…of the Shuttle program. With Atlantis in orbit for the last time, I have some cautionary words over at Popular Mechanics.
[Update a few minutes later]
I also have a piece up over at National Review Online — is the era of big-government space programs over?
[Update at 12:45 PM PDT]
I’m on hold right now with Bill Carroll on KFI-AM in LA, who’s been bewailing the end of the space program. Don’t know if I’ll get on or not.
[Update a while later]
The segment ended before he got me on, but I’ve been in an email discussion with his producer, so maybe next week.
Rand, my main concern about your Popular Mechanics article is your response about “reusable launch vehicle not cost effective.”
Is your argument that the Shuttle is not fully reusable hence any proposition on full-reusable hasn’t been tested? I think you article suggests that the non-reusable parts — the External Tank and Solid Rocket Boosters (OK, OK, partially reusable on those) — are not the problem. Your are suggesting (as many other do) that the Orbiter itself is the hanger queen.
So then, how does the problem with the Orbiter have anything to do with the “cost cutting measures back in the ’70’s”?
Don’t have the title handy, but there is a book out following the history of Shuttle designs and how-they-arrived-at-the-Shuttle-as-it-turned-out. The Shuttle was originally to have been a fully reusable two-stage-to-orbit vehicle consisting of a Booster and an Orbiter spacecraft. Yes, the External Tank and SRB’s were a cost-saving compromise, but the original Shuttle concepts (there were several) were simply huge vehicles, both Booster and Orbiter requiring reentry thermal protection. Their size speaks to technological uncertainty.
The other part of the put-money-into-a-full-reusable-design vs how-the-Shuttle-turned-out camps is that the full reusable design will cost more up front, and do you ever get that money back spread over enough flights? That is, can you fly often enough to justify the up front costs?
I think the Way Forward to come out of the Shuttle experience is to develop a reusable spacecraft, perhaps a blunt reentry vehicle instead of a wings-and-wheels Orbiter, and simply launch it on an expendable booster, perhaps working on recovering the first stage down the road. That is what Elon Musk is doing with Dragon and Falcon, no?
What NASA probably should have done, subject to the constraints of NASA-ness, was to have had Big Aerospace bid on whatever spacecraft fit on top of an EELV Heavy of some stripe, and have had Big Aerospace bid on whatever type of “human rating” of the booster was required.
Failing that, they should have gone with one of the proposals to use existing ET’s and SRB’s, essentially develop one vehicle, something along the lines of Shuttle C, Direct, whatever you call it, that is, if the purpose of the Shuttle is to preserve jobs in Utah. That one vehicle would have been undersized for launching the Antares Moon lander and the leave-Earth-orbit stage, way oversized for the crewed spacecraft, but the bulk of the booster or the rocket stage is never the cost driver — there is that example from the MCD (minimum cost design or “Big Dumb Booster”) people that for Project Discoverer/Corona, the Agena upper stage cost way more than the big dumb Thor rocket.
It is not so much fully vs partial reusable vs expendable, multiple launches vs heavy lift booster, winged reentry vs capsule and parachute. It just seems that owing to political and institutional constraints, NASA cannot design its way out of a wet paper bag.
There is one might-have-been in connection with the Shuttle, and that is Max Faget’s straight-wing Orbiter vs going to the delta wing with high cross range that they ended up with. It was been said that the delta wing was driven by DoD polar orbit missions that never materialized, and that delta added all kinds of weight and complication.
Max Faget’s “DC-3” Shuttle proposal or whatever it was called, was not only a fully reusable Booster/Orbiter spacecraft pair, not only was it straight-winged on both Booster and Orbiter, it was also much less payload capacity — was it something like a mere 10,000 lbs? The idea was that the Shuttle would be, well, a shuttle — kind of like my Mom telling me, “don’t carry all that stuff from the car all at once, make multiple trips!” In other words, a Shuttle would be tied into the idea of frequent but small payloads to orbit and some form of on-orbit assembly of cis-lunar and interplanetary spacecraft and of (ta dum!), Rand’s drumbeat about propellant depots.
But the idea that intrigues me is the straight-wing reentry vehicle and that I don’t think people ever understood it. The knock on the straight-wing orbiter is that straight wings are dangerously unstable in the hypersonic reentry regime.
Max is no longer with us, but he was one smart guy, at times called the American Korolev, although in his modesty, Max asserted his contribution was never that big. But the man knew a thing or too about the reentry problem, about ballistic blunt bodies, and about bodies with a small amount of lift (Gemini and Apollo RV’s).
My understanding of the straight-winged orbiter was not that the straight wing would “fly” at hypersonic reentry speeds in any manner that the X-15 flew on such wings. Rather, the orbiter would reenter belly first, essentially pancake into the atmosphere. In the hypersonic flight regime, the bottom of the Orbiter and its straight wing would be as if you took a cookie cutter to an Apollo heatshield. Think also of those Road Runner cartoons where Coyote (wearing roller skates and powered by a sky rocket pushing on his back) smashes front-on into a brick wall, leaving a Coyote-shaped outline in the bricks.
Hence, the straight-wing orbiter would reenter just like an Apollo spacecraft — as a low-lift blunt body, and Max probably had enough experience with that mode of reentry to get the straight-wing Orbiter to work.
The only hangup I can think of us that the Orbiter comes plowing into the atmosphere belly first with those straight wings in a full stall, and once at subsonic speed, it needs to recover from that stall and start flying on those wings. But again, Max probably knew more about aerodynamics and I ever will or perhaps many other people.
I’m in complete agreement Rand, but publishing an article like that almost demands that you stick your neck out and offer an opinion on how you think reusable should be done, and how big a launch vehicle is needed?
X-15 or SS-1 are examples. Perhaps X-37, but the launcher is currently disposed.
I admit, scaling is required, but then X-33 was on the track, so was DC-X.
DC-X and SpaceShip-1 proved that reusability isn’t a fantasy, nor an unattainable goal. The trick for RLV SSTOs is to maintain a high enough mass fraction and a high enough ISP to give useful payload performance. I think long term the best chance for this is with LOX/Methane and liberal use of composite materials, though there will be steps to reusability long before then.
Regardless, it shouldn’t require presenting an alternative RLV design to debunk the idea that a vehicle which is overhauled in a months long process and requires newly built fuel tanks and boosters for each flight is any test of reusability, or that a mismanaged government program that attempted to use multiple bleeding edge technologies simultaneously is either.
I forget who it was, Henry Spencer perhaps, who called this practice of abandoning an entire design after one failed execution as “Wile E. Coyote engineering” but the term is quite apt.
I stole that from Henry for my first piece in The New Atlantis, and he told me that he got it from Mitchell Burnside-Clapp.
1. The Shuttle Proved That Reusable Launch Vehicles are Not Cost Effective
I think we can all be grateful that Elon Musk took this “false” lesson to heart.
I don’t expect to see Rand taking Musk to task for it on the Popular Mechanics site any time soon, though. 🙂
“The Final Launch” was right on. One of your best. I thought your false lessons was not up to your usual standard.
Now that I’ve said that I should probably say why or what about it?
The shuttle may not have proved anything, but that almost suggests there are no lessons. But then you did go on to point out the most important lesson we did learn. So I guess that’s ok.
Death of astronauts seems a bit of a strawman. …and I think it is bad form to knock grieving even with your explanation which I actually agree with. They were volunteers that had a better conception of the risk than most and would do it again no doubt if offered the chance (evidenced by the fact that we sent others up.)
I know my justification for dissatisfaction is weak. Just sayin’.
I think we can all be grateful that Elon Musk took this “false” lesson to heart.
What is that supposed to mean?
> I think we can all be grateful that Elon Musk took this “false” lesson to heart.
Huh? Reusablity is one of Musk’s big goals.
Yours,
Tom
“Reusablity is one of Musk’s big goals.”
With the Falcon launch vehicles? Is that going to be an improvement on the reusability of the SRB’s?
“I admit, scaling is required, but then X-33 was on the track, so was DC-X.”
The original TSTO shuttle designs never got off the drawing boards because they were deemed too expensive, SSTO with rockets all the way gives lousy PL ratios, air breathing Skylon is supposed to cost 12 billion, what’s the bet that price tag’s going to grow? What orbital launch system is reusable and affordable, my favorite is a large version of Mitchell Burnside-Clapp’s APT, but other ideas are also possible without a huge and expensive winged first stage.
Or am I looking at this all wrong, does a change to private rather than state control mean that even discussing such things is pointless?
@Paul Milenkovic:
*reaches behind my head*
Probably “Space Shuttle: The History of Developing the National Space Transportation System” by Dennis R. Jenkins.
I was using the book as a pillow prop as I’m surfing the Internets laying down, which lets my cat use me as a heating pad.
Years ago I read an interesting engineering argument that we were trying to do reusability backwards. Given the rocket equation, adding mass and components for reusability incurs vastly less performance penalties when you focus on reusing the first stage, then perhaps the second stage.
On the first stage an extra pound toward reusability will add a little over a pound to the launch weight, whereas adding an extra pound to the orbital vehicle adds dozens of pounds to the launch weight.
But of course your crew capsule necessarily returns intact (maybe with some scorching), so it makes sense to reuse or refurbish that. Re-using a giant payload bay was probably a performance killer.
Anyway, speaking of weird configurations, has anyone ever looked at using a biplane? During re-entry the top wing would be in the shadow of the lower wing, so it wouldn’t need much thermal protection at all. To get even weirder, make it a flying boat so the landing gear and gear doors disappear.
Reusablity is one of Musk’s big goals.
I think Musk has the whole reusable/expendable question correctly prioritized.
I think Musk has the whole reusable/expendable question correctly prioritized.
Again, meaning what?
Well, why DC-X and X-33 are SSTO reuseable design, perhaps the issue is why SSTO? SS-1 and X-15 are not what I would call SSTO, but then they weren’t orbital. I do think there is plenty of room and capability to build something between SS-2 or X-15 and what might have been X-33. It may need a re-useable launch platform that starts at altitude. SSTO is a sexy one piece, but sexy doesn’t get you as far as being practical.
…reusability incurs vastly less performance penalties when you focus on reusing the first stage, then perhaps the second stage.
Second stages generally go to orbit. So refueling should make them reusable as EDS. So what we need is a low maintenance and small labor reusable first stage.
The first stage has to get through atmosphere… that buoyant stuff. Hmmm…
Spending a bit more than 2 or 3 minutes there means not designing for max-Q either… I see BALLOONS.
Balloons could also replace parachutes or at least the drogue chute for recovery of the F9 first stage. You would not have to worry about parachutes not inflating at high speeds in thin atmosphere. They could wait until the balloon provided enough drag before deploying the main chutes.
I would not be surprised if SpaceX next big achievement is recovery of the first stage.
Again, meaning what?
Which word is giving you trouble, Rand? I think I was being clear. If Musk never recovers and reuses a stage or an engine he can be competitive and his business can succeed. Reusability is not in the critical path. Contrast that to the schemes mentioned upthread where it certainly was.
And yet, he still says that if he doesn’t reuse the vehicle, at least eventually, he will have failed.
So I continue to miss your point. Obviously he is not counting on it in the short term. That is simply common sense. That doesn’t mean that he considers it unimportant.
And I also fail to see what this has to do with what I wrote at PM.
And the plan for Dragon-with-crew is still to use parachutes only in backup situations.
And yet, he still says that if he doesn’t reuse the vehicle, at least eventually, he will have failed.
And he still says he wants to go to Mars. Maybe he thinks he will have failed if he doesn’t make it. He sets high goals for himself. That’s very admirable. In any case I think posterity will have a more generous verdict than failure.
So I continue to miss your point. Obviously he is not counting on it in the short term.
That’s another admirable quality Musk has. He grasps the difference between what can be accomplished in the short term and what might only be possible in the long term. Many can’t.
That is simply common sense.
And yet I suspect if he had asked your advice 10 years or so ago you would have advised something much more ambitious.
That doesn’t mean that he considers it unimportant.
Just way down the priority list.
And I also fail to see what this has to do with what I wrote at PM.
I think Musk ultimately found the arguments of his staff against the cost effectiveness of RLVs to be persuasive, despite his personal enthusiasm for them. Indeed, I think the arguments went further, questioning the feasibility of RLVs. And I think the example of the shuttle and other RLV projects, both public and private, loomed large in those arguments.
I had this same discussion with the late Len Cormier a number of times. He claimed that launch vehicle development was hijacked by the missile lobby a half century ago but yet was effusive in his praise of Musk for developing what was effectively another missile. I thought that was inconsistent.
And yet I suspect if he had asked your advice 10 years or so ago you would have advised something much more ambitious.
I would have advised something different. Whether or not it was “more ambitious” would have been in the eye of the beholder. I’m glad that he’s been successful in his approach, even though it would not have been mine, and I am also glad that he continues to recognize the importance of reusability, regardless of the route by which he plans to get there.
I think Musk ultimately found the arguments of his staff against the cost effectiveness of RLVs to be persuasive, despite his personal enthusiasm for them.
Then why is it that he continues to pursue a reusable vehicle?
A quibble. I think “reusable” is too broad and vague a term for something as technical as aerospace engineering.
We don’t call automobiles or commercial aircraft “reusable,” since we don’t mentally catagorize them as something that can be used “probably more than once, if not dozens of times. Yet the troublesome term covers concepts from that, where you just top up the fluids every now and then, to the Shuttle.
Our earlier non-reusable space capsules were probably more reusable than the Shuttle, had we chosen to pursue it, and we did reuse one Gemini capsule. In fact, our earlier capsules would certainly have a shorter turn-around time and vastly lower man-hours than servicing the Shuttle. But we never called them “resuable.”
I think a more appropriate term for the Shuttle would be “overhaulable.” Though that’s not a word, perhaps it should be, yet that too doesn’t quite describe the situation. Any craft can be overhauled after every flight, but only the Shuttle has to be overhauled after every flight.
NASA calls it “servicing”, as if they just pull it into a service station and have Billy Ray give it a quick once-over, which fails to convey the task. If you pull into a service station and half of Ford’s entire workforce spends two months working on your car, it’s not being serviced, it’s getting a more thorough tear-down and examination than the Men In Black would give to a crashed UFO at Roswell.
As an engineering parameter that was obviously badly missed on the Shuttle, any concept of “reusable” needs to be very closely broken down and examined early and often in the design process. With a vehicle designed so close to the margins on so many critical systems, there needs to be a very realistic assessment of the tear down, inspection, maintenance, refurbishment, and sign-off on the components.
The Shuttle’s designers can rightly blame political externalities for its expensive configuration, but the staggering error in their turn-around times and operational costs are probably in large part due to lack of any prior attempt to reuse a rocket, especially one as wildly complicated as the one they built. The aerospace community didn’t have some a maintenance manager who could answer the question, “What do you go through to check out a reusable rocket that’s been launched into outer space and recovered?”
For example in early testing the SSME’s would destroy their turbopump bearings within 2.3 seconds, each and every time the engine was fired. It took NASA and Rocketdyne a long time to get the problem corrected by discovering and stopping a type of swirl mode unbalanced flow condition by adding some little bump in the casing to trip the fluid flow. Yet at the same time that NASA was selling the Shuttle concept as cheap and frequent access to space, someone should’ve thrown up a red flag and said there’s no way its returning engines would be wiped down, kicked, and declared good to go. They were going to be taken out, disassembled down to the bearings and seals, put under a microscope, x-rayed, reassembled, reinspected, and retested almost every single flight.
The false lesson from NASA, that reusable isn’t cheaper, has never been tested in one critical sense. What NASA showed was that “overhaul required” isn’t cheaper. Third shift assembly line workers can put an engine or other critical space systems together because all the component parts they’re using are known good and within specs, and each person can do a tiny step in the process. Continually reverifying that each of those components are still within tolerance, involving disassembly and inspection, and the vastly larger decision tree, expertise, and knowledge base at each step of that process, is NOT necessarily cheaper.
The Shuttle’s yearly operational costs never dropped below its design and construction costs, even though the whole concept was high front-end costs to get low operational costs. The failure to rigorously and realistically nail down what “reusable” really meant, component by component, and designing a system at the cutting edge where its difficult to provide even ballpark estimates, is probably the reason.
So when we discuss Elon Musk’s opinions on reusability, I think we should recognize that he might have to break new ground one something the Shuttle designers didn’t quite grasp, as if “reusable” was a marketing term and not a difficult engineering question where the wrong answer can leave an organization as large as NASA trapped by its own hype and then ensnared by the politics produced by its own unanticipated new workforce.
The day it was announced that NASA would be demanding zero reuse of Dragon capsules, even for cargo resupply, was anyone surprised? I know I wasn’t. It’s not just that they don’t care about cost.. it’s that they specifically never want costs to go down. It would change the status quo.
If NASA is demanding zero reuse of Dragon capsules, that means SpaceX can offer refurbished capsules to non-government users for lower prices, without running afoul of federal contracting regulations (which demand the government not be charged more than any other customer).
I would have advised something different. Whether or not it was “more ambitious” would have been in the eye of the beholder.
Just what would you have advised? This beholder is very interested.
Then why is it that he continues to pursue a reusable vehicle?
I think he pursues a reusable vehicle because that is a lot cheaper than actually committing to build a reusable vehicle. Much like Boeing or Airbus continuing to pursue an SST while continuing to design and build subsonic airliners.
Just what would you have advised?
I would have advised him to get to orbit through suborbit.
I think he pursues a reusable vehicle because that is a lot cheaper than actually committing to build a reusable vehicle. Much like Boeing or Airbus continuing to pursue an SST while continuing to design and build subsonic airliners.
One of these things is not like the other. In multiple ways. In fact, I see no similarities whatsoever.
@Paul D:
That sounds like the crazy law prohibiting our whiskey distillers from reusing bourbon barrels. As a result Kentucky has a thriving bourbon barrel cooper industry and Scotland, Ireland, France, and other countries get really cheap once-used barrels (for making Scotch, wine, brandy, and tequila), not to mention all the barrels people around here use for bar decorations, tables, and potted plants. ^_^
I’m sure museums would prefer non-reusable capsules because it makes them easy to get, easy to describe (this is the capsule that flew mission 123A), and easy to display, since a capsule can fit in a corner, whereas a Shuttle takes up a whole hanger).
Of course the other benefit of non-reusable capsules is keeping the production line open and active, making it vastly easier to incorporate major design revisions and introduce entirely new models.
An interesting parallel to the Shuttle’s effect on space vehicle development popped up in computer simulations of biological evolution. When evolving CG organisms were made immortal, the rate of evolution plummeted because a few very fit breeders stayed around forever, dominating the gene pool from that point forward, reducing genetic diversity and essentially halting change and progress.
“Overhaulable” is a word George. Words are defined by people not dictionaries. If nobody has ever used that word before then the honor is yours. It’s about time we stop fawning over false intelligence and respect the true.
Are Jim’s statements another example of confirmation bias? He knows Elon has changed his mind when perhaps Elon himself doesn’t fully know? Elon’s great talent (among many) is humility (the ability to learn.)
I’m very happy Elon chose the path he did. He’s sent quakes through the whole industry (not shutting up the naysayers, but allowing the rest of us to be confident of their idiocy.)