Hacker News Reader

Comment by jmyeet | Hacker News Reader

I don't understand the online obsession with nuclear power in spite of all the evidence that it's simply not economical. Canada needs new power now. Not 15-20 years from now, which is how long it takes to build a new nuclear power plant. And it can be done today, incrementally with renewable sources and before anyone screams "baseload", that's what batteries are for if it really comes down to it.

Nuclear power is the highest cost source of electricity in LCOE terms [1]. We just need to look at Hinkly Point C ("HPC") in the UK. HPC was proposed in 2010, approved in 2016, began construction in 2018 and is scheduled to completion currently somewhere between 2029 and 2031 for the first reactor with the second following 1-3 years after (IIRC). From an initial cost estimate of 15 billion pounds in 2015, it's ballooned to 31-35 billion and may well exceed 50 billion [2][3].

The contracted price per MWh is linked to inflation and currently pushing 140 pounds, about 50% more expensive than offshore wind that could be built in a fraction of the time.

So there is a 35 year contract period for power but HPC has a lifespan of 60 years. What happens after? Market rates. Many will argue it'll get cheaper as the plant is paid off. If that's the case, why hasn't electricity from nuclear sources gotten cheaper as the existing plants have aged?

The answer is the same with any nuclear criticism: "this time it'll be different". Fukushima? "This time it will be different." Chernobyl? "This time it will be different." Spiralling costs? "This time it will be different." Massively delayed completion dates? "This time it will be different."

And we haven't even touched the negative externalities yet. That is, the uranium fuel cycle. Processing uranium ore produces waste. Using fuel rods produces waste. We don't really have a good solution for dealing with that waste. There's a lot of hand-waving about "just store it underground and centuries from now we'll hope they've figured it out". Storage, particularly for the first decade or more is not as easy as the hand-waving makes it out to be. It requires cooling ponds because the waste still produces significant heat. So you need infrastructure from that. UF6/UF4 from procesing aren't a solved problem either.

I will never understand why so many otherwise smart people keep trying to make nuclear happen in their minds.

[1]: https://en.wikipedia.org/wiki/Levelized_cost_of_electricity

[2]: https://www.world-nuclear-news.org/articles/edf-announces-hi...

[3]: https://www.telegraph.co.uk/business/2026/02/20/hinkley-poin...

View on HN

> I will never understand why so many otherwise smart people keep trying to make nuclear happen in their minds.

I don't really get this either. I've come to think that it comes down to two pieces. The easy piece is that some people don't seem to realize just how good renewable power sources have gotten in the last 10-20 years. Nuclear has simply been outcompeted in so many ways. But this happened pretty quickly, so not everyone has gotten the message.

The other one is more subtle. For decades there were a lot of bad attacks on nuclear as a technology. (And a few good criticisms, but for some reason those never seem to get the attention, even though they should -- they're pretty strong arguments!) There's a certain type of person who loves to debunk these bad arguments, and there's plenty of that type of person around here. And that can get you emotionally invested into the thing you've been defending (perhaps rightfully: they were crappy arguments against it), and might keep you promoting it after its natural time has passed.

(To be clear: I don't think nuclear plants are worthless, and I think keeping the ones we've got operating smoothly as base load stations is probably an excellent idea. But I don't think it makes a whole lot of sense to be building more of them these days.)

I see that France has the most nuclear heavy grid and also some of the cheapest energy costs and lowest CO2 emission per unit energy in the world. When I see that matched by a solar / wind focused grid I will believe the cheap renewables hype.

And even when I see that, the low energy density still has its own problems. The amount of resources needed for the panels and batteries is massive in itself. And the land area requirements are going to turn vast swathes of wild land into something like this: https://www.instagram.com/reel/DSUY5dhiVF6/

France has higher prices than several EU countries.

Spain in particular has low prices thanks to their solar and wind, and the Nordics thanks to hydro.

And the French cannot seem to replicate the putatively low price they paid for their first nuclear rollout.

Spain has 3x the emissions intensity of France. The Nordics (some of them) have energy that is cheap and clean like France. That's because they have base load that doesn't emit CO2 like France.

And the germanics have higher price than France, which can benefit from importing cheap spanish power (when not in outage) and reselling it at 5x to germanic countries.

Spain has better weather, more hydro, cheap African gas and still nuclear. And emissions are worse

Probably it depends on what part of the world you are and on what is your goal, what you want to optimize for.

In many countries there are usual systematic weather events where all renewable production goes to basically nothing for few days or even 2 weeks. You can not solve that by improving renewable sources, there isn't enough raw energy they could capture.

Storage for that long is currently impossible and even if it would be, it would be prohibitively expensive. So what you can do, build gas or coal plants. Building those, having people on call all the time, and the opportunity cost is probably many times more expensive than the building cost of renewables themselves.

And you still need to buy and store fossil fuels, you are still dependent on geopolitical issues, and you still produce a lot of CO2.

If your goal is environment protection or reducing climate change, then nuclear is probably better. If your goal is to reduce energy cost then probably renewables + short term battery storage + gas backup is the winner if you use an appropriate electricity pricing model.

Nuclear seems to be the old, known, stable thing, while renewables are the new and shiny thing that solves everything cheaply (and that sounds like it has huge catch). When you are building such critical infrastructure as the electrical grid, then staying safe and choosing the known, but expensive solution might seems to be the right choice for many people.

It is a political choice. Pro-nuclear propaganda in Australia is all about the long time frames, and the fossil fuels needed until they start coming online. Climate targets get to be pushed back, scrapping 2030 targets in favor of 2050 targets. It keeps coal, gas and oil money flowing for another generation. And the problem of actually building and paying for the nuclear power plants is also next generations problem, as they are expected to all be over cost and delayed, and not a priority once all the new gas plants are online. Everybody knows all this, but nuclear still gets traction because when you put lipstick on it and take all the most optimistic estimates from the salesmen, it looks like a pro-environmental policy. One that the right and far right can get behind, because it is not what the greens are saying needs to happen and anything those communists want must be bad.

I don't know if it is similar in Canada. Solar is less viable, relying more on wind. And they have more experience building and running nuclear power plants.

Canada has own supply chain and all recent refurbs were on time and budget.

Ironically, if SA contracted Korea for a npp (and if nuclear was legal there) at the same time as UAE, it'll probably be a net exporter of low carbon power by now.

Australia is expanding gas exploration even under current govt. Climate targets seem to be pushed back regardless

All forms of generation have downsides.

> Canada needs new power now. Not 15-20 years from now,

Building nuclear doesn't stop you from building whatever else you want. Though I assume that Canada being Canada, it'll take 15 years just to complete the requisite negotiations with every indigenous tribe and to arrive at a settlement with whatever environmental and assorted NIMBY groups are already warming up their lawsuit-filing laptops right now.

Also, you're predictably citing a couple of bad nuclear accidents, over like 70 years of nuclear generation. Both are actually pretty well understood. If we applied that risk management logic to forms of transport, you wouldn't even be allowed to walk anywhere.

Chernobyl was almost the largest disaster in all of history. I'm not saying nuclear reactors are unsafe now, but the reality is that a true disaster at a nuclear power plant literally means the end of huge amounts of land, enough to end entire countries or large parts of continents. You can't say things like that about walking or other types of transport...

Chernobyl's reactors were fundamentally unsafe designs from an engineering perspective, to say nothing of the perverse incentives at play because of the Soviet political system. We've learned a lot since the RBMK was designed in the 1960s.

Not convinced. The problem is with the human layer of managing large complicated projects.

Nuclear could become less unsafe once humanity has found ways not to go commity horrble violence every other generation.

>Not convinced.

What, if anything, would convince you?

Compact, mass produced nuclear energy projects with no nuclear proliferation risk and radioactive waste management time limited to less than one human generation's professional career span. That seems like a decent baseline to me.

Not sure if fission will ever be able to reach that. Fusion perhaps? I'd certainly like to see that researched with high priority.

In the short to medium term at the very least, I see more economic potential in simple, modular tech. Cheap generation using solar, wind and water. Matching supply and demand better through storage and interconnects.

I'd also be very interested in actual research on how to actually lower demand, in beating the Jevons paradox.

Some of the GenIV designs would be compact and easily mass-produced.

You'll never get waste management below about 300 years with fission, because that's basically what you get from the fission products. But the really long-term stuff is plutonium and other transuranics. Those are unburnt fuel. Fast reactors and some molten salt reactors are supposed to eliminate that. Bury the fission products for 300 years and they're back to the radioactivity of the original ore.

As an American this seems like a long time to me, but when I lived in Germany it didn't so much. We had a brewery in town that had been operating continually for 800 years.

Proliferation resistance gets complicated but some designs are a lot better at it than others. Almost everything requires at least some enriched fuel for startup, even if unenriched works after that. CANDU reactors don't require enriched fuel at all but they don't achieve the waste requirement. Some designs let you extract usable weapons material from reactor fuel (including current CANDU reactors), with others there's no way to extract fissile that's easier to enrich than natural uranium ore.

It might be doable to centralize startup fuel production in nuclear powers, and use reactors that take unenriched fuel after startup, have no way to extract weapons-grade material, and consume the transuranics.

Fusion of course would fix a lot of this. D-T fusion does produce a lot of neutrons that you could use to make plutonium, but you need those neutrons to make more tritium. You get activated reactors parts but those fit your time requirement.

> You'll never get waste management below about 300 years with fission, because that's basically what you get from the fission products.

That's at least one and probably more like five wars over the reactor territory.

I'm mildly optimistic about fusion's potential...

Bear in mind the decay is exponential, so most of the radioactivity is in the near term. There's no fissile, and by mass it's about 1% as much as we're getting now from conventional reactor spent fuel. Encase it in glass and bury it, and you're probably good.

I'm mildly optimistic about fusion as well. One big advantage is regulatory; fission reactors can be safe if you design them well, but fusion reactors can't do much damage even if they're terrible. The NRC has already decided on a much lighter regulatory regime for them.

Proliferation is not much related to plants. It's cheaper to build a weapon with distributed enrichment capacity for uranium in some caves.

> The problem is with the human layer of managing large complicated projects.

I guess we should stop having large, complicated projects. Potable water mains, road and rail networks, the power grid, the internet, bridges, medicine, etc, are all too complicated for humans to manage.

I mean, nuclear is only the safest form of energy generation that humanity has ever produced, but you're absolutely right.

What's the absolute worst that could happen when a water mains breaks? What's the absolute worst that could happen when a train derails? What's the absolute worst that could happen when a backhoe snacks on a fiber trunk?

Now, what's the absolute worst that could happen when a nuclear reactor spirals out of control?

> What's the absolute worst that could happen when a water mains breaks?

People drink contaminated, unpotable water and die.

> What's the absolute worst that could happen when a train derails?

People die.

> What's the absolute worst that could happen when a backhoe snacks on a fiber trunk?

Life-critical infrastructure that depends on the communication fails in a bad way and people die.

> Now, what's the absolute worst that could happen when a nuclear reactor spirals out of control?

People die.

Nothing in life is without risk.

Nuclear reactors spiraling out of control have killed fewer people per KWH generated than any other source of energy that human beings have come up with.

Hypothetically, a train could derail, the train was carrying nuclear waste, the derailment occurred in a highly populated area, near a Virology Lab. The lab was damaged, which released a deadly form of Smallpox, which spread to every corner of the Earth, killing every single human. That would be pretty bad, but not sure if it would be the absolute worst.

You don't need the nuclear waste in that, the train could derail, be carrying a lazy courier transporting a deadly bio-hazard, and unleash a deadly virus and kill literally everyone. From a human-centred perspective that is probably the worst case.

If we're talking non-human it is a bit harder.

What's the absolute worse if a dam breaks? Oh, only up to a quarter of a million dead [1]. And yet we propose to build more dams because they are a great renewable source of energy.

[1] https://en.wikipedia.org/wiki/1975_Banqiao_Dam_failure

> I guess we should stop having large, complicated projects. Potable water mains, road and rail networks, the power grid, the internet, bridges, medicine, etc, are all too complicated for humans to manage.

I'd rather see this simplified and improved than stopped.

> I mean, nuclear is only the safest form of energy generation that humanity has ever produced, but you're absolutely right.

Ground mounted solar is clearly superior in terms of safety.

The problem with Chernobyl was that (1) it didn't have a containment dome, and (2) it was designed so as the temperature increased, the reaction sped up. It was fundamentally unstable.

Neither of these problems is true of more recent reactors.

We don't make bridges safe by getting humans to cooperate better and cross bridges one car at a time. We make them strong and stable so humans can drive however they like and the bridge is fine. That's how all engineering works, and it applies to nuclear reactors just like anything else.

Chernobyl was triggered by humans but the ultimate problem was it's design. And it shows - Fukushima was triggered by tsunami, but ultimate problem was again the design. But since the design was much better, the impact is much milder vs Chernobyl

> Fukushima was triggered by tsunami, but ultimate problem was again the design.

And honestly it wasn't, not really. It was hit with a record-breaking earthquake and a record-breaking tsunami... and more lives were lost due to evacuation than to the plant shutdown.

The design was bad in the sense they backported generator placement from US BWR projects in basement. It makes sense in US but risks in Japan are different

Yes, radiation killed zero. Evacuation - thousands

To be fair Chernobyl was designed what, 15 years after the invention of nuclear technology? Even discounting all the politicial and management control problems, the engineering and scientific knowledge of nuclear reactor design was still in its infancy. Imagine if we judged the safety of automobiles on pre-Model-T cars. Or steam boilers and engines on the first 20 yearrs of their invention.

What's the worst accident involving a Model T, maybe a dozen dead? Early steam boilers aren't going to be much worse either. Nuclear accidents are essentially unlimited in size. Nothing else can do that kind of country-sized - let alone it being permanent.

Chernobyl showed the potential impact. Fukushima showed that even several decades down the line things can still rapidly run out of control. All the knowledge and experience in the world isn't going to save you when something unexpected happens and things are just waiting to spiral out of control.

Steam boilers have a lot of regulation, all written in blood. Entire boats would explode killing most/all onboard.

https://en.wikipedia.org/wiki/List_of_boiler_explosions

Do you understand the scale of Chernobyl if they had failed to stop the meltdown? Half of eastern europe would be uninhabitable. It would've killed 100s of millions.

It's even worse: the problems of that reactor type were, in fact, well-known at the time.

Which is one of the reasons that reactor would never have gotten an operating license in any western country. Oh, and not having any containment.

Isn't that a little hyperbolic? Sure cancer rates will be elevated wherever the fallout blows but it's not going to end anything.

In terms of severity, Chernobyl was a long way from the worst case.

If the core had melted down to a body of water, the steam flash could have vaporized it & ejected it high into the atmosphere.

That's city-ending, if not quite "continent rendered uninhabitable".

The problem is more about Chernobyl affecting global nuclear deployments imo

> Chernobyl was almost the largest disaster in all of history

Not at all hyperbole when you consider how badly it poisoned the well for future nuclear projects.

When ranking Chernobyl accident for death toll (95–4,000+ deaths) it's very far behind Failure of Banqiao Dam (26,000–240,000), behind 2023 Derna dam collapse (11,300), behind the world's worst industrial disaster - Bhopal disaster (3,787–16,000), behind 1979 Machchhu dam failure (1,800–25,000), about as deadly as Halifax Explosion (1,950 deaths).

Most tragic thing is that Chernobyl accident could have been prevented.

https://en.wikipedia.org/wiki/List_of_accidents_and_disaster...

>Building nuclear doesn't stop you from building whatever else you want.

It kind of does though, since it demands pretty lavish subsidies to be built at all and those subsidies would give WAY more bang for the buck if used on pumped storage, batteries, solar and wind.

You also have to cap liability in case of nuclear disaster. Private insurers won't touch nuclear power with a barge pole unless taxpayers are forced to pay for disaster cleanup. As a taxpayer Id rather not have that liability.

We're not allowed to flood valleys anymore, so pumped storage is not cheap. Maybe that would change if there was a climate emergency cough

They could mass replicate https://en.wikipedia.org/wiki/Drake_Landing_Solar_Community and drop country-wide fossil fuel consumption about 1/3 and save money, but there's no big company pushing that.

What if, we actually instead overbuild power gen and distribution, that it actually supresses price?

Imo, the answer should always be, yes, build, please.

Heavy industries (which support blue collar growth) come to places with cheaper power.

> Building nuclear doesn't stop you from building whatever else you want.

If you build the solar and wind you don't need the nuclear. That's the point.

> Also, you're predictably citing a couple of bad nuclear accidents, over like 70 years of nuclear generation.

Here we go with hand-waving away all the uncomfortable counterexamples.

It's hard to get exact numbers because of plant decmossioning and that some nuclear reactors don't produce electricity (eg they are breeder reactors for plutonium or isotopes for medicine) but an estimate of somewhere between 400 and 440 worldwide seems reasonable. I've also read that fewer than 700 nuclear reactors have ever been built. Not a single one without significant subsidies I might add. Of those 440 (for argument's sake), we've had 3 serious accidents:

1. Chernobyl. The absolute exclusion zone for Chernobyl remains at 1000 square miles ~40 years after the accident with no end in sight. The estimates of the accumulated cleanup costs seem to be at least $700 billion [1];

2. Fukushima. It'll likely take more than a century to clean this up and the cost may well exceed $1 trillion [2];

3. Three Mile Island. Far less significant than the other two but still involved a core meltdown.

Do you have any idea how much renewable power generation $700B and $1T could've bought instead?

But it gets worse. The US nuclear energy doesn't pay insurance representing the true potential cost of a nuclear disaster. The Price-Anderson Act limits liability to (in 2026) $500 million in primary insurance, $15 billion in secondary insurance from an industry-wide fund paid in by operators and there's also another limit I forget on incidents that cover more than one reactor [3]. So how do you get from $15B to $700B or $1T? Why the government of course, which means the taxpayers.

[1]: https://globalhealth.usc.edu/wp-content/uploads/2016/01/2016...

[2]: https://cleantechnica.com/2019/04/16/fukushimas-final-costs-...

[3]: https://en.wikipedia.org/wiki/Price%E2%80%93Anderson_Nuclear...

> If you build the solar and wind you don't need the nuclear.

Don't forget the enormous battery arrays for winter, cloudy skies, or wildfire smoke. Hope you have enough batteries. But you won't, so ok, now you need gas reactors to fill in the blanks. Isn't that what we're trying to get away from?

Ah yes, the ridiculous strawman engineering of saying batteries would be used for seasonal storage.

Then what would you use for seasonal storage?

> Then what would you use for seasonal storage?

Like Germany: coal and the energy provided by the rest of the EU. So Nordic countries hydro and nuclear, and some French nuclear.

While lobbying to make all those interconnections increase domestic prices for the providers.

> Like Germany: coal and the energy provided by the rest of the EU.

So, from actual non-intermittent sources like checks notes nuclear?

No, nuclear is terrible for covering seasonality, since unless it's used with high capacity factor the cost per kWh skyrockets.

> No, nuclear is terrible for covering seasonality,

wat

How is nuclear bad at covering the seasonality of, for example, winter in the Nordics?

It's best for covering a constant demand, not a seasonal component of demand.

If renewables grow to the point where there is no remaining near-constant unmet demand, nuclear has no place at all.

No, it's ok if you schedule maintenance during summer. And with more AC adoption you'll partially reduce this heavy swing. Nuclear at 60%cf is still economical. The problem arises if your plant takes 20y and 20bn to build

In the short term, one would burn natural gas in turbines. The marginal cost of displacing this by using nuclear instead would lead to an enormous cost per unit of CO2 avoided, so high that most other uses of fossil fuels would be eliminated first (like, all use in ground vehicles).

In the long term, either non-fossil fuels burned in turbines (e-fuels like hydrogen or biofuels), or bulk thermal storage of renewable electricity. These both have lousy round trip efficiency (maybe 40%), but that's still cheaper than using batteries, because the capex per unit of storage capacity is far lower, and the cost of the RTE is low when there are so few charge-discharge cycles (as happens with seasonal storage); cost of seasonal storage is dominated by capex, which is why using high-capex batteries for it is such a bad idea.

Personally, I consider bulk thermal storage of cheap DC-coupled PV the most promising approach, as being pursued by Standard Thermal. They claim to be able to deliver 365/24/7 heat at 600 C for $3-5/GJ, which is competitive with Henry Hub natural gas.

I would prefer to reduce emissions using technology that exists today, I know it works, and I have seen it operate at national grid scale, not speculative future tech.

You do highlight something there: the case for nuclear requires one to assume that the competing technologies stop their rapid advance. If not, the 40 (or 60, or 80) year investment horizons needed to partially shore up the bad economics of nuclear become utterly absurd.

(The criticism that renewables don't last as long as nuclear suddenly looks like praise when viewed in this light; renewables don't need those very long time horizons to pay out.)

But making this bet, that renewables will suddenly come up short, that the experience curves will suddenly break their historic trends on the log-log plot, has never worked out well.

Something like hydrogen seems guaranteed to be available if needed. Realize that green hydrogen is needed even in a nuclear-powered world, because of existing hydrogen demand that is currently satisfied by steam reforming of fossil fuels (mostly natural gas). So lots of hydrogen will be made; it doesn't require new technology to make some more.

I'll add that if you are sticking to currently available commercial technologies, nuclear is a loser, since burner reactors are far too fuel-inefficient to last very long on existing estimated uranium resources. The current estimate of uranium resources at 3x current price would provide the world's current rate of primary energy demand for just 5 years, if burner reactors were used.

> the case for nuclear requires one to assume that the competing technologies stop their rapid advance.

1. No, it doesn't

2. Other tech has to actually show this rapid advance, and not be the permanent state of fiction

3. You assume that nuclear is incapable of advances

> But making this bet, that renewables will suddenly come up short, that the experience curves will suddenly break their historic trends on the log-log plot, has never worked out well.

Renewabl;es do come short in one very specific area: they are intermittent, and to account for that they have to be very extremely overbuilt and all available large scale storage is very short-term.

> Something like hydrogen seems guaranteed to be available if needed. Realize that green hydrogen is needed

Speaking of technologies that are permanent fiction. We don't even know how to reliably store it at required scales. All known methods are either extremely complex and volatile, or require large amounts of energy to release hydrogen back, or cannot store much hydrogen to begin with: https://www.sciencedirect.com/science/article/pii/S025405842...

> I'll add that if you are sticking to currently available commercial technologies, nuclear is a loser

something something assuming no rapid advances or something

> 1. No, it doesn't

It does, for the reason I gave. You didn't give a reason why not.

> 2. Other tech has to actually show this rapid advance, and not be the permanent state of fiction

Incredibly, you seem unaware of just how rapidly the cost of solar and wind and batteries have dropped.

If we project the demonstrated experience curve of PV forward another five doublings or so, PV energy will be delivered at under $0.01/kWh. This is basically impossible for nuclear to compete with.

> 3. You assume that nuclear is incapable of advances

Unlike renewables, nuclear hasn't demonstrated a good experience curve. If anything, it has shown a negative experience curve.

But in any case, even if nuclear were capable of rapid advance, this would still argue against assuming 40 (or 60, or 80) year lifetimes for nuclear power plants when calculating their economics. The power plants would be obsolete and uncompetitive long before that time span ended.

One cannot have it both ways: both assuming rapid advance, and assuming long economic life.

> Renewabl;es do come short in one very specific area: they are intermittent, and to account for that they have to be very extremely overbuilt and all available large scale storage is very short-term.

One can model to determine the effect of intermittency and renewables still come out on top. This is why renewables are being installed globally and nuclear largely isn't. Listen to the market when it's sending you such a strong signal.

> Speaking of technologies that are permanent fiction. We don't even know how to reliably store it at required scales.

Yes we do. We store it just like we store natural gas, in underground caverns. This is demonstrated technology, and would be very cheap (capex < $1 per kWh of storage capacity). There's a well-advanced project to do this in Utah, for example. The salt formation there could store enough hydrogen to power the entire US grid for something like a day.

> something something assuming no rapid advances or something

I'm pointing out your requirement that no advances be considered also rules out nuclear. I'm willing to consider nuclear advances, I just note that nuclear hasn't been very good at delivering them quickly or economically, unlike renewables and storage.

> nuclear hasn't been very good at delivering them quickly or economically, unlike renewables and storage.

Kind of circular reasoning / begging the question here, since in North America we essentially stopped building nuclear over 35 years ago.[1] We stopped because people were scared of nuclear, and thus a ton of regulatory roadblocks were increased, making it uneconomic. Note that the changes in regulations post-1986 were not new regulations to improve the safety of plants, rather, they were increased environmental review burdens, state-level moratoria and voter approval requirements -- populist measures designed to do exactly what they did do: kill nuclear.[2]

So, we've chosen to not build any reactors anymore, which means comparatively little advancement is happening (since who would invest in that when no one is soliciting bids for a new plant), and then we're saying "See, nuclear isn't advancing, so we shouldn't invest in it."

[1] https://insideclimatenews.org/wp-content/uploads/2021/03/USN...

[2] The 1990s saw much lower growth of electricity demand, too, so few new plants were needed during that decade, and by the time more capacity was needed, cheap shale gas drove the rest of the nails in the nuclear coffin. Of course, anti-nuclear activists who are also anti-carbon-emissions activists shouldn't count that as a win.

> We stopped because people were scared of nuclear,

This is the usual line nuclear advocates use, but it doesn't fit the facts. Nuclear stopped because it became clear it wasn't competitive with alternatives. There were large cost overruns, increased risk from demand growth slacking, and large new sources of power becoming available from PURPA and natural gas.

As the famous Forbes cover story said:

“The failure of the U.S. nuclear power program ranks as the largest managerial disaster in business history, a disaster on a monumental scale. The utility industry has already invested $125 billion in nuclear power … only the blind, or the biased, can now think that most of the money has been well spent.”

Forbes: Nuclear Follies, February 11, 1985

> It does, for the reason I gave.

You didn't give reasons. You presented this as an undeniable fact. And the whole reason is "but advances".

E.g. you literally listed green hydrogen as a viable storage solution even if we literally don't know how to store it reliably at required scales. Oh, wait. Your answer to that is "we just store it in underground caverns" lol. Even though it's very, very different to storing natural gas. For example, it takes 16 times as much energy to compress hydrogen as methane. Or that hydrogen embrittlement is a thing (I'm pretty sure you didn't know about this and think that we just pump hydrogen or, indeed, natural gas into empty caverns underground).

And so on and so forth.

> Unlike renewables, nuclear hasn't demonstrated a good experience curve.

Could it be almost 40 years of fear mongering and no advances in nuclear? Whereas France with its nuclear reactors has been busy keeping Germany afloat after it shut down its plants. And whereas China is went from 9 constructions in 2000 to 36 in 2025, 42 new ones proposed, and over 140 on the roadmap, 6-7 years construction time per reactor.

> One can model to determine the effect of intermittency and renewables still come out on top.

Ah yes. So on top, that once there's winter all "on top" countries end up importing energy from countries with stable power generation in form of nuclear and hydro.

> This is why renewables are being installed globally and nuclear largely isn't.

Nuclear isn't installed due to 40 years of nuclear fear-mongering and anti-nuclear policies. China has no issues installing both nuclear and renewables.

Many countries are now reverting their stance on nuclear precisely because diversifying energy sources is a good thing, and we literally don't have more stable sources of electricity than nuclear. We're literally discussing this under a post about Canada finally admitting that nuclear is a good thing actually why don't we build more of it. Following most of Europe, for example. And Asia has never been shy of nuclear energy, with China busily building reactors all over Asia and Middle East.

> I'm pointing out your requirement that no advances be considered also rules out nuclear.

Note how you again invent requirements for no advances, assume this is a fact, and pretend that it is I who requires this. Imagine if you ever had an argument in good faith.

1. Lot of nuclear fission products from Chernobyl catastrophe have already decay ed away. There was mapping done for the long term plan of shrinking the Chernobyl exclusion zone.

"In the long term, the Ukrainian radiation protection authorities can use the BfS measurement data as a planning basis for reassessing the size of the exclusion zone. The data can be used to assess which areas of the exclusion zone could be reopened for use."

https://www.bfs.de/SharedDocs/Pressemitteilungen/BfS/EN/2022...

https://www.bbc.com/news/science-environment-47227767

The Russian invasion of Ukraine has halted the reassessment of Chernobyl exclusion zone, Ukraine has currently much bigger problems than Chernobyl. One could also say that, the decline of nuclear power in Europe because of Chernobyl accident caused much stronger dependency of Europe on Russian fossil fuels and indirectly supported the Russian invasion of Ukraine by bringing a lot of European money to Russia.

2. They got the currency symbol wrong in the cleantechnica article. "First estimates included costs as high as ¥1 trillion (US$13 billion), as cited by the Japanese Prime Minister at the time, Yoshihiko Noda "

"In 2016, Japan's Ministry of Economy, Trade and Industry estimated the total cost of dealing with the Fukushima disaster at ¥21.5 trillion (US$187 billion)"

https://en.wikipedia.org/wiki/Fukushima_nuclear_accident_cle...

Lot of missing nuclear electricity production after 2011 in Japan was replaced with electricity production from imported LNG. Because of impacts Iran war on LNG gas delivery Japan is now rapidly moving to restart nuclear power plants.

https://www.politico.com/news/2026/04/18/japan-nuclear-fukus...

3. Three Mile Island was very costly destruction of power generation asset without impacts on the public health, but it caused mass panic amplified by the simultaneous release of the The China Syndrome movie.

https://en.wikipedia.org/wiki/Three_Mile_Island_accident#Act...

Other industries also don't pay insurance representing the true potential cost of a large disasters.

"US law requires payment of 8 cents per barrel of oil to the Oil Spill Liability Trust Fund for all oil imported or produced. In exchange for the payment, operators of offshore oil platforms, among others, are limited in liability to $75 million for damages, which can be paid by the fund, but are not indemnified from the cost of cleanup. As of 2010, before payouts related to the Deepwater Horizon drilling rig explosion, the fund stood at $1.6 billion.

The hydroelectric industry is not generally held financially liable for catastrophic incidents such as dam failure or resultant flooding. For example, dam operators were not held liable for the 1977 failure of the Teton Dam in Idaho that caused approximately $500 million in property damage."

https://en.wikipedia.org/wiki/Price%E2%80%93Anderson_Nuclear...

You can see how you dont need nuclear with your strategy today in Germany https://app.electricitymaps.com/map/zone/DE/live/fifteen_min...

TMI cleanup costed 1bn, so far lower than the secondary insurance amount. Fukushima cleanup will take less. Most exclusion areas were lifted (with delays but still). JP already wants to extract unmelted fuel elements to gain xp to later do it with the rest. Ukraine is less lucky - recent russian strikes destroyed part of the equipment meant to extract corium.

Germany spent so far on EEG alone double the cost of entire french fleet. You can see the results. Not just financial https://link.springer.com/article/10.1007/s10640-025-01002-z

You think they shouldn’t negotiate with native tribes?

If they're not building reactors on the land allocated to native tribes, why should they?

That's the thing, they will be on unceded land. As I understand it Canadian settlers signed treaties which allowed indigenous people to retain rights to the land. Canada then violated those treaties and built on land they didn't own. Today Canada is trying to respect the original treaties while also appreciating that they can't undo what's already been done.

First Nations have treaties with Canada with constitutionally protected land use rights that have implications beyond tiny reserves. Rights to hunt and fish can be implicated by heavy industrial land use which compels a duty to consult. Doesn't mean that First Nations can veto a project, but also doesn't mean that all this can be ignored.

All of this is more complex in British Columbia where in many places treaties were never signed and so the land is unceded and under unresolved land claim.

Did the native tribes negotiate with whoever they last defeated? Or did Europeans literally invent the concept of conflict and war, and all the tribes of North America just filed into America in an orderly fashion and never battled each other?

I'm not saying every square mile of land in Canada, America, Europe has been controlled by at least two distinct tribes or nations, and been fought over at some point since the dawn of man, however I would say that most desirable land probably has. Many cultures have been partly or fully wiped out over human history. It's sad in all cases, but I'm not convinced there's any purpose served by arbitrarily elevating some of those cultures as noble and deserving of special veto power.

Note: I acknowledge that regardless of the above, Canada has formally granted them such powers anyway, so, yep, they get to live in that system they've created.

It could take much less in Canada. And recent refurbs being on time and budget are encouraging. We'll see in fact Canada's capacity to deliver sooner, but not in Canada - Romania already signed for Cernavoda expansion to a full 4 unit plant.

> Nuclear power is the highest cost source of electricity in LCOE terms [1].

The graph actually suggests something different - you can see how coal (a mature and well -understood technology) has basically flat-lining costs that increase very slowly over time as we mine out the easy fuel. That is pretty much what we'd expect for a mature technology.

Gas, Solar and Wind have rapidly decreasing cost curves following some sort of asymptotic pattern which is what we'd expect for new and exciting technologies.

Nuclear has the most bizzare cost curve of any new technology where every year it costs more than the year before; a pattern which makes effectively no sense and is really only explainable by the heavy and effective political attack that nuclear has been under in the US and EU. On a technical basis it is probably going to be cheaper than coal and if allowed to innovate likely much cheaper than solar and wind (the too-cheap-to-meter line is plausible, we've seen that sort of market in networking).

> The answer is the same with any nuclear criticism: "this time it'll be different". Fukushima? "This time it will be different." Chernobyl? "This time it will be different." Spiralling costs? "This time it will be different." Massively delayed completion dates? "This time it will be different."

That sounds like an extremely reasonable answer? It was different after Chernobyl and Fukushima. We've never seen a plant melt down that was designed & built around the 1970s. And again, project budgeting is mostly about politics not the technology involved. If costs are consistently X the technical estimate, planners will add in a factor of X unless there is a political reason not to.

> We don't really have a good solution for dealing with that waste.

Seems to be a solved problem? We've been doing this for 50 years now and despite their best efforts the anti-nuclear crowd haven't managed to come up with a concrete example of what the problem is that isn't easily ignored. Society produces a lot of toxic waste already and it really isn't that big of an issue. I did the calcs once a long time ago for a HN post and we're often talking about a few shipping containers worth of material in these conversations; ie nothing.

We haven't figured out how to deal with the toxic byproducts of solar panels either and that is largely a non-issue. Plan A is to dump the waste somewhere and Plan B is to go with a better option if one turns up. Problem solved.

> Nuclear has the most bizzare cost curve of any new technology where every year it costs more than the year before; a pattern which makes effectively no sense and is really only explainable by the heavy and effective political attack that nuclear has been under in the US and EU.

Or by generally exploding costs of megaprojects. Look at e.g. high-speed-rail in UK, France, Germany, ... . The first projects were the cheapest, after that it only got more and more expensive.

A lot of those rising costs are also due to a (much) heavier regulatory environment.

And the much heavier regulatory environment exists for a bloody reason.

The US alone spent billions to clean up superfund sites on the taxpayers dime (because companies created a huge mess in pursuit of profit and unhampered by regulation in the 20th century).

Everything that exists for a reason; the question is whether the reason is good. We've spent decades being told that climate change is going to cost us percentages of GDP to avoid, and your first counterexample is the suggestion that mitigations might be in the billions for the US over the course of a century.

One of those numbers is bigger than the other; and it favours nuclear pretty decisively. The regulations set up an environment where business as usual appears to be worse than if actual unlikely nuclear catastrophe occurred, all the air pollution in the interim and the reduced access to cheap abundant energy are real problems that have real consequences.

Whoever accepted those reasons has blood on their hands, so well might they be called "bloody reasons". The consequences have been serious and terrible even before getting in to oil-related resource wars and the like.

it's one thing to have regulation that makes sense and it's another thing to have regulation that doesnt do much. UK is prime example with HPC

Or by the technology being heavily subsidized and its flaws papered over until they became expensively unignorable.

But no, it must be the extremely selective omnipotence of the greens that did it. /s

> I don't understand the online obsession with nuclear power in spite of all the evidence that it's simply not economical.

Independence from China and the US. Once you have your reactor engineering set and can churn them like China almost everything can be sourced either locally or you have multiple providers. Solar and wind? China. Batteries? China.

When you get in a spat with China you suddenly have to setup those industries from 0 at home. And that won't be just 15 years to ramp-up.

So the best is to start building nuclear reactors, silicon fabs, rare earth processing etc. now instead of having the exact same argument we had 20 years ago in 2045.

It's because you dont want to. Nuclear env impact is low https://unece.org/sites/default/files/2022-04/LCA_3_FINAL%20...

France has lower prices vs Germany and now, during peak summer heatwaves you can see who does best https://app.electricitymaps.com/map/zone/DE/live/fifteen_min... . Germany spent on EEG alone so far double the cost of entire french fleet. Results are clear

HPC is not a rule but exception. And it'll work for more than 60 for sure. Existing nuclear is in fact cheap https://www.kkg.ch/de/uns/geschaefts-nachhaltigkeitsberichte... or check Lazard data for it

>LCOE

Is bunk. You should be using LFSCOE instead.

https://davidturver.substack.com/p/lcoe-levelised-cost-of-en...

Which is a metric having one source throughout all weather, coupled with 2018 battery storage as per the study showcased in the blog.

Not sure what the relevancy is.

Here, a modern article modeling "System LCOE". In other words, the whole grid including transmission backup and everything else. It starts by giving new built nuclear power the benefit of doubt, having it cost 40% less than Flamanville 3 and 70% less than Hinkley Point C. Since no one would ever be stupid enough to greenlight a project like that again.

https://www.sciencedirect.com/science/article/pii/S036054422...

It finds that for Denmark, a country with very low insolation and awful winters that renewables are 53% cheaper than the nuclear system.

>Canada needs new power now. Not 15-20 years from now.

Those can both be true. Canada will likely need more power in 15 years too. It's called long term planning.

I don't know if this is true; I'm not making any claim; weren't renewable energy figures also not economical before we invested a ton of money in them? In other words, is there a situation where nuclear becomes economical because we build a lot of it before it's economical?

What if it becomes urgent to reduce CO2? There's a lot of places without hydro or geothermal power, and if you needs gobs of power for, say, making aluminum you need as much as you can get power wise.

>What if it becomes urgent to reduce CO2?

What?! It has been urgent for years.

It if "becomes urgent" (it already is), spending a decade and a half building a reactor won't exactly be helpful, will it?

Another other things nuclear power plants don't take 15-20 to build in sensible economies. You also cannot use wind & solar + batteries to replace nuclear power.

I wouldn’t say you cannot but I also wouldn’t say it is proven that you can.

My prediction is that in the not to distant future solar/wind + storage will be able to replace nuclear in most areas on Earth. The growth of solar has historically been underestimated [1], and it will continue to be underestimated. Even if nuclear gets cheaper, solar will get cheaper faster.

The development of storage has a long way to go. Outside batteries, there are other options, such as pumped storage. Even then, battery prices might go down enough to make other forms of storage uneconomic.

I also predict that a revolution is yet to happen in the transport of energy. For those areas that can't be self-sufficient in solar/wind, it may turn out to be cheaper to capture renewable energy elsewhere then transport it to where it needs to be used (we already do that with fossil fuels).

[1] https://www.sciencedirect.com/science/article/pii/S136403212...

Cannot with our current level of technology. You are not going to provide the required level of power in Canada during the winter with wind or solar with todays battery technology.

I asked Claude:

"If combined wind+solar output drops to ~10% of nameplate during one of these (a standard threshold), a ~77 GW fleet sized to meet average winter demand produces ~7.7 GW against a ~22 GW cold-snap peak — a 14 GW shortfall that storage alone has to cover. That works out to roughly 340 GWh for a 1-day lull, ~1 TWh for 3 days, ~1.7 TWh for 5 days, ~2.4 TWh for a week, and ~3.4 TWh for 10 days. Ontario's entire current and under-construction battery fleet sits in the single-digit GWh range, so even a mild 3-day lull needs ~100-200x what's actually being built, and a serious week-plus event needs 400-600x — which is why lithium-ion batteries work fine for hourly duration but make no economic sense at the multi-day scale these lulls demand."

One of my pet peeves is that people keep quoting numbers about solar costs oblivious to location, time of year, etc. No wonder some people are sticking their fingers in their ears and saying "neener neener neener".

Battery storage for diurnal variation in favorable locations looks feasible, battery storage for annual variation looks absurd. Maybe you can overbuild solar by a 3x factor in some places, I've gotten cost numbers from 'a little less than what an AP1000 is claimed to cost' to 2x more with back of the envelope calculations that probably aren't worth anything. Then there's Dunkelflaute.

It would help if you could find a good use for the excess energy but the capital cost of anything you don't use all the time is multiplied.

Household batteries work wonders for residential consumption. It is interesting what happened once subsidies for batteries was introduced in Australia. The uptake was huge (because free or cheap electricity in off peak periods). Average install size went up, covering about 24 hours of winter usage. Subsidies needed to be tweaked, to reduce the number of 50+ kWh installations. It is not unreasonable to use current technology to have 24 hours or maybe 48 in most or all residences, with an investment payback time to consumers of around 5 years. With dynamic pricing, most consumption switches to non-peak. All distributed, rather than large scale battery facilities. As long as you are prepared to import from China, manufacturing is available. What is needed is political backing to make it a good investment for consumers via subsidies, and loans to ensure people without spare cash can also benefit. And maybe the numbers work out well, with less subsidies going to fossil fuel generation.

Pre-Fukushima, the Koreans were able to pop out a gigawatt every 5 years or so. Things dramatically slowed down afterwards, so even they are not immune to whatever it is that makes constructing nuclear powerplants slow as all hell around the world.

The Barakah plant in the UAE, built by the Koreans, took 9 years.

Barakah took about 8y/unit. But local projects are expected to take less. Part of the reason is precedent govts were willing to ditch the industry entirely like Germany

> Nuclear power is the highest cost source of electricity in LCOE terms [1].

That's utterly incorrect. For a country like Canada (or Germany), the priciest form of energy is solar. Wind is close second.

And no, I'm not hallucinating. The key here is _guaranteed_ power during wintertime. There are no generally feasible renewable solutions for that.

You’re missing the point which is to create jobs, it’s what the Canadian government is pushing really hard for now, with all the infrastructure projects it’s launching.

Something that will need people working on building for 15 years sounds about right for what government is doing now.

> You’re missing the point which is to create jobs

I sure hope that the ultimate point of a government push to build nuclear powerplants is in fact getting nuclear powerplants on the other side, not just jobs along the way. The latter seems responsible for so many ills in today's Western societies.

Nuclear plants would reduce Canada’s energetic reliance on other countries but - is there any, really? Last I checked, Quebec at least exports power to other provinces and the US.

Sure with more power generation Canada has more to sell and any country would be happy to have more energy, but it doesn’t sound like something the country _needs_ as much as, say, more housing. Or deep health care system improvements and staffing. Or … jobs.

Cheaper energy unlocks all of those things. All human problems are energy problems, in the end.

Sounds like California high speed rail, where the state government is actually touting the number of jobs created as the measure of its success, even without a single mile of working track.

Sounds like Canadian high speed rail :)

you can create even more jobs with ren. But not all jobs are the same. With ren you compete with China. With nuclear - you are somewhat protected since many don't like the idea of a hualong near them

What’s ren? What’s hualong?

Ren= renewables. Hualong is the domestic Chinese reactor design

It cuts both ways. Solar and wind are great but intermittent, and the storage issue seems to be treated as a solvable ergo solved problem. Add a sprinkle of "overcapacity", gas peakers and demand shaping and we can have a fully green grid.

So why didn't this happen anywhere - except perhaps two of the sunniest and windiest places in the world, Australia and California, where energy demand (AC) also matches production? Where are the seasonal battery storage facilities that places like Europe or I guess most of NA would need?

My only conclusion is that renewables are also far more expensive than the sticker price, due to the needed grid investment, batteries and frankly unsolved problems of seasonal storage.

I don't mind being wrong, but status quo seems to be, let's not build nuclear because it's too expensive, we're sort of building renewables, but CO2 emmissions, never mind levels, keep on increasing.

It doesn't seem to add up to a coherent story.

What doesnt add up? Almost all western industrialized nations are on a downward trajectory (or flat) regarding electricity use.

So there is simply little economic incentive to "greenify" electricity quickly because demand is already met by existing infrastructure.

Lots of people are completely unwilling to pay more for energy just to decrease emissions quickly (you might be surprised about peoples selfishness!).

But if you look at countries where electricity demand grows, you can clearly see renewables overtaking everything else; China had more growth in solar PV energy (GWh/y) in the last 2 years than nuclear power in 2 decades (and they're a pretty nuclear-friendly environment, too).

These Western countries are also still exporting their manufacturing and energy use to China. Meanwhile, Chinese CO2 emmissions are still increasing, regardless of how much renewables they are building. This would mean that the marginal cost of burning coal is still lower for them than the "dirt cheap" renewables, when accounting for everything. Either that or China can't count, which I doubt.

In any case - displacing fossil fuels is cheaper than operating a fully renewable grid - because you have the luxury of simply dialling back gas or coal production when it's windy and sunny. The proble starts when you dont rely on these at all - this is my point. I haven't seen this happen anywhere or anywhere close to it either.

It's one thing to provide some marginal power generation in a grid based predominantly on fossil fuels, and another to do the same thing without that backup. The typical solar PV plant doesn't care at all about energy storage - it's someone else's problem, and hence cost as well.

Why would you expect people to go for fully renewable grids right now? You'd need no pre-existing dispatchable power for that to be appealing to start with, and this is the case approximately nowhere (excluding countries that get free hydro from their geography here).

Electric storage to get rid of the last percent of dispatchable fossils in the grid is invariably gonna be the last thing that happens because its just not appealing economically; future overprovisioning is gonna eat into your margins, capex for batteries is non-negligible and you are basically making a longer term investment into rapidly improving tech which is always rough.

But just consider a single household right now: You can just slap panels on the whole roof for peanuts and put 100kWh of batteries somewhere (that's basically a single chunky car-battery)-- that's pretty much autarky right there, and it is very feasible (but if you rely on a bit of dispatchable fossil grid power instead that's still cheaper and easier for now).

In my opinion all that we would need to accelerate this tremendously is like a $300/ton in carbon tax, then just re-emit the gains into mainly lower income tax brackets (poor-ish people might even come out ahead!) and tax literally every import comparably that can't demonstrate a clean chain of the same CO2 taxation for its inputs. But people would wine endlessly, because suddenly flying into vacation 3 times a year or similar BS becomes actually expensive, oh the horror...

We can't generate power out of thin air and the coal/natural gas powerplants got shut down what do you propose?

China, Canada, Sweden and others, are not stupid. We really don't understand how it is that all the experts say that Nuclear needs to be parts of the equation but all of you "online activist" keep insisting that, they are just idiots and industry shills. It is the same playbook the anti-vaxers use.

The same China that started construction on at least 10 reactors last year?

Canada needs new power now. Not 15-20 years from now

Canada won't need new power 15 years from now? Did a time traveler tell you about a coming Dark Age?

I think the assumption is that anything that you can build now, you can build more of later. Unless you think there is some reason you can't?

Reply on news.ycombinator.com