Hacker News Reader

Comment by dvh | Hacker News Reader

If we just assume the over-budget of $14 billion that would give us 13 GW of solar, and nuclear fanboys still thinks nuclear is viable option? And the nuclear plant is only 2.3GW. Seriously nuclear has to die, it is ridiculously expensive even in the best case scenario, it's a no brainer.

View on HN

"13 GW of solar" is going to have a way lower capacity factor, as well as a shorter life-span.

At the very least, you should compare life-cycle $/MWh, not $/GW.

The Energy Information Administration regularly produces forecasts for levelized cost of electricity ($/MWh). You can see the present forecast here, which makes projections for plants entering service in 2026 (see table 1b):

https://www.eia.gov/outlooks/aeo/pdf/electricity_generation....

Looking at the older report from 2018, we can see costs for plants entering service in 2022:

https://web.archive.org/web/20181227232223/https://www.eia.g...

In 2018 the LCOE for new nuclear built in 2022 was estimated at $92.60/MWh while that from solar PV was estimated at $63.20 (before tax credits) or $49.90 (after tax credits).

The newer report with projections for 2026 shows that both nuclear and solar PV get tax credits, and at that time the nuclear tax credit is actually larger. The after-tax LCOE is $30.43 for solar PV and $70.59 for nuclear.

The earliest report in this series was from 2016. The AP1000 reactors currently under construction in the US started building before 2016, back when solar was considerably more expensive. The decision to build new AP1000 reactors was economically rational given the then-current costs of solar and the then-projected costs of new reactors. The reactor projects went tremendously over budget and utility scale PV costs subsequently plummeted, so building new reactors in South Carolina today would be a much more dubious choice, but I don't fault the decisions made back then.

In 2020 in the United States, utility scale solar PV had a capacity factor of 24% while nuclear power had a 92% capacity factor:

https://www.eia.gov/electricity/monthly/epm_table_grapher.ph...

Not to mention the amount of land required to generate that much power.

Generating solar or wind power does not take the land out of use.

Both are compatible with agriculture. Agricultural productivity under partial solar cover is higher than without, for less water input. Pasture is probably the easiest to make compatible, and there is a very great deal of pasture, although little of it is irrigated. The prospect of less irrigation need will drive solar onto more intensive cropland.

Maybe in an ideal world, but I think it would require way more coordination and upkeep than you are giving credit for in order to do that at the scale required. Way easier to maintain a couple dozen distributed nuke plants with wind and solar to handle daytime peaks and residential, personal use where appropriate.

What matters is cost, and the cost picture is crystal-clear: Nukes are very expensive, and not getting any less so. Solar/wind/storage cost is much lower and still plummeting.

It's a no brainer, right. Until you need to turn on the lights at night, that is.

Reply on news.ycombinator.com