Comment by bryanlarsen
2 days ago
Actually, that report is stronger than you're implying.
It's saying solar + batteries is enough to supply 97% of power cheaper than any other way in sunny locales.
It's possible to get 99.99% of your power with solar + batteries, you'd just need a lot of batteries. The news is that batteries have got so cheap that you're better installing enough batteries to hit 97% and leave your natgas peakers idle 97% of the time. That number used to be a lot lower, and that 97% number will be higher every year.
The other cool thing about that report is that it gives a number of 90% for non-ideal places. Sure solar is cheap in sunny locales, but that solar is cheap in places that aren't sunny is far more exciting to me.
The other thing the report isn't saying is that those numbers improve a lot if you have power transmission or other forms of power generation (say wind). They're calculating things as if you're a datacenter in a single location trying to yourself without any grid connection.
A small amount of other power generation whose output isn't correlated with the sun overhead should do a lot to make the last few percent (which come up when there's many cloudy days in a row) cheaper.
Solar's just knocking it out of the park at this point. Building out anything else new (as in you haven't already started) doesn't really make sense.
It is possible to get >100% from solar + batteries. All energy needs can be handled using only a small fraction of solar radiation reaching the planet’s surface.
That said, using it in aircraft (and a number of boots/submersibles) economically is an unsolved problem, but many other places can use it.
Using it in aircraft cheaply is an unsolved problem. We know how to turn CO2 and water into jet fuel with enough energy input. It's just an order of magnitude more expensive than the fossil alternative.
What a koinkidink, I just saw a news about a research platform for exactly that (okay, it's for ships but still) starting now, with the idea being to use surplus offshore wind electricity which otherwise would go unused: https://www.dlr.de/en/tt/latest/news/2025/synthetic-fuel-fro...
What's the energy and ecological cost of producing and transporting the batteries?
As with everything, an upper bound on the energy cost, how many n-kWh does it take to produce a battery that stores 1-kWh-per-cycle-times-m-cycles, is the $ cost of that {1 kWH, m cycles} battery divided by the $ cost of 1 kWh of energy.
E.g. if a {1 kWh, 1000 cycles} battery costs USD 50 to make, and it's made using electricity that costs USD 0.1/kWh, (USD 50)/(USD 0.1/kWh) = 500 kWh. If it needed more energy than that, they would be getting sold at a loss. As a bonus point, this upper bound naturally includes the entire supply chain including the personal purchases of the people working in the factories that make the batteries, all the way up to any waste from e.g. unnecessary private jet flights made by unwise billionaire owners of the battery companies.
This example battery then allows you to time-shift 1000 kWh of electricity from day to night before it needs replacement or refurbishment.
But note the difference between "energy" and "electricity". This kind of calculation is made more complicated by the actual energies used being quite diverse in cost and type, e.g. Pacific-crossing cargo ships are mostly fossil fuelled, the stuff the mining company uses could be any mix of electric or fossil, the aluminium is extracted from ore electrically but any steel probably isn't, etc.
The ecological cost is also strongly dependent on how far the world has gone in greening itself before that battery was made. The first Li-Ion batteries were made in an industrial base that was mostly fossil powered, new ones in China are made in an industrial base that gets 35% of its electricity from renewables.
way less than for transporting similar amount of energy in coal, oil, natural gas or building nuclear power plant.