DarkenedOne":1r2zwr4r said:
It is true that the highly technical labor is expensive, but then again so is the airline industry. So why is access to space so expensive?
[...]
So why don't we make launch vehicles single stage and reusable like aircraft. The reason is that it takes far to much energy and thrust to make it to orbit than any single stage reusable craft using chemical propulsion could produce.
Agreeing with your overall point, and I underestimated the savings of small-well launching, but exploring some details for edification...
First, why are spaceship launches so much pricier than airplane flights, even disregarding fuel costs and disposable hardware?
* Gravity-drag -- Not only do you have to spend delta-V to increase altitude/velocity, you have to spend delta-V to bear the weight of the vessel (airplanes are supported "for free" by the atmosphere). The longer you take to reach orbit, the more load-bearing delta-V you spend. The need for fast acceleration has to be met with much more powerful engine designs (which require more maintenance and development costs and have less error tolerance).
If the Earth had one-sixth its gravity (with, magically, its normal atmosphere) this would still make orbital flight more expensive than atmospheric flight, even disregarding fuel costs.
* Acceleration tolerance -- With greater acceleration you need stronger load-bearing structures (again, this is not just a matter of weight, but the costs associated with building & maintaining a more advanced design with less error tolerance). You also need a pretty sturdy launch platform (which itself requires more maintenance).
If launching from the moon, the concern for gravity drag (and hence the need for rapid acceleration) is reduced. But it's still there, and larger payload launches are likely to need stronger load-bearing structures than would be present on an airplane (and, obviously, they can't ride up on the moon's atmosphere).
* Pressure (vacuum) tolerance -- Airplanes only need to be pressure-rated to a certain height, and are error-tolerant even with passengers. A launch vehicle, even unmanned, has to operate in very thin atmosphere, and any part that makes it to space (everything if it's entirely reusable) has to work in a vacuum. Again, this requires high-performance technology.
* Guidance demands -- A fixed-wing passenger plane is, to a certain degree, self-stabilizing. An upward bound rocket is not, and thus requires more sophisticated navigational hardware (and monitoring). Furthermore, anything which goes into Earth orbit has to be guided away from 30,000 potentially violent collisions (requiring more personnel and equipment on the ground). Airplanes need traffic controllers too, but they move much slower, and-- more importantly-- every airplane is taking directions from ground control, whereas space junk tends to ignore orders to change its course.
* Atmospheric tolerance -- An airplane that goes 500 mph needs to withstand wind speeds of ~600 mph (allowing for a gust of headwind). A vehicle going into orbit has to take much more; and if it's reusable, it has to take it again when landing.
This is one reason I underestimated the savings of moon launches. Obviously, there's no atmosphere, and no need for those carefully-monitored ceramic tiles. But on the down side, a reusable vehicle launched from the moon will have to carry all the fuel it needs to slow down and descend; it can't get free deceleration from atmospheric drag.
* Amortization of development cost -- There are ~10,000 satellites in orbit, plus ~20,000 pieces of junk. So I'm going to estimate ~50,000 orbital launches in all of human history. By contrast, there are some 18 million commercial airplane flights each year, leading me to an historical estimate of about 10,000 airplane flights to every one orbital launch.
Second, A single-stage, completely reusable craft is feasible. Note that the STS re-uses the most expensive piece of hardware (the orbiter), as well as both SRBs. But most of NASA's non-shuttle launches, and all commercial rocket launches use expendable hardware because it's actually cheaper. So how do they differ from airplanes, which are clearly a better bargain when reused?
* Because of gravity drag, it's profitable to jettison anything you don't need as soon as you don't need it. If you wanted to reuse it, then, you'd have to haul it back and put it back together (which also necessitates making nice, clean screws instead of exploding bolts-- more construction cost).
* You have to load up on reentry armour to make anything reusable. I think this is the biggest reason that unmanned launch vehicles are disposable. They want to avoid the direct expense (buying and sticking on the armour) as well as the weight cost.
* The combined violence of high-speed atmospheric forces and rapid acceleration requires stronger material (and/or more refit time) if the craft is to be reused. A rocket which is scrapped after one launch can be made more cheaply.
* Because of low demand rates, the specter of technological obsolescence reduces the potential return on any reusable system. If a new airplane is expected to go obsolete in five years, it can still provide thousands of flights. But the Atlas III rocket, to use a demonstrative example, was used only six times between its maiden voyage in 2000 and its retirement in 2005.
So to sum up: A lunar-launched vehicle would eliminate nearly all of the reuse problems, would not require atmospheric tolerance, and could operate with much less fuel. Considering those factors, it would be significantly cheaper than an Earth-launched vehicle.
It would still require sophisticated navigation and vacuum-operational systems, making it more expensive than an airplane, but not astronomically so. The lack of atmospheric braking is a wild card, which could drive up the cost significantly depending on how its used.
The biggest factor in the short term would be development and design, which would make its initial cost many times more than that of an airplane.
DarkenedOne":1r2zwr4r said:
First of all what you must understand is that I, as well as many of the supports of that destination, do not advocate that the such a moon base should be made completely self-sustaining.
Instead you would start with water, which is simple to mine and process from the frozen lakes of the Moon. This water would provide water for people, oxygen for people, and fuel for hydrogen rockets.
I probably agree with you, but I think it depends on what we're considering as the long-term goal. I can certainly get behind the idea of mining lunawater to support lunabases, but that's assuming that lunar exploration and living is a worthwhile goal in itself.
I'm still not sold on taking advantage of the lower gravity
as a means to provide launchable resources for missions further into space. It's not exactly a question of whether or not its self-sustaining, but whether or not the resources produced outweigh the resources expended (which we'll have to keep expending, in great degree if the base is not at least partially self-sustaining).
Consider this: The Constellation program is estimated to cost $230 billion (in 2004 dollars) through 2025. The
mission plan includes about 10 ISS rotation flights (which would only be ~$10 billion at shuttle prices), but the big-ticket item is a lunar base that's expected to be finished by 2024. That base includes a currently-hypothetical
In-situ resource utilization, but this is not an industrial-sized device, it doesn't include deep-shaft mining capability, and there is no launch pad or delivery system to carry extracted resources into lunar orbit for rendezvous.
So assuming everything goes perfectly, that $230 billion gets them a regular supply of drinking water. How much more do you think it would cost to send up enough hardware (and do enough engineering) to turn that concept into a fuel-supplying station for outbound craft from Earth? They'll need more (and/or bigger) ISRU machines, a certain amount of mobility, a means to turn water into fuel (electrolysis is only part of the process), a launch pad & ship (to make the fuel available without requiring a Mars-bound ship to land on the moon), plenty of storage tanks, regular crew rotation, and either (a)a continuos supply of food & energy, or (b)the necessary infrastructure to provide those themselves.
Let's be generous and assume they get all of that for a mere $70 billion. Let's say, unrealistically perhaps, that $300 billion provides a steady supply of ready-to-burn rocket fuel that can be parked in lunar orbit for refueling.
Now the alternative. Our current payload cost for LEO is ~$10,000/kg. If you started doing thousands of launches per year, I bet that cost would go down. Furthermore, if you were building up a supply of hydrogen & oxygen fuel in orbit, you could tolerate lots of risk for even more savings (the tanks are all identical, therefore interchangeable, and there is no environmental impact from spreading oxygen & hydrogen). But let's be conservative and say the cost stays the same.
At that rate, for the same $300 billion, we could put 30,000 metric tons of fuel into orbit for use by outgoing spacecraft.
Which means that's our break-even point. But the problem is, even if the lunar base
can produce 30,000 metric tons of fuel, it's not worthwhile (as justified solely by resources provided to outbound spacecraft) unless we can also
use that much. Three hundred tons of fuel should be enough for a Mars mission, considering its already in orbit. Can we afford to develop, build, and launch into lunar orbit one hundred such missions?
Eventually, perhaps, but now we're on the scale of a century again (instead of just a decade), and chances are good that before we get that far, a better solution to launch costs will arise. We could have a robot working on the martian polar ice cap (same principle, true, but with a much more definite source); we could nudge a comet into Earth orbit for easy mining; or, my personal favorite,
maglev ramp launches. Maglev rails are a fully-proven, operational technology, and require only some engineering to be applied to orbital launches. Even better, the investment can be scaled to predicted frequency of use. For a dozen launches a year, it's not worthwhile to build; for several hundred launches a year, a track of several miles can be profitable; for ten thousand launches a year, a track could be hundreds of miles long and extend to high altitude.
And finally, an investment in improved Earth-based launches gives us something that no lunar system can: The ability to put things into Earth orbit, which is where the vast majority of useful satellites go, and where the first generation of space tourists will likely be content to visit.
Cheerio,
Jason