Steve Lindsey, head of the astronaut office, makes the latest attempt to defend their disastrous architecture choices, in the Houston Chronicle.

So what should we replace the space shuttle with? We need a replacement spacecraft and launch system that can fulfill two roles. First, it must be a vehicle that can support the space station. Second, it must be a system that can take us beyond low Earth orbit and back to the moon in accordance with NASA’s exploration plans, then possibly on to more distant destinations.

Note that he makes no mention of commercial services for supporting ISS, which is the primary purpose of COTS. With COTS D, this can include crew transfer (and potentially even lifeboat duty — something that the Shuttle cannot do). But if NASA develops a new vehicle to “replace Shuttle” (even though it’s not a replacement at all, in terms of functional requirements), it will be competing with the private sector for that work, and reducing the potential market that it needs for success. If NASA is going to develop a new vehicle at all, it should be focused on missions beyond LEO, as was the original intent of the VSE.

Perhaps this spacecraft will be a core building block for Mars missions that will happen in our future. If this new spacecraft has no capability beyond performing station missions, then the system will become obsolete when station operations end. We would have to start over to design and build yet another costly new spacecraft. Instead, we need a system that can serve as a bridge between the end of one program and the beginning of the next.

Again, the exploration vehicle is what the focus should be now, not a vehicle to get them through a few years of ISS operations that competes with the private sector.

The architecture NASA has chosen to replace the space shuttle has its roots in components and concepts from both Apollo and the shuttle. The rocket, called Ares I, consists of a two-stage launch vehicle based on an upgraded shuttle solid rocket booster and an improved upper-stage engine from the Apollo program. Both are proven, reliable, human-rated components.

This is simply false. The SRB was never “human rated,” because the Shuttle was never “human rated.” And even if it had been, the SRB that will be used as the first stage of the Ares is essentially a new system, never flown, with additional subsystems and five segments instead of four. And the new J-2 is a new J-2, again, with so many upgrades and mods it might as well be almost a new engine, and one can’t conclude from the fact that its ancestor was “human rated” (whatever that means in the context of an engine) doesn’t make it so. It won’t be “human rated” until it’s…human rated.

As I’ve said many times, the use of the phrase “human rated” almost always obfuscates rather than enlightens, and its most common use by NASA is to baffle with BS. Following Humpty Dumpty’s lead, it means whatever NASA wants it to mean on any given occasion.

NASA spent several years studying architectures and researching every available commercial rocket to find the design that could best accomplish those dual mission objectives in the safest, most reliable way. The conclusion we reached was that a space shuttle and Apollo derived vehicle was the best choice.

Yes, and then after all those years of studies, they completely ignored their results, and did a sixty-day study rigged in advance to give them the answer that the new administrator wanted. They have never provided the basis of the trade studies, and the assumptions behind it, that resulted in their conclusion.

Many have challenged this decision and still are advocating the use of existing commercial launch vehicles. Those have been studied extensively and fall short of our requirements for several reasons. First, existing commercial launch vehicles can lift only a fraction of the mass required for station and lunar missions in a single launch.

Kind of like Ares 1, which cannot do a lunar mission in a single launch.

Second, existing commercial launch vehicles have been designed and built to carry unmanned payloads; they would need to be heavily modified to meet our human rating requirements.

They have been designed and built to carry very expensive unmanned payloads. They have been designed to a state of high reliability, because both their customers and their insurers demand it. To quote the NASA administrator from Congressional testimony a few years ago:

“What, precisely, are the precautions that we would take to safeguard a human crew that we would deliberately omit when launching, say, a billion-dollar Mars Exploration Rover (MER) mission?” he asked. “The answer is, of course, ‘none’. While we appropriately value human life very highly, the investment we make in most unmanned missions is quite sufficient to capture our full attention.”

The Atlas 5 and Delta 4 EELVs, he noted, have a specified design reliability of 98 percent, in line with experience with the premier expendable vehicles to date. If such a vehicle was used to launch a crewed spacecraft equipped with an escape system of just 90 percent reliability, he noted, the combined system would have a 1-in-500 chance of a fatal accident, “substantially better than for the Shuttle.”

So I’ll ask Chief Astronaut Lindsey the same question I asked Dr. Griffin a few days ago. What changed? When did these vehicles suddenly become death traps? As usual, I have no expectations that a reply will be forthcoming.

A core goal of the Astronaut Office is that the next launch vehicle should be an order of magnitude safer than the previous vehicle. That’s a goal the Ares-Orion architecture can meet. Current commercial launch vehicles, even if they could lift the mass needed, cannot meet that goal without extensive modifications in structural strength, the addition of new launch abort systems, significant alterations of flight termination systems, addition of cockpit command and control of systems, addition of manual control, addition of redundancy and robustness in several critical systems and subsystems, and an entirely new second stage to provide adequate lift and abort performance during ascent.

What “modifications in structural strength” are required? The launch abort system, last time I checked, was part of the Orion program, not the Ares 1 design, so it doesn’t have one, either. What “alterations of FTS” are required? What does he expect to “command and control” from the cockpit on a Delta or Atlas? Where will he add “redundancy and robustness” where it doesn’t already exist to maximize the probability of a successful satellite delivery to orbit? As for the second-stage issue, that might apply to a Delta, but there’s no problem with the Centaur of which I’m aware. And even if all these things were true, how can he justify the notion that they would cost more than designing an entirely new vehicle from scratch?

In our final Exploration Systems Architecture Study, the shuttle-Apollo derived launch vehicle, besides being the only design that could meet our mission requirements, was the highest-rated for crew safety. In fact, it was about twice as safe as any other option. This option also proved superior to others in terms of cost and schedule.

Again, until they show us the study, and the assumptions, these are not facts, but assertions, and from a source with a conflict of interest.

Beyond the engineering discussion, there is an even larger question; why do this? Why continue to invest in human space exploration? I believe this is an investment our nation can’t afford to pass up. President Kennedy’s 1962 speech at Rice University said it best: “We choose to go to the moon in this decade and do the other things, not because they are easy, but because they are hard.”

Funny, I choose to do things because they are useful. The notion that simply being hard can justify a multi-billion-dollar program, and particularly one in which the chosen solution makes things even harder (and more expensive), is one of the great false myths of the space age. As I wrote years ago:

“Because it is hard” has long become a convenient mantra for the current way of doing business.

When things don’t go right, “because it is hard” always provides the people doing them with a convenient excuse for failure — even forty years on, and even in the face of obvious management disasters. They can ask for billions of dollars for a new program, “because it is hard.” And when it screws up, they can say, “see, we told you it was hard — we just proved it. Apparently, you have to give us even more money.”

It makes it harder to get other funding sources, or try other approaches, as well. “Because it is hard” means that only a government agency can do it, and any investor who puts money into a private space venture might as well throw it on the table in Vegas, or onto the compost pile.

“Because it is hard” means that very few get to go into space, and that the only way to do it is the NASA way — study your math and science, figure out what kind of personality traits and characteristics they want, apply to be an astronaut and then hope that, against all odds and the other hundreds or thousands of applicants, you’re accepted. Then hope that they eventually get from a three-person station to a six person station and you actually get a chance to fly sometime before you have grandchildren and retire.

But NASA continues to perpetuate it, and too many people continue to buy it.

[Early evening update]

This quote from Mike Griffin above, is worth repeating, because it makes another point about ‘human rating” the Shuttle.

“What, precisely, are the precautions that we would take to safeguard a human crew that we would deliberately omit when launching, say, a billion-dollar Mars Exploration Rover (MER) mission?” he asked. “The answer is, of course, ‘none’. While we appropriately value human life very highly, the investment we make in most unmanned missions is quite sufficient to capture our full attention.”

I once asked the question: if we had decided to build the Shuttle without the capability of carrying crew (assuming that it’s meaningful to talk about a Shuttle without a crew, since carrying crew and acting as a temporary space station was one of its fundamental features), how would we have designed it any differently for more reliability? The answer is, probably not in any way at all (other than putting in actuators for nose-wheel steering, dropping the gear, and braking). It would have had just as much redundancy and structural margin as for a crewed version. (I recall, early in my career, getting into an argument with a structural engineer who was working for me who wanted to put in lower margin than 1.4 on an unmanned reusable vehicle, because he had been taught that there are different margins for manned than unmanned systems, not understanding the underlying economics).

Why? Because each one of the vehicles cost on the order of two billion dollars, and would probably cost more than that to replace, because a lot of the subs were shut down once the fleet was built, and would have had to be reactivated to build a replacement (one reason that procuring “structural flight spares” prior to Challenger was a foresighted decision — it allowed us to build a replacement at reasonable cost). We couldn’t afford to lose them. We want to get the vehicle back every time, regardless of whether or not anyone is aboard. Unreliable reusable vehicles are unaffordable.

I’ve gotten criticized in the past for pointing out this basic truth, and it will probably happen again, but it remains true. We have many astronauts, and we can produce more as needed at much less marginal cost than two billion dollars per crew complement, but we don’t have very many orbiters.

We never “human rated” the Shuttle, because it was supposed to be intrinsically safe, because it was important to get the vehicle back. And the Shuttle never met (and still doesn’t meet) the criteria for human rating, because (when you can pin them down on what it means) one of the things NASA considers essential for human rating is zero-zero (that is down to zero altitude and zero velocity) abort capability. The Shuttle has no abort capability from launch until two minutes into ascent, after SRB separation, and no practical means to provide it. But it was (at least until Challenger) considered “safe.”