Transterrestrial Musings  


Amazon Honor System Click Here to Pay

Space
Alan Boyle (MSNBC)
Space Politics (Jeff Foust)
Space Transport News (Clark Lindsey)
NASA Watch
NASA Space Flight
Hobby Space
A Voyage To Arcturus (Jay Manifold)
Dispatches From The Final Frontier (Michael Belfiore)
Personal Spaceflight (Jeff Foust)
Mars Blog
The Flame Trench (Florida Today)
Space Cynic
Rocket Forge (Michael Mealing)
COTS Watch (Michael Mealing)
Curmudgeon's Corner (Mark Whittington)
Selenian Boondocks
Tales of the Heliosphere
Out Of The Cradle
Space For Commerce (Brian Dunbar)
True Anomaly
Kevin Parkin
The Speculist (Phil Bowermaster)
Spacecraft (Chris Hall)
Space Pragmatism (Dan Schrimpsher)
Eternal Golden Braid (Fred Kiesche)
Carried Away (Dan Schmelzer)
Laughing Wolf (C. Blake Powers)
Chair Force Engineer (Air Force Procurement)
Spacearium
Saturn Follies
JesusPhreaks (Scott Bell)
Journoblogs
The Ombudsgod
Cut On The Bias (Susanna Cornett)
Joanne Jacobs


Site designed by


Powered by
Movable Type
Biting Commentary about Infinity, and Beyond!

« Didn't He Actually Read The Story? | Main | Maybe Brits Don't Make Suicide Bombers After All »

Wrong Lesson

There are many lessons to be drawn from the programmatic disaster that is the Shuttle program. Unfortunately, many of the most popular ones are wrong. Alan Boyle has apparently succumbed to one of them, when he writes:

Of course, it takes a lot of innovation to replace a manned spaceship capable of putting almost 30,000 pounds of cargo into orbit. I’m betting that Rutan would favor another approach, sending the payload and the people on separate breeds of spacecraft. It turns out that NASA’s new management tends to feel the same way — a big reason why the shuttle fleet’s days are numbered.

I suspect that to the degree that Burt has given much thought to the matter (probably not a lot--he's focused on his own markets, not NASA's), he would do things the same way that aviation does them--a single aircraft design for both cargo and passengers. The notion that the problem with the Shuttle was mixing crew and cargo is one of those popular myths of the old space age, but it's really a red herring.

There's nothing intrinsically wrong with mixing crew and cargo. The reason that people think that there is, is due to the fact that the Shuttle hasn't turned out to be as reliable or safe as many imagined, or hoped, at the program inception. After Challenger, the conventional wisdom became that, because flying into space is intrinsically hazardous, we shouldn't risk the lives of astronauts for cargo missions that don't require them.

That's fine, as far as it goes, but it misses the point. The fact that the Shuttle isn't safe or reliable doesn't mean that safe and reliable space transports can't be built (the foundation of many false lessons from the Shuttle lies in the fundamental mistake of drawing generalizations about a class from a single flawed example). A well-designed space transport would be sufficiently reliable so as not to necessitate a crew escape system (which was in fact the philosophy of the Shuttle--it was just poorly executed due to pinch-penny development). And regardless of whether it carried passengers or not, it would have to be designed for high-reliability recovery, because the cost of the asset itself is so high.

An aircraft guy like Burt would probably recognize that safety and reliability are paramount in the design of a reusable space transport, just as they are in a reusable air transport, because otherwise a reusable vehicle is not economically viable. It's only in the space business that we would come up with such a nutty idea as throwing hardware away every flight, a philosophy that lies at the core of our reliability problems. The only difference in design between a 747 and a 747F is the cargo door and lack of windows in the latter. It is a single aircraft design, capable of handling both passengers and cargo safely (even on the same flight), and there's no intrinsic reason that we couldn't do the same thing with a space transport.

Unfortunately, NASA has apparently decided (learning more wrong lessons from the Shuttle, and other failed attempts of the agency, such as X-33 and X-34) that it's not possible to build vehicles that are either reliable or affordable, so instead they've gone back to the future, and are redoing Apollo, which over the long term was neither affordable or sustainable.

Posted by Rand Simberg at July 17, 2005 08:12 AM
TrackBack URL for this entry:
http://www.transterrestrial.com/mt-diagnostics.cgi/4033

Listed below are links to weblogs that reference this post from Transterrestrial Musings.
Comments

I find NASA's position amazing. If it's not possible to build launch vehicles that are reliable or affordable, then manned space is itself pointless. So why does NASA even exist?

VSE was defensible when it was still believable that it could lead to markets that could encourage private launch firms, but that dream is more and more resembling wishful thinking. I think VSE will end up like the shuttle and like the station, producing much less than the vague optimistic initial PR had suggested. Maybe at that point the inherent failure of central planning will finally be inescapably obvious.

Posted by Paul Dietz at July 17, 2005 10:04 AM

I would imagine that if you were to launch cargo and crew in separate ships, and land them in separate ships, maybe not now but in the near future, you could send many types of cargo much quicker and even rougher journeys than you would send humans on.

Not that I'm trying to defend those guys or anything.

Posted by Taoye at July 17, 2005 03:15 PM

The shuttle was once sold to us as a space truck. I think that's ironic, since the trucking industry might have something to teach. Trucks come in two parts, the tractor, then one or more trailers (yes, I'm ignoring other paradigms for the sake of my point.)

There's absolutely no reason for the cargo containers (if they were designed as such) to be able to reenter. That's overdesign and that's the problem with the space shuttle. A crewed tractor isn't neccessary either.

So what's left?

If you're not worried about cargo reenty, then you only need a craft big enough for people to return in. That doesn't have to be much. It could be a DC-X or SpaceShipOne or some other design... but it certainly doesn't need to be an redneck RV.

Posted by ken anthony at July 17, 2005 06:58 PM

I think Taoye may have a good point - if, say, high-g launch and recovery profiles are cheaper to achive than low-g profiles. I seem to recall that Energia hit 12 G's just before MECO. Could it be that you don't need throttlable engines for cargo? Spash- and thump-downs are also more reasonable for cargo (specifically fuel upload flights).
That being said, I agree that the cargo/people mix was one of the shuttle's few strengths - there was a recent quote by one of the Old Timers morning the ban on commercial payloads on the orbiters. So long as they charged a fair price (marginal cost + a percentage of fixed) it might have helped develop a services paradigm, and spur competition on the schedule side, thus promoting reliabilty, and thence lowering prices.
Or maybe not.

Posted by Duncan Young at July 17, 2005 07:13 PM

Ken,
It might be argued that it is better come back with a large empty internal volume behind your cross-section, as it decreases your density and makes re-entry gentler. The big overdesign with the shuttle was the 6000 km crosstrack requirement while returning huge undeployed national security payloads.

Posted by Duncan Young at July 17, 2005 07:24 PM

First off, Ken, it sounds like you're postulating a launch vehicle that is separate from the "payload" vehicle, which implies that you've decided upon expendables as the one true way to begin with.

Second, I thought the bit about high g-forces was a myth: that although you get less gravity losses that way, you also wind up with greater structural mass to deal with both the g forces and the atmospheric drag; I vaguely remember Gary Hudson pointing this out, and that there's a feedback loop involved...

Posted by Phil Fraering at July 17, 2005 07:25 PM

Ken: Truax made that point years ago, for the launching propellant. He said you should always launch propellant in expendable tanks, since making those tanks be able to survive reentry from orbit was far more expensive than just building new tanks.

Posted by Paul Dietz at July 17, 2005 07:26 PM

Duncan: having a lower ballistic coefficient doesn't make the reentry gentler if by that you mean having a lower acceleration. To first order, acceleration during reentry is independent of ballistic coefficient, and depends only on the L/D ratio of the vehicle.

A lower ballistic coefficient does decrease the heat/area that must be dissipated, but it also increases the area that must be protected from that heat. A wider, but thinner, heat shield may be more vulnerable to catastrophic damage than a narrower, thicker one.

Posted by Paul Dietz at July 17, 2005 07:29 PM

Paul,
Thanks for the clarification. I was refering (opaquely) to the heat dissipation problem. The big problem with the current shuttle is that the relativily high ballistic coefficient pushs the temperature curves into regimes where you need exotic and fragile materials like RCC and ceramic tiles as your TPS. If you can drop below that point, to where you can use alloys, a wider heatshield may become more acceptable in your cost/risk trade space.

Posted by Ducan Young at July 17, 2005 07:54 PM


> do things the same way that aviation does them--a single aircraft design
> for both cargo and passengers.

Or at the very least both cargo and crew, who add greatly to the safety of the system. It's time we recognize that the safety of cargo and airframes also matters, and yes, it is okay to "risk" pilots to ensure their safe arrival -- just as Federal Express "risks" pilots every night to ensure the safe arrival of cargo flights.

(And please, can someone get NASA to stop using the term "crew" to mean both flight crew and passengers? That semantic confusion does not help in discussing the issue.)

> A well-designed space transport would be sufficiently reliable so
> as not to necessitate a crew escape system

One of the alarming things about CEV are statements that it will be safe simply because it has an escape system.

There seems to be no recognition of the fact that even "successful" aircraft ejections often have long-term medical consequences. "Attempted suicide to avoid certain death." In aviation, ejection is the last resort, after a pilot has exhausted all other options. The ELV guys think it's a first resort, something you can do instead of giving the pilot options. (Which is not really surprising, given that most of them would rather not have a pilot in the first place.)

Posted by Edward Wright at July 17, 2005 08:00 PM

I'm not convinced metallic TPS is workable. The alloys involved are dense, and if you go to a low ballistic coefficient design the metal must be quite thin and may not be sufficiently robust.

For operability, let me suggest transpiration cooling. This should be easier on a high ballistic coefficient vehicle, since you need fewer coolant orifices and each can be larger.

Posted by Paul Dietz at July 17, 2005 08:04 PM


> There's absolutely no reason for the cargo containers (if they were
> designed as such) to be able to reenter.

There is you're paying for the containers.

> A crewed tractor isn't neccessary either.

Have you seen a container coming down the road without a crewed tractor lately?

If you did, I bet you ran for life. That's the reason for the crew -- safety.

> for the launching propellant. He said you should always launch propellant
> in expendable tanks, since making those tanks be able to survive reentry from
> orbit was far more expensive than just building new tanks.

But if you use the tank 20 times (or 100 or 1000), you can afford to spend far more on it than a tank which is only used once.

The Shuttle's external tank hasn't contributed any measure of economic success.

Posted by Edward Wright at July 17, 2005 08:09 PM

I'm not convinced metallic TPS is workable. The alloys involved are dense, and if you go to a low ballistic coefficient design the metal must be quite thin and may not be sufficiently robust.

Fair point - the X33 is the definitive example.

For operability, let me suggest transpiration cooling..
But doesnt this imply a lot of tankage and thus a lower ballistic coefficient? I dont know how much fluid is required.

One TPS that would definity work better with a high ballistic coefficient would seem to be a replacable ablator.

Posted by Ducan Young at July 17, 2005 08:18 PM


Also, remember that with an expendable tank, each flight will be a first flight for that tank. If you had to install brand-new fuel tanks on a 747, you would want to do a post-maintenance flight test before putting it back into revenue service. With expendable tanks, you can't do that. Every revenue flight becomes a post-maintenance test flight.

Posted by Edward Wright at July 17, 2005 08:19 PM

Okay, to reveal some (of my) ignorance here, I'll pose a question.

What is keeping a space launch vehicle - say, 100 tons to LEO - from being as reliable as a 747F?

I mean - obviously, they're not. Why is it that way?

(By 'reliable', I include ease of maintenance, turnaround time, everything that makes a 747F as useful as it is. The '100 tons to LEO' bit is utterly arbitrary, but does refer to the Shuttle. If that's its job, it's doing a damn poor job of it. A follow-up question might be, 'Is that what we should be shooting for?')

Posted by Noah D at July 17, 2005 09:38 PM


> What is keeping a space launch vehicle - say, 100 tons to LEO - from being
> as reliable as a 747F?

The learning curve. Fewer than 500 people have flown in space, and right now the death rate's about the same as it was for the first 500 airline passengers. Reliability improves with experience, but with trivial flight rates, we've never gained much experience.

Posted by Edward Wright at July 17, 2005 09:48 PM

"There's nothing intrinsically wrong with mixing crew and cargo. The reason that people think that there is, is due to the fact that the Shuttle hasn't turned out to be as reliable or safe . . "

Since there isn't any rocket powered space launcher that's proven even 98 percent reliable, your premise is false, and hence your conclusion as well. Your conclusion requires a hypothetical near perfect space launcher that, in spite of 40 years of trying, has yet to be built. Until such time, it is transparently obvious that mixing cargo and people makes the cargo far more expensive and the people much less safe.

Posted by K at July 17, 2005 11:33 PM


> Since there isn't any rocket powered space launcher that's proven even 98 percent reliable,

What makes you think it's impossible to build a rocket that's more than 98% reliable?

The X-15 completed 199 flights with only one fatality -- and that was a research program, where loses are expected.

> Your conclusion requires a hypothetical near perfect space launcher that, in spite of 40 years of trying

Of course it doesn't. Aircraft are far from perfect. So are boats, trucks, cars, and trains. We don't remove crews from imperfect vehicles to "ensure safety." We put crews onto imperfect vehicles to enhance safety.

What hypothetical near-perfect vehicle do you think has been tried for 40 years? ELVs? If you're drawing statistics from those, it's no wonder you come to faulty conclusions.

You don't draw conclusions about the safety of aircraft based on the accident rate for model airplanes.


Posted by Edward Wright at July 18, 2005 02:56 AM

K, you're making exactly the logical error that I described in the post.

Posted by Rand Simberg at July 18, 2005 04:04 AM

Duncan: transpiration cooling shouldn't require tanks that are so large as to greatly affect the ballistic ratio. It's similar to ablative cooling, except that you're expending a liquid to generate insulating vapor rather than pyrolyzing a solid heat shield to produce that vapor. It's analogous to the relationship between liquid and solid fueled rockets.

Posted by Paul Dietz at July 18, 2005 04:44 AM

--
If you're not worried about cargo reentry, then you only need a craft big enough for people to return in. That doesn't have to be much.
--

Sounds like an F-111 design...you'd think NASA had already considered it. Perhaps the design wasn't sexy enough for the then current administration to approve it.

Posted by CJ at July 18, 2005 09:32 AM


> If you're not worried about cargo reentry, then you only need a craft big enough for people to
> return in. That doesn't have to be much.

If you have a craft big enough for people (plural) to return in, you can also do cargo flights just by offloading the passengers and their seats.

> Sounds like an F-111 design...you'd think NASA had already considered it.

The F-111 ??!! You are kidding, right?


Posted by Edward Wright at July 18, 2005 11:06 AM

Two comments: First, I think the biggest reason to have different vehicle (or vehicle configurations) for cargo and personell is that people need oxygen, room to wander around, bathrooms, shaving kits, etc. At $5,000 per pound, sending up the minimum requirements is probably better.

Second, (and to at least partially invalidate the first...) the reason no one likes metalic tiles is because they are too dense, so you need to make them thin. The reason you care about density is because orbit is $5,000 per pound. But, completing the circle, the reason it is $5,000 per pound is because everyone is so worried about density. Perhaps it would make sense to double the fuel load to half the reuse costs?

Posted by David Summers at July 18, 2005 12:09 PM

It may be a little bit late to chime in, but I had a post on my blog about two or three weeks ago on this topic....where was it....ah:

http://selenianboondocks.blogspot.com/2005/06/crew-and-cargo.html

Posted by Jonathan Goff at July 18, 2005 12:11 PM

--The F-111 ??!! You are kidding, right?

Nope...not kidding.

Admittedly, I extremely ignorant of the design challenges for a space worthy craft. However, the original post that I referred to by Ken mentioned not worrying about cargo reentry. The F-111 was designed to eject the entire crew cockpit to deny access of sensitive equipment to the 'enemy'...the crew was expendable I suppose.

Thus my reference to the F-111 design. Fly 1 single piece up...do what you need to do...disengage the cargo portion (heck, preprogram a splashdown for cargo container) and the command craft glides right on in under human control.

My main point was I figured that NASA probably already thought about it and discarded it for numerous reasons....I don't even know if it's possible even it sounds reasonable (to me).

Posted by CJ at July 18, 2005 01:07 PM


> First, I think the biggest reason to have different vehicle (or vehicle
> configurations) for cargo and personell is that people need oxygen, room
> to wander around, bathrooms, shaving kits, etc.

You've never seen a fighter cockpit, I take it?

> At $5,000 per pound, sending up the minimum requirements is probably better.

That's fine if you want a minimal space program. At $5,000 per pound, you can't afford to do anything of value. This is a circular argument: We aren't doing much of anything in space because the cost is too high, and we don't do anything to reduce the cost because we aren't do much of anything in space anyway.

Posted by Edward Wright at July 18, 2005 01:16 PM


> The F-111 was designed to eject the entire crew cockpit to deny access
> of sensitive equipment to the 'enemy'...the crew was expendable I suppose.

No, the F-111 was designed to eject the cockpit to save the lives of the crew. That didn't work very well, which is why later aircraft went back to ejection seats. Expending crew members was a bug, not a feature.

> Thus my reference to the F-111 design. Fly 1 single piece up...do what you
> need to do...disengage the cargo portion (heck, preprogram a splashdown
> for cargo container)

That means you're probably not going to be able to reuse your cargo containers. "Splashdown" sounds gently, but it's not, and salt water is nasty. It also means a recovery fleet including big helicopter carriers, etc.

It's much easier just to have the pilot fly the container back to base.

> My main point was I figured that NASA probably already thought about it
> and discarded it for numerous reasons

Irrelevant. NASA has given up on the idea of ever having a significant number of humans in space, so NASA is no longer significant to human spaceflight.


Posted by Edward Wright at July 18, 2005 01:31 PM

"A rocket a day" concept i.e. cheap mass produced lower-than-normal-reliability expendables would still be desireable for bulk cargo launches.
The reliability could be improved in the long run like currently various production lines are improving yields ( chip and electronics manufacturing etc ) by tuning the manufactoring process.

But this doesnt mean that vehicles designed for human payloads should not be able to lift various valuable cargoes like high-tech cadgetry of electronics etc whose cost more per pound is in the range of their launch costs.

Posted by kert at July 19, 2005 04:00 AM

IOW, relatively cheap ( on ground ) cargoes can be lifted by any unreliable rocket as long as its dirt cheap, and because its unreliable it would be pointless to build it reuseable.

Of course, if one day RLV flight costs will approach a small multiple of propellant costs it would be end of the business for such expendables, but i dont think this day is close yet.

Posted by kert at July 19, 2005 04:09 AM


> "A rocket a day" concept i.e. cheap mass produced lower-than-normal
> -reliability expendables would still be desireable for bulk cargo launches.

Even if NASA could afford to build and destroy a rocket a day (it can't), why would that be desirable? Just to employ lots of workers?

> The reliability could be improved in the long run like currently various
> production lines are improving yields

No, the final "yield" for the production process is not a rocket, it's a rocket flight. As long as the flight rate is low, reliability will also be low, and cost high, and -- okay, what's the point again?

> IOW, relatively cheap ( on ground ) cargoes can be lifted by any unreliable
> rocket as long as its dirt cheap,

If it's unreliable, it can't be dirt cheap. Reliability and low cost go hand in hand. Ask Southwest Airlines.

> and because its unreliable it would be pointless to build it reuseable.

If it's unreliable, it's *impossible* to make it reusable. You can't reuse something that explodes into a million bits and pieces. And if it isn't reusable, it's impossible to make it reliable, because you can't test it thoroughly.

What is the point of deliberately making rockets unreliable and nonreusable? To keep spaceflight expensive? To create make-work jobs?


Posted by at July 19, 2005 01:41 PM

> Even if NASA could afford to build and destroy a rocket a day (it can't), why would that be desirable? Just to employ lots of workers?

Why do we buy disposable pens? Because it ends up being cheaper. Right now, expendable orbital launchers are cheaper than the (partially) reusable orbital launcher we have, and NASA has concluded that developing a new reusable wouldn't save money. They're probably right, at the flight rates being imagined.

The high cost of entry for reusable orbital launchers (slideware notwithstanding) is not something that advocates can just handwave away.

I suggest reusability, when it arrives, will arrive incrementally, by reusing increasingly large components of expendable launchers.

Posted by Paul Dietz at July 20, 2005 07:29 AM

"Even if NASA could afford to build and destroy a rocket a day (it can't), why would that be desirable?"

I was referring to the "A Rocket a day keeps the high costs away concept".
http://www.fourmilab.ch/documents/rocketaday.html

"the final "yield" for the production process is not a rocket, it's a rocket flight. "
Exactly. And improving yields in such configuration means removing weak spots in manufacturing process and tweaking it ( while keeping the basic rocket design the same ) so that flight failure rates will go down. The same was basically done with V2 rockets that actually were mass produced.

"As long as the flight rate is low, reliability will also be low"
A launch a day is not LOW flight rate by todays standards. You can even increase flight rate artificially by making your rocket smaller ( and thus also easier to produce/transport/launch )
I have been thinking about this concept several times over. I will give you these numbers for instance: 500KG equatorial LEO payload, once a day, with initial failure rate of 50%. Payloads: propellants, food, water, raw construction materials. Water costs basically nothing on earth, it inevitably costs about $2000/lb right now on orbit, because of launch costs on market. So that half a ton of water would be worth about three million on LEO. If you could build two such rockets a day, costing you a million each, fail half of them while launching, you'll still be turning in a profit. That, of course, assumes you have a customer ...
Dont you think there could be a market for such delivery service ?
Cost depends only on how simple you can build such a rocket (pressure-fed, ablative engines etc ) and thus also how complex and expensive manufacturing line you need. Price obviously would be determined by demand.

"If it's unreliable, it can't be dirt cheap"
Huh .. somehow im forced to think about Korean cars

"What is the point of deliberately making rockets unreliable and nonreusable? To keep spaceflight expensive?"
The point would be to allow for mass production methods of such rockets. Mass production, after all, is one of the cornerstones of modern consumer world. And there are some paylodas that can be launched with unreliable rockets, and then some that cant. Im not proposing putting humans or satellites on top of such rockets.

Posted by kert at July 20, 2005 08:31 AM


Post a comment
Name:


Email Address:


URL:


Comments: