Hacker News Reader

Comment by doublespanner | Hacker News Reader

The part I hate about the math used in this argument, is that really we should be working with a goal of much cheaper energy production, to enable other green technology.

Yeah, if you use standard new construction capacity planning in some cases solar + wind wins. If you target a much lower average/maximum cost per GW (and higher consumption) nuclear wins.

Things like EVs, electric furnaces for recycling, greener chemical plants and carbon capture mechanisms all become more viable with consistently cheap electricity.

View on HN

> Yeah, if you use standard new construction capacity planning in some cases solar + wind wins. If you target a much lower average/maximum cost per GW (and higher consumption) nuclear wins.

I'd love to see your sources for this. To the best of my knowledge it isn't even close and solar is several times cheaper that nuclear. They used to be more comparable a decade or two ago, but solar costs have dropped dramatically since then.

Mostly the viability studies in the French reactor program.

It heaviy depends on how you set up the comparison. If you look at most current energy markets and say "how can I make money with these rules" the answer is almost always build a small amount of renewables. If you say, how should a government invest to retire coal power and achieve a low and stable energy cost, then nuclear can be viable (in some places).

I've seen several studies, none that reached the conclusion you are putting forward. The closest was one that said a lower, but still high percentage nuclear power in France is optimal for reducing CO2 emissions given the nuclear infrastructure that already exists there.

Do you have any specific studies in mind I may have missed?

Anything French on nuclear is simply suspicious, they have a massive interest in selling it - to then double or treble prices during construction, as seen with Hinckley C.

> To the best of my knowledge it isn't even close and solar is several times cheaper that nuclear.

Only if we build reactors in the modern way rather than like the French did in the 1970s. (The reasons why its so much more expensive are complex, but mostly a regulatory ratchet and an tolerance for risk so low that if applied to the rest of life we'd close down parks as too dangerous)

Ah, you mean back when French wages were much lower?

Nuclear (and construction in general) is a victim of the Baumol Effect https://en.wikipedia.org/wiki/Baumol_effect , where the cost of something increases over time if it does not see labor productivity improvement, simply because other sectors of the economy do see labor productivity improvement.

Inflation adjusted wages in France increased by 33% from 1991 to 2023. During that time the inflation adjusted cost of nuclear power plant construction has gone from around 1500/kWe to 4000/kWe.

It's not just Baumol.

Keep in mind that solar and wind alone can't power a single city. You need something to compensate, something like coal/natgas or storage. The amount of storage you need, depends on geography and local weather conditions. If your storage comes short, even a bit, the amount of conventional power stations you need to keep the lights on is exactly the number if power stations you would have to operate if you never had invested into wind or solar in the first place.

This is usually missing in typical cost calculations for solar or wind.

Nuclear needs the same compensation. The high fixed cost low variable cost model lends nuclear power to only run at 100%.

Take the California grid, peak energy usage is 2x minimum. Nuclear plants are insanely costly when ran at 100%. Imagine running at much lower capacity factors. Say the peaking plants run at 50%, that means the cost for consumers would be ¢2.4-4/kWh. [1]

Logically this entails that if we can solve a nuclear grid then we can solve a renewable grid since they impose the very similar constraints on the grid operators.

[1]: https://www.lazard.com/research-insights/2023-levelized-cost...

>If you target a much lower average/maximum cost per GW (and higher consumption) nuclear wins.

It loses every way. Its LCOE is 5x higher. The PR campaign to save it was about neither its cost nor the environment but economically buttressing the nuclear military industrial complex.

It's SO much more expensive in fact that it's actually cheaper to use wind/solar to electrolyze hydrogen, store it underground in a salt cavern and burn that to generate electricity.

>Things like EVs

Things like EVs are even less suited to nuclear power because they dont need constant power and can charge while electricity is cheap. Ditto electric heating.

The problem with the whole nuclear vs. renewables argument is that we don't have the luxury of choosing anymore. We need a huge amount of carbon-free electricity right now, not just to meet current demand but to actively decarbonize our industry.

The only reason we can realistically get to net zero with batteries and renewables is because we export our polution abroad by having China produce everything. And we then ship it back to us using incredibly carbon-intense modes of transportation.

If we had to onshore all that production and actually count it towards our own emissions we'd have no hope of meeting our climate goals with solar panels and wind power.

This argument is clearly bogus. There's a huge set of preposterous ways of generating electricity. No one is going to say we need to do all of them. So why is nuclear not also in that set? You can't just assume it isn't.

> while electricity is cheap

Electricity is cheap mostly when there is more base load than demand; i.e. at night. I don't think you can have that concept if you want to remove base load and just make electricity when the weather lets you.

Reply on news.ycombinator.com