Good but it should probably be mentioned that solar power production and transmission from geostationary orbit (where the sunlight is nearly constant) would work better and have the same positional advantage. Costs have to be considered, of course.

I still kinda like the idea of dropping a few nukes on crater rims at one of the poles to reduce the height necessary to get to perpetual sunlight though.

I know there’s methods for transmitting solar from orbit to ground, but this made me think about the sci-fi idea of orbiting mainframes that transmit the data to simple terminals.

I think space based solar is probably the way to go. Starship will make launch so cheap it’ll be worth getting the panels into perpetual sunlight.

According to the article author it requires costly infrastructure on the moon, compared to their proposal of beaming energy through microwave from Earth.

Space based solar would also use microwaves to beam the energy to the moon.

Or can you just put some space mirrors in to redirect sunlight to panels on the moon? Maybe even concentrate it?

Might be still simpler at that. 'Specially if you build the solar panels on the moon from lunar materials, and there is lots of silicon there, yup.

Well, in The Moon is a Harsh Mistress, they had a bit of a revolution on the moon and started throwing rocks. Having the power generators situated on earth would keep that scenario at bay.

Still, a good space navy & sufficient sensor coverage is the only real mitigation against people playing with KEWs.

Still I wonder how dangerous a mass driver launched rock from the Moon is as long as the mass driver is originally configured for cargo launch, so "just" a few km/s.

Still could do quite a bit of damage if it hits something but if it does not do any.manuevers &could be detected, then there should be ~days for arranging something to collide with it and turn it to plasma or dust.

But I guess if they could spam it & be creative with the resulting trajectories. Still the mass driver is fixed on the Moon surface, so it might not survive for long until it is itself hit by something in retaliation.

Even having ~days would be pretty impractical if it turns out they target a major metropolis. Let's say they target NYC, which had on the order of 18 million inhabitants in 2022. Moving 18 million people in a few days is going to be a major undertaking that few governments would be able to manage without significant prep time. Just imagine the traffic jams you get, exacerbated by cars stopping in the middle of the highway without fuel because the current fuel infrastructure is not going to manage the demand of several million cars passing by. Apart from the amount of people killed, the property damage would be insane.

IIRC the early Cold War civil defence plans (pre ICBM) envisioned just that - quickly truck/bus people out from urban centers to dispersal locations when nuclear bomber stream has been detected. It apparently worked to some degree and was regularly rehearsed in some areas. Then ICBMs with <30 minute warning windows kinda made it moot.

Being fair to people also prevents revolution.

Come to think of it, the tidally locked orbit kind of makes such a supply seem obvious, doesn't it? I mean it's like a big fat "send power here" sign.

Too bad it's not also in geosynchronous orbit, or we could just use an extension cord.

I'm not an expert or anything, but is this making the assumption that there is 0 loss? You put 1Kw into a microwave transmitter on earth and expect to get 1Kw out of the receiver on the moon?

That's preposterous - I would be surprised if you get more than 5% of the power considering the losses in the transmitter, receiver (and related circuitry), not to mention the atmosphere itself plus whatever little bits get in the way (like, I don't know, the thousands of Starlink satellites and other space junk).

I think the assumption is that Earth power is much cheaper than lunar, so the inefficiency doesn't matter.

The two parts article does acknowledge inefficiencies and losses, but dismisses them

> Current space hardware development costs range from $100,000/kg for basic, boring LEO hardware up to >$1m/kg for deep space robots such as the Mars Science Laboratory rover. A 50 T lunar power system would then cost between $5b and $50b

This assumption seems highly dubious. We're not even defining what this "lunar power system" consists of (solar? nuclear?), but the vast majority of that bulk is going to be repeating elements (solar panels, batteries, turbines) that only need to be designed once, meaning the development costs will scale sub-linearly.

This is one of those late-night-dorm-conversation topics that is fun but pretty much hypothetical. We will not have any bases on the moon for a long, long time. We can't even manage powering our society here on earth.

> We can't even manage powering our society here on earth.

Those are political problems, not engineering ones.

Ah yes, so they’re unsolvable, rather than merely difficult.

> We can't even manage powering our society here on earth.

It's orders of magnitude less power so that doesn't matter.

We’ll have one within 20 years. NASA has plans (annd money) and spacex starship has an even more ambitious plan and schedule (and wants money). We’ll probably just tag along for the ride as starship is going by on the way to mars.

How infeasible would it be to deploy an HVDC cable that actually connects opposite sides of the moon and sidesteps most of the storage issue? The longest one on earth is 2500km so it seems within the realm.

You could hypothetically split the conductors east/west, eliminating the need for most insulation. Conductors could potentially be deployed from space and stitched together with robots or something.

The weight of it would be fantastic.

The temperature on the dark side of the moon is -130 degrees Celsius. The hottest super conductor we've build is ~-140 degrees Celsius which isn't too far away. I wonder if you can use HVDC to transfer the energy from the light side to the dark side over a shorter distance and then use super conductors to transfer it "for free" all over the dark side...

Just about every solution you can think of is feasible with enough engineering effort. We need to think about what we want to optimize.

We need a solution which involves a logistical tail to Earth as tiny as we can, being very light on the mission [ = don't cost much weight/fuel ], and dealing as little as possible with lunar, and in general non-terrestrial, surfaces [ = less uncertainty, easier to fix] . Your suggestion fails the last two, and probably the first if we ever need to replace the cable (unless someone does come up with those ultra-special robots half of these proposals suggest).

These considerations strongly lean towards external power supply solutions. I hadn't considered the variant the article suggested. Neat, but I suspect it's politically infeasible - all nations on Earth will demand a vote, and just look at climate negotiations to see how long this takes and how expensive this will get.

IMHO, the obvious solution given our tech level involves geostationary satellites, requires no tech beyond our current capabilities, covers 100% of the surface and involves relatively little burden on the mission itself or dealing with non-terrestrial surfaces. If we think anything is at risk of going wrong, we can just launch more satellites.

3400KM is the diameter at the equator, lesser at latitude, I would say do-able once aluminum is available in qty and can be extruded, losses shouldnt be too horrible at 1MV. I like the out of the box thinking here.

How about microwave relay instead?

Apologies to Casey if this was in the article and I missed it, but what about nuclear power?

It was in the article.

He didn't explain why we can't make a "beefy" nuclear option. It's also the option I would have expected, especially when you think of nuclear subs.

I didn't think of nuclear subs. It would seem ideal for running a small moonbase. But I don't know how self-contained those reactors are, or how much they weigh. Surely a lot.

There are also the nuclear batteries in some of the longer lived space probes. Those batteries have the major advantage of being very simple - a lump of radioactive material that gets warm - attach a thermocouple and voila! Electricity!

LFTRs can scale down to closet sized, and most of the fuel (Thorium) is stable aside from the ignition uranium plug, but there's a LOT of materials to solve for long service.

What about storing solar thermal for in situ resource utilization : https://www.esa.int/Enabling_Support/Preparing_for_the_Futur...

Storing thermal energy is a good idea (heat up big insulated tanks of lunar regolith and heat the base off them at night). Not sure I’d want to store thermal and convert back to electric.

Maybe we wait for fusion instead?