2024 is the year it became clear that we're actually going to do battery energy storage. That it was time to stop soberly reminding anyone too excited about solar that storage is an unsolved problem. In fact I personally became very excited about solar, and briefly thought that this year's wave of battery installations was just the beginning of building Solartopia, where we run on solar power all day and batteries all night. I no longer think we're quite there, and I have a new set of sober reminders, but I'll save those for the next post. I'll start by showing you the data that finally got me on board the ship of solar futurism, and how I learned that Elon Musk has been a solar futurist this whole time.

This is Tesla's battery energy storage sales by quarter, from their 2024 Q3 report:

If it helps you get a sense of scale, a typical nuclear reactor produces about 1 GW of power. It would take 12 GWh to provide that power all night (assuming night is 12 hours long). More than that was installed in just six months. Or, to describe a more typical use of battery energy storage, it takes 4 GWh to provide that power for the four hours or so between sundown and lights out, which has been the biggest issue for solar power. That's what they've been installing every quarter, and in recent quarters about twice that. Of course Tesla isn't the only battery energy storage provider, but I won't get into the rest of the market and just note that at least as of 2023 they had the greatest market share for battery energy storage. I think the sales by GWh are mostly of their Megapack, introduced 2019, which is a shipping container-sized battery energy storage module. Utilities buy dozens or hundreds of them, attach them to a concrete slab, and that's a battery energy storage facility.

But why is Tesla, a car company, supplying these units for the electric grid? Rewinding to 2015, when the first Tesla Energy battery energy storage products were introduced. There was no Megapack yet, but there was a home storage product, and a larger unit that was used in early battery energy storage facilities. You might think that it was just something to do with excess battery production. But, quoting from an article in Nature:

At a press conference in Los Angeles on 30 April, the company’s charismatic founder Elon Musk said that the firm’s lithium-ion batteries would enable economies to move to low-carbon energy sources. Solar energy sources are erratic — but by storing their energy and then releasing it when required, batteries could solve that problem, he said.

A year later, Tesla Energy purchased SolarCity, a solar installation company in which Musk was already heavily invested. So Tesla has long been in the solar energy business, though their "Solar Roof" was a flop. It was easy to ignore Tesla Energy until the recent success of the Megapack.

While Tesla Energy was a significant pivot for a car company, conceptually it is consistent with the original plan. In 2006, Elon Musk published "The Secret Tesla Motors Master Plan (just between you and me)", which began:

As you know, the initial product of Tesla Motors is a high performance electric sports car called the Tesla Roadster. However, some readers may not be aware of the fact that our long term plan is to build a wide range of models, including affordably priced family cars. This is because the overarching purpose of Tesla Motors (and the reason I am funding the company) is to help expedite the move from a mine-and-burn hydrocarbon economy towards a solar electric economy, which I believe to be the primary, but not exclusive, sustainable solution.

Part four of the plan is "While doing above [selling cars], also provide zero emission electric power generation options".

As someone who wasn't aboard the solar futurism ship even last year, I've been speculating about why Musk would have been on it in 2006. Of course he wasn't the only one who saw the solar panels of the time as a near-future solution for generating electricity without emitting carbon dioxide. And within the tradition of futurism, which is not necessarily concerned with existing technology, Drexler noted in Engines of Creation that "the energy we use totals less than 1/10,000 of the solar energy striking Earth", and I'm sure there are many similar numbers in futurist writing.

I'm going to speculate here that Musk, who is said to have a "first principles" approach, may have been exactly the kind of person who would take those futurist calculations seriously. For an example of what "first principles" refers to here, consider SpaceX's project of propellant densification. Rockets—unlike jet engines which combust fuel with oxygen in the air—carry their oxygen in liquid form. Liquid oxygen must be below its boiling point, but can be cooled all the way down to its melting point, and is more dense at lower temperatures. The increase in density is about 10% (calculated in a Mathematica notebook):

Quoting from the Isaacson biography:

In his relentless quest to conquer gravity, Musk had redesigned the Falcon 9. The new version packed more liquid oxygen fuel onto the rocket by supercooling it to minus 350 degrees Fahrenheit, which made it much more dense. As always, he was looking for every way possible to cram more power into a rocket without significantly increasing its size or mass. “Elon kept hammering at us to eke out a tiny percent more efficiency by chilling down the fuel more and more,” says Mark Juncosa. “It was ingenious, but it was giving us a real pain in the ass.” A few times Juncosa pushed back, saying it would present challenges with valves and leaks, but Musk was unrelenting. “There is no first-principles reason this can’t work,” he said. “It’s extraordinarily difficult, I know, but you just have to muscle through.”

Here's my own variant on the calculations showing the potential of solar power. How much of the area of the US would have to be covered with solar panels to fully power it? Assuming panels with efficiency 20% and capacity factor 20%:

Not just less than one percent, but less than a tenth of one percent. There's a lot that I'm ignoring here, like of course the area of the solar farm is larger than the area of the panels, but you're going to get a small number no matter what. And of course even 1% is a lot if we're talking percent of US land area, it's a large country, but it's much less than what we use for farming.

Even now that seems hard to approach. Our solar farms are not yet visible from space; we don't yet have patches of desert turning black. But in 2006 it sounded not just difficult but fantastical, and all ideas for energy storage at scale were purely theoretical. But, still speculating, maybe Musk would have said there's no first principles reason why it can't work.

So here we are, at the end of the year of the big battery, and it looks like the solar enthusiasts were right all along. Looking ahead to my next post on this topic: last year I thought that solar installation had to slow down soon. What's the point of building a solar farm when the grid is already flooded when the sun is shining? But what they're doing is installing more solar panels to charge batteries which discharge for a few hours in the early evening. But if we can supply power in the early evening, why not in the middle of the night? The answer, of course, is that electricity is expensive in the early evening, and that's how you pay off your investment in the batteries. Batteries will get cheaper, and maybe one day cheap enough that they can pay for themselves by discharging at midnight. But in my next post I'll try to get across why even at current prices, batteries have completely shifted how I think about solar. They address an anti-solar argument which seemed very plausible to me until this year: that solar actually makes power more expensive because you still need all the non-solar generation after the sun goes down but you leave it idle during the day.