Hacker News Reader

Comment by Arnt | Hacker News Reader

«“The Germans are telling us: we will be very pragmatic on the issue of nuclear power,” said a senior French diplomat involved in the talks. This meant that “all the biases against nuclear power, which still remain here and there in EU legislation, will be removed.” “This will be a sea-change policy shift,” said a German official.»

Will it really?

https://www.pv-magazine.com/wp-content/uploads/2024/10/Level... is a little unusual, it shows the cheapest nuclear power to be as cheap as the most expensive PV+battery. Still, it's difficult for me to see how this policy change will really change anything.

A real change would require a commitment against market-based production/sale of electricity, e.g. a quota such that power generators using PV/batteries were allowed to produce at most x% of the power in the EO.

View on HN

> Will it really?

Yes, it will. As they are currently drafted, the EU energy regulation forces France to invest in renewable and shift away from nuclear it already has to avoid missing the mandated European target as nuclear is not considered renewable. Currently France is being routinely fined despite providing Europe with a ton of clean energy.

The situation is beyond silly.

Good point. That is a real change, I agree with that, and it makes a great deal of sense for France to want it.

I was only thinking about new investment.

> as nuclear is not considered renewable

Btw, not "all" nuclear is not considered renewable. Making use of the full fuel life cycle through breeder reactors would count as renewable nuclear energy.

> Btw, not "all" nuclear is not considered renewable.

There are perhaps two things being considered here, the EU Renewable Energy Directive [1] and the EU taxonomy for sustainable activities [2]. The former results in France being fined for using nuclear power (as nuclear is not mentioned as being renewable), the latter is a tool for investment categorisation, but weights towards closed cycle and Gen III+ reactors.

[1] https://energy.ec.europa.eu/topics/renewable-energy/renewabl...

[2] https://www.europarl.europa.eu/RegData/etudes/BRIE/2022/6989... page 5

China installs about as much new solar capacity each year as the rest of the world combined. They manufacture the most PVs and have access to the cheapest PVs for that reason. Yet, they also have installed more new nuclear capacity (by far) than the rest of the world combined. Not only they do that, but they also have more nuclear plant being built right now, and more being already approved. Nuclear power is expensive because of the initial construction cost, and that is expensive because we forgot how to build. China didn't.

China is building more nuclear power than the rest of the world, but it's not even remotely keeping up with their renewable buildouts [1]. Mostly this is because renewables like PV modules are built in a factory using standardized mass-production techniques, and nuclear plants (currently) are not.

[1] https://cleantechnica.com/2024/01/12/nuclear-continues-to-la...

In terms of scaling PV, there's some lag time before grid-scale issues appear (due to percentage of total) and battery / nuclear et al. are need to solve them.

China is coming from a huge (and still growing) base load of coal. Still 60%+ of electricity mix. [0]

So even with PV's growth, it's not a systemically problematic component.

Compare that to Germany's mix [1], and you realize that China is aiming their reactor building at a problem they'll have in 10 years. (Depending on how aggressively the rest of the world makes China and India stop burning coal...)

[0] https://www.iea.org/countries/china/electricity

[1] https://www.iea.org/countries/germany/electricity

I don’t think the percentage of nuclear was ever planned to be high enough to serve as a replacement for storage, and China is badly behind those plans.

Nuclear isn't storage: nuclear is base generation and frequency intertia.

I.e. the thing that makes the grid as a whole reliable even with fluctuating renewables

For that, only a fraction of the total is needed.

I mean, unless we're turning electricity into Uranium, nuclear obviously isn't storage.

But when it comes to reliability, storage absolutely is the alternative to generation like nuclear. China already has 57GW of pumped hydro storage (about equal to their whole nuclear fleet), and they added 42GW of new 2.5hr battery storage just in 2024 alone. Given the declining price of both types of storage, it's likely that these will dominate nuclear when it comes to short- and medium-term grid reliability.

What you are saying is correct, but you are not contradicting me. If nuclear was economically completely uncompetitive, China would build zero, or near zero. Yet China has 31 GWe under construction right now. Yes, it's much less than the new solar capacity for a single year, but it's a number that is not zero, and it is also going up rather than down. So China thinks that going forward nuclear is a good investment, most likely because they are climbing the nuclear construction learning curve, and are able to build nuclear for cheaper and cheaper. Just like PVs were 100 times more expensive 30 years ago, there is a possibility that nuclear could be many times cheaper in the future. There is no law of nature that prevents that.

China's nuclear plans are based on targets set by the government: they are not necessarily economically competitive. They initially planned to have 18% of their electricity generation produced by nuclear in 2060, which they estimate as 400GW. Those (percentage) targets don't seem realistic anymore given how relatively quickly their renewable deployments are moving. But even if we just consider the 400GW target, they'll need to speed up nuclear construction by an order of magnitude: for the past five years, nuclear additions have been averaging between 1.2-2.3GW per year for the (down from higher numbers before), which is really not enough to meet those targets.

As for the construction learning curve: the main element that made PV inexpensive was the ability to build it in factories at scale. For nuclear the equivalent is SMRs. China is building experimental versions of those too! To the extent that China perfects SMRs and brings the costs way down, they could definitely put nuclear back on track. In a world where cheap SMRs may be a few years away, I question whether building additional AP1000/EPRs is a cost-effective move (even ignoring the possibility of dirt-cheap storage.) China's relatively slow construction rate also makes me wonder if they've come to the same conclusion.

There is zero chance that Germany will build or reopen nuclear power plants. Neither citizens nor power companies want that.

> nor power companies want that.

Important Context: Power companies were paid by the state to shut down their nuclear power plants, of course they want free money for not producing anything.

Shutting down nukes costs money to do right, it's not money for nothing.

The money for shutting down the nuclear power plant and replacing it with another would have eventually been paid anyway, the money from the govt was for missing out on the income from actual energy production.

That was a one time payment and wouldn't stop them for reopening (accept for the dismantling process) or building new ones

The Greens will take care of that (kind of ironic, really)

The Greens are out of power and I think they're going to stay out of power for a while. The road is free for anyone who wants to invest. (We even have precedent that the state will pay if an operating license is withdrawn, which might happen if the Greens return to power.)

> Neither citizens nor power companies want that.

It looks as if some citizens want it. https://anschalt-konferenz.de/english/

While that conference is partially made up of "normal" citizens, there's also an energy consultancy behind it that specifically advocates for nuclear power - so far from just citizens.

https://anschalt-konferenz.de/english/about-us/

> so far from just citizens

And so we come back to the point of the story. Hopefully Germany will stop indirectly funding anti-nuclear organisations that seek to influence other countries. https://www.europarl.europa.eu/doceo/document/E-9-2023-00217...

It isn't immediately obvious to me that the French School of Economic Warfare is a trusted source on these topics, especially since you just linked to a parliamentary question - I'll have to read up on that. I just wanted to make sure your link isn't taken as purely a citizens initiative, because it isn't one.

You quote a PV-lobbying firm as your source, what do you expect?

Also LCOE is just not adequate to compare these two, as you have additional storage and transmission costs for intermittent, weather-dependent sources, whereas most nuclear power plants can be online providing full capacity more than 90% of the time.

LCOE does not account for that.

This IEA report uses a metric that includes those system costs (value-adjusted LCOE, VALCOE) and it shows nuclear energy is definitely competitive. Especially if managed well and power plants aren't prematurely shut down due to political reasons, like they were in Germany.

https://www.iea.org/reports/projected-costs-of-generating-el...

Competitive now? Or competitive in the near future? All of the components of renewables, including PV modules and storage, seem to be following an inverse exponential in price. This does not seem to be true of nuclear. It's easy to make conservative assumptions that leave you with a bunch of financially unsustainable infrastructure. This would be a total disaster.

> All of the components of renewables, including PV modules and storage, seem to be following an inverse exponential in price.

Definitely not true of offshore wind. https://news.ycombinator.com/item?id=44012502

Also it seems like a fallacy of composition to believe that a cheap system is formed from cheap individual components.

A cheap system is a combination of cheap components, land, and the labor required to assemble it. But in terms of minimized labor costs, you just don’t get a lot better than PV and batteries. And there’s still room for cost reductions from technological improvements and more efficient PV.

> A cheap system is a combination of cheap components, land, and the labor required to assemble it.

The point is that cheap components may require the use of other components to compensate for its deficiencies, and so on.

Solar panels need batteries, synchronous condensers, more transmission capacity, electrolysers, hydrogen storage, backup power stations, demand side response, etc.

And batteries are plummeting in cost as well.

In Australia it was calculated that the extra grid upgrades for renewables comes out to $15B. That is less than the subsidies a single nuclear reactor needs.

To electrify society and industry we need to expand our grids by 1.5-2.5x. Stringing an set of wires, using thicker ones or stepping up the voltage higher is a minuscule cost when already having to uprate the grid.

Gas turbine emergency reserves would be $1B.

> In Australia it was calculated that the extra grid upgrades for renewables comes out to $15B.

AEMO report from 2024 [1] page 13 note 2 very carefully draws the line around what the extra 16B AUD will pay for.

    This value is the net present value of capital costs for transmission augmentation up to 2049-50 only, and does not include the cost of commissioned, committed or anticipated projects.

Committed and anticipated ISP projects :- Project EnergyConnect cost has ~doubled to 4.1B AUD [2] Central West Orana REZ is around 5.45B AUD [3] CopperString has escalated to 14B AUD although that might be partly due to general inflation [4]

> Gas turbine emergency reserves would be $1B.

Gas turbines are now about 2400 USD/kW (3750 AUD/kW), so 1B AUD results in ~267 MW of capacity.

[1] https://aemo.com.au/-/media/files/major-publications/isp/202... [2] https://reneweconomy.com.au/deja-vu-on-transmission-project-... [3] https://www.energyco.nsw.gov.au/sites/default/files/2024-05/... [4] https://www.energyco.nsw.gov.au/sites/default/files/2024-05/... [5] https://gasoutlook.com/analysis/costs-to-build-gas-plants-tr...

Yes. The grid needs to be expanded by 1.5 - 2.5x no matter what to electrify society and industry.

The extra costs are what is needed when more distributed generation is added compared to the baseline.

It also not like the current renewable supply will stop existing.

> The CEO of NextEra Energy said that gas turbines have a multi-year backlog, leading to soaring costs for new gas-fired power plants. Renewables “are cheaper and available right now.”

> “There is a lot of demand for gas turbines right now. You have to get in a long line. It has pushed the prices up,” NextEra Energy CEO John Ketchum said at CERAWeek. NextEra built 16 gigawatts of gas-fired power over the past two decades, and operates a fleet of 26 GW of gas capacity. It also builds renewable energy.

> “We built our last gas-fired facility in 2022, at $785/kW. If we wanted to build that same gas-fired combined cycle unit today…$2,400/kW,” he said. “The cost of gas-fired generation has gone up three-fold.”

Ahh yes, gas turbines are expensive because a supply crunch. Not because they inherently are expensive.

Will this get fixed in the 20 years it would take to build nuclear power? Yes. Will likely even see several cycles of boom and bust before said nuclear plant comes online.

Nuclear power simply is completely out of step with the current energy industry. Too expensive and too slow to build.

> Yes. The grid needs to be expanded by 1.5 - 2.5x no matter what to electrify society and industry.

It is unclear that the grid needs to expand by that much if energy is consumed close to where it is produced, or if energy can be consumed at a less variable rate.

> It also not like the current renewable supply will stop existing.

There are examples of generation technologies getting dropped from the grid (by bans).

> Not because they inherently are expensive.

Before the crunch CCGTs were perhaps $1B/GW ($.7B/GW to $1.15B/GW for peakers according to Lazard 2024). So the $1B for backup isn't enough.

https://pris.iaea.org/pris/worldstatistics/threeyrsenergyava... doesn't show many >90% numbers.

> doesn't show many >90% numbers

The US achieves over 90% capacity factor, as does Finland, and Germany used to.

Also capacity and availability are different things; getting pre-empted by solar should not count against the nuclear plant.

> getting pre-empted by solar should not count against the nuclear plant.

But it makes an already horrifically bad economic calculus even worse.

Reply on news.ycombinator.com