Hacker News Reader

> That's a bit disingenuous. Those units were brought online in 1985 and 1986 and are licensed to operate through 2025. That's forty years - which is the life span of a power plant, especially a nuclear power plant. So, they're not being "switched off", they've hit end-of-life.

I doubt that is true https://en.wikipedia.org/wiki/Diablo_Canyon_Power_Plant#2021... talks about a report that states "that keeping Diablo operating until 2045 would save the state a cumulative $21 billion," and I doubt they'd make a projection like that if the plant would be worn out in 2025.

It's probably not end of life, it probably just needs maintenance and recertification. This nuclear plant (https://en.wikipedia.org/wiki/Turkey_Point_Nuclear_Generatin...) has some reactors that were built earlier, and are apparently now licensed to operate for a total of 80 years.

> Building a new nuclear power station is logistically impossible. Without getting into it, in essence the U.S. has lost the ability to construct nuclear power stations. People cite the regulatory nightmare, but that's not the real issue - that's an excuse. The real issue is much more complicated.

You make this claim, but you don't really support it. Why is this a "lack of ability" rather than a lack of political will?

> Why is this a "lack of ability" rather than a lack of political will?

Even in countries that are massively pushing for nuclear energy and where the utilities are sometimes outright government-owned, like in France, Finland or the UK, nuclear reactor projects have continuously hit massive cost and deadline overruns: Finland's Olkiluoto 3 had 12 years delay and ended up costing over three times as expected, French Flamanville will end up at over six times the estimated budget and at least eight years delay, and UK's Hinkley Point can't even be reliably estimated.

The only ones being able to build out plants is China, and for what it's worth I think they will manage to blow up at least one of them due to shoddy construction.

150 plants in 15 years. There will definitely be at least one meltdown.

>China’s Climate Goals Hinge on a $440 Billion Nuclear Buildout

https://www.bloomberg.com/news/features/2021-11-02/china-cli...

I’m suspicious of the appropriateness of the even in this phrase. Massive bureaucracies with no competition produce late, cost-ineffective megaprojects, and a nuclear reactor is essentially megaproject bait even if does not actually have to built as one (unlike, say, a hydroelectric station). I’m reminded of how Higgs physics on the LHC is done on two completely independent detectors, ATLAS and CMS, because only building one would have ended up more expensive.

The problem at the root cause is the public's (reasonable!) expectations about safety. Humanity has experienced twice just how much damage a nuclear reactor blowing up can cause, not to mention how brittle they are and how lax people treat safety - just look at how many accidents happened at Sellafield - or the terrorism risk. Many of the experimental design ideas add the question of nuclear weapons proliferation on top.

Accounting for all these risks takes up serious amounts of money, enough to make the construction of a new plant so expensive it literally cannot ever turn a profit against incredibly cheap solar energy and ever-cheaper battery or molten-salt storage.

One might even say that nuclear energy was never profitable because the only way it appeared to be cheap was that the general public took up an extremely risky bet that the reactors would not go boom and because the costs for teardown and storage were never really accounted for.

The other problem is the public's unreasonable expectations about safety. Given the amount of radiation already emitted into the atmosphere by fossil fuels, a nuclear power plant meltdown cannot be considered to be a remarkable addition to that balance.

Except humanity has a very incomplete picture here that suffers cognitive biases. Coal outputs way more cumulative radiation than nuclear, amongst all kinds of other issues, but it’s slow and a little bit at a time so people don’t go up in arms. One or two booms gets all the attention.

There is nothing reasonable about the public expectation of safety when it comes to nuclear energy.

Coal plants cause orders of magnitude more premature deaths, but since they happen in hospitals and cancer treatment centers rather than homes, nobody cares. You can't make a 'Chernobyl' miniseries about cancer and emphysema patients scattered over an entire country (or planet.)

> Humanity has experienced twice just how much damage a nuclear reactor blowing up can cause

We haven't, really. We were quite lucky.

What if the Chernobyl disaster had reached the Pripyat river, and in turn contaminated the Dnieper? We are all familiar with them geography of Ukraine now.

What if a radioactive cloud from Fukushima had reached Tokyo?

What would have been the cost of a temporary evacuation of Tokyo?

I mean.. in both cases you would have to wait for the radiation to dissipate. Have people remain indoors and wait a few days for the cloud to pass. In the case of contamination of a water source, you would want to filter the water before drinking until the radioactive fallout burns itself out. Plus water is exceptionally good at stopping radiation so a nuclear accident into a body of water is like the best case scenario.

Nuclear radiation isn’t a boogeyman, and isn’t the most dangerous type of contamination by a long shot. I would be way more worried about a chemical spill or a industrial accident that released toxic gas.

What you described is not the lack of ability, it's the lack of political will.

It is remarkable that the UK does not trust Chinese involvement with our 5g network but we will let them design and build nuclear power stations here.

I think people confuse the lifetime of the plant with the lifetime of the license to operate it. The NRC issues a 40 year license which can then be renewed if approved.

We have come to understand quite a lot better how long it is safe to operate nuclear plants. And less than 40 years is not that old yet for one. Clearly possible not to be at end of lifespan.

> Those units ... are licensed to operate through 2025. ... So, they're not being "switched off", they've hit end-of-life

But that's not necessarily accurate, either. My dad worked in the nuclear industry all his career and specifically within the licensing group for maybe five or ten years. His job was, for a time, to relicense the plant that he worked at so they could continue after their 40 year initial licensure period.

Maybe Diablo happens to be end of life due to other issues, but saying that the license is up is not the same as it being end of life.

That's true. There are many nuclear stations that are more than 40 years old. PG&E announced in 2016 they weren't renewing the operating license because California has put emphasis on renewables and if the nuclear power station isn't guaranteed to provide base load then it's too expensive to operate. Otherwise there shouldn't be an issue of running the station until 2035 or 2045. I'm not a resident of California so I don't know the state politics and what all is driving this decision.

> if the nuclear power station isn't guaranteed to provide base load then it's too expensive to operate.

If nuclear can't guarantee baseload, then how is it a better investment than much cheaper renewables?

I don't believe they're saying that the nuclear plant can't provide the base load, it's that the California government won't commit to buying the base load from the nuclear plant.

Nuclear can guarantee base load, PG&E is arguing the preference for dispatching base load will be given to renewables which would cause Diablo to be dispatched at an expected rate of 50% of what it's being dispatched today. Such a scenario is financially untenable to continue further operation, hence the decision to not renew the operating license - or at least a factor in that decision anyway.

That makes sense. Thanks for the details.

> at least a factor in that decision anyway.

What were the other factors? Were the plummeting LCOE of renewables and storage part of it? Also what about the (previously low) price of natural gas?

Now that natural gas is expensive, it seems like keeping Diablo Canyon operational for another 20 years is back in the table if federal funds can be used to subsidize it:

https://www.utilitydive.com/news/analysts-differ-on-feasibil...

Other factors I'm sure PG&E is considering:

* Cost to extend the operating license of the plant. The NRC will likely mandate modifications be made in order to keep the plant running safely for another 20-40 yers.

* Public Relations. There's strong public opposition to nuclear power. PG&E already has a poor image in California. Closing the plant would be expected to generate some goodwill toward PG&E.

* Soaring insurance costs/increased liabilities. California has made it abundantly clear to PG&E that if Diablo were to have any problems affecting the public in any manner whatsoever then they are totally on the hook for the financial impacts. Look at what happened with the wildfires caused by arcing transmission equipment. Now imagine the costs associated with a nuclear "incident." That's simply too much risk.

Between the high financial risk and decreased dispatch that's a one-two punch that makes continued operations untenable.

As my sibling comment also queries, I wonder what you mean by "Building a new nuclear power station is logistically impossible." Can you expand on this further?

Cost overruns, quality control issues, and the inability to manage large, industrial construction projects lasting a decade or more. I'm reminded of a quote I recently read for why the U.S. continues building aircraft carriers even we arguably no longer need them: it's so we can remember how to build them and keep all the supply chains in place. Nuclear power stations represent the limit for what private enterprise is capable of building. I've always we should have adopted the French model where there's a singular design everybody adheres to, it's approved and methodically modified and everybody in the industry knows how to build them. Instead ours are all snowflakes - which increases the complexity that much more.

"Impossible" might be a bit strong of a word, but recent attempts have proven that the initial estimates of time and cost are not very useful for evaluating actual time and cost of completion. And completion is not a guaranteed outcome, as ballooning costs, corruption, or market conditions can cause failure and loss of billions of dollars of labor and materials.

* VC Summer in South Carolina: failed due to corruption and construction/design incompetence

* Vogtle in Georgia: might finish, but every few months gets delayed more for a grab bag list of reasons. See, for example:

https://thecurrentga.org/2021/12/03/construction-quality-iss...

* Flamanville in France: who knows what went wrong here

* Hinkley Point C in the UK: not looking great

* Olkiluoto in Finland: it finally finished and works! Yay!

https://thecurrentga.org/2021/12/03/construction-quality-iss...

Even in France's first generation of building, of the same design, costs rose as they built more. Nuclear just can not seem to find its groove of being reliably constructed or planned.

You're kind making OP's point here, no? In essence, there are some huge failures to resolve these issues at the level where those solutions would matter in terms of moving the needle. The responsibility is being abrogated at the governmental level, and being kicked down to individuals where the cost to benefit ratio is far worse, both for the individual and the environment.

If the US is no longer able to create nuclear power stations the reason for the failure is to be found in the regulatory overload, not in the technical ability. The solution therefore is to prune the regulators, not in building new gas-firing plants to replace nuclear power stations whose life can be extended while saving billions of dollars while keeping emissions at the same (very low) level as they've done since they were put on-line.

Here in Europe eyes seem to have been wrenched open and the resistance against nuclear power is being seen for what it is by many, a form of Luddism which is partly pushed by fossil fuel industries. Several countries - the Netherlands and the UK among them - are now planning to build new nuclear power stations. France seems to have been saved from the brink of switching their mostly nuclear power fleet to non-nuclear and is now planning to continue its reliance on nuclear power. Then there is Germany, the black sheep (probably due to it grazing next to a coal-fired power station) of Europe where 'die Grüne' (the Green party) and their ideological allies have managed to increase Germany's emissions substantially by forcing the closure of all nuclear power stations. Germany is one of the better examples on how not to manage power production and distribution.

Europe seems to have the exact same problems with building nuclear power plants that the US does, which suggests that the problem is not with individual regulators who need to be replaced.

Consulting the current list of nuclear power plants under construction around the world by a pro-nuclear power agency here: https://world-nuclear.org/information-library/current-and-fu...

I see the following as the only ones still under construction in all of Europe (not counting Belarus and Russia): 1) Mochovce 3 in Slovakia. Construction started November 2008, originally scheduled to complete in 2012, now hopefully complete later this year, so 15 years total, 10 years late. 2) Flamenville 3 in France. Construction started in 2007, originally scheduled to complete in 2012. Hopefully complete in 2023, so 16 years later, 11 years late. 3) Mochovce 4 in Slovakia. Construction started November 2008, original scheduled to complete in 2013, now hopefully complete in 2023, so 16 years total, 10 years late. 4,5) Hinkley Point C1 and C2 in the UK. Construction started in roughly 2008[1], originally expected to be online 2022 or so ("early 2020s" is the best I can find with Google now, and that's for both C1 and C2 to be online). Now C1 is expected to be complete in 2027, and C2 in 2028. So 19-20 years total, 6 years late.

(The US has two reactors on the list, Vogtle 3 and 4, started in 2009, originally expected to finish in 2016 and 2017, now expected to finish in 2023.)

South Korea has 4 reactors under construction right now (plus two they are building in the UAE). India has 8, China has 19, and Russia has 3 (plus 4 they are building in Turkey, 1 in Belarus, 2 in Bangladesh). Argentina has their first indigenous design under construction- and that appears to be it, as far as nuclear power plant construction goes around the world.

One or more of those countries might have things figured out, but whatever problems the US has seem to be happening across the developed world (minus potentially South Korea), which suggests that it really is not a case where firing a few rogue regulators will fix things.

[1]: The Hinkley Point story is a complete flustercluck of nonsense.

Can you point me to any resources that go into more detail on why the US cannot construct nuclear power stations? I'd love a cliff's notes version if you want to write it up, but I don't want to take up too much of your time

> The real issue is much more complicated.

Come on now, don't leave us hanging like that!

> The last nuclear generation station brought online was in 2016 and took 43 years to build! It also cost billions of dollars. Nobody has the appetite for that kind of endeavor.

It doesn’t have to be this way.

Anybody who says it does should not be allowed near anything in even the lowest levels of government/bureaucracy

That's saying nothing about the extreme amount of gas that will still be burned to heat the food now.

Gas being burned to heat something is way more efficient than gas being burned to turn turbines to make electricity to be transported hundreds of miles to a restaurant only to be used to heat coils at 1500 watts to boil water. I am guessing at least 15 times more gas would be used.