private vehicle in space. NASA’s unwillingness to allow its astronauts to take similar risks bespeaks a lack of seriousness about our national space endeavors. When Columbia was lost in 2003, the nation’s mourning was made bitter by the realization that the astronauts died after circling the Earth rather than actually exploring beyond it.

Recommendations for the Augustine Next Inevitable Committee

Just as war is too important to be left to the generals, man’s future in space is too important to be left to NASA. After President Reagan proposed the creation of a national missile defense system in 1983, it became clear that the U.S. Air Force was not properly organized or motivated — and so a new agency was created to pursue the president’s vision. The new agency, today called the Missile Defense Agency, was very innovative and made great progress because it could focus on its one goal. Along those lines, the Bush administration might have done well to establish an Office of Space Development (with “exploration” being merely a means to an end) that could draw on other federal resources — not just NASA, but the Departments of Defense and Energy — as well as the private sector.

Of course, an independent space development organization with such power would be politically unfeasible. But that is part of the problem: our sclerotic space agency is subject to forces of legacy politics; it protects existing bureaucratic structures and emphasizes jobs over achievement; and it perversely rewards failure with more funds and punishes success with budget cuts. Short of an independent entity, the Augustine committee should at least revisit the Aldridge commission’s recommendation of converting the NASA centers to FFRDCs.

Assuming, though, that NASA in roughly its present form is here to stay, what should the Augustine committeeone recommend to put the agency back on the right course?

First, there is great irony (as space blogger Clark Lindsey has noted) in the fact that NASA has not successfully developed a launch system in decades, with many failed attempts, whereas it has developed many techniques and technologies for orbital assembly and operations — and yet it is pouring billions of dollars into the former and neglecting the latter. Critics often bemoan NASA’s abandonment of Saturn rocket technology upon the end of the Apollo era. But to abandon the orbital assembly and operations technology developed during the shuttle era — as the Constellation architecture implicitly does; it doesn’t even call for an airlock on the new crew capsule for the crew to conduct extravehicular activities — would be a much greater tragedy, because unlike the Saturn infrastructure it actually offers a path to a future of abundant low-cost space activities.

Donald Rumsfeld, the former Secretary of Defense, infamously remarked that “you go to war with the army you have.” NASA should have planned on going to the Moon with the launch vehicles it had and not those it wanted to have; in retrospect, the agency should have been explicitly forbidden from developing a new launch system. Billions have already been wasted in developing a redundant launch capability when the focus should have been on getting beyond low Earth orbit. The space agency must finally, after half a century, be a good customer, and provide a market not for cost-plus contractors to build hardware at their direction, but for private transportation services. The Commercial Orbital Transportation Resupply Services (COTRS) program should be revitalized with additional funding, new entrants should be invitedcontinued, and its role should be broadened far beyond the current charter to service the International Space Station — to supporting exploration itself. In addition, COTS DCommercial Crew (for delivery of crew to the International Space Station in addition to cargo) should be immediateproperly funded, to provide redundant means of getting passengers to and from orbit and the space station on American hardware. A robust COTScommercial program, in combination with a requirement that companies begin to deliver hundreds of tons of propellant into orbit each year, would provide enough traffic and competition among launch providers to finally start to drive down the cost of access to space. This would be a welcome change from the stagnation of high launch costs over the past few decades, and an improvement over the promise of still higher costs from ConstellationSLS/Orion. The aim should be to develop architectures that are not dependent on any particular launcher but that are redundant both in their ability to get to orbit and to travel between nodes beyond Earth.

Third, the savings from avoiding the development of unnecessary new launch systems should be spent on resurrecting the Research and Technology program initiated by Admiral Steidle. Specifically, NASA should work on developing the tools and techniques needed to store and transfer cryogenic propellants in orbit. The agency should begin to define requirements for (redundant) propellant depots, and perform studies on optimal locations for such depots. NASA should perform experiments in propellant handling at the International Space Station, and it should lease space in a Bigelow orbital habitat at low inclination as a testbed for orbital transportation support operations. The agency should do with its space transportation needs what the U.S. Postal Service did with its airmail needs back in the thirties: create a vibrant new transportation industry. And it should provide the kind of technology development support that NASA’s predecessor, the old National Advisory Committee on Aeronautics, did for aviation in the first half of the twentieth century.

Let us finally abandon our race with the Soviet Union, the race we won four decades ago against an adversary two decades vanquished and vanished. We don’t need to remake Apollo; we need to open up the new space frontier the way the old American frontier was opened. Let us unleash private enterprise and create not just jobs but true wealth. Let us innovate and find new ways for free men and women to use new resources. And let us work hard and risk greatly in the pursuit of our individual dreams — for it is those dreams, and our countless failures and triumphs along the way, that will determine man’s destiny beyond the Earth.

27 thoughts on “The SLS Frenzy”

      1. Autistic? That is a new one. Actually claiming after the fact that you’re just kidding is pretty lame. Accusing people of psychosexual disorders is also pretty weak.

        1. I wouldn’t have bothered to have to make such a “lame” claim if you, like others, had been able to naturally discern sarcasm. And who knew that autism was a “psychosexual disorder”?

          1. I was of course referring your revival of Missile Envy as a diagnosis of people who support big rockets.

  1. Caldicott was wrong (about a great many things). And Rand isn’t serious when he talks about the phallic symbol thing. The bottom line reason for SLS is to keep NASA’s workforce swollen and spread across some key congressional districts. It certainly doesn’t make sense from an engineering or economic perspective.

  2. By the way, considering that rumors that Musk is working on an even bigger rocket, what does Rand’s Freudian theory say about him?

    1. My understanding is that Mr. Musk will not be destroying the rocket immediately after every flight and then building an entirely new one for every mission, but rather will be reusing it over and over, quite unlike SLS and Orion. Oh, and it will be bigger and more powerful as well. Plus it will actually go to Mars I heard.

      And possibly build a city. What does your government have?

  3. I remember the first time I read that post and even though it had meatier content the line that always got me was…

    …lots of people wandering around and peering over the next rill…

    1. The comment over there that got me again was, “But it’s chump change compared to all the other things the government does.” My rebuttal (though I don’t bother over there) has changed to, “So if a movie ticket is now $100.00, you will buy it anyway because it is chump change to the rest of your expenses?”

      1. The problem is that while SLS is a money pit, almost everything else the government does is too. Focusing on cost alone isn’t going to convince people who like NASA but don’t know the details.

        So, in order to convince people SLS is bad, the arguments have to be based on the technical details. Flight rate is a big one but people will respond that we just launch more often. Flight rate problem solved! You have to go even deeper to show how the flight rate can’t be increased.

        A technical take down of SLS by a knowledgeable Alpha Nerd would be more persuasive, if it wasn’t too terribly long to read, IMO.

        One of the best arguments on cost though, is pointing out the opportunity cost of the alternative and not comparing NASA’s budget to other agency budgets. One SLS launch = X Falcon 9 launches and as a bonus, Falcon 9 is flying right now.

        1. I recently pointed out on reddit:

          This year’s SLS budget is about 3 billion. You could instead do ALL these:

          * Rebate 1 billion to the taxpayers
          * Purchase 20 (!) Falcon 9 launches
          * Spend 500 million to make 10 Mars/Europa/Lunar/Asteriod probes (launch already paid for)
          * Spend 500 million on Space Station experiments and astronauts (launch already paid for)

      2. We have a name for doing something to someone without there consent. It’s called rape and SLS qualifies as a rape of taxpayers. So a chump change argument is just, but “it’s only a little rape, quit yer complaining.”

        If I reach into your wallet and take some money, how much before it’s theft?

        Once they decide any little amount is not theft or rape or is simply justified for the greater good we’ve opened the door to taking everything. Obama Senior believed 100% taxation was fine because they would give you everything they decided you need.

        Rapists and thieves, the whole bunch of them.

          1. That there’s a whole can of worms, Ken, and not just about SLS. Once that can gets opened, nobody will pay much attention to the tiny part that is SLS.

  4. I agree wholeheartedly about SLS, but…

    Rand, do you usually respond to yourself on twitter? 🙂 (that’s how your twitter quotes look anyway)

Comments are closed.