Meta to power its bit barns with energy from space

With AI demand growing, Facebook parent Meta is looking for new ways to power its datacenters, with one ambitious project pledging to send solar power down from orbit. Another agreement offers Meta the opportunity to store enough power to keep its bit barns going, even when the grid is over capacity or down. Zuck-corp says the …

[Anonymous Coward]

An impractical plan to bamboozle naive investors

The numbers don't work because of cost and complexity. The numbers will probably never work as long as getting mass into orbit is subject to...gravity.

Cheaper to just deploy solar panels and batteries down here on Earth. We should be talking about the impact of obstructionist red tape on building solar installations or upgrading the grid. We need clean energy for a lot more than AI, and right now, solar and grid construction get held up by years in lawyer-driven permitting nightmares.

[Anonymous Coward]

Re: An impractical plan to bamboozle naive investors

Well, at least in space you're safe from Trump's "I want this country to run on oil and coal" whims.

That been said real estate in the geosynchronous orbit is rare and valuable, and there will be dense swarms of satellites (all those new humongous constellations in LEO) interrupting the beam all the time. Not sure this will work as well as they hope, but I wish them luck.

[Anonymous Coward]

Re: An impractical plan to bamboozle naive investors

Uncorrected negative externalities and fossil fuel subsidies undermine the value proposition of clean energy wherever it is.

Why is it easier to lease public land for fossil fuel extraction than building a solar installation? That's dysfunctional real estate allocation down here on Earth, where it exists in large excess. As you note, orbital real estate is more limited, and it would be a shame to squander it on lower-value uses because Trump's policies have undermined an effective and abundant terrestrial substitute.

[DS999]

Re: An impractical plan to bamboozle naive investors

I want this country to run on oil and coal

I'll bet if you asked Trump "when we build a base on the Moon should the astronauts mine for coal or run exclusively on solar and nuclear power" he would say "absolutely we should mine for coal!" because there's no way he knows there's no coal on the Moon (let alone why) or why you can't burn coal on the Moon.

[Doctor Syntax]

Re: An impractical plan to bamboozle naive investors

"the impact of obstructionist red tape on building solar installations"

The most important solar installation with the longest history is agriculture.

I don't think I've come across anyone planning to give up their share of the product in order to replace it with electricity generation.

[Anonymous Coward]

Re: An impractical plan to bamboozle naive investors

We have sufficient agricultural land which has been growing more and more productive with technological advances. The problem there is farmers who lack sufficient access to capital to embrace new agtech, as well as a distribution system which overlooks the poor and hungry.

Solar isn't a threat to the food supply. Climate change, however, is.

Solar installations don't prefer to compete for prime agricultural land. Where agriculture is displaced, farmers get paid well for repurposing marginal agricultural land consistent with the merit order effect.

But that's not necessary most places. Solar can be built on land which isn't suitable for agriculture, and in many cases, is currently doing absolutely nothing.

The build-nothings still oppose development in places like deserts, contaminated land, rocky locations, etc.

[retiredFool]

Re: An impractical plan to bamboozle naive investors

I've seen a couple examples where solar actually helps ag. Shade. Sheep liked it and I thought I saw cows grazing the area as well. It might even help the grass as full scorching sun all day long may be too much. The video I saw gave a row of panels about a row and a half of open space and then another row of panels. The panels were high enough off the ground for the sheep to go under.

[nobody who matters]

Re: An impractical plan to bamboozle naive investors

"Solar can be built on land which isn't suitable for agriculture"

Unfortunately for that argument, (in the UK at least), solar is currently being built almost exclusively on highly productive agricultural land (grade 3 or higher), and not on poor quality land (grade 4 or below).

[nobody who matters]

Re: An impractical plan to bamboozle naive investors

I would love to know what exactly the two downvoters think is wrong with what I posted - it is a demonstrable fact, as is plain when comparing the siting of existing and planned large scale solar farms in the UK and a map of agricultural land grading.

[Phil O'Sophical]

Re: An impractical plan to bamboozle naive investors

Ah, but it allows them to show off their 'green' credentials, while building a large gas/diesel "backup" facility. They will then "temporarily" use that backup facility to power the datacenter until the clean energy comes on-stream, while knowing full well (and not caring in the slightest) that it's unlikely to ever happen. "We tried, it's not our fault our supplier couldn't deliver what they promised". Pure virtue-signalling.

[Yet Another Anonymous coward]

Re: An impractical plan to bamboozle naive investors

>"the impact of obstructionist red tape on building solar installations"

I imagine if you have a satellite that can beam 1 GW of energy into a focused infrared beam you can deal with ground based obstructionist red tape rather easily.

In pretty much the same way that the Soviet 1st Army Front dealt with Berlin's one-way system

[steelpillow]

Just not heating this feckin' icebox of a planet fast enough

Why don't they go into space nearer where the energy comes from? No need for them to come back.

[Clausewitz4.1]

Re: Just not heating this feckin' icebox of a planet fast enough

You had quite an idea… megapacks of batteries up there recharging. The logistics to swap them, tough, dunno.

[steelpillow]

Re: Just not heating this feckin' icebox of a planet fast enough

Manufacture them on the dark side of the Moon. Simples.

[elsergiovolador]

Project

one ambitious project pledging to send solar power down from orbit

I bet codename of the project is "Sun".

That said, it sounds like nice cover for zapping machine. Facebook knows where you are and now they'll have ability to vaporise you in an instant, from space.

[cyberdemon]

Re: Project

Yes I saw a documentary about this, it was called "Die Another Day"

But the project codename was "Icarus"

[EricM]

Re: Project

In the 80's we called this idea "SDI".

Imagine China announcing they'd develop a "1GW energy beam from space" - of course only to power an AI data center...

[Chloe Cresswell]

Re: Project

I'm thinking of Real Genius....

The CrossBow project: there's no defense, like a good offense

[Aladdin Sane]

Re: Project

How many solar panels can you fit on a Death Star Planetary Ore Extractor?

[IGotOut]

Meta....

...looking at Musk's bullshit "promises" and going "we'll have some of that"

[Wiretrip]

Re: Meta....

Hahaha yes and he claims to be a physicist!

[Excused Boots]

I do have to question the opening statement of this article, "With AI demand growing,......’.

Just who is demanding this growth?

[MonkeyJuice]

Investors.

[Excused Boots]

Ah yes, OK, should really have known.

[munnoch]

"focused onto existing solar farms"

Whats the efficiency of PV panels in converting solar irradiation into electricity. Pretty pitiful iirc. And they're doing this twice. Once in space and again on the ground.

And how narrow a beam would you need from geostationary orbit to get most of your energy into something the size of a field at ground level? That's some Death Star level shit right there.

The panels on the ground won't be contiguous so more losses there.

Its a cute idea but seems like the usable output will be tiny.

[cyberdemon]

Re: "focused onto existing solar farms"

Well the one in space might just be a focusable array of mirrors (and er, smoke, from all the rockets), but nevertheless the peaceful utility of this idea is dubious, if not entirely spurious.

A bit like autonomous robots, it has conspicuously more utility as a weapon (oh no! We would never ever do anything like that!) than any "peaceful honest capitalist" endeavour..

[Yet Another Anonymous coward]

Re: "focused onto existing solar farms"

>it has conspicuously more utility as a weapon

Except it's in an orbit over the equator. It would only be useful against targets in the tropics - it would be useless against the threat of Greenland

[John Robson]

Re: "focused onto existing solar farms"

Anywhere from 20 to 30% - which isn't pitiful at all, given that the input is plentiful and renewable.

Doing it twice *is* painful, but if they're looking for 100 hours of storage then there's no reason not to use a terrestrial renewable mix (wind, solar, geothermal) as your power source.

[munnoch]

Re: "focused onto existing solar farms"

"given that the input is plentiful and renewable"

Except that the giant solar repeater Death Star is neither plentiful nor renewable. For every kW on the ground it needs to beam at least 3kW from space. Assuming it all hits the target, which it won't. Pitiful starts to look generous...

[John Robson]

Re: "focused onto existing solar farms"

It think you might be mistaking my "not pitiful" for a comment on space based solar beaming down to earth... That's not what I meant at all...

It's perfectly possible do this with terrestrial resources.

Running through two sets of PV is not a good idea.

[Kevin McMurtrie]

ROI

Geosynchronous orbit is way out there. Does a solar collector ever return as much energy as was needed to place it there?

I know it's the same argument as ground solar 20+ years ago, but overcoming gravity sounds like a tougher problem to crack than slicing silicon thinner.

[Clausewitz4.1]

Re: ROI

” Does a solar collector ever return as much energy as was needed to place it there?”

It can be collected by large solar panels and beamed as a high-powered laser 24/7. Problem is the safeguard of the birds, planes, environment in general. I gave this idea in a forum a few years back, likely here.

[Anonymous Coward]

Re: ROI

Problem with that is beam divergence. Physicists and engineers know what that means. Gullible investors don't.

[Yet Another Anonymous coward]

Re: ROI

An executive order overrules the 2nd law of thermodynamics

[DrewPH]

SimCity 2000!

With Disaster mode on...

[Androgynous Cupboard]

“Carbon based storage”

Powered by light from space? I’d almost be amused if Zuckerberg found he’d invested in a coal mine

[[e-mail]]

It's a scam

Its a shame that investors don't read old technical magazines like the IEEE Spectrum, 1978 or so. They would see that this idea came up about 50 years ago and wasn't feasible. It still isn't. Zuck is just fleecing another set of investors.

[That Badger]

"With AI demand growing"

Where's this demand coming from? Surely not from users...