Space Based Solar Power For the Moon

[Warren Platts]

A variation on the traditional concepts of SBSP occurred to me the other day. Ordinarily, we envision a big array in GEO that beams to a fixed spot on the Earth. Alternatively, some people envision a solar station built on the Lunar surface out of Lunar materials that would beam energy back to Earth.

What I propose is that we reverse the direction of power: set up the solar PV array at an Earth-Moon Lagrange point, and then beam the energy to the Moon in order to power a Lunar research/propellant station.

The motivation is that producing respectable amounts of Lunar propellant is supremely energy intensive: e.g., to produce a mere 10,000 mT of ISRU propellant per year, it would probably require on the order of 20 megaWatts to get this done. Such an array would rival the largest PV arrays that have been built right here on Earth. It would be heavy and would require numerous landings to the surface of the Moon. In addition, building the array in a zone of perpetual sunshine isn't really much help because the angle of the sun is so low, the solar panels will wind up shading each other. The power requirements are probably the closest thing to a true showstopper when it comes to Lunar ISRU.

The relatively NSSO study on SBSP makes the case that relatively small arrays might be economically worthwhile for certain applications like beaming power to a remote army Forward Operating Base where supply lines are thin. The idea is to focus the 1st-generation SPSB station on a place with high strategic value where energy costs are already at least an order of magnitude more expensive than normal electrical costs.

Well, a Lunar station would fit that bill nicely. The research station would have a strategic value to the US, and it's electricity requirements would be very expensive.

So it might make sense (be cheaper) to build a 20 megaWatt array at a Lagrange point, and then beam the energy (microwaves or laser?) to the Lunar station.

Pros:
24-7 sunlight
lighter weight
much reduced launch costs

Cons:
have to build rectenna on Moon
this is possibly complicated by locating base at polar latitudes

What do you guys think?

[savuporo]

such schemes have been proposed before.
a really simple variant of which would be using a regular solar powered rover in permanently shadowed lunar craters, by illuminating its panels from lunar orbit with a laser.

[Warren Platts]

such schemes have been proposed before.

That a Lunar base should be powered by a multi-megaWatt SBSP station? I don't think so. But if you've got a reference, I'd love to see it.

[Hop_David]

A variation on the traditional concepts of SBSP occurred to me the other day. Ordinarily, we envision a big array in GEO that beams to a fixed spot on the Earth.

One of the problems with GEO SBSP to earth's surface is the 36,000 km distance. You need large rectenna receiving stations as well as large satellites.

What I propose is that we reverse the direction of power: set up the solar PV array at an Earth-Moon Lagrange point,

EML1 and 2 are nearly twice as far from the moon's surface than GEO is from earth's surface. EML4 and 5 about ten times as far.

For these you would need even bigger rectennas and space power sats.

Establishing large rectennas on the moon would also require a daunting amount of upmass.

[Warren Platts]

A variation on the traditional concepts of SBSP occurred to me the other day. Ordinarily, we envision a big array in GEO that beams to a fixed spot on the Earth.

One of the problems with GEO SBSP to earth's surface is the 36,000 km distance. You need large rectenna receiving stations as well as large satellites.

What I propose is that we reverse the direction of power: set up the solar PV array at an Earth-Moon Lagrange point,

EML1 and 2 are nearly twice as far from the moon's surface than GEO is from earth's surface. EML4 and 5 about ten times as far.

For these you would need even bigger rectennas and space power sats.

Establishing large rectennas on the moon would also require a daunting amount of upmass.

I just found an interesting paper that proposes a hybrid laser-microwave SBSP system to provide emergency power for disaster relief or military FOB's. They would broadcast power to a blimp or something in the stratosphere (~20 km up) using a laser, which would then rebroadcast microwaves to a portable rectenna on the Earth's surface.

The thing is, the reason microwaves are chosen is because they're more or less unaffected by the Earth's weather; the problem with higher frequency implementations is that they get eaten up by atmosphere, and especially raindrops. Microwaves can work in all weathers.

Since there's no weather on the Moon, there's no need for microwaves and big rectennas. Just using lasers should hopefully result in beam spot sizes that would be manageable.

As for L4 and L5, I guess I was under the mistaken impression they were about the same distance as L2. No matter, for a Whipple Crater base, since it's a little on the Dark Side, the SBSP station would have to be @ L2.

http://ursigass2011.org/abstracts/ursi/CHGBDJK-3.pdf

[Andrew_W]

such schemes have been proposed before.

That a Lunar base should be powered by a multi-megaWatt SBSP station? I don't think so. But if you've got a reference, I'd love to see it.

http://www.nss.org/settlement/ColoniesInSpace/colonies_chap06.html

The colony will earn its economic keep by building power satellites. The first of these will be maneuvered toward the moon. Forty thousand miles above the near side, it will be stabilized in position, its power beam directed to the lunar base. There, solar power will at last replace nuclear power.

Nuclear plants will serve for the early years of the lunar base as a ready source of power. But it is no more desirable to rely on nuclear power for the long run upon the moon than upon the earth. Just as on the earth, solar power from a satellite will be the long-term supply which ensures the permanence of the lunar base.

Prior to the arrival of the power satellite, the moon-miners will prepare for it. They will build large trough-shaped reflectors of aluminum to concentrate and gather the microwaves to be beamed from space. Possibly they will build a small aluminum plant there to meet the needs of the lunar base and render themselves that much less dependent upon the earth.

[DarkenedOne]

I think it would be many times more complex, more expensive, and significantly less reliable than simply using nuclear power as NASA was planning to do for its moon bases.

You can bring up a reactor the size of a trash can, and set it up on the lunar surface.  It would be able to provide many megawatts constantly for decades.

[DarkenedOne]

such schemes have been proposed before.

That a Lunar base should be powered by a multi-megaWatt SBSP station? I don't think so. But if you've got a reference, I'd love to see it.

http://www.nss.org/settlement/ColoniesInSpace/colonies_chap06.html

The colony will earn its economic keep by building power satellites. The first of these will be maneuvered toward the moon. Forty thousand miles above the near side, it will be stabilized in position, its power beam directed to the lunar base. There, solar power will at last replace nuclear power.

Nuclear plants will serve for the early years of the lunar base as a ready source of power. But it is no more desirable to rely on nuclear power for the long run upon the moon than upon the earth. Just as on the earth, solar power from a satellite will be the long-term supply which ensures the permanence of the lunar base.

Prior to the arrival of the power satellite, the moon-miners will prepare for it. They will build large trough-shaped reflectors of aluminum to concentrate and gather the microwaves to be beamed from space. Possibly they will build a small aluminum plant there to meet the needs of the lunar base and render themselves that much less dependent upon the earth.

". But it is no more desirable to rely on nuclear power for the long run upon the moon than upon the earth."

Sounds like typical anti-nuclear propaganda to me.  Nuclear power is ideal for outer space as there are practically no environment to worry about. 

Reactors here on Earth have to be heavily shielded in order to protect the environment and people.  Since space is already highly radioactive ships and buildings have to have heavy radiation protection anyway.

Also one must consider the point that the moon has more uranium than Earth does.

[savuporo]

such schemes have been proposed before.

That a Lunar base should be powered by a multi-megaWatt SBSP station? I don't think so. But if you've got a reference, I'd love to see it.

just one most obvious example
http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19890017428_1989017428.pdf

[Andrew_W]

". But it is no more desirable to rely on nuclear power for the long run upon the moon than upon the earth."

Sounds like typical anti-nuclear propaganda to me.  Nuclear power is ideal for outer space as there are practically no environment to worry about. 

Reactors here on Earth have to be heavily shielded in order to protect the environment and people.  Since space is already highly radioactive ships and buildings have to have heavy radiation protection anyway.

Also one must consider the point that the moon has more uranium than Earth does.

Sounds like a typical nuclear addicts rant to me 

But since you've brought up the all-things-nuclear-are-wonderful point of view*, how easy would it be to run a nuclear station on the moon during daylight hours? the lunar surface gets pretty hot when the sun's up, and large nuclear requires large heat sinks, on Earth of course we can just run a river through the plant or use evaporative cooling, on the Moon you're restricted to radiators, which are ok (only ok, still going to be a system with a lot of mass) at night, not so great during the day.

* I'm actually politically neutral in the nuclear vs other systems debate, to get anything done in space cost has to come first.

[DarkenedOne]

". But it is no more desirable to rely on nuclear power for the long run upon the moon than upon the earth."

Sounds like typical anti-nuclear propaganda to me.  Nuclear power is ideal for outer space as there are practically no environment to worry about. 

Reactors here on Earth have to be heavily shielded in order to protect the environment and people.  Since space is already highly radioactive ships and buildings have to have heavy radiation protection anyway.

Also one must consider the point that the moon has more uranium than Earth does.

Sounds like a typical nuclear addicts rant to me 

But since you've brought up the all-things-nuclear-are-wonderful point of view*, how easy would it be to run a nuclear station on the moon during daylight hours? the lunar surface gets pretty hot when the sun's up, and large nuclear requires large heat sinks, on Earth of course we can just run a river through the plant or use evaporative cooling, on the Moon you're restricted to radiators, which are ok (only ok, still going to be a system with a lot of mass) at night, not so great during the day.

* I'm actually politically neutral in the nuclear vs other systems debate, to get anything done in space cost has to come first.

"nuclear addicts"

Now that is an interesting term.  You find plenty of people who are against nuclear, but how many people who tolerate nuclear power are against things like wind. 

I am for the best power source whatever that may be.  In this case nuclear power is clear a better power source.  On Earth power failures are tolerable, but in space a prolonged power failure results in life support failure, which results in death.   Satellites of course are exposed to all sorts of threats including attack by enemies, micrometeriods, radiation bursts, and etc.  A nuclear plant would be by comparison far less vulnerable.

[Andrew_W]

In this case nuclear power is clear a better power source.

Why?

[gbaikie]

A variation on the traditional concepts of SBSP occurred to me the other day. Ordinarily, we envision a big array in GEO that beams to a fixed spot on the Earth. Alternatively, some people envision a solar station built on the Lunar surface out of Lunar materials that would beam energy back to Earth.

What I propose is that we reverse the direction of power: set up the solar PV array at an Earth-Moon Lagrange point, and then beam the energy to the Moon in order to power a Lunar research/propellant station.

The motivation is that producing respectable amounts of Lunar propellant is supremely energy intensive: e.g., to produce a mere 10,000 mT of ISRU propellant per year, it would probably require on the order of 20 megaWatts to get this done. Such an array would rival the largest PV arrays that have been built right here on Earth. It would be heavy and would require numerous landings to the surface of the Moon. In addition, building the array in a zone of perpetual sunshine isn't really much help because the angle of the sun is so low, the solar panels will wind up shading each other. The power requirements are probably the closest thing to a true showstopper when it comes to Lunar ISRU.

Mere 10,000 tonnes per year?
Anyhow, on the Moon, you put solar panel vertically. Particularly if one is talking about a large array. You want to put vertically because you would get higher percentage of sunlight.
And you could rotate the tower or panels on tower.
So, as guess if you want power in the range of 10 to 20 megawatt range, one might have tower heights of 1000' feet.
Whereas for more modest power needs the towers might 100' or less.

So it might make sense (be cheaper) to build a 20 megaWatt array at a Lagrange point, and then beam the energy (microwaves or laser?) to the Lunar station.

Pros:
24-7 sunlight
lighter weight
much reduced launch costs

Cons:
have to build rectenna on Moon
this is possibly complicated by locating base at polar latitudes

What do you guys think?

Or send power from earth to L-1 and bounce it to Moon.


The L-points are single point but are also a very large area/volume- and therefore if one has rectenna or whatever not at the point but instead region to the north or south of that point one could be able to send a signal to either the northern or southern lunar polar region.
Also I believe the point of L-1 is somewhere around 75,000 km from the Moon. But again the region of L-1 [or any L-point] is very large- and therefore one could still be in L-1 and be a lot closer to the Moon. Maybe say 30,000 km from the Moon. And/or by using delta-v [equal or similar as is used station keeping for LEO or GEO] one could probably get closer than 30,000 km.

Something related to what I am saying:
"For lunar communications and navigation, Ely recommends spacing three satellites 120º apart in the same elliptical orbit at an inclination of 51º. Each satellite in turn would go screaming down past periapsis (closest approach to the lunar surface) only 450 miles (700 km) above the north lunar pole, but would each linger fully 8 hours of its 12-hour orbit at 5,000 miles (8,000 km) above the horizon over the south lunar pole. In this configuration, two of the three satellites would always be in radio line-of-sight from a South Pole moonbase."
http://science.nasa.gov/science-news/science-at-nasa/2006/30nov_highorbit/
If rather than be 8000 km one could go even higher, say twice as high or say 20,000 km. Whether you dive toward the moon, or drift towards L-1 or L-2- the point is you spend a lot time sort of loitering directly above a pole [your orbital speed is very slow that far from the Moon. And/or said differently, according to above ref, you can't even have stable lunar above 750 km above the Moon- one certainly can't have a stable orbit 20,000 km away from the Moon

[MP99]

Anyhow, on the Moon, you put solar panel vertically. Particularly if one is talking about a large array. You want to put vertically because you would get higher percentage of sunlight.
And you could rotate the tower or panels on tower.
So, as guess if you want power in the range of 10 to 20 megawatt range, one might have tower heights of 1000' feet.
Whereas for more modest power needs the towers might 100' or less.

Those guys looking for perpetual sunlight locations on the Moon's surface had it all wrong. Build a tall enough tower and everywhere at the poles can get perpetual sunlight!

cheers, Martin

[MP99]

such schemes have been proposed before.
a really simple variant of which would be using a regular solar powered rover in permanently shadowed lunar craters, by illuminating its panels from lunar orbit with a laser.

The point here is to collect ice from the permanently-shadowed Lunar craters. These craters are the coldest places in the solar system, so they're getting less than Pluto's surface, ie <1/1500th the Moon's normal insolation. This is necessary for the ice to accumulate.

You laser is obviously intended to be more intense than normal sunlight. If the beam happens to traverse the crater by accident, what will happen to the ice in those craters that you'd been hoping to harvest?

Suspect same goes for microwave beams.

cheers, Martin

[DarkenedOne]

In this case nuclear power is clear a better power source.

Why?

I just explain from a reliability perspective.  I do not see how space based solar power using satellites can be nearly as reliable as nuclear reactors on the surface. 

On top of that satellites are probably one of the few things more expensive than nuclear power.

[JohnFornaro]

WRT the OP:

such schemes have been proposed before.

Point that thing the other way. 

This in response to a suggestion for SBSP, pointed at the Earth.  If yer gonna build one of these things, point it at the Moon, where power is needed, not at the Earth, where there is plenty of power available.

You can bring up a reactor the size of a trash can, and set it up on the lunar surface.

Altho this is the most simple conceptual plan for lunar power that I've ever heard, it will not happen, because there are no plans whatsoever, unless in some skunk works somewhere, to make such a "nuclear battery".  However, as some wag pointed out to me in a very helpful fashion:  It's a free country, go ahead and build one.

[Bill White]

In this case nuclear power is clear a better power source.

Why?

I just explain from a reliability perspective.  I do not see how space based solar power using satellites can be nearly as reliable as nuclear reactors on the surface. 

On top of that satellites are probably one of the few things more expensive than nuclear power.

I continue to favor concentrated solar power for use on the lunar surface. 

http://en.wikipedia.org/wiki/Concentrated_solar_power

Concentrated solar power (CSP) systems, are systems that use mirrors or lenses to concentrate a large area of sunlight, or solar thermal energy, onto a small area. Electrical power is produced when the concentrated light is converted to heat which drives a heat engine (usually a steam turbine) connected to an electrical power generator.

Concentrate the sunlight with mirrors fabricated from Mylar - perhaps even via inflatable heliostats.

Hmmmm . . .

I wonder if an inflatable heliostat could be made to turn and track the sun via selective inflation and deflation?

The launch mass of such a system would seem far less than either nuclear or photovoltaic.

[Warren Platts]

The problem with both thermal solar and nuclear is cooling. There are rivers of running water you can use to cool down your steam. The only thing you could do, perhaps, is an underground system of piping that would cool things conductively, but now you're adding a whole other layer of mass and complexity.

[Warren Platts]

Also one must consider the point that the moon has more uranium than Earth does.

False.

I just explain from a reliability perspective.  I do not see how space based solar power using satellites can be nearly as reliable as nuclear reactors on the surface.

Solar panels have worked reliably in space for decades. Nuke plants have not. Reliability is not the issue. Cost and politics are. There were people like Michio Kaku protesting the Cassini launch because they thought that was too risky.

such schemes have been proposed before.

That a Lunar base should be powered by a multi-megaWatt SBSP station? I don't think so. But if you've got a reference, I'd love to see it.

just one most obvious example
http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19890017428_1989017428.pdf

Thanks savuporo. Very interesting paper, especially the discussion about apertures, etc. It also shows that lasers aren't a panacea because the conversion of electricity to laser light is less efficient than converting to microwaves.

Those guys looking for perpetual sunlight locations on the Moon's surface had it all wrong. Build a tall enough tower and everywhere at the poles can get perpetual sunlight!

cheers, Martin

Right, ideally, you'd want one big panel up on the mountaintop that could spin 360 degrees; thus no solar panels would shade each other. The problem is the size of the panel would be measured in hectares. For 20 megaWatts, assuming 35% efficiency, the panel would be 4 hectares in area: 200 meters on a side if it's square. Thankfully, there's no wind on the Moon to blow it over, but still that's a pretty stupendous construction feat.

[savuporo]

such schemes have been proposed before.

That a Lunar base should be powered by a multi-megaWatt SBSP station? I don't think so. But if you've got a reference, I'd love to see it.

just one most obvious example
http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19890017428_1989017428.pdf

Thanks savuporo. Very interesting paper, especially the discussion about apertures, etc. It also shows that lasers aren't a panacea because the conversion of electricity to laser light is less efficient than converting to microwaves.

This is just one of the classic papers by G. Landis, there is quite a bit more research around on this.
I suggest trawling NTRS and scholar.google.com. For a backgrounder, look here http://www.geoffreylandis.com/laser.htp

[Andrew_W]

The problem with both thermal solar and nuclear is cooling. There are rivers of running water you can use to cool down your steam. The only thing you could do, perhaps, is an underground system of piping that would cool things conductively, but now you're adding a whole other layer of mass and complexity.

I doubt the underground piping would work too well by itself as thermal conduction through rock and soil isn't that high, though maybe with just a little more complexity an underground cold store might work, with another set of piping to the surface that runs coolant only at night.

That SBSP is looking better and better.

[Warren Platts]

That SBSP is looking better and better.

It would be a 3 for 1 technology-wise: (1) it would provide plenty of power for a major Lunar ISRU station; (2) it would be the world's first practical SBSP system; (3) for the SEP folks, 20 megaWatts and 4 hectares of solar panels are probably a minimum that would be necessary for an SEP-powered, human-carrying MTV. RoboChris could probably correct me if I'm wrong about that.

Re: with the latter, thus the DDT&E costs for a major SEP system would largely be absorbed by the Lunar station costs.

[JohnFornaro]

I continue to favor concentrated solar power for use on the lunar surface.

Which is pretty much the proposal that I have left open for discussion on that Spudis and Lavoie thread.  There have been some suggestions that a Sitrling or maybe an Ericsson cycle engine might be prefereable as the prime mover, but there is not the body of knowledge on these heat engines that there is concerning steam turbines.

Steam turbines.  That's what you invest in.

[Warren Platts]

I continue to favor concentrated solar power for use on the lunar surface.

Which is pretty much the proposal that I have left open for discussion on that Spudis and Lavoie thread.  There have been some suggestions that a Sitrling or maybe an Ericsson cycle engine might be prefereable as the prime mover, but there is not the body of knowledge on these heat engines that there is concerning steam turbines.

Steam turbines.  That's what you invest in.

Only one problem: how do you cool your steam once you recover it?

[gbaikie]

In this case nuclear power is clear a better power source.

Why?

I just explain from a reliability perspective.  I do not see how space based solar power using satellites can be nearly as reliable as nuclear reactors on the surface. 

On top of that satellites are probably one of the few things more expensive than nuclear power.

I continue to favor concentrated solar power for use on the lunar surface. 

http://en.wikipedia.org/wiki/Concentrated_solar_power

Concentrated solar power (CSP) systems, are systems that use mirrors or lenses to concentrate a large area of sunlight, or solar thermal energy, onto a small area. Electrical power is produced when the concentrated light is converted to heat which drives a heat engine (usually a steam turbine) connected to an electrical power generator.

Concentrate the sunlight with mirrors fabricated from Mylar - perhaps even via inflatable heliostats.

Hmmmm . . .

I wonder if an inflatable heliostat could be made to turn and track the sun via selective inflation and deflation?

The launch mass of such a system would seem far less than either nuclear or photovoltaic.

Focusing on reflectors, has it's merits. It seems rather silly not to use any reflectors. One can use reflector to merely increase the electrical output of solar panels. One could use them to simply provide light in a dark crater. For days the sun stays in one spot in the sky- and so it's a lot different than on earth. But perhaps having the ability of having reflector which are mobile- and how they can be moved should be considered. All kinds of things are possible, hanging them from wires, rail tracks, rocket powered, or wheeled.

[Warren Platts]

It seems rather silly not to use any reflectors. For days the sun stays in one spot in the sky-

No.... The sun apparently moves ~13 degrees every (Earth) day.

[DarkenedOne]

You can bring up a reactor the size of a trash can, and set it up on the lunar surface.

Altho this is the most simple conceptual plan for lunar power that I've ever heard, it will not happen, because there are no plans whatsoever, unless in some skunk works somewhere, to make such a "nuclear battery".  However, as some wag pointed out to me in a very helpful fashion:  It's a free country, go ahead and build one.

Nasa was developing one when they were preparing to go to the moon for Constellation.

[DarkenedOne]

The problem with both thermal solar and nuclear is cooling. There are rivers of running water you can use to cool down your steam. The only thing you could do, perhaps, is an underground system of piping that would cool things conductively, but now you're adding a whole other layer of mass and complexity.

On the moon you have the moon itself as a thermal mass.  Or you can just radiate it out into space like NASA wanted too. http://www.engadget.com/2008/09/11/nasa-looking-to-go-nuclear-on-the-moon/

[JohnFornaro]

Steam turbines.  That's what you invest in.

Only one problem: how do you cool your steam once you recover it?

Now you're just being polite.  There's actually many problems to solve in this dozen F9 launch scheme I've cooked up.  Cooling the steam is one of them, for sure.  I've considered piping the hot water back into the crater for use in melting.  But I've been stuck at the Stirling engine analysis for a month or two.  Lately, I've been reading up on steam.  Great stuff, largely because of the body of practical experience in steam turbines, when compared to the body of practical experience in Stirling engines.  And, if the initial estimates are correct, there's plenty of water to make steam with up there.

To Darkened One:  I'm sure NASA was developing a nuclear battery.  Are you just pointing this out as a general observation?

[DarkenedOne]

Steam turbines.  That's what you invest in.

Only one problem: how do you cool your steam once you recover it?

Now you're just being polite.  There's actually many problems to solve in this dozen F9 launch scheme I've cooked up.  Cooling the steam is one of them, for sure.  I've considered piping the hot water back into the crater for use in melting.  But I've been stuck at the Stirling engine analysis for a month or two.  Lately, I've been reading up on steam.  Great stuff, largely because of the body of practical experience in steam turbines, when compared to the body of practical experience in Stirling engines.  And, if the initial estimates are correct, there's plenty of water to make steam with up there.

To Darkened One:  I'm sure NASA was developing a nuclear battery.  Are you just pointing this out as a general observation?

Nuclear battery is typically reserved for radioisotope generators.  What NASA was planning on developing was a full-fledged, self-contained reactor that could generate many kilowatts of power with minimal setup. 

[Warren Platts]

The problem with both thermal solar and nuclear is cooling. There are rivers of running water you can use to cool down your steam. The only thing you could do, perhaps, is an underground system of piping that would cool things conductively, but now you're adding a whole other layer of mass and complexity.

On the moon you have the moon itself as a thermal mass.  Or you can just radiate it out into space like NASA wanted too. http://www.engadget.com/2008/09/11/nasa-looking-to-go-nuclear-on-the-moon/

It's not that simple. The link you provided describes a 40 kilowatt system--about 3 orders of magnitude less than what would be required for a major Lunar propellant operation. So you're trading hectares of PV panels for (probably more I'm guessing) hectares of radiative panels (which are going to be in the sunlight for much of the time).

Geothermal-in-reverse is no panacea either. The nuke plant will heat up the rock faster than the rock can lose the heat to surrounding rock. It will work at first, but eventually, you'll heat up the rock to the temperature of the steam. The longer you make your pipe system, the longer the reservoir will last, but now you're talking about bringing a fully fledged drilling operation capable of drilling thousands of feet (not to mention the thousands of feet of casing that would be required).

I think having such a drilling capability will be useful in the long run. My own calculations suggest that water could exist in a liquid phase at about 9 km down below the surface on average. Fractured basalt can actually make a decent reservoir rock (it's sometimes used to store natural gas on Earth). Who knows? Maybe there's even live organisms living down there in Lunar aquifers.

But the point is building up such a drilling capacity is going to take many 20-mT cargo flights (probably around 10 or 20 at a minimum judging from the land rigs I've been around).

It would be nice if you could drop a 20 megawatt nuke plant in a single cargo flight and have it work, but it looks to me the auxiliary equipment that would have to go along with it (massive radiative panels or massive drilling equipment) would require many more than a single cargo flight.

And it's the number of cargo flights that determine the economics of a 1st generation facility, which is after all what we're talking about.

Nuke plants might make sense for a second or third generation facility, especially if some steel making capability could be developed (so you could make your own drill pipe and casing). But for a first generation Lunar station, it's not at all clear that a nuke plant (or solar thermal) would be more economical than a PV system, whether in orbit or on the surface. 

[gbaikie]

The problem with both thermal solar and nuclear is cooling. There are rivers of running water you can use to cool down your steam. The only thing you could do, perhaps, is an underground system of piping that would cool things conductively, but now you're adding a whole other layer of mass and complexity.

On the moon you have the moon itself as a thermal mass.  Or you can just radiate it out into space like NASA wanted too. http://www.engadget.com/2008/09/11/nasa-looking-to-go-nuclear-on-the-moon/

It's not that simple. The link you provided describes a 40 kilowatt system--about 3 orders of magnitude less than what would be required for a major Lunar propellant operation. So you're trading hectares of PV panels for (probably more I'm guessing) hectares of radiative panels (which are going to be in the sunlight for much of the time).

I think think system that provide 40 kilowatts is adequate for lunar mining in the first 4-5 years. Though it wouldn't be a  major Lunar propellant operation. I don't see any need for lunar mining operation to begin with using large scale operation.
The problem now and the problem then is the "shortage" of market.
I think we need to provide a market, and then grow the market.

If you want to "artificially" create a market- and you have govt or some really rich dude, there are ways of doing this which could be more effective [require less time and less money].

[DarkenedOne]

The problem with both thermal solar and nuclear is cooling. There are rivers of running water you can use to cool down your steam. The only thing you could do, perhaps, is an underground system of piping that would cool things conductively, but now you're adding a whole other layer of mass and complexity.

On the moon you have the moon itself as a thermal mass.  Or you can just radiate it out into space like NASA wanted too. http://www.engadget.com/2008/09/11/nasa-looking-to-go-nuclear-on-the-moon/

It's not that simple. The link you provided describes a 40 kilowatt system--about 3 orders of magnitude less than what would be required for a major Lunar propellant operation. So you're trading hectares of PV panels for (probably more I'm guessing) hectares of radiative panels (which are going to be in the sunlight for much of the time).

Geothermal-in-reverse is no panacea either. The nuke plant will heat up the rock faster than the rock can lose the heat to surrounding rock. It will work at first, but eventually, you'll heat up the rock to the temperature of the steam. The longer you make your pipe system, the longer the reservoir will last, but now you're talking about bringing a fully fledged drilling operation capable of drilling thousands of feet (not to mention the thousands of feet of casing that would be required).

I think having such a drilling capability will be useful in the long run. My own calculations suggest that water could exist in a liquid phase at about 9 km down below the surface on average. Fractured basalt can actually make a decent reservoir rock (it's sometimes used to store natural gas on Earth). Who knows? Maybe there's even live organisms living down there in Lunar aquifers.

But the point is building up such a drilling capacity is going to take many 20-mT cargo flights (probably around 10 or 20 at a minimum judging from the land rigs I've been around).

It would be nice if you could drop a 20 megawatt nuke plant in a single cargo flight and have it work, but it looks to me the auxiliary equipment that would have to go along with it (massive radiative panels or massive drilling equipment) would require many more than a single cargo flight.

And it's the number of cargo flights that determine the economics of a 1st generation facility, which is after all what we're talking about.

Nuke plants might make sense for a second or third generation facility, especially if some steel making capability could be developed (so you could make your own drill pipe and casing). But for a first generation Lunar station, it's not at all clear that a nuke plant (or solar thermal) would be more economical than a PV system, whether in orbit or on the surface. 

Wait first of all lets talk about the power required.  Where do you get the megawatt figure.  As with most things it would be wise to start out small than grow.  An operation with a few tens of kilowatts is more than enough IMHO to get things started. 

Secondly lets talk about the power usage.  Radiators are only needed to convert the heat generated from the nuclear reactor into electricity.  A lunar propellant operation would require heat probably more than any other form of energy.  Heat to keep machines and humans warm, to melt lunar ice, and heat for use in high temp electrolysis. 

Thirdly lets talk about ISRU.  Solar panels are require advanced technology, expensive facilities, and expensive materials to manufacture like other semiconductor technology.  Pipes and radiators on the other hand are a simple technology that requires cheap facilities to manufacture, and can be constructed from cheap, abundant materials.  In fact I would not even both with producing anything.  I would just use a tunnel boring machine like the ones used to run pipes, and cement the sides.

4.  The nuclear reactors on subs are trash canned size and deliver tens of megawatts of power.  The limitation is not so much the reactor itself, but the process of generating electricity.

 

[Patchouli]

I continue to favor concentrated solar power for use on the lunar surface.

Which is pretty much the proposal that I have left open for discussion on that Spudis and Lavoie thread.  There have been some suggestions that a Sitrling or maybe an Ericsson cycle engine might be prefereable as the prime mover, but there is not the body of knowledge on these heat engines that there is concerning steam turbines.

Steam turbines.  That's what you invest in.

Only one problem: how do you cool your steam once you recover it?

I would not use water as the working fluid instead I'd go with a molten salt reactor design and a Brayton cycle turbine with gaseous a helium loop.

Even if water was the primary coolant I still would go with a MSR design.

It would be best to have multiple power sources as each has it's own strength and weakness.
Solar wins hands down during the lunar day but during the lunar night nuclear is by far the best option.

If I designed a lunar base it would use solar during the day and at night since the thermal environment is better the reactor would throttle up and be the primary energy source.

As for beamed power it's probably not going to happen anytime soon due to the masses involved and conversion efficiencies.
I certainly would not want it on the critical path.

[Andrew_W]

I would not use water as the working fluid instead I'd go with a molten salt reactor design and a Brayton cycle turbine with gaseous a helium loop.

Even if water was the primary coolant I still would go with a MSR design.

It would be best to have multiple power sources as each has it's own strength and weakness.
Solar wins hands down during the lunar day but during the lunar night nuclear is by far the best option.

If I designed a lunar base it would use solar during the day and at night since the thermal environment is better the reactor would throttle up and be the primary energy source.

As for beamed power it's probably not going to happen anytime soon.
I certainly would not want it on the critical path.

So you're choosing the cost and complexity of two systems over one.

Even looking at just the L1 SPS vs surface based power, I don't see any reason to think surface is cheaper, it'll require an extra 2.4 km/s to land it, it'll need to be structurally stronger (heavier) for working in a gravity field and turning to track the sun. The surface based rectenna, which is the only bit of a L1 power system needed on the Moon, would be very light.

Certainly an L1 SPS system isn't useful until you get to a multi-MW demand, but once there I don't think either surface solar or nuclear could be competitive, having to use both? Nope.

Edit: Thinking about it some more, the area of the rectenna if using microwaves is going to be substantial, so its mass could be considerable if its mass/area isn't very low.

[Proponent]

Wait first of all lets talk about the power required.  Where do you get the megawatt figure.  As with most things it would be wise to start out small than grow.  An operation with a few tens of kilowatts is more than enough IMHO to get things started.

I'm inclined to agree; Spudis & Lavoie envision producing about a tonne of propellant per year per kilowatt of electrical power.

Secondly lets talk about the power usage.  Radiators are only needed to convert the heat generated from the nuclear reactor into electricity.  A lunar propellant operation would require heat probably more than any other form of energy.  Heat to keep machines and humans warm, to melt lunar ice, and heat for use in high temp electrolysis.

Lunar soil is a poor conductor of heat.  Given that the equipment and people, if any, will be surrounded by lunar soil and vacuum, I doubt that keeping them warm will figure prominently in the energy budget.  If it does, we're doing something wrong.

Thirdly lets talk about ISRU.  Solar panels are require advanced technology, expensive facilities, and expensive materials to manufacture like other semiconductor technology.  Pipes and radiators on the other hand are a simple technology that requires cheap facilities to manufacture, and can be constructed from cheap, abundant materials.  In fact I would not even both with producing anything.  I would just use a tunnel boring machine like the ones used to run pipes, and cement the sides.

Surely for power production on the order of just 100 kW there's little point in developing in situ manufacturing of any kind.  Just ship the power plant from earth.  Besides, the low conductivity of lunar soil makes cooling pipes difficult.  Methinks we're looking at radiators if we want either nuclear power (as NASA proposed) or high-temperature electrolysis.

[JohnFornaro]

I would not use water as the working fluid instead I'd go with a molten salt reactor design and a Brayton cycle turbine with gaseous a helium loop.

Maybe that would be the theoretical best thing.  But the salt and the helium are not up there already; water is abundant.  My line of thinking is to use what resources are available. I had no idea what the efficiencies might be, so I took a quick look at Moran & Shapiro "Fundamentals of Engineering Thermodynamics".

The first thing I noticed is the additional complexity required by the addition of a compressor in front of input heat exchanger, run off of the turbine axle as is done in jet engines and similar gas turbines.  According to Moran & all, "a relatively large portion of the work developed by the turbine is required to drive the compressor".  The Brayton cycle works best, that is, its thermal efficiency increases when there is an "increasing pressure ratio across the compressor".  This is related to temperature as well, but there is an upper limit of about 3060 degrees R on the turbine inlet, due to metallurgical limitations.  There are theoretical gains in efficiency which cannot be made due to this limitation.  Further, the cycle is such that the trade is mass flow for power output, and mass flow requires, well, more mass, in the prime mover, but I don't know how that compares to masses for a steam turbine of similar power output.  I studied some of the example problems, but couldn't get a good handle on efficiencies in five minutes.

In short, a closed Brayton cycle looks to be an appropriate type of prime mover for power generation on the Moon, with a solar heat input.  My limited analysis relies on the simplicity of a steam based vapor system, and the ready local availability of a working fluid.

Even looking at just the L1 SPS vs surface based power, I don't see any reason to think surface is cheaper, it'll require an extra 2.4 km/s to land it, it'll need to be structurally stronger (heavier) for working in a gravity field and turning to track the sun.

As always, I'm less concerned about the 2.4km/s.  The trade I'm seeing is a lunar base of the same population no matter the power system.  If people aren't landing and taking off, and prop isn't being shipped to the depot, the choice of power systems becomes far less important.  The way I'm seeing a polar parabolic mirror array, or a polar linear PV panel array working, is that the focal axis of the array, aligned with the sun, rotates round its vertical axis at the leisurely rate of one r per lunar day.  Some power would have to be drained from the system to drive the rotation mechanism, and structural mass would be required to support the rotation mechanism, and all of this would have to be landed on the surface and assembled.

I would say that the assembly of the surface structure would be easier than the zero gee assembly of the space structure of similar net power output.  By "net power output", I mean the power at the plug, available for the base or the cracking plant, or whatever, assuming either system is fully implemented.  Obviously, the rectenna and its associated power conversion mechanism would have to be landed at 2.4km/sec.  Without a more detailed design, I couldn't comment on which one would be cheaper.

One system which I briefly considered would be a constellation of six power sats in a heliocentric polar orbit, which transter power between themselves such that the sat which is over the trackable rectenna is always beaming power to the base/plant.  That's a fair amount of complexity when compared to a rotating parabolic mirror.  Another system would be a heliocentric power sat at L1 or L2, beaming power to a perpendicular rectenna at the poles, which would have to rotate around its vertical axis in the same fashion as the rotating parabolic mirror.  Either of these systems is not readily characterized as "cheaper" than a ground based parabolic mirror or PV array, without a good bit more detail.

I'm pretty sure that cooling would have to be radiative, from an area of perpetual shadow.  The scheme I've been cooking up would feature a large circular area in perpetual shadow, which eventually would cool down to a chilly temperature, and be the home of the cooling fins.  I haven't yet calculated the size of this thing however.

Pesky detail.

[Warren Platts]

Edit: Thinking about it some more, the area of the rectenna if using microwaves is going to be substantial, so its mass could be considerable if its mass/area isn't very low.

From the formula from the savuporo link, Diameter of rectenna = 0.61 * d * wavelength. If highest frequency microwave is used, (wavelength = 1mm), and assuming 60,000,000 m from L-points, then your're looking at a rectenna with a 36 km diameter: much larger than Whipple Crater itself (~10 km).

Clearly, we want to go with lasers. For a wavelenth of 2*10^-7 m, then the receiver would only be the size of about 10 meter diameter.

[Bill White]

Supercritical CO2?

The density of the material allows much smaller turbine blades, which need to be shipped from Earth in any event.

Also, the Moon has abundant O2 and the C can shipped as fuel and run through an ordinary generator to produce electricity and CO2 as a by product.

Sandia Labs News Releases

March 4, 2011
Supercritical carbon dioxide Brayton Cycle turbines promise giant leap in thermal-to-electric conversion efficiency

ALBUQUERQUE, N.M. — Sandia National Laboratories researchers are moving into the demonstration phase of a novel gas turbine system for power generation, with the promise that thermal-to-electric conversion efficiency will be increased to as much as 50 percent — an improvement of 50 percent for nuclear power stations equipped with steam turbines, or a 40 percent improvement for simple gas turbines. The system is also very compact, meaning that capital costs would be relatively low.

https://share.sandia.gov/news/resources/news_releases/brayton-cycle-turbines/

and this

Research focuses on supercritical carbon dioxide (S-CO2) Brayton-cycle turbines, which typically would be used for bulk thermal and nuclear generation of electricity, including next-generation power reactors. The goal is eventually to replace steam-driven Rankine cycle turbines, which have lower efficiency, are corrosive at high temperature and occupy 30 times as much space because of the need for very large turbines and condensers to dispose of excess steam. The Brayton cycle could yield 20 megawatts of electricity from a package with a volume as small as four cubic meters.

http://www.physorg.com/news/2011-03-supercritical-carbon-dioxide-brayton-turbines.html

= = =

Also too, supercritical CO2 handling capability will allow astronauts to clean their clothing without water and clean clothes will improve morale.

Seriously . . .

Cleaning with supercritical carbon dioxide

[JohnFornaro]

My doctor had me cut back on supercritical CO2 last year, so I'm not up to date on the latest developments.

However, that brief news blurb from Sandia is promising.  Today, steam turbines would be available faster.  Of course, I have no idea if the tech is so simple that a development program for the supercritical lunar CO2 prime mover system could be successful within ten years, say.  Given the same amount of development dollars and time, I'd suggest, but wouldn't insist, that steam would win.

It would be a closed system, yes?  The heat source would be the mirror array, yes?  You'd have to trade carbon mass for turbine mass, among other things.

[Bill White]

It would be a closed system, yes?  The heat source would be the mirror array, yes?  You'd have to trade carbon mass for turbine mass, among other things.

Yes, but carbon mass could be scavenged from other applications, including the organic waste stream (including plastic containers) of the lunar base.

Incineration of the waste stream with lunar oxygen will generate CO2

Control the O2 levels and the same facility can make CO which can be used for carbonyl digestion of metallic asteroid fragments (Mond Process).

 

[Bill White]

The problem with both thermal solar and nuclear is cooling. There are rivers of running water you can use to cool down your steam. The only thing you could do, perhaps, is an underground system of piping that would cool things conductively, but now you're adding a whole other layer of mass and complexity.

What is needed is a heat sink.

Pump heat into the sink during the lunar day (have the sink be the "cold end" of a heat engine) then during the lunar night, reverse the process and extract heat from the sink with the sink being the "hot end" of the same heat engine.

Using supercritical CO2 (or helium or ? ? ?) as the working fluid rather than water avoids many of the drawbacks of steam including the availability of a wider band of useable temperatures.

= = =

Volatiles newly extracted from a cold trap can also be heated by this system, helping to cool the working fluid.

= = =

Whether a heat sink system would be cheaper or easier than space based solar power at an L point strikes me as a "trade study" question that cannot be answered without rigorous crunching of numbers.

Also remember, a solar event will radically degrade photovoltaic panels at an L point forcing their replacement. That is another factor to work into comprehensive cost estimates.

[JohnFornaro]

Forgot about the solar events and their effect on PV panels, which would also hold true on the lunar surface.  My preference, today, is for a reflective solar power concentrator, driving a steam turbine/generator combo.  But backing up to a carbon based Brayton cycle for a sec.  True that later on, there would be a "carbon cycle" at the base.  But that wouldn't be true at first.  The prime mover/generator thingy must be landed in fully operable condition, whether that is one nuclear battery "chunk", or a multiple series of landings of a complicated electric generation facility.

[A_M_Swallow]

Any steam based system for the Moon (or space) needs to be leak proof since it is difficult to replace water.

The current landers, launched on an EELV, can probably put 200 kg - 300 kg of cargo on the Moon.  That puts a limit on the mass of the generators, solar collectors and radiators.

[Warren Platts]

Any steam based system for the Moon (or space) needs to be leak proof since it is difficult to replace water.

The current landers, launched on an EELV, can probably put 200 kg - 300 kg of cargo on the Moon.  That puts a limit on the mass of the generators, solar collectors and radiators.

Current landers? There are no current landers, and 200 to 300 kg is certainly not the limit. You're only off by two orders of magnitude. Try reading the ULA lunar/depot-based architecture papers please.

[Proponent]

Volatiles newly extracted from a cold trap can also be heated by this system, helping to cool the working fluid.

Prompted by a discussion with MP99 several posts up the thread, I was thinking about the energy flow at each stage of the propellant-production process:  melting or sublimating ice, electrolyzing it and liquefying the resulting hydrogen and oxygen.  Electrolysis, which involves about 16 MJ/kg at 100% efficiency (50-ish percent seems more likely) dominates.  To an order of magnitude, the other two steps, each involving about 1 MJ/kg if the ice starts at a temperature of 30 K, cancel each other out.  If the feedstock is pure ice, then, the amount of waste heat associated with electrolysis (generating electricity and possibly heating the water) exceeds the heat needed to warm ice by an order of magnitude, and we can't dump much heat that way.

On the other hand, if the concentration of water ice is a just a few percent, then maybe it becomes feasible to dump heat by dumping warm slag some distance away.  This becomes easier if the the feedstock contains volatiles aside from water that you're not interested in trapping (although I suspect you'd probably want to keep most of the volatiles).

EDIT:  "Orders of magnitude" -> "an order of magnitude."

[A_M_Swallow]

Any steam based system for the Moon (or space) needs to be leak proof since it is difficult to replace water.

The current landers, launched on an EELV, can probably put 200 kg - 300 kg of cargo on the Moon.  That puts a limit on the mass of the generators, solar collectors and radiators.

Current landers? There are no current landers, and 200 to 300 kg is certainly not the limit. You're only off by two orders of magnitude. Try reading the ULA lunar/depot-based architecture papers please.

I am not talking about the ULA proposals, they are many years in the future.  I am talking about the Lunar Lander test bed NASA is actually test flying.
http://morpheuslander.jsc.nasa.gov

[Andrew_W]

Well, this discussion has affirmed my belief that planetary surfaces are lousy places to build high energy demand industries.

[Warren Platts]

Wait first of all lets talk about the power required.  Where do you get the megawatt figure.  As with most things it would be wise to start out small than grow.  An operation with a few tens of kilowatts is more than enough IMHO to get things started.

Prompted by a discussion with MP99 several posts up the thread, I was thinking about the energy flow at each stage of the propellant-production process:  melting or sublimating ice, electrolyzing it and liquefying the resulting hydrogen and oxygen.  Electrolysis, which involves about 16 MJ/kg at 100% efficiency (50-ish percent seems more likely) dominates.  To an order of magnitude, the other two steps, each involving about 1 MJ/kg if the ice starts at a temperature of 30 K, cancel each other out.  If the feedstock is pure ice, then, the amount of waste heat associated with electrolysis (generating electricity and possibly heating the water) exceeds the heat needed to warm ice by an order of magnitude, and we can't dump much heat that way.

On the other hand, if the concentration of water ice is a just a few percent, then maybe it becomes feasible to dump heat by dumping warm slag some distance away.  This becomes easier if the the feedstock contains volatiles aside from water that you're not interested in trapping (although I suspect you'd probably want to keep most of the volatiles).

EDIT:  "Orders of magnitude" -> "an order of magnitude."

The 16 MJ/kg is what I got based on the standard enthalpy of formation of water. I've heard that 50% efficiency is about what can be expected, so that's 32 MJ/kg.

I agree that electrolysis is the big consideration, although Jim here has called the liquification process "an energy hog". From what I understand, it's not a simple matter of running it through a refrigerator and radiating the heat: it requires several steps of compression, and compressing air takes a lot of energy. But your main point stands: we can figure out a rough order of magnitude energy requirement by focusing on the electrolysis step.

Now let's consider what order of magnitude production would be required. What we want is an ISRU station that can make a difference. It should be more than a mere demo. It should be more than a self-licking ice cream cone. It should not slow down the Mars project; rather, it should speed it up, or at least make it more viable. It should not merely duplicate what a Mars Direct architecture can do; it should do it better, safer, faster, cheaper.

So what we really want is enough extra propellant to enable a completely reusable Mars architecture.

Keep in mind that to stock an L2 depot, 60% of the LH2/LO2 is spent just lifting it to the depot (leaving enough left over so that the reusable ACES-71 propellant tender can make it back to the Lunar station). So it takes 10 kg of production to deliver 4 kg to L2.

Production Levels:

10 mT/year ---> will lift a couple of ascender modules per year
100 mT/year ---> will make a dent in the cost of running the Lunar station
1000 mT/year ---> will render the base self-sufficient for propellant
10000 mT/year ---> will enable a reusable Mars architecture

So total energy required to electrolyse 10000 mT = 3.2 X 10¹⁴. Let's round that up to 5 X 10¹⁴ to account for the melting and liquification processes plus miscellany energy expenditures.

Half a quad/year is the basic energy requirement.

Thus the minimum power running full blast 24/7 = 5 X 10¹⁴ J/yr / pi X 10⁷ sec/yr = 16 MW

So round that up to 20 MW. That's how I got that figure. So if significant down time due to Lunar night is a factor, then the power would have to upped even more.

To calculate the area of the SBSP array, we can use the following formula:

A = P / C / eff_{L2 array} / eff_{Laser conv} / eff_{Lunar rec}

where A is the area of the L2 SBSP array and C is the solar constant (1366 W/m²)

Conservatively, if we assume the efficiency of the PV arrays (including the receiver on the Lunar surface is 35% and the efficiency of converting electricity to laser light is only 10%, then for 20 megawatts at Lunar surface:

20 MJ/sec / 1.366 KJ/sec/m² / 0.35 / 0.10 / 0.35 = 120 hectares = 1.2 km²

Optimistically: 50% efficiency for the PV arrays and 65% for laser conversion (e.g., with free electron lasers), then we get 9 hectares. So the size of the solar array can vary by an order of magnitude depending on the efficiency. (With 35% PV efficiency the area = 18 hectares.)

For comparison, if the solar panels were emplaced on the Lunar surface, then (assuming 40 MW for peak power), the area would be 8.4 hectares (35% PV efficiency).

[Bill White]

If vapor phase pyrolysis of lunar regolith can allow the extraction of LOX, then concentrated passive solar can provide much of the energy input needed without the need for a 20 MW power plant.

http://www.uapress.arizona.edu/onlinebks/ResourcesNearEarthSpace/resources08.pdf

Advantages of "brute force" thermal pyrolysis:

(1) Flexibility in primary energy sources

(2) No need to import chemical reagants

(3) Metallic by-products

[Hop_David]

If vapor phase pyrolysis of lunar regolith can allow the extraction of LOX, then concentrated passive solar can provide much of the energy input needed without the need for a 20 MW power plant.

http://www.uapress.arizona.edu/onlinebks/ResourcesNearEarthSpace/resources08.pdf

Advantages of "brute force" thermal pyrolysis:

(1) Flexibility in primary energy sources

(2) No need to import chemical reagants

(3) Metallic by-products

Do you know many joules it'd take to make a tonne of oxygen? Glancing at the pdf, I see big temperatures. So I'm guessing getting oxygen from lunar regolith via pyrolysis would take more joules than electrolyzing water.

There are schemes to provide joules in thermal form by concentrated sunlight. But I haven't seen figures on thermal watts per kilogram for mirrors.

[A_M_Swallow]

{snip}


There are schemes to provide joules in thermal form by concentrated sunlight. But I haven't seen figures on thermal watts per kilogram for mirrors.

From the Wikipedia page on solar sails "5 micrometre thick Mylar sail material mass 7 g/m²".  That includes the reflective surface.
http://en.wikipedia.org/wiki/Solar_sail

The supports for the mirror will increase the mass.  A deployment mechanism may also be needed.

[Warren Platts]

Well, this discussion has affirmed my belief that planetary surfaces are lousy places to build high energy demand industries.

Floating in zero g is even worse. Maybe we should just give up on space industries....

If vapor phase pyrolysis of lunar regolith can allow the extraction of LOX, then concentrated passive solar can provide much of the energy input needed without the need for a 20 MW power plant.

http://www.uapress.arizona.edu/onlinebks/ResourcesNearEarthSpace/resources08.pdf

Advantages of "brute force" thermal pyrolysis:

(1) Flexibility in primary energy sources

(2) No need to import chemical reagants

(3) Metallic by-products

It's not that easy. The way it works if first you have to find ilmenite enriched regolith. Even then you'll only get 4 or 5 percent O2 at best. Then you have treat it with imported hydrogen (since you don't want to go for the ice). This makes water. So then you're back to square 1 and have to electrolize the water anyways. Sorry. No dice.

[Andrew_W]

Well, this discussion has affirmed my belief that planetary surfaces are lousy places to build high energy demand industries.

Floating in zero g is even worse. Maybe we should just give up on space industries....

Right, you asked for it...

[Bill White]

It's not that easy. The way it works if first you have to find ilmenite enriched regolith. Even then you'll only get 4 or 5 percent O2 at best. Then you have treat it with imported hydrogen (since you don't want to go for the ice). This makes water. So then you're back to square 1 and have to electrolize the water anyways. Sorry. No dice.

Nope, vacuum pyrolysis of lunar oxygen does not require hydrogen and it can be done entirely with concentrated solar thermal energy Also, the end product is oxygen not water and therefore no electrolysis is required.

Vacuum pyrolysis can be done without catalysts or consumed reagents and without a multi-mega watt space solar power beaming facility. 

But this sub-thread is off thread so I started a new one . . .

[Proponent]

I agree that electrolysis is the big consideration, although Jim here has called the liquification process "an energy hog".

Good point.  If you have a heat sink that's colder than the boiling point of the substance to be liquefied, then pumping isn't necessarily needed for liquefaction.  But even if we dump heat into feedstock at 30 K, I can see that we are likely to need some pumping to liquefy hydrogen.  Perhaps in principal one could construct radiators that would see only the very low temperature of the sky, but that's gotta a pretty big undertaking.

All of this reinforces the conclusion that dumping heat into the slag isn't likely to help a lot in eliminating waste heat.

Now let's consider what order of magnitude production would be required. What we want is an ISRU station that can make a difference. It should be more than a mere demo. It should be more than a self-licking ice cream cone.

I can see that at the scale needed to support a Mars transportation system, lots of electricity will be needed.  I guess for starters, I'm thinking more modestly, along the lines of Spudis & Lavoie, who initially enable merely lunar exploration.  Doesn't it make sense to start on this scale, iron the wrinkles out, and then move on to large-scale production?

[Warren Platts]

I agree that electrolysis is the big consideration, although Jim here has called the liquification process "an energy hog".

Good point.  If you have a heat sink that's colder than the boiling point of the substance to be liquefied, then pumping isn't necessarily needed for liquefaction.  But even if we dump heat into feedstock at 30 K, I can see that we are likely to need some pumping to liquefy hydrogen.  Perhaps in principal one could construct radiators that would see only the very low temperature of the sky, but that's gotta a pretty big undertaking.

All of this reinforces the conclusion that dumping heat into the slag isn't likely to help a lot in eliminating waste heat.

Dissipating the heat leftover from the liquification process is something I have glossed over so far, I must admit. Need to do some more research and figure out a rough estimate of the total heat energy that must be dissipated.

Now let's consider what order of magnitude production would be required. What we want is an ISRU station that can make a difference. It should be more than a mere demo. It should be more than a self-licking ice cream cone.

I can see that at the scale needed to support a Mars transportation system, lots of electricity will be needed.  I guess for starters, I'm thinking more modestly, along the lines of Spudis & Lavoie, who initially enable merely lunar exploration.  Doesn't it make sense to start on this scale, iron the wrinkles out, and then move on to large-scale production?

It depends strongly on the landers you use. The Spudis and Lavoie proposal assume landers that are underpowered and a lot more expensive to operate than the DTAL landers that ULA proposes. As a result, they have to put ISRU propellant production on the critical path for human landings. As such, their plan represents a worse-case scenario. It only gets better from theirs as launch costs go down and better landers are evolved.

Me personally, my thinking is that time = money, so it's best to get to full production as soon as possible, leapfrogging incremental steps wherever possible. The thing isn't going to happen overnight in any case--we're talking 10 or 20, perhaps 30 years, so there will be plenty of time to evolve designs as we go along.

I'm just thinking we need to integrate our Lunar and Mars architectures as much as possible. If there was a clear commitment by NASA to get enough ISRU Lunar propellant to enable a reusable Mars architecture, I believe that politically, this would unify the space community by getting more Mars Society types to support a Lunar ISRU effort.

[Solman]

 I was wondering if a Kraft Ehricke "Solletta" type solution might be worthwhile - a series of large thin mirrors in polar orbit reflecting onto large ground based mirror or PV arrays at the poles to increase insolation and thereby make the ground based solar more effective. This would also allow regions of the Moon away from the poles to be lit up at night.
 The mirrors would have to be able to change orientation as they orbited and this would be a challenge for large mirrors but I wonder if a distributed combination of thrusters and mechanical actuators could do the job.
 Making the entire Moon, not just the poles, much more open to settlement and industrialization would be the payoff.

Sol

[Proponent]

I confess I'm unfamiliar with Ehricke's Solletta proposal, but I worry about mirrors in orbit reflecting sunlight to the ground, simply because the size of the illuminated spot on the ground is at least 0.01 times the distance from the mirror to the ground, 0.01 being the apparent diameter of the sun in radians.  If you manage to place the mirror in a lunar orbit at an altitude of 10 km -- and it would be tough to keep it there for very long, given lunar mascons -- then the spot size is a minimum of 100 m if the mirror is passing directly overhead.  And most of the time, it's not going to be even above the horizon, much less overhead.

[Hop_David]

I confess I'm unfamiliar with Ehricke's Solletta proposal, but I worry about mirrors in orbit reflecting sunlight to the ground, simply because the size of the illuminated spot on the ground is at least 0.01 times the distance from the mirror to the ground, 0.01 being the apparent diameter of the sun in radians.  If you manage to place the mirror in a lunar orbit at an altitude of 10 km -- and it would be tough to keep it there for very long, given lunar mascons -- then the spot size is a minimum of 100 m if the mirror is passing directly overhead.  And most of the time, it's not going to be even above the horizon, much less overhead.

In an earlier thread you had written:

  D_i = (L_i/L_s)*D_{s .}

I had asked how that was derived. Well, I've derived it on my own.



Light rays from opposite limbs of a light source hitting a point diverge by an angle, I call it alpha.

Since angle incidence equals angle reflection, the light rays reflected from the mirror also differ by an angle alpha. This makes the two cones similar. Similar cones gives:

D_i = (L_i/L_s)*D_{s .}

Edit: copied and pasted wrong equation.

[JohnFornaro]

Hey.  Good detective work there, Watson.

[Warren Platts]

Since angle incidence equals angle reflection, the light rays reflected from the mirror also differ by an angle alpha. This makes the two cones similar. Similar cones gives:

D_i = (L_i/L_s)*D_{s .}

You could run it through a Fresnell lens after reflecting off the mirror, and than would straighten out the beam some.

[Solman]

I confess I'm unfamiliar with Ehricke's Solletta proposal, but I worry about mirrors in orbit reflecting sunlight to the ground, simply because the size of the illuminated spot on the ground is at least 0.01 times the distance from the mirror to the ground, 0.01 being the apparent diameter of the sun in radians.  If you manage to place the mirror in a lunar orbit at an altitude of 10 km -- and it would be tough to keep it there for very long, given lunar mascons -- then the spot size is a minimum of 100 m if the mirror is passing directly overhead.  And most of the time, it's not going to be even above the horizon, much less overhead.

 Actually I was thinking of something at higher orbital altitude - maybe 100 km and providing 1/100th of a sun or so over a much larger area for general illumination and perhaps emergency PV power for a base that used PV away from the poles. Of course 1% of the daytime power might not be worth bothering with I'll admit.

[Hop_David]

Since angle incidence equals angle reflection, the light rays reflected from the mirror also differ by an angle alpha. This makes the two cones similar. Similar cones gives:

D_i = (L_i/L_s)*D_{s .}

You could run it through a Fresnell lens after reflecting off the mirror, and than would straighten out the beam some.

For an example, I'll use a geosynch mirror beaming to earth. A 336 kilometer mirror 36000 kilometers high would subtend .5 degrees, the same as the sun. Not coincidentally, using the equation proponent provided, 336 km is the minimum spot on earth.

So to provide full sunlight, the mirror must subtend at least the same angle as the sun. (That's assuming the mirror's perfectly reflected)

Is it possible to have a mirror that's just as bright as the sun but subtending a smaller angle?



Above is a mirror that subtends half the angle as the original light source. It's area is 1/4 that of the original light source. To match brightness, it must be 4 times as hot.

I emailed a university professor noting I seem to have recalled a demonstration such a mirror could be used to power a perpetual motion machine. He replied:
"There is a very easy reduction to a perpetual motion machine. It goes like this: The temperature of the sun's surface is about 6000 K, if it were
possible to concentrate sunlight to a mathematical point, the
irradiation would tend to infinity, and so would the temperature of
any object placed at that point. Thus heat, in the form of solar
light, would be flowing from 6000 K into an object at a higher
temperature, which is forbidden by the 2nd law of thermodynamics
(Clausius statement). This would hold even if the temperature isn't
infinite (it's enough if it surpasses 6000 K) or only a fraction of
the light is collected (any sunbeam would do)."

I believe it would take the powder of ground up unicorn horns to get a smaller D_i than Proponent's equation gives.

[Proponent]

For an example, I'll use a geosynch mirror beaming to earth. A 336 kilometer mirror 36000 kilometers high would subtend .5 degrees, the same as the sun. Not coincidentally, using the equation proponent provided, 336 km is the minimum spot on earth.

I think you mean 360 km rather than 336 km.

I emailed a university professor noting I seem to have recalled a demonstration such a mirror could be used to power a perpetual motion machine. He replied:

"There is a very easy reduction to a perpetual motion machine. It goes like this: The temperature of the sun's surface is about 6000 K, if it were possible to concentrate sunlight to a mathematical point, the irradiation would tend to infinity, and so would the temperature of any object placed at that point. Thus heat, in the form of solar light, would be flowing from 6000 K into an object at a higher temperature, which is forbidden by the 2nd law of thermodynamics (Clausius statement). This would hold even if the temperature isn't infinite (it's enough if it surpasses 6000 K) or only a fraction of the light is collected (any sunbeam would do)."

I know little about non-imaging optics, but according to the relevant Wikapedia entry it is actually possible to produce a spot that is brighter than the surface of the source.  What remains impossible, and prevents a violation of the 2nd law, is to illuminate an object over its entire surface with an intensity greater than that of the source.

Unfortunately, as far as I can tell, non-imaging optical technology seems to work only over short distances.

[Hop_David]

For an example, I'll use a geosynch mirror beaming to earth. A 336 kilometer mirror 36000 kilometers high would subtend .5 degrees, the same as the sun. Not coincidentally, using the equation proponent provided, 336 km is the minimum spot on earth.

I think you mean 360 km rather than 336 km.

I call diameter sun 1.4e6 km and distance to sun 1.5e8. Perhaps these roundings have led to our different results. I attached the spreadsheet I whomped up for this.

I emailed a university professor noting I seem to have recalled a demonstration such a mirror could be used to power a perpetual motion machine. He replied:

"There is a very easy reduction to a perpetual motion machine. It goes like this: The temperature of the sun's surface is about 6000 K, if it were possible to concentrate sunlight to a mathematical point, the irradiation would tend to infinity, and so would the temperature of any object placed at that point. Thus heat, in the form of solar light, would be flowing from 6000 K into an object at a higher temperature, which is forbidden by the 2nd law of thermodynamics (Clausius statement). This would hold even if the temperature isn't infinite (it's enough if it surpasses 6000 K) or only a fraction of the light is collected (any sunbeam would do)."

I know little about non-imaging optics, but according to the relevant Wikapedia entry it is actually possible to produce a spot that is brighter than the surface of the source.  What remains impossible, and prevents a violation of the 2nd law, is to illuminate an object over its entire surface with an intensity greater than that of the source.

I will have to read that article carefully. I suspect it will cause me to change my opinions/models.

Unfortunately, as far as I can tell, non-imaging optical technology seems to work only over short distances.

[Proponent]

Yes indeed, the angular size you use is more accurate and somewhat less than 0.01 radians.

I agree with your spreadsheet.  By the way, for the very small angles involved here, you can approximate tan x as x, for x in radians.

I will have to read that article carefully. I suspect it will cause me to change my opinions/models.

As I read it, it's not going to change things much.  It sounds like non-imaging techniques can make for somewhat cheaper or more-efficient solar collectors and radiators, but as far as I can tell they're not going to produce a big boost in achievable efficiencies or temperatures.

[Hop_David]

By the way, for the very small angles involved here, you can approximate tan x as x, for x in radians.

(looking at triangles....) Yes I can see that. Length leg adjacent gets closer and closer to length hypotenuse as the right triangle looks more and more isosceles. And sin(x)/x --> 1 as x--> 0.

Now I can see why you like to call the sun's angle .01 radians.

[JohnFornaro]

I just checked out a bit of Hop's page, and noticed the "sinus" drawing of the guy's nose.

A memnonic that I remenber from Junior High School regarding these angles is the famous Indian Chief: Soh-Cah-Toa.

Sine = Opposite over Hypotenuse
Cosine = Adjacent over Hypotenuse
Tangent = Opposite over Adjacent