Hacker News Reader

Comment by bildung | Hacker News Reader

The term of trade is "Levelized cost of electricity", https://en.wikipedia.org/wiki/Levelized_cost_of_electricity

Coal is about 2x, gas about 1,3x and nuclear is about 3.6x the LCOE of wind, in 2020, without differentiating between onshore and offshore wind.

View on HN

Lazard’s Levelized Cost of Energy V15.0: https://www.lazard.com/media/451905/lazards-levelized-cost-o...

In the meantime, the prices for natural gas and coal have doubled: https://tradingeconomics.com/commodity/coal

That metric disregards _when_ the electricity is actually produced - which is actual blocker. If the raw amount of generated electricity was the real problem, then just build solar. The problems are when wind is still and sun is shining for average an hour a day, like in Poland in December.

Just as in most other parts of central Europe winds seems to be perfectly anticyclical to solar, e.g. higher when solar is lower: https://weatherspark.com/y/87583/Average-Weather-in-Warsaw-P...

It's far from covering the needs in December.

That’s why you want a network of interconnected grids to smooth out the distribution.

The metric also disregards the true cost of nuclear: the government gives them as an industry a ~$13 billion liability cap, when there are are plausible trillion dollar+ disaster events (one time the great lakes were almost ruined and that is comparatively minor).

Liability caps sounds fun. I wonder how much liability the government spends on hydroelectric power in the case of disaster, like say flooding of whole cities downstream.

It would make for a fun thought experiment to take a large one, say Grand Coulee Dam, and calculate the total cost if all that water would suddenly flood downstream.

Why not both? We had nuclear plants in the US downstream of dams that had to be refitted after Fukushima because they realized they could lose backup power from flooding in a dam failure and potentially lead to meltdown.

If we're going to talk about liability, that makes nuclear look better compared to fossil fuels not worse, as nuclear disasters are so rare and deaths from something like coal are extremely common.

Why is nuclear, middle-of-the-pack in terms of safety, worth singling out? The only thing I can think of is the human preference for people to die gradually in subtle hard to attribute ways than for a far smaller amount of people to die in dramatic nuclear meltdowns.

If what you say is in fact realistic, nuclear plants would be insurable without capped liability. Insurers are not driven by public opinion.

It’s much harder to buy a contract that pays $1B on a 1:1m event, than $1M on a 1:1k event, even though the EV is the same from both contracts.

Very infrequent events are hard to quantify and so the error bars are wider, making insurance more expensive and just riskier for insurers.

Here is the canonical source I site for power death rates: https://ourworldindata.org/grapher/death-rates-from-energy-p...

How is $13B at all a reasonable liability number for the whole industry?

That's on the order of only the reactor cost itself. The reactor owner may even get compensated instead of the population since some have done leaseback schemes with the operator.

You got it wrong. It’s $13B that the whole industry will pay for a single reactor accident. Every nuclear power plant operator will have to pay $131M per reactor if there is one accident. Each reactor also has a mandatory insurance of $450M. This is to cover liability only.

Separate from that is a mandatory insurance imposed by the NRC of $1.06B per site to cover cleanup costs

[1] https://www.nrc.gov/reading-rm/doc-collections/fact-sheets/n...

Reply on news.ycombinator.com