Our manned space infrastructure is extremely fragile--even brittle.

NASA just found a problem with the Space Shuttle Orbiters Atlantis and Discovery. They have hairline cracks in some jackets in the main propellant lines.

These aren't cracks in the lines themselves, so there's no danger of a leak of oxygen or hydrogen, but they could result in a small piece of metal getting ingested into the engines, which could potentially cause anything from a premature engine shutdown to a turbopump explosion. Considering that, for the brief few minutes that it operates during each flight, each of the three fuel turbopumps generates about 70,000 horsepower (or about fifty megawatts--equivalent to a small power plant or dam), that could be a pretty spectacular show.

If it happened during flight it would be comparable to the Challenger disaster. If it happened on liftoff, it could take out the whole pad, along with the Shuttle and crew. Even in the event of a simple engine shutdown early in the flight, a Return To Launch Site (RTLS) abort would be exciting, and dangerous, and it's never been attempted. No matter what, it would make for a very bad day for all aboard.

NASA doesn't yet understand how the cracks got there (my own "recovering engineer" guess is fatigue from thermal cycling as they're repeatedly soaked in very-cold liquified gases over several years), how long they've been there, and whether or not the other two vehicles have them. As a result, they have prudently grounded the entire Shuttle fleet indefinitely. Indefinitely, in this case, means not that it will be necessarily a long time, but that they can't say how long it will be, which means that Columbia's mission to the space station next month will almost certainly be delayed until they do have some answers.

I've written before about the high costs of space due to lack of economies of scale, but our minimal activity level causes other problems as well. It makes it difficult to afford a robust and resilient space transportation infrastructure.

In 1979, when a DC-10 literally lost an engine and crashed in Chicago, the whole McDonnell-Douglas DC-10 fleet was grounded, but it didn't shut down the airline industry, because there were hundreds of aircraft of many other makes and models that weren't affected.

In contrast, we learned in the Challenger breakup the danger of relying on a single launch system, with a small number of vehicles, when grounding it means putting all activity on hiatus. A loss of an Orbiter would constitute the loss of a quarter of our fleet. The loss of another one after that would be another third of the remainder. And grounding the fleet to avoid this may result in more delays to the beleaguered space station program.

NASA has studies underway to look at solutions to this problem, such as the Space Launch Initiative, or the Alternate Access to Space program. But these programs seem to be stuck in the same mode of thinking that gave us Shuttle. People talk about "the" Shuttle replacement, or "the" next-generation launch system, as though there will be only one, because no one can imagine market or funding for more. And all the focus remains on technology and vehicle concepts, which are beside the point.

No one in the government seems to recognize our real problem, which is the currently infinitesimal market size for space transportation. NASA continues to pay the traditional aerospace contractors for traditional solutions, and ignores the fact that we need a diversity of approaches and providers. Such a diversity can only be supported by a large, vibrant and growing commercial demand for space transportation services.

There is an old tale, about "for lack of a nail...a kingdom was lost."

As long as we, as a nation, refuse to acknowledge the problem with our space markets and approaches, we will remain in our current state of fragility, in which the fate of a multi-billion-dollar space station, that, for all of its cost, can only support three people, is held hostage to the whims of microscopic slivers of metal in frigid propellant ducts.

[Update on Thursday morning]

I screwed up--Columbia is flying a Spacelab mission next month, not an ISS mission. Columbia is too heavy to get to ISS.

Rand, is RTL even possible? Do you know of any tests, or is this stuff all theoretical?

I don't know. I'm not familiar with that TLA...

What is RTL?

There's a rumor floating around that a complete replacement of the liners could ground much of the orbiter fleet for two to three years, mainly because the tooling doesn't exist for making them any more. Does that seem plausible to you?

Yes, it wouldn't surprise me at all. There are many component in the Orbiters for which the vendors went out of business long ago, but I though most of them were for primary structure, like spars and keel.

Many of them might be makeable out of new NC machine tools, but I don't know what it would take to get them through NASA's certification process.

What's the source of the rumor? JSC or USA? Or Boeing?

Rich Kolker passed it on to me. He might have gotten it from JSC.

I just wanted to point out an inaccuracy in your article. STS-107 was not going to go to the International Space Station. Columbia does not have an ODS (Orbiter Docking System) and therefore cannot dock with the station. STS-107, Columbia's next mission, was going to be a Spacehab mission.

Sorry I missed your correction. The largest constraint to Columbia not going to the station is the lack of the ODS though, not that it's too heavy. Although it is heavy and would not be able to take up close to the amount of payload the other orbiters can.

Yes, I should have said, "too heavy to get there with useful payload." Though even that's apparently not quite true, since I was informed in sci.space.policy that is in the manifest for at least one ISS mission in the next couple years. I don't look at the manifest that much, myself.

RTLS, is it possible? Has it been tested with payload weights?

Inquiring minds want to know!