If you’re going to ship it from Earth, why not use copper?
Well, it would be less copper (by volume), so a good bit thinner. Even thinner it may be a better conductor.
OTOH, weren’t we talking about smelting the metal out of lunar regolith? If so, aluminum is what we would have.
Sure, if we’re going to make it out of local materials, aluminum it is. Bit of a chicken-and-egg problem perhaps though: we need a lot of electric power to refine it from the regolith… 
Realistically of course, even if we can eventually set up industrial extraction of useful materials from lunar sources, it’s going to take a LOT of ‘seed capital’ in the way of shipments of equipment and other resources from Earth.
Aluminum is a better conductor per unit weight. Copper is only useful when space is at a premium (say, when packed densely in an electric motor) or when some of the physical properties are useful (like that it tends to make more reliable connections in household situations). But if you’re just going for a simple run of wire and you’re trying to save mass, aluminum beats copper.
Sodium is even better (again, worse by volume but better by mass), but has some obvious downsides. I think electric airplanes should use liquid sodium wiring, which would double as a heat transfer medium.
I agree that making it from regolith would be the end goal. But you probably need a seed cable to power the initial base. You can expand the wire over time with more strands.
I’m not getting the point of the wire. This is just to provide distribution of solar power around the Moon? Or is a current somehow induced in this wire? It seems like this is limited in that all users would have to be at the same latitude, And laying such a wire from orbit or the ground seems like a monumental task. But maybe I’m not understanding the issues.
This seems like overkill for a fairly simple problem. For one thing, nuclear power is a no-brainer for the Moon, and NASA is working on small nuclear power sources for thr Moon. Second, the really energy intensive stuff, if solar powered, could simply run on a 50% duty cycle - run it when the sun is shining, and shut down when it isn’t. For lower power needs like life support, Kilopower nuclear is more than adequate.
One of the problems with nuclear on the moon has been weight - nuclear reactors are heavy, and we didn’t have a lot of ability to put heavy objects on the Moon. But with Starship, that problem goes away to some extent, and makes it way easier to design nuclear power for the Moon as the weight constraints are less.
NASA wants to have a 40kW reactor ready to place on the Moon’s surface by the end of this decade. That would power a small base indefinitely. And you could put dozens of these things on the Moon, placed exactly where the power is needed. And I’m guessing the most power will be needed at the poles, if we’re going to harvest water.
I’m very skeptical about space-rated reactors, but we’ll see. “This decade” seems very optimistic. And 40 kW is not much for heavy industry.
The point of the wire is so you can have small solar installations spaced equally by longitude, which means half will always be illuminated and you don’t have to deal with 15 days of storage.
Consumers would have to be close to the line, but you can always run additional wires to tap off. Or vary the latitude to intersect interesting places.
You’d lay the wire from the ground with a semi-automated rover. It could be controlled from Earth easily enough; the latency is low enough. 7500 km is not very far; you could drive it in a few months. And really, it makes sense to go to an even high/lower latitude, if you’re really looking to extract water from polar craters. That makes the whole process faster and cheaper. You might get away with something as short as 1000 km.
At the poles there are ridges in permanent sun.
Whatnis your skepticism about nuclear power on the Moon? The ground is a fine heat sink for dispersing excess heat, and the Kilopower units used radiators anyway.
Worried about nuclear waste? Why? The entire surface of the moon is bathed in deadly radiation.
Nuclear thermal rockets do have issues with thrust to weight and heat dissipation, but none of that matters for ground power.
The design and construction of Kilopower-type units is already underway. They use Stirling engines to convert heat to electricity. 40kW per unit is fine, because the system is designed to be modular. A single material processor might have one of them dedicated to it, and a base might have ten.
So what are your concerns?
Oh, I was going to mention that one of the advantages of nuclear power is that it opens up the outer planets and asteroids in a way that solar can’t. And the wire around the moon trick for solar power can’t really be applied elsewhere, so it’s a bit of a dead end. But if we develop nuclear power for the moon, we can use it for ISRU on Mars, and to power rovers and helicopters on Titan or ice melters on Europa. Nuclear Thermal Rockets greatly expand what we can send to other planets.
Nuclear for both spaceships and ground power should be a core enabling technology for the space program, and the better we get at it, the more capability we will have to do big things in space. If we ever want to build so,ething like a telescope that ises the sun’s gravitati9nal lensing to observe exopanets, we will have to use nuclear for both propulsion and power, as there is almost no solar power available at 100 AU.
NASA should be focused on things like that and get out of space lift - which is what they are slowly doing.
Development, mainly.
Starship has changed the way I look at the development process. Lots of examples here, but stainless vs. carbon fiber is the most obvious one.
Iteration time matters (much faster to build up rings of stainless sheet in parallel vs. one thing at a time for CF). Reducing the cost of your prototypes matters (SS is cheaper than CF). Ease of fabrication (welding), using well understood materials (CF under cryo still has lots of unknowns), being able to build things out in the open, etc. all matter. It’s the only way to make rapid progress on hard problems with lots of unknowns.
Nuclear reactors don’t have these advantages. They use expensive materials, are hard to fabricate, are hard to test on Earth, have all kinds of safety issues, and so on. It’s expensive to iterate, and worse, slow. This is what causes projects to balloon in cost and time.
That’s a big reason why I favor solar systems. Again: made from cheap materials, easy to fabricate, easy to parallelize, doesn’t require sophisticated handling, can be made in parallel without specialized equipment, and so on. All the factors you want for a cheap and fast development process. It might be “worse” in some narrow ways, but it greatly makes up for that in everything else.
I agree with that (well, maybe not the asteroids), as I mentioned in the fusion thread. But it’s going to come after we have a foothold on the moon and Mars. We have a few decades to go before that happens.
To be clear, I think NASA and others should be working on nuclear power. I just don’t think it’ll be ready in time for what we want to do in the next decade or two.
OK, so don’t deal with 15 days of storage. Just run your aluminum-smelter for 14 days, and then shut it down for 14 days. At least for the first installations, the smelter will be, for practical purposes, the only consumer of power. And then for the later installations that add other significant consumption, most of them can be shut down at night, too. It’s not until you have human habitation that you need continual power.
I’m not so sure. A lot of machinery will not like the extreme temperature cycles and might prefer heaters be kept on during the lunar night.
Separately, any machinery we do deliver up there will be relatively small, wimpy, low capacity stuff. At least compared to it’s Earthbound brethren. So getting a decent annual throughput from any given machine will do a lot better on a 100% duty cycle than on a 50% duty cycle. As the colony (whether inhabited or not-yet-inhabited) grows, the productive capacity will presumably increase. But so will the size of the marginal demand for production.
My overall take is that planning to waste 50% of the very valuable working time versus having power available via either transport from the day side or via non-solar local generation will be a practical necessity. However expensive building a Moon colony will be, doing it leisurely will really ramp the total financial cost.
Me neither. I can’t see how there would be any induced current in the wire. So why not have a number of solar farms around the moon, connected by a fairly conventional grid. Probably high voltage DC for long distance transmission?
I think you are correct about that. But the moon’s orbit does wobble quite a lot: I wonder how many sites are viable for constant solar generation?
That is exactly what the wire the poster proposed was doing. As they explained in a subsequent post.
Given the Moon’s very slow rotation, it was just a way for the currently sunlit side to generate and carry power to the currently dark side.
Presumably most of the solar generation facilities and also the power consuming facilities would be arranged more or less along one line of latitude just to minimize the total amount of wiring needed to connect everything up.
Or simply a hybrid energy system, where you use solar power for industrial activity when the sun is shining, and Kilopower nuclear for the dark periods where you just run life support and such, or for high powered vehicles like mining machines where solar panels are not likely to work.
It might be easier to double-build manufacturing and run it at a 50% duty cycle (only runs when the sun is shining) than to try to deliver power from one side of the Moon to the other.
Or, my preference of powering almost everything with nuclear, woth solar being used where it makes sense.
For example, with a nuclear powered harvesting machine carrying its own power source, you could run it over the regolith 24/7, using microwaves to extract oxygen and water and other volatiles, leaving sintered brick behind for building rocket launch sites or whatever.
With solar power, you’d have to have rovers and harvesters with batteries, and a charging infrastructure. It would be slower and less efficient.
I posted earlier a recent study that found that there was enough water in the lunar regolith in the northern latitudes to fill a 12 oz bottle of water from every cubic meter of regolith. That was the concentration in Clavius where the Chang’E-5 lander sampled the soil. That’s at 78 degrees south.
Another lab found that 55% of the water could be removed by hitting the regolith woth 250W of microwave energy for 25 mins. Call it 120 Wh. For that, they extracted a maximum of 3.3g of water. So for each kg of water extracted, we’d need about 36.3 kWh, plus the power to run a machine that can shovel up and replace regolith at whatever microwaving rate we can achieve.
To,power such a mixrowave heater, we would need about 30 square meters of solar panels (well, more because of inefficiency and losses). That would produce 2kg of water per hour, assuming the regolith was already there. Collecting the regolith required would likely take as much or more energy. You’d need about 3 cubic meters of regolith for every kg of water.
A machine powered by nuclear power harvesting the regolith and delivering it to a solar extraction site which produces water when the sun shines would work. On the other hand, it’s probably more efficient for a machine to scoop up regolith, microwave it internallymto collect the water, then dropmit back in place. That would require a large amount of solar cells which may not be viable on a moving vehicle.
Good thoughts all.
Directly solar-powered land vehicles make no sense as you say.
Battery-powered land vehicles might. Not as wildland rovers but as machines working a fixed facility like a mine, material refinery, or factory. Even if, like the open seam coal mining in Europe, the mineworks slowly walk across the landscape eventually covering miles and miles.
As you say, nuclear has great potential. Waste heat rejection is always going to be an issue with machinery working in a vacuum. I don’t know how much waste heat comes off the sorts of reactors we might be able to build and deploy on a reasonable timeframe at reasonable cost. That might be one of the longer poles in the tent. I know I don’t know enough to handicap this race; I’m just asking questions.
I will suggest that by terrestrial standards, getting a pound of machinery in place and working will be very expensive. And at first the actual productive capacity will be pitiful versus the demand to finish building out the hab / facility / colony. So letting any piece of equipment sit idle for 2 weeks of 4 in darkness for lack of power seems like a losing plan. Whether it runs on batteries, nukes, or solar power carried by “grid” from the far side, stuff will need to be working all hours of the lunar day and night to make its ROI requirements.
Exactly. And to be honest, I’m only being like 70% serious here. It would work, but I don’t exactly expect it to be built.
The advantage over a purely conventional grid is the simple observation that every circuit needs to be a closed loop, which normally entails having separate outgoing and return wires. But you can eliminate that waste if you stretch the loop to be the size of the entire moon!
And since the long days imply the need for solar plants at all points of longitude, you naturally need a loop that big anyway.
If you wanted redundancy, you could run two or more wires, which would still gracefully degrade to one-wire mode in case of a break.
I have the strong feeling that our current battery tech is many-many years from being used up there…
- heavy as f#uck, for the amount of energy they store
- its freezing cold up there and will kill/render useless batteries
- if not, its f#cking hot up there and will kill/render useless batteries (our current batteries are designed to work well at 20°C … move up or down just measly 20° from there and you kill them in no time
- battery advances have been awfully slow - our current top-of-class lithium batteries are pretty much the same they were 30 years ago, advances have been incremental, but by no means order-of-magnitud
- rather short total-lifecycle (low 1000s of charge-discharge cycles + “shelf-life” issues)
- the complexity of potentially running 10s or 100s of 1000s of cells in a system
I am not too optimistic on running heavy-industry-stuff off of batteries in highly adverse environments in the next 20-50 years