That's assuming that prices stay the same for the materials needed to build the grid. Have you looked into the amount of copper required to build this grid? It is many multiples of the current annual production, this would cause prices to rise enormously.
> Have you looked into the amount of copper required to build this grid? It is many multiples of the current annual production
Yes, I have, and it is, even including aluminium (surprisingly a better choice for various reasons). I think I'm one of the main proponents of this on Hacker News, so you might have even seen my previous comments calculating this.
WRT economics, not my field; all I can say is that higher demand increasing prices will encourage (but not necessarily ultimately create) an increase in supply, which has a difficult to forecast feedback relationship when planning for a specific total output.
At least the initial rampup would be expensive - the cheapest copper is being mined already, so prices would necessarily need to rise to mine the more expensive copper too. How much it would rise long term once the new mines are running is geologically dependant.
Aluminium has different issues, in that it is very very very energy intensive to refine. It may require new coal/nuclear power generation buildout near the mine for smelting the bauxite to make it even feasible at a large scale, otherwise you run into a chicken/egg problem with your grid idea since the grid doesn't exist yet. This is because aluminium smelters can't be shut down at night, they have a single turn on/off in their lifetime otherwise they need a full disassembly and rebuild, more likely scrap. So it can't just be run intermittently on local solar. To some degree it can be ramped down, so maybe a smaller base generation assisted by solar during the day, but it still needs significant power at night somehow.
Aluminum smelters have traditionally only been able to modulate a few percent, and many monitized that to sell demand response to the grid in return for cheaper power.
But its not a fundamental limitation, it was just the cheapest way to take advantage of the existing power system.
Set up somewhere with cheap hydro they can't fully use and ship out the energy as metal, running at nearly full power all the time.
Since there was no benefit to running at a higher variable draw, only costs, no one bothered.
Now there are benefits, so someone made a way to take advantage of them. And smelters who use it will be able to have access to cheaper energy, in more locations:
Aluminum ore is super-abundant. There is no possibility of any country anywhere on Earth running short.
The only per-ton expense to produce it is for electricity, which price will be falling very, very fast. So, no, we will not run short of aluminum, even at many times current production rate. Aluminum production process from ore + electric -> metal is extremely mature. Factories to make the factories have already been running for most of a century.
They will operate at night on wind and, during lulls, drawing down storage. They could keep hot by burning powdered aluminum. New ones will be built preferably where wind is reliable, or (as now) hydro power is cheap. Or they might be built near a cliff, and operate their own pumped hydro.
So, no, intermittent sources are not a problem for them or really for anybody, because storage is easy and cheap.
Absolutely, no shortage of aluminium, as I said, copper is the one where the mining is the bottleneck, for aluminium it is energy for the smelting.
Night time energy storage for the smelters is not a solved issue still and significantly increases the costs beyond just adding sufficient PV panels adjacent to the smelter. Storage is not "easy and cheap" as you assert, otherwise there would be no reason to build the longitudinal grid connections GP was talking about.
Grid connections will be used when they are cheaper than drawing down storage. You need them anyway, and want them as a place to dump excess power when local storage is full.
Storage cost is falling even faster than renewable generation did, coming from behind. Now the factories for storage are under construction. By the time we have enough renewable generation capacity to charge storage from, storage will be very cheap. It would be stupid to build out more storage than you have excess to charge it from, beyond some batteries for load smoothing.
Unsurprisingly, people responsible are doing non-stupid things. (Except placing solar farms in the desert, that is indefensible.)
> (Except placing solar farms in the desert, that is indefensible.)
You should blog or vlog about this. I've seen you write this enough times over the last few years(?), and linking to a longer-form argument may be worthwhile.
(I should do the same for the global power grid; there are some wrinkles I want to explore with that also).
It's trivial, really: heat and dust drag down conversion efficiency and panel life. There are just so many other, better places to put them that may continue being used.
