undefined | Better HN

0 pointstsimionescu3y ago0 comments

Past advancement is not necessarily an indication of future advancement. Battery tech is still exploring ways to improve Li batteries, but capacities have stopped increasing dramatically as far as I know. Even worse, even at the theoretical maximum capacity for these types of chemical batteries, there just aren't enough resources on Earth to build enough to cover the storage requirements of a huge grid.

Now, is it possible that a new type of battery will appear and take over the market by 2030? Yes. Will there be enough of them to compete with the stabilizing power of a nuclear power plant? Maybe. Will there be enough nuclear + storage so that the best use of that storage will be to shut down the nuclear plant? Absolutely not.

Basically, we need massive amounts of new electrical power (so we can transition transportation to electrical), and we need to shut down all fossil fuel power generation yesterday. Since we can't have enough storage yet for an exclusively green grid, nuclear shouldn't be competing with green energy in this discussion. Our choices are not "all green by 2030 VS green by 2030 + nuclear", they are "green + coal/gas by 2030 vs green + nuclear + less coal/gas". If we're instead talking about 2100, or 2200, then yeah, well probably be able to shut down nuclear as well.

0 comments

5 comments · 1 top-level

ncmncm3y ago· 4 in thread

Lithium batteries are adapted for cars, where weight matters. Anything happening with lithium has no relevance to utility power.

The best use of a nuclear power plant is not to build it at all. Having not built it, the storage built instead at a tiny fraction of the cost will certainly be more than enough to replace it.

Building any nukes means, for each, building several times less new renewable generating capacity than could otherwise be.

tsimionescuOP3y ago

Then what storage of some good few hundreds of megawatts would you build for a tiny fraction of the cost of a nuclear power plant, that also fits where a nuclear power plant could be built?

ncmncm3y ago

Space requirements for storage are not dictated by optimal nuke construction site criteria. In particular, no storage method has the potential to render a whole region uninhabitable, and does not need the square-mile exclusion zone.

There are many alternatives to batteries, for longer term storage. Tanked anhydrous liquified ammonia may be common, particularly for smaller and more isolated utility districts, in part because tankage can be topped up from imports if it runs low.

Underground or sea-floor compressed air are practical, as is buoyancy, in many places. Many more places can use pumped hydro than have an existing dam, or watershed; either up a hill or down to an underground cavity.

The method used will be whatever is cheapest in the place and time constructed. Cost for most methods is falling fast.

Building a nuke is by far the most expensive baseline alternative, provided it can even be approved and built on a useful schedule.

1 more reply

pfdietz3y ago

Let's look at simulation/optimization results as to what it would cost to use renewables + storage to produce "synthetic baseload" to compete with nuclear, using real historical weather data.

(The "synthetic baseload" comparison is favorable to nuclear, as dealing with variable demand can only help renewables as it already requires storage, and that storage can serve double duty in the variable demand case. Similarly dispatchable demand and transmission between countries also only helps renewables.)

https://model.energy/

It's difficult to get nuclear to compete with wind+solar+batteries+hydrogen in the 2030 cost scenarios. Nuclear does least worse in places like Poland then.

viernullvier3y ago

Thermal storage, for instance: https://en.wikipedia.org/wiki/Thermal_energy_storage

j / k navigate · click thread line to collapse