White House Decides to Outsource NASA Work

[MSimon]

Really? Seems simple to me. Take up six or maybe 20 coils, bolt them together. Plug in the power leads and cooling lines. I mean it is not as if you have to provide a vacuum chamber or anything. Well, maybe a real thin one for operations in near planet orbit.

You need a chamber even in space (depending). Only about 10% of the fuel flow gets burned. It may not matter. But it is a consideration.

[93143]

Well, I would think a dozen or so medium ELV and a large number of KITE/HASTOL launches. Of course, your "inflatable" may reduce the number of ELVs significantly.

Why would you launch a dozen EELVs when you could launch two Jupiters and be done with it? It's cheaper, easier, faster, more reliable...

By the by, how heavy would one coil of an SC Polywell be? We may actually have to start planning for a 15-25Mg tether launch system.

Assembling a Polywell on orbit strikes me as a really difficult and complicated endeavour. Also, Polywell is not a given. Suppose you need a fission reactor?

Really? Seems simple to me. Take up six or maybe 20 coils, bolt them together. Plug in the power leads and cooling lines. I mean it is not as if you have to provide a vacuum chamber or anything. Well, maybe a real thin one for operations in near planet orbit.

Ugh, no. Bolt them together? Haven't you been paying attention to the heating issues with WB-7? Also, bolts are bad for arcing; you might want to weld and polish instead... You will want a vacuum chamber to catch unburned fuel and deflect micrometeoroids (one good ding could mean the difference between a working reactor and an arc discharge demonstration toy), and it should at least have a strongback structure capable of withstanding the magnet loads. Then there's the direct conversion system... unless you want a thermal plant, which is probably outright impossible to assemble on orbit... I'm pretty sure an operational Polywell would be a hell of a lot more complicated to put together than you make it sound. Doing it in vacuum and zero gee with spacesuits and robotic manipulator arms, when you don't have to, is insane.

Oh, and a lead shadow shield for a 6 GW core is about 100 mT... but I suppose that doesn't have to all go up in one piece...

Originally I simply intended to challenge the widespread assumption that assembling a mission in space from little bits, launched with rockets we already have, is obviously better than launching it in a few big chunks using a larger rocket. It's not.

I would certainly accept MAY not be.

Read carefully. The hedge is already inherent in the phrasing.

But assembing such a mission on orbit is the way to go if the infrastructure is there.

Sounds like an unexamined article of faith to me. Why would you assemble something on orbit instead of on the ground unless you had no other choice?

Remember, right now, heavy lift is more cost-effective even if you disregard the advantages of ground integration, so you're essentially paying more to make your mission harder...

Plus there's no guarantee that whatever system eventually gives us cheap LEO access (if any) will be mass- and volume-limited to the degree you assume (most of the current proposals are, I grant you...). Like I said, I suspect that shielding requirements and the 7th-power scaling will make a Polywell-powered SSTO fairly large...

I am tired of NASA stifling space enterprise.

NASA hasn't done anything to stifle space enterprise since 51L. They've just withdrawn and done their own thing. What has space enterprise accomplished in the quarter-century since then?

...oh, you meant you want NASA to subsidize space enterprise with taxpayer dollars. That's different. I don't disagree, but it is different.

However, space-craft are so expensive partially because launch costs are literally astronomical. Who in their right mind would try to save $50M on a space-craft when launch costs approach $150M for that craft? Demanding that 99.999xx% reliability is what costs. If the launch costs were $1.3M rather than $150M I suspect you would see the price of the space-craft plummet!

That's part of the problem, but I suspect you're overstating things. How much does an airliner cost? How about a fighter jet? Launch costs are zero for these things; they operate from the ground. And they're mass produced. Sure, there are savings, but space is hard, and there's a lot we don't know and can't do - most of the cost of spacecraft is functionality, not weight scrubs. I don't see spacecraft getting vastly cheaper any time soon, ESPECIALLY if we have to assemble a Mars mission in zero gee out of a dozen free-flying 10 to 25-ton chunks and over a hundred free-flying inert (!!) 5-ton chunks...

This is why I'm a fan of Sea Dragon. Simple, cheap, and it minimizes the orbital assembly problem... unfortunately it really is a bit too large for our current budget...

Besides, your example doesn't hold together. I said spacecraft are MORE expensive than the launchers. If a spacecraft costs $300M and the launcher is $150M, saving $50M on the spacecraft could land it on a $250M launcher, yes - but even if it didn't, and you saved the $50M, the spacecraft is not all that much less expensive. If the launch were free, and the $50M were saved on the spacecraft as a result, the mission would go from $450M to $250M - not exactly a paradigm shift...

[Skipjack]

If we had a working polywell, then I would say we would just build an LV that uses it (as proposed by the late Bussard), right? That would probably be the best way to get a working polywell into space, no?
I think the decision to axe NASAs cost plus launcher programmes was a good idea. Ares IX was a fraud, if you ask me and an expensive one at that. There was very little hope that this attitude was going to change within NASA and their contractors.

I think that they are probably opening some champaign at SpaceX and Orbital. Best decision Obama has made so far(that and the new laws regarding stem cells)

[KitemanSA]

Well, I would think a dozen or so medium ELV and a large number of KITE/HASTOL launches. Of course, your "inflatable" may reduce the number of ELVs significantly.

Why would you launch a dozen EELVs when you could launch two Jupiters and be done with it? It's cheaper, easier, faster, more reliable...

Assuming you HAVE a Jupiter. And true, the alternatives don't necessarily exist at present, but their development cost can be spread over a MUCH bigger potential market.

[KitemanSA]

Originally I simply intended to challenge the widespread assumption that assembling a mission in space from little bits, launched with rockets we already have, is obviously better than launching it in a few big chunks using a larger rocket. It's not.

I would certainly accept MAY not be.

Read carefully. The hedge is already inherent in the phrasing.

Only if you read it in one specific way. If you read the "Its not" as "its not obviously better", then ok. Hower it can also be read "its not better" and I don't necessarily agree with that. I think we basically agree. The issue isn't quite settled, one way or the other.

[KitemanSA]

But assembing such a mission on orbit is the way to go if the infrastructure is there.

Sounds like an unexamined article of faith to me.

Article of faith? Assuredly. Until the infrastructure is there, it must be an article of "faith". The sun coming up tomorrow is an article of "faith".
"Unexamined"? Not so much. Historically speaking, the sun has come up pretty much every day for just about ever. Historically speaking, each "step further out" occurs when the current step has had its infrastructure developed enough to support the next step. Attempts by "King and Country" to leapfrog that condition has always failed in the long run. I am eager to be shown wrong on this. If you have ANY data where this has NOT been the case, please show me.

[KitemanSA]

I am tired of NASA stifling space enterprise.

NASA hasn't done anything to stifle space enterprise since 51L. They've just withdrawn and done their own thing. What has space enterprise accomplished in the quarter-century since then?

...oh, you meant you want NASA to subsidize space enterprise with taxpayer dollars. That's different. I don't disagree, but it is different.

Perhaps I should reply to you like you did to me. However, I will admit to the possibility of being unclear. I did not intend to state that I had just recently gotten tired of, but that I am and have been for a significant while, tired of, NASA stifling private enterprise. Admittedly it has been a while since I was a space fanatic/L5/NSS type, but...
When NASA decides to investigate a technology that is being developed by private enterprise, it stifles that private enterprise. What investor would provide $1M for an investigation when NASA just provided one of its favorite few $5M to research and develop a direct competator? Then they drop it and the technology has died. Oh whell.

[MSimon]

...oh, you meant you want NASA to subsidize space enterprise with taxpayer dollars. That's different. I don't disagree, but it is different.

Air mail got us airlines.

[pfrit]

But assembing such a mission on orbit is the way to go if the infrastructure is there.

Sounds like an unexamined article of faith to me.

Article of faith? Assuredly. Until the infrastructure is there, it must be an article of "faith". The sun coming up tomorrow is an article of "faith".
"Unexamined"? Not so much. Historically speaking, the sun has come up pretty much every day for just about ever. Historically speaking, each "step further out" occurs when the current step has had its infrastructure developed enough to support the next step. Attempts by "King and Country" to leapfrog that condition has always failed in the long run. I am eager to be shown wrong on this. If you have ANY data where this has NOT been the case, please show me.

If the plan is to send MANY missions to {Moon, Mars} then building many reusable peices makes a lot of sense. A rocket to LEO, a ferry from LEO to GEO or higher orbit, a ferry for GEO to {Lunar, Martion} orbit, A ferry for {lunar, Martian} orbit to landing. If it is a stunt mission, then you have the Apollo disposable system.

[Skipjack]

Stunt missions dont make sense. Apollo was ultimately a failure. It was a great demonstration of engineering and the US capability to dominate anything they wanted to, but it did not bring uns any further in the long term. Now, more than 40 years later, we still have not been back. Why? Because there was no sustainable infra structure. It is still as expensive to do today as it was back then.
Going to Mars just for a stunt, makes no sense in these times.

[pfrit]

Stunt missions dont make sense. Apollo was ultimately a failure. It was a great demonstration of engineering and the US capability to dominate anything they wanted to, but it did not bring uns any further in the long term. Now, more than 40 years later, we still have not been back. Why? Because there was no sustainable infra structure. It is still as expensive to do today as it was back then.
Going to Mars just for a stunt, makes no sense in these times.

I certainly wasn't suggesting that we do a stunt mission. I am saying that we should be clear up front whether a specific mission is a stunt or a long term mission. If you want to see someone go to Mars in the next 20 years, then guess what? It will be a stunt. Columbus' mission to the Americas was a stunt. Personally, I am not convinced that there is a good reason to send a lot of people to the moon or mars. It reeks of manifest destiny. Sending a few explorers is different. Especially robotic explorers... I am inclined to think that we will colonize the solar system when individuals want to go themselves and not before. Historically, the only other way to do it is by the state sending criminals or religious zealots...

[93143]

Hmm... maybe I stepped over the line there a little...

Perhaps I should explain my position, rather than being reactive and trying to prove my opponent comprehensively wrong at all points (he isn't, so that's a good way to paint myself into a corner)...

I think NASA should start exploring. We have sufficient technology to start now; the question is money, and the United States can easily afford it. There's room for philosophical disagreement here, but I don't think we should sit around waiting for technological improvements that may not happen, and are probably less likely to happen if we sit on our hands waiting for them rather than going out and needing them...

I think NASA should use the most efficient methods to mount exploration missions. Compromising efficiency in pursuit of secondary goals will just get the program cancelled, or squeeze the budgets of other parts of NASA that need money too (plenty of advanced projects were cancelled to make way for Ares). If this cuts out commercial because their methods aren't efficient enough, tough beans. If not, great. (Commercial is already plenty good enough for Earth-orbit operations, but manned exploration is another story.)

Right now, it looks to me like the most efficient method for LEO access for exploration purposes that's available either now or with high confidence in the near term is Jupiter. Clean sheet would be better, but the political support isn't there for a NASA clean sheet, SpaceX BFR isn't near-term high-confidence (much as it pains me to say it; I'd actually prefer to use it rather than Jupiter), and I think we need to seize any political opportunity to get out of LEO; they're rare and valuable. Lunar mission costs:

Lunar Class Missions (annual LV Program cost in millions):

# Jupiter EELV (Atlas-V)

1 $2,348 $1,892
2 $2,866 $3,006
3 $3,345 $4,035
4 $3,800 $5,012

Now, Ross' numbers, when vetted by independent experts with no axe to grind, tend to turn out to be pretty good, even conservative. If I'm not mistaken, these numbers include amortized development costs (his methodology in the past has been such).

What this says is that the high fixed costs of the HLV operating from LC-39 make it cheaper to use EELVs for one lunar mission per year - but if you want to do more than that, it's better to use the HLV because the incremental cost of a launch is so favourable. It's easy to see that a Mars mission's much higher IMLEO requirement skews the balance even further towards the HLV.

All else being equal, ground integration of modules is to be strongly preferred over in-space assembly, because it is easier, cheaper, faster, and produces a safer and more reliable spacecraft.

As a result, any light lift system that requires more orbital integration for a deep space mission must have enough of an advantage over the heavy lift system in some other area (cost, probably) to overcome this disadvantage. The size of the necessary advantage depends on what you're trying to assemble. For instance, Skylon could capture the world market for propellant on orbit (provided there is one; enter NASA's exploration program), but I don't think it would be suitable for flying a nuclear reactor - still, the cost advantage per kg to LEO is potentially so large that I could be mistaken.

Currently, though, with expendable rockets, all else being equal heavy lift is cheaper per kg to orbit...

Off-world resource utilization - building things in space without having to launch the pieces - could well be the way to go if the infrastructure is there. This is very different from the scenario where everything comes up from Earth; in that scenario, ground integration is always preferred, and there has to be a very good reason to choose orbital integration instead.

@Skipjack: Apollo's science return exceeded that of all unmanned exploration of the moon to date, by a large margin. IIRC Apollo 15 is still considered the most successful manned space mission ever. That's ignoring the secondary science and technology dividends, which were considerable (large temporary increase in science/engineering grads, for example). The only "failure" was the U.S. government cancelling the program and squeezing NASA's budget down into the noise. NASA's mandate is not something that can be accomplished by commercial enterprise (no short-term profit, you see), but it is something that needs to be done, no matter how much harder it is than anything else we've ever done.

Fortunately we are now in a position to explore again without it being so much of a horribly expensive and dangerous easily-cancelled stunt...

@pfrit: I'm not suggesting a stunt. I'm suggesting a sustainable manned exploration program that can teach us things we need to know about humans living and working on other planets and in interplanetary space. My current understanding is that DIRECT's plan (Jupiter + PDs + commercial LEO and propellant services) is the most likely candidate for such a sustainable program, especially when U.S. politics are taken into account.

Also, robots are no substitute for humans, and they won't be for a long time yet. I seem to recall that after one of the Mars rovers was done its initial 90-day mission, they suited up a geologist, gave him the necessary tools, and let him loose in a patch of ground made up to look like the area the rover had explored. He was done in 40 minutes... and that's not even taking into account all the things he could have done that are still far beyond state-of-the-art robotic capabilities; there's really no comparison at all.

[pfrit]

Also, robots are no substitute for humans, and they won't be for a long time yet. I seem to recall that after one of the Mars rovers was done its initial 90-day mission, they suited up a geologist, gave him the necessary tools, and let him loose in a patch of ground made up to look like the area the rover had explored. He was done in 40 minutes... and that's not even taking into account all the things he could have done that are still far beyond state-of-the-art robotic capabilities; there's really no comparison at all.

There is a comparison worth making. How much does it cost. A human can do x amount of science per dollar. A robot can do y amount of science per dollar. For exploring a mall x >> y. For exploring Antartica, x > y. For a volcano x ~= y. For the moon or mars x < y. Robots give a whole lot of bang for the buck the more extreme the environment. They do not, however, give more Buck Rodgers for the buck. That is where the "stunt" comes in...

[93143]

For the moon or mars x < y.

That's what I'm telling you - it's not true.

The MER program cost about a billion dollars. A human Mars mission, starting from now, would cost on the order of a couple hundred billion.

Four geologists doing surface science for a year and a half, in eight-hour-per-day shifts, would be able to get between a couple hundred and a couple thousand times the amount of science done that the rovers have managed since mission start.

So far the rovers aren't looking too good.

Now launch another manned Mars mission, for a few billion more dollars. You've just doubled the science return, without substantially increasing the price.

You can't do this with the Mars rovers, since the support network can't handle a fleet of them, so you have to do things slowly, one or two at a time. (Which is an adequate counter to your argument all by itself.) And the price doesn't go down nearly as far - another rover would be a substantial fraction of the cost of the whole program to date.

And we haven't even considered the fact that humans can do things robots cannot. Humans can adapt to situations they weren't designed for, and do tasks that were not in the mission plan. A robot might have seen the Genesis rock, but it wouldn't have been able to do much with it... Also, humans can dig, and pick up all sorts of objects for later analysis in a lab. It's surprisingly hard to do this with robots...

Furthermore, launching robots tells us nothing about human habitation in deep space and on other planets. We're going to have to start learning this stuff eventually...

[pfrit]

For the moon or mars x < y.

That's what I'm telling you - it's not true.

The MER program cost about a billion dollars. A human Mars mission, starting from now, would cost on the order of a couple hundred billion.

Four geologists doing surface science for a year and a half, in eight-hour-per-day shifts, would be able to get between a couple hundred and a couple thousand times the amount of science done that the rovers have managed since mission start.

So far the rovers aren't looking too good.

Now launch another manned Mars mission, for a few billion more dollars. You've just doubled the science return, without substantially increasing the price.

You can't do this with the Mars rovers, since the support network can't handle a fleet of them, so you have to do things slowly, one or two at a time. (Which is an adequate counter to your argument all by itself.) And the price doesn't go down nearly as far - another rover would be a substantial fraction of the cost of the whole program to date.

And we haven't even considered the fact that humans can do things robots cannot. Humans can adapt to situations they weren't designed for, and do tasks that were not in the mission plan. A robot might have seen the Genesis rock, but it wouldn't have been able to do much with it... Also, humans can dig, and pick up all sorts of objects for later analysis in a lab. It's surprisingly hard to do this with robots...

Furthermore, launching robots tells us nothing about human habitation in deep space and on other planets. We're going to have to start learning this stuff eventually...

MER has cost a tiny bit over 1 billion dollars and is part of Bush's Vision for Space exploration. As far as the cost of a Mars mission goes, we simply cannot design it in such a way as to have any reasonable hope for the astronauts surviving at any cost. The dealing with radiation alone is currently beyond our means. A couple of hundred billion dollars won't get you started. And the martian environment is so lethal (again radiation being the worst hazard) as to make the idea of staying there as a colonist fairly laughable. We are talking trillions of dollars spent over 30 years here. Now, I don't necessarily disagree with the idea that we act like all the problems will be solved in time as that is what we are doing right now. But the idea that we send people who in the best case scenario will be condemned to die at a very young age to do some science that could be done safer and cheaper is a stunt. BTW, the idea that we count on the problems being solved by the time to need it goes for the idea of sending robots who perform better than humans on mars as well.