First new US nuclear reactor in decades enters commercial operation in Georgia | Hacker News Reader

First new US nuclear reactor in decades enters commercial operation in Georgia

apnews.com

265 pointsCharlesW357 comments

The Nuclear Regulatory Commission was established in 1975. Since then, no plant license that was initially submitted to the NRC has started operations.

Plant Vogtle was approved by the Atomic Energy Commission (the predecessor to the NRC). Their license was grandfathered in. Building this reactor required a new reactor license (not plant license). Shortly after the reactor design was approved and construction started, the federal government added new rules about containment vessels being resilient to passenger aircraft impact. The NRC applied these rules retroactively, causing the containment vessel to be redesigned and construction to be halted.[1] The companies working with the NRC are reluctant to criticize regulators, as they fear retaliation from the NRC. The NRC supervises and approves each step of nuclear reactor construction, making it very difficult to schedule work with contractors and suppliers. Honestly, it's amazing this plant was built at all.

1. https://www.ans.org/news/article-1646/root-cause-of-vogtle-a...

The NRC was established just about when the first nuclear buildout collapsed under the weight of its own foolishness. Massive cost overruns did nuclear no favors whatsoever, clearly inadequate safety (especially on those first generation BWRs; hello Fukushima), and (most devastatingly) the deregulation of the US electricity grid, with PURPA (in 1978) and later steps opening grids to non-utility providers. Nuclear projects that would make sense for a monopoly utility (hey, let's boost the capital spending to increase our regulated earnings) no longer made any sense in a competitive market.

Fukushima shows that even early nuclear reactors are incredibly safe. You take an aging plant based on an outdated design, construct it in an unsafe location, then hit it with the fifth-largest earthquake ever recorded, then hit it with a 20 foot high tsunami, and then it explodes... and what were the casualties? One worker and maybe 50 early deaths due to cancers. A typical coal power plant kills more people every year from the air pollution it spews out. And don't forget that the cause of the meltdown was an earthquake and tsunami that killed 20,000 people!

If we judged other forms of power generation the way we judge nuclear, we simply wouldn't have electricity.

It shows that the people who considered the location and plant to be safe enough to be wrong.

If your standard for “safe enough” is that no one will directly die or be injured, ever, then the only feasible solution is no electricity at all.

People often fail to compare the opportunity cost of alternative methods to nuclear power. At that time the alternative to nuclear was coal power plants. Radioactive carbon isotopes have harmed many times more people than Nuclear.

Coal dust, sans radioactive stuff, and coal burning emissions also exact a terrible toll on respiratory health. A lot of people have died, or had much worse lives, thanks to the widespread use of coal.

Don’t leave out the coal miners. The history of oppressive companies, grueling work conditions, constant possibility of death through any number of horrible circumstances, and the lifelong damage of unprotected mining on the lungs has harmed generations of families should never be forgotten.

Meanwhile as a population they’re voting for people who want more coal mining. They are 6 figure jobs in a depressed economic area.

And yet, what if that labor went into nuclear construction instead? How many 6 figure construction jobs is many multi-billion $ nuclear plants?

What percent of coal miners do you think could make that transition?

Most of them; deisel mechanics, electricians, explosives techs, train drivers, riggers, machine operators, etc.

What, tell me you don't think coal miners have pick axes and chip coal out of tunnel walls into sacks.

Coal miners are human too, they can adapt, people can pick up new skills.

It seems like you're implying that many of them cannot make the transition. Why do you think they cannot?

Well for one there have been gov efforts before under Obama but it didn’t result in a big shift (for whatever reason). But also I think it’s very elitist and with a lot of hubris to suggest an entire population should easily be able to shift from one thing they’ve done for decades with all of the structural support in a community of it, to something else. Most people are not so fluid, both in capability but also receptive to change. Throughout history change gets resisted.

Coal contains no significant radioactive carbon. All the radioisotopes of carbon have half life many orders of magnitude smaller than the age of coal deposits.

Coal contains other radioactive elements, though. However, the quantity of these is still usually less than the quantity that needs to be mined to power a light water reactor.

Yes but the nuclear fuel is carefully contained during and after use. Not so much for coal.

The decay products of uranium are not well contained at uranium mines. They are in massive tailing piles, just as the radioactive elements in coal end up in ash.

Less gets said about these mine-related radioactive elements because they are typically in other countries, often countries where the people have darker skins.

Yes the radioactive decay products of the naturally occurring ore already in the ground naturally are still in the tailings. But they were there already before. With the removal of the uranium the net radioactive material left would be lower the the initial levels though no?

The daughter products will eventually decay away, but it will take a very long time. In particular, radium has a half life of 1600 years.

In any case, these are exactly the same daughters as from the decay of uranium in coal, so the argument you are making there is equivalent to an argument that could be made about coal ash.

but in coal they are burning those decay products and putting them into the air to spread around not so with the uranium ore where those decay products go back to the same ground they were dug out of

Almost all the radioactive elements in coal go into ash, which is not emitted into the air. Do you have some vision of old coal plants where emissions were not controlled?

Well that depends. 57% of the world's uranium is mined with in situ leaching:

> In situ leaching (ISL), also known as solution mining, or in situ recovery (ISR) in North America, involves leaving the ore where it is in the ground, and recovering the minerals from it by dissolving them and pumping the pregnant solution to the surface where the minerals can be recovered. Consequently there is little surface disturbance and no tailings or waste rock generated.

https://world-nuclear.org/information-library/nuclear-fuel-c...

Over 82% of uranium production is outside of Africa, in countries like Kazakhstan, Canada, Australia, and Russia...

The quantity of radioactivity mined is of little relevance. It's the quantity of radioactivity emitted to the environment that matters.

The point has always been that coal dumps far more radioative substance out to the environment that nuclear ever has, to tamp down illogical radiation scare information about nuclear power.

And the point I make is that if you don't give uranium mining a free pass, then with current thermal reactors your statement is false.

I'm being a bit pedantic here, but it's the heavy metals contained inside of coal - cadmium, uranium, thorium - that are the main pollution concerns, not the carbon from the coal itself. Coal is definitely one of the dirtiest fuels, though, besides biomass.

How many people have died in coal mines, had black lung, died from coal pollution etc. It's really no comparison. People are just flat out afraid radioactivity from nuclear because of movies and bad information campaigns. It -is- scary, invisible, can give you a death sentence in moment if you get a LOT of the right kind. I think human civilization dying out is a alot scarier to me.

Absolutely agree with this take. Coal power plants are a public health disaster on the scale of leaded petrol and no one talks about it. Even wind turbines have a terrible impact on birdlife. The arguments against nuclear are irrational; future generations will wonder what was wrong with people in the late 20th century.

The number of nuclear plants is so small that any example is statistically anecdotical.

Yes, Fukushima did not kill much. Is it because the nuclear is in itself super safe, or is it because we were lucky? We would need ~100 Fukushima before even having a rough estimation.

I think risk cannot be counted as "number of deaths", but as "capacity of losing control". Cars and cows kill a lot of humans. Yet, their risks are controllable: it is not true that if just few things change or if the circumstances are slightly unlucky, suddenly, they will do 1'000 times more deaths. With this way of thinking, we can understand better why scientists are worried about pandemic rather than common cold, even if before the last pandemic (and maybe during, whatever) common colds was killing more.

And it works with coal too: unpredicted effects of global warming are a way of losing control, and it is why scientists are so adamant about stopping using coal.

I don't think it is true that we have judged the other forms of power generation differently than the way we judge nuclear. We judge the same way: based on our current understanding of how easy it is to keep or lose control.

Sure, coal is dangerous and has unpredicted effects. But the reason it was treated differently is because the capacity of losing control with coal were not at all obvious from the start, while they were written in black and white from the start for nuclear.

Nuclear plants produce 68% of the energy in France. If that’s not statistically relevant, doubling the number wouldn’t be statistically relevant either.

For that matter, even in the USA, nuclear power produces about 10% of all electricity. Multiply that by 10 and that’s the entire population as a sample.

That's not at all the point.

Imagine the following game: you throw three dices. If the three dices are not ending on 1, you win 1 dollar. If they are ending on three 1, you lose 100'000 dollars.

If you throw the dice ten times, you will, with high probability, not have three 1 (the probability of appearing is 0.5% per throw). You will probably need to throw more then 10 times before loosing. But when loosing, you will lose a lot too, which makes the bet intrinsically a bad idea (the win does not justify the risk if unlucky).

France has 56 nuclear plants. 56! It's totally unreasonnable to draw conclusion on intrinsic safety of the nuclear plants on such a small number. The fact that you cover the full population with a small number of plants does not change that: it does not change the fact that you cannot conclude scientifically. The same way it is incorrect to say "I only need 3 throws, so it means that the 1-1-1 bet is intrinsically a good idea".

(PS: please do not answer about "loosing 100'000 dollars, but with nuclear disaster, it's not that bad", it's not the point. The point is that statistics is a large number game, and that we just don't have a large number of nuclear power plants to do statistics with them)

Note that powering the world (all uses of energy, not just the grid) with nuclear will require ~6000 3 GW(th) nuclear reactors, to provide the 18 TW of primary energy the world currently consumes. About two orders of magnitude more than France has. At the historical rate of world nuclear accidents this would yield > 1 meltdown a year.

There are hundreds of nuclear power plants that have been in operation for decades. Most operating nuclear power plants run with dramatically higher uptimes than what they were originally conservatively designed to meet. Nuclear power plants are operating in an active war zone right now! Six reactors were the site of a military battle and ended up sustaining damage beyond anything foreseeable by the designers. Every one of them shut down successfully and at no point was there any evidence that there was ever a safety risk to the public unless a malevolent actor wanted to cause such an accident.

This body of evidence strongly supports the view that nuclear is not only safe but incredibly safe. We have incredible knowledge about how to run nuclear plants safely.

Again, you miss my point.

"hundreds of reactor in operation for decades" is NOTHING. These plants have a large number of differences, and are located and operated often very differently. At the end, you have something like "10 plants operated for ~10 years under conditions A", "10 plants operated for ~15 years under conditions B", ...

SIX! Six reactors were the site of a military battle and ended up sustaining damage beyond anything foreseeable by the designers. Six is NOTHING. You cannot say "I trust that nuclear is intrinsically well-design to resist to military conflict" based on SIX cases. The damage may have been lucky. Or even may have been unlucky (I'm not saying that nuclear is unsafe, I'm just saying the stats says you cannot conclude).

The only people who pretend we have incredible knowledge about how to run nuclear plants safely are either people who also demonstrate they don't have any idea of what they need to check to "demonstrate safety", or they are experts that are on the left side of the Gaussian curve: they are as numerous as as-much-qualified experts who pretend nuclear plants are proven very unsafe (in other word: they are cherry-picked experts, and do not represent the consensus which is: nuclear plants are too complex, too diversified, too sensitive to surrounding circumstances and not yet used enough so that we can scientifically conclude)

How do you reconcile this kind of thinking with driving a car?

Are you seriously thinking that someone is incapable to drive a car if they are not incorrectly convinced that an incorrect math computation is correct?

The large majority of things in life is "impossible to statistically prove". I had no proof that I will not burn my toast for breakfast this morning. Yet I had breakfast without problem AND I did not pretend that "since I've used this toaster 5 times and did not burn a toast, it implies that burning a toast will never occur" (this sentence is incorrect: the observations are compatible with the fact that my setting is very good and reduce the risk of burning the toast very low, but also compatible with another realistic hypothesis: my toaster burn a toast about 1 time over 10, and so far I was just "lucky").

But why would I not make a toast or take my car? I just don't know if it will work or not, but I also don't know if it will not work or yes. You cannot say "if you are not sure of X, you should act as if non-X is sure", because if you set Y=non-X, you will say "if you are not sure of Y, you should act as if non-Y is sure", and you end up saying you should both act as if both X and non-X are sure.

Nobody here is saying nuclear is not a solution (I mean: if you think I'm saying that, you are dead wrong). But it is tiring to see pro-nuclear people making pseudo-scientific claims to entertain their own belief (belief that may turn out to be right). If you are so confident in your belief, just say so: "I know we cannot properly tell that nuclear is 'safer' than X or Y, it does not make any sense to do so, but yet I believe it is still a good option. You are free to disagree with me, but you also have no ground to pretend it is 'less safe' than X or Y, so your position is as legitimate as mine".

Global warming is 100% snake eyes when you get down to the bottom line of "why nuclear?"

Human deaths aren't the only measure of safety.

Pollution into the ocean from Fukushima was huge, unfortunately.

Just because it's not as bad as Chernobyl (which was as worst-case as it gets) doesn't mean it's safe.

Why this is downvoted? Propaganda from Japanese Ministry of Foreign Affairs?

Thank you for CCP propaganda

Please don't do this here.

It's huge, but it's dispersed across the massive volume of water in the pacific. Charts like these are scary, until you read the legend and realize that even the bright orange areas indicate concentrations only 0.1% higher than the typical pre-fukushima concentrations of radioactive isotopes: https://www.fisheries.noaa.gov/west-coast/science-data/fukus...

Eat a banana and you'll receive a dose larger than anything related to Fukushima.

Maybe, but i think the argument in this thread seems to be that NRC creation basically killed the whole industry. Which is complicated, because you still want safety improvements over Fukushima, but not at the cost of basically halting any new development for decades if not ever.

"If we judged other forms of power generation the way we judge nuclear, we simply wouldn't have electricity."

Yes, of course. I can only imagine the deaths and destruction that tsunami would have caused if it were a wind or solar plant.

/SARCASM you idiot!

I recall seeing some statistics that the death per kilowatt ratio of different means of energy production is higher for all other forms of electricity than nuclear.

No need for name-calling here.

Have you considered the costs to contain and clean up Fukushima? The financial costs and loss of lands for decades is a heavy hit for a country to take.

Japan passed a special tax to pay for the disaster (https://www.reuters.com/article/us-japan-fukushima-anniversa... ).

From wikipedia (https://en.wikipedia.org/wiki/Fukushima_disaster_cleanup) >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),...

The Soviet Union, in part, collapsed from Chernobyl.

There is a huge fat tail risk on nuclear power. Roughly and a bit sadistically speaking, coal kills people predictably at a "death-from-car-crashes" rate. There is no fat-tail, you just have a high rate of cancer,etc. It is like letting people driving cars and accepting the 40,000/year deaths from accidents as the price.

> maybe 50 early deaths due to cancers

No long term deaths have been recorded yet.

They will never be recorded, because they are lost in an ocean of cancers that normally occur from other causes. That doesn't mean regulators can ignore them. Radiation is not a criminal defendant who must be presumed innocent until proven guilty.

It’s brings a lot confidence that a nuclear power was built through a legal loophole.

Because the regulatory environment is inhospitable to new plants?

I mean it makes sense, if the Atomic Energy Commission already licensed the plant site, why shouldnt they be able to reuse it?

Laws have changed a lot since the AEC existed.

without looking up we can easily bet that anti-nuclear activists did their share of lawsuits to try to get those new laws applied to the plant, so there have been reviews.

One very important milestone was the Calvert Cliffs decision in 1971, applying NEPA to nuclear construction. This was while the AEC still existed, though.

https://en.wikipedia.org/wiki/Calvert_Cliffs%27_Coordinating...

If the NRC is the problem, then why do we see identical cost overruns and schedule delays for France, Finland, and the UK?

I don't think we can blame the NRC for this. There's something deeper.

Similar regulatory regimes started by the same groups of people, which work in similar ways? AKA think of every possible thing that could go wrong and mandates designs with have engineering margins far in excess of what the a reasonable analysis of the risks/etc would call for.

We could do similar things to renewables by simply mandating that they have carbon free backup sources sufficient to guarantee four of five nines of reliability.

So you think this is a realistic possibility for France, a nation that loves nuclear?

Have any evidence of these regulations? What regulations can we change?

I've looked and looked and never found somebody saying "these regulations are unnecessary" but it's trivial to find examples of management not taking the regulations seriously and then paying for it when they cut corners (basically all the welding for the AP1000s in Georgia and South Carolina).

It's also easy to find management errors that ballon costs and cause massive delays.

I'm completely unconvinced that we can pin this on the NRC.

If you look at the cost overruns in GA quite a number of people have come out and said that no one wants to criticize the NRC, seemingly because they can make your life hell.

And change orders, and the like were listed as one of the reasons for the overruns, how much of that do you think is some engineer who designed the plant waking up one day and deciding something needs to be changed, or is it a case of the NRC nitpicking everything, and just loading the whole process up with so much paperwork that everyone is frozen, and any little mistake in the paperwork ends up costing the project. Sure blame the plant mgmt for not having the correct paperwork and delaying the start by 6 months, but who is asking for that paperwork to begin with? A lot of it sounds like a game of gocha, create a problem so complex that its nearly unsolvable and most of the work isn't even welding pipes but putting a thousand people in the chain, everyone of which can stop the line at any time, largely with no repercussions.

Sure, contractors mess up, but most contracts are written so that its the contractor who pays for it, not the organization hiring the contractor. So, again, contractor messes up, comes up with a workable mitigation/rework plan, but you sit there for 6+ months arguing about whether its allowed.

I've seen first hand the mess of regulation related paperwork it takes to run a normal NG/etc plant, it takes teams to of people to review it every year and assure compliance. Layering the NRC/etc into this is how you end up spending insane amount of capital and producing nothing but paperwork (see nuscale) on a design that isn't particularly ground breaking in the face of existing plants. I can't even imagine how much it would take to certify something like a navy reactor in that kind of environment.

None of the exposed I have seen mention anything about NRC specific regulations. It's just bad management processes dictating sloppy first attempts at design, leading to literally unconstructable designs, followed by even worse management processes for getting feedback and changes, or any sort of streamlining to allow agile response.

Imagine a huge waterfall design, with the architects being part of a completely different company than the implementers, and months long turn around to get clarification on any aspect or to get the designers to sign off on the necessary changes for anything to happen.

People are not afraid to criticize the NRC, it falls out of people's mouth so easily. But if they are actually afraid of the NRC, why are they so willing to provide critique, but unwilling to suggest the exact things that are the problem?

More likely you have a bunch of substandard folks unwilling to take any sort of responsibility for their actions or for management and instead blame a third party.

Suppose that any one of these companies spent a tiny tiny fraction of their money of proposing changes to regulations that would allow them to build effectively? Why don't we see any sort of proposals like that?

As for nuscale, this is a perfect example of using the NRC as a scapegoat and being unwilling to look at the true source of their problems:

https://www.power-eng.com/nuclear/report-claims-serious-prob...

I am still waiting for some sort of example of the NRC layering on something unnecessary, after six years of begging people making the claim to produce one shred of evidence.

That's fair. But what about South Korea, China, India, Turkey?

In China you can't probably trust the official cost numbers (I'm not even trying to google for them). But you can't deny the astonishing pace they've built in the last decade: in 2012 they had 12 GW of nuclear capacity, in 2022 this went to 52 GW. 40 GW in one decade. France built 41 GW of capacity between 1980 and 1990, and the US 44 GW between 1970 and 1980, so this is not unprecedented, but it'd darn impressive in post-Fukushima era.

[1] https://world-nuclear.org/information-library/country-profil...

Those countries have costs of labor far far below Western countries. There are also huge questions about corruption and actual work performed. For example, Sourh Korea, the least corrupt country there with the highest cost of labor, put executives in jail for their fraud on safety certifications.

My personal hypothesis is that nuclear is only kind of affordable for a very narrow band of technological advancement, where labor is still cheap enough but tech capability has not advanced too much. After your technological capacities advance far enough, labor is better spent on other tech than building massive cathedrals.

As for China, try comparing nuclear builds to a single quarter of solar or storage output. It's a big country, and China's investment in nuclear is proof that even in a country known for excellent skills in managing massive construction projects, it's not really giving renewables a run for their money.

> Those countries have costs of labor far far below Western countries

That could be for some labor, but not all. The US is still in the game when it comes to automobiles. Well, when it comes to Tesla, it looks to me Tesla is eating everyone's lunch. The US is still in the game for other very high tech things, like military hardware. Or industrial trucks (think Caterpillar).

Where the US and other Western nations are a bit hopeless is the costs of megaconstructions, be they nuclear power plant or high speed rail, or simply skyscrapers.

But nuclear power plants could be more like Caterpillar than like "massive cathedrals", as you call them.

I worked with an engineer that came from the Vogtle project and he would talk about the hoops they had to jump through to get anything done. Even the most basic unassuming weld would require tons of paperwork and cooperation from a lot of people.

There is no doubt that NRC needs to be completely revamped. Maybe we can use a more productive model like the French have.

One project in 30 years doesn't benefit from economies of scale. Nuclear is the only hope we have to make up for the gaps in generation for solar and wind. Solar and Wind should be used whenever possible, but they are not 100% of our solution mix, even if they are the majority.

If we want to mitigate the impact of climate change, we need to invest in decreasing Nuclear costs and building many more plants in the US.

Nuclear is dope and we should build tons of it, but it ain’t our only hope; plenty of other alternatives to build at the same time.

Eg hydro, overprovisioning solar or wind, transmission to remove local weather variations, coupling wind and solar, demand flexibility.

Fervo just started its first full-scale new-gen geothermal plant, for instance; 24/7 firm power. You might like David Roberts interview with Tim Latimer about it: https://www.volts.wtf/p/enhanced-geothermal-power-is-finally...

Hydro isn’t clean (extreme ecosystem damage, both up and down stream); and if we dammed every river it’s a drop in the bucket of what we need.

I don’t know how far solar/wind can get us. But hydro sucks more than most people realize.

True but PV+wind make our existing dams go much further - recall most of them operate as giant annual batteries, they refill in the rainy season once per year, then dispense that energy over the dry season.

Adding PV means hydro can save much more of its water, just dispensing to “fill in the gaps”. This is already how the hydro in Norway and Sweden operates, you can see it daily if you look at the hourly power breakdowns by generation type.

But I agree, I hope we will get to a point where we can decommission the big dams..

> recall most of them operate as giant annual batteries, they refill in the rainy season once per year, then dispense that energy over the dry season

That’s not quite how it works.

Consider the biggest hydro project in US, the Columbia river, with its 14 dams.

The system does fill up in the rainy season, but the crucial thing is that it’s not the dams that do fill up, but rather the whole watershed, meaning things like snowfall and ground water. This means that we have very limited amount of control over when we let the water through. We can’t just dam the river for an extended period: if we don’t use it for generating power, we must spill (waste) it. This means in practice that the dams are not batteries: to a large degree, it‘s use-it-or-lose-it.

To make these into batteries, we’d need to somehow refurbish the dams to tremendously increase the power generating capacity on each dam, so that instead of assumption of continuous flow (either through turbines or spillways), we make the flow more intermittent, so that we can make up for closed times by pushing more water through during open times.

This is tremendously difficult in practice: dams are simply not designed to allow for refurbishing with many more turbines or much more flow through them than they were originally designed for, the upstream reservoirs are not designed for quickly varying water levels, etc.

Point is, we can’t just “save the water”.

> That’s not quite how it works.

Yeah in reality there are two forms of hydropower: the classic reservoir dam, and 'run of the river' powerplants. You're absolutely right, even reservoir dams don't really fulfill the 'annual battery' idea, since they must maintain some minimum outflow for downstream consumers and can't shut it off entirely if it better suits the power generation goal.

> Point is, we can’t just “save the water”.

The storage level at Lake Mead has an annual spread of ~100ft; the dam is currently at historic low capacity and could hold about twice its current level: https://www.usbr.gov/lc/region/g4000/lakemead_line.pdf

Reservervoir dams certainly can store water and dispense it when appropriate.

Like I said, this is happening every day already, you can see it in the hourly data on wind/hydro generation share, and you can see it in the annual storage capacity data for Norwegian hydro.

Yes, we can do this for some dams, but not enough to move the needle. Hoover Dam’s generating power is quite tiny relative to the total storage needs, and if you add up all the other viable dams, you’ll find that it doesn’t add up to a lot.

To put it in context: Columbia River basin produces 40% of US hydroelectricity. Out of 14 dams on Columbia River, 12 of them produce (each individually) more electricity than Hoover Dam, most by a factor of 3x or more. Very little of it can be turned into storage as of today, what does not get used must be spilled.

This is not to say that using dams for energy storage is a bad idea, it’s not. It just will not work well in practice with most of currently existing dams, at least without huge retrofits and/or screwing over downstream water consumers and upstream reservoir users.

> But hydro sucks more than most people realize.

There's a reason Russia blew up the Kakhovka hydro dam. The down-stream impacts of the flooding were more devastating than what they could reasonably accomplish with conventional weapons.

No.

It killed <75 people.

They blew up the dam because it was also a bridge and removing it allowed them to redeploy troops to the east.

I'm as far from russia apologist as can be - but they blew the bridge part few months earlier, in November. https://www.youtube.com/watch?v=xxHLImMbnAw

They dented and rendered the road inoperable months earlier.

But the thing about dams is that they're over-engineered and massive. Even with high explosives it's difficult to demolish one. Hence why dam busting bombs were large (>4000 kg) and still required underwater detonation to boost their power [0].

It wasn't unreasonable for the Russians to expect Ukraine might be able to seize a bridgehead on the far side, rebuild an operable road over the dam, and then rush armor (and logistics) across it.

Much harder and more time consuming when there are no longer any load bearing remnants available.

[0] https://en.m.wikipedia.org/wiki/Operation_Chastise#The_attac...

>It wasn't unreasonable for the Russians to expect Ukraine might be able to seize a bridgehead on the far side, rebuild an operable road over the dam, and then rush armor (and logistics) across it.

It was, as much as it was unreasonable for russians - that's why they ran over to that side of river.

Looking this up.

https://www.usbr.gov/lc/hooverdam/faqs/powerfaq.html

> Hoover Dam generates, on average, about 4 billion kilowatt-hours. And the lake covers an area of 1,495,806 acres

Rough estimate, 2GW worth of solar produces that much in a year. And requires I think 12000 acres of land. (Open to have phat fingered the calcs)

My take away is it's not even economic anymore and requires flooding 100 times more land then a solar plant.

Dams still get built for non-power reasons, to manage water supplies, so adding turbines (and floating solar PV) to them might make sense, but yes I think we're now at the point where new hydro purely for energy production is going to struggle to make a case for itself against just deploying renewables and batteries and interconnects.

edit: to add some context re the Hoover dam:

> Upon becoming Secretary of Commerce in 1921, Hoover proposed the construction of a dam on the Colorado River. In addition to flood control and irrigation, it would provide a dependable supply of water for Los Angeles and Southern California.

It's not exactly relevant how much solar produces per year - it's very important _when_ the energy can be produced and how much you can control that.

Absolute nonsense. Your attitude is actively harmful to decarbonization.

Ecosystem change =/= ecosystem destruction. The lake produced by a dam is a far more beneficial ecosystem to a much broader range of life than the river that preceded it

Source?

I like hydro but this is not at all what people I know working in this field believe to be true.

Canada's energy mix is roughly 60% hydro, and some provinces are well past 90%. You ought to check your numbers.

No offense, but Canada isn’t relevant on the global scale.

Hydro use is growing, especially in developing countries. However it’s a shrinking percentage of total energy generation. There is an absolute cap on theoretical hydro energy production, and it isn’t enough.

Hydro is low carbon and renewable. But it isn’t green, and it’s not enough.

Canada is second-largest country by area, occupying around 6.5% of Earth's land surface. It's only 37th by population, with 0.5% of world's population. It's very non-representative example.

My assumption is that rising CO2 levels in the atmosphere pose a much greater existential threat to humanity than localized ecosystem damage resulting from river blockages. But maybe I'm wrong.

Speaking of hydro, I do think pumped hydro-storage ought to be looked at a lot more for energy storage (esp. versus giant lithium-ion battery banks), especially as we transition to inconsistent renewable sources like solar and wind. I'd assume that creating new, isolated bodies of water wouldn't incur as much ecological damage as blocking off existing rivers or greatly increasing our mining of rare earth minerals)

But it isn't "localized ecosystem damage". Building dams absolutely fucking destroys the ecosystems upstream and downstream. It's the civil engineering of burning the crops when you retreat. The "new" ecosystem has nothing in common with the prior ecosystems.

I'd rather some species of fish die out than the ocean boils, personally

Yeah. And dams can be removed (and as far as I can tell, nature is relatively adaptable and can recover).

Hydro is only source of renewable power that generates >50% of any country's electricity grid: https://en.wikipedia.org/wiki/List_of_countries_by_renewable...

Wind caps out at 42% with Denmark, solar at 15% in Australia. Many countries have nearly 100% of their electricity coming from hydroelectricity. Besides nuclear power, hydroelectricity is one of the few non-intermittent sources of renewable energy - sure, rainfall does technically make hydro intermittent in a sense, but it's not going to change output on a dime when the sun goes down or the wind stops blowing.

Well, you also have geothermal power, but that's even more geographically constrained than hydro.

That’s a non sequitur.

I never said hydro wasn’t renewable. I said it wasn’t clean. Most people think of hydro as renewable and low carbon and clean. Two are true, one is false. Hydro is dirty and, imho, we should phase it out the same way we should phase coal out.

Nuclear isn’t renewable by most definitions. But it is low carbon and clean.

Nuclear was good 40 years ago. Solar is getting good. Wind I don’t know much about. Hydro is an ok bootstrap but is much dirtier and more problematic than most people realize.

Hydro is indeed clean in that it does not emit greenhouse gases. Sure, it disrupts life around the dammed waterway. But that's way less of an impact than climate change.

Hydro emits huge amounts of GHG when flooded organic matter decomposes btw.

This is a one-time overhead emission, and I doubt it's "huge" compared to even a small amount of fossil fuel use.

In comparison to FF it’s small of course. But it’s billions of tons, globally more than Germany’s footprint. Which is larger than most peoples assumptions, hence the “huge”.

Nuclear is renewable and generates more than 50% of many countries electricity.

Every serious analysis in green energy has never been able to make it work without nuclear.

This is a baldfaced lie (unless you are engaging a No True Scotsman dismissal of the large number of studies that say otherwise.)

https://youtu.be/bkIK4Xo1GRM

Not wasting 100 minutes of my life on that. I'd read a paper defending such a claim. In the meantime, here are a large number of abstracts of peer reviewed papers showing 100% RE is feasible:

https://web.stanford.edu/group/efmh/jacobson/Articles/I/Comb...

Here's a review article on 100% RE studies:

https://ieeexplore.ieee.org/document/9837910

"The main conclusion of most of these studies is that 100% renewables is feasible worldwide at low cost."

I’m not so sure with overprovisioning, batteries and controllable load. Also pumped hydro and other gravity batteries. We barely have any storage on the grid right now.

> Also pumped hydro and other gravity batteries

pumped hydro is not a winner. the locations that could be used are few and far between and require massive amounts of water and wreak ecological nightmare on a wide area.

The locations where pumped hydro could be used are extremely abundant. Remember, it doesn't have to be on a river. It can be out in a desert! Here's an example of a project being built in Nevada. Basin and Range geography, more vertical relief than you can shake a stick at. Look how small this thing is for the capacity!

https://www.cityofelynv.gov/pdf/CityCouncil2021/cc1-28-21/Wh...

https://www.whitepinepumpedstorage.com/

(the whole thing could be sped up; that's a general problem in the US)

The problem with pumped-hydro is - and all storage based solutions is - how low can you let the storage get before it's an emergency?

Electricity is a vital service: completely vital. Without it, modern civilization halts. It might be annoying being unable to make a cup of coffee, but municipal water and sewage need electricity to work. You go without power for a week, and the entire wastewater infrastructure will start shutting down. Refrigeration and food storage fails. Even backup fuel storage becomes a liability because you need electricity to pump it around.

So the question is, how low can you let the reservoir get? Because it's not about how long you could run going from 100% to 0% - it's how much of it can you use. And we have a model for this, in the form of another service: city townwater supplies.

In Australia, water restrictions go into effect when we hit <50% water capacity in the dams. That's the level at which usage cuts are applied to try and ensure we don't run out. At <40% we increase the severity. But this sort of resource exhaustion is also slow - we lose storage capacity over the course of months, not days.

And this is a resource which is dependent on electricity to supply (we also have a desalination plant, so we have some guaranteed capacity).

So within that context then - i.e. imagine you're planning a nation-state electricity supply, what are your risks? - how good does pumped hydro - or any storage-based solution - look, when your requirement is "the power cannot go off - ever". Put on your systems engineering hat, treat it like a software deployment - what level of redundancy and overbuild would you want when you're told "this is a mission critical, safety-critical system consuming an intermittently available resource". How much capacity and overbuild would you believe is necessary to have confidence, or even decision-making capability, when pressured?

You seem to be assuming pumped hydro is the only storage technology used.

This is a common problem in the anti-renewable arguments. You pick a particular design for an energy system, argue it doesn't work, then (wrongly) claim no renewable energy system can work. But to reach that conclusion, you have to show that no combination of elements can make a system that works.

It makes sense to have multiple storage technologies with different performance characteristics. You want efficient, if somewhat expensive, technologies for short term storage with large numbers of charge/discharge cycles. You want low capital cost systems for ultimate backup, even if those systems are not as efficient.

For example, one could back up the entire grid with combustion turbines burning an e-fuel like hydrogen. These are massively cheaper per unit of power output than nuclear. Because we are not using them very often, the low round trip efficiency doesn't matter much. You want guarantees this won't run out? Make the storage caverns larger. This is already what we do with natural gas -- we store a good chunk of seasonal demand and count this being sized large enough to not run out.

You are proposing build 4 times the power generation: Renewables for the daytime load, Renewables for charging the night time storage, the night time storage, and then a whole extra set of power plants for emergency standby.

Why would hydrogen combustion plants - which don't exist at the moment, don't have turbines on the market, don't have a fuel supply pipeline - be cheaper then current coal fired powerplants?

So your cost of generation already is - at minimum - at least as expensive as a coal fired powerplant, in terms of fixed costs for maintenance (and investment - who's building these when they can't sell the power from them?)

Hydrogen combustion CC plants will certainly have a capital cost less than coal fired plants. They will be almost identical to natural gas fired CC plants (just the combustor will be different). These are known to be quite inexpensive, costing maybe $1/W.

There need not be any fuel pipeline, since the plant can be built at the hydrogen storage site. The electrolysers will be there also.

There have been industrial turbines that burn hydrogen for decades. It's not some sort of exotic technology.

Yes, there are many parts here. And it's still cheaper than nuclear. Nuclear is pathetic in that way.

Reading through the linked article, I don't think that's the way it's designed to be used. It can completely empty itself in 8 hours. That sounds like it's specifically designed to augment renewables at night and charge during the day, cycling every 24 hours.

That's absolutely how it's going to be used because it's cost efficient for the owner.

But if you're electrical grid doesn't mostly have always on sources to backstop it, if you were all renewables and storage, then how much storage would you need to guarantee supply - 24/7/365 days a year.

So, you're saying it would be stupid to use 8 hours capacity pumped hydro for long term storage.

Thanks for the observation, Mr. Obvious.

You also seem to be implying longer term storage of some form isn't feasible. If so, you are incorrect.

Note that I pointed to this project to debunk the falsehood that the locations for PHES are scarce, not to claim that PHES is good for long term storage.

Pumped hydro can work anywhere that has hills. It's just a lot cheaper in some locations than others, due to favorable geography.

Water use just has to keep up with evaporation on average.

I'm not sure why building a reservoir needs to be an "ecological nightnmare" except in the sense that it's a sudden change to an environment.

I was lumping dam infrastructure into the "other gravity batteries" which does cause ecological nightmare situations. if they only meant the fringe gravity batteries like block elevators and the like then it's not as bad a thing but you still require a large amount of fresh water which is an issue in a lot of places too.

Pumped storage, even in a desert, consumes at least an order of magnitude less water than a nuclear power plant with the same energy output.

All the coal and gas storage currently exists. We need to stop using that for non-storage uses, replacing with solar and wind, before we need to replace its storage function once we get down to the last 15% or so of grid electricity being fossil based.

Counterpoint just because someone has to go against the pro-nuclear orthodoxy here: economy of scale won't fix nuclear power. You will never get the cost (or risk) down far enough with existing technology, and none of the advanced technologies have panned out so far. And large scale renewables + battery storage are good enough.

> And large scale renewables + battery storage are good enough.

I wish this were true, but I haven't seen convincing evidence that it is. Up here in Minnesota, we heat our homes with natural gas. Once that's converted to electric, that's a _lot_ of energy to generate and store, and it has to be absolutely reliable for six straight months or you're talking mass death. Nuclear seems like a perfect fit for this scenario. I think it's a poor choice to take it off the table.

Generating green hydrogen with renewables and then burning it in combined cycle plants would be cheaper than Vogtle-level nuclear.

Maybe. The thing about nuclear reactors is that the fuel cost is nearly zero on account of how little of it you need. Even with reprocessing, it is still <$0.01/kWh. So nearly all cost with nuclear is upfront. Mass production of nuclear power plants will drive that down dramatically. Nuclear, if done at scale, is can be very cheap.

So having a stable source of cheap electricity that can be built in nearly any location is still a good idea. Even in a world with super cheap green hydrogen.

With nuclear you're burning your powerplant, essentially. That the fuel is comparatively cheap is irrelevant, since that's not what's dominating the cost of power.

That's what I said. It's the cost of the powerplant that drives nuclear's cost. But you can reduce that dramatically via mass production of power plants. It will be a cheap source of electricity if done right. And we don't want just one energy source for everything anyways.

It's not at all clear mass production of nuclear power plants will reduce cost. This has not been demonstrated. It runs counter to economies of scale from large power plants.

You pick one reactor design and make dozens of them. That will be much cheaper than building a bespoke reactor type every time.

Again, this has not been demonstrated. Nuclear in France showed negative experience effects.

A lot of that is just anti-nuclear propaganda. Nuclear in France does not have major problems.

I'm sorry, you don't get to dismiss inconvenient facts by calling them "propaganda".

There's no such thing as a perfect solution. The problems people keep on bringing up are minor and do not qualify as serious problems.

The problem is that you are comparing a hypothetical with a proven system. Nuclear works, and works in large scale in a lot of places - heavy users include France, Lithuania, Sweden and Belgium.

One can argue about costs, but costs are at least low enough to be viable, otherwise these countries couldn't exist as they do. You can say that the costs are being externalized to taxes or some other place, but these societies are being able to absorb these costs in aggregate. Nuclear might not be cheaper than gas and oil, but it's possible to build a modern industrial society with nuclear.

Now contrast with green hydrogen generation and battery storage, for instance. These approaches aren't working in country-level scales anywhere. We compare hypotheticals with systems that, despite problems, costs and limitations, are known to work.

France has massive problems with their reactors and leads the world with only 62.6% from nuclear. A comparable number to Denmark that has almost no hydro power. There are several countries with higher percentage of renewables up to 100%.

Costs aren't the only problem, new nuclear reactors simply cannot be build fast enough to counteract the climate crisis.

> France has massive problems with their reactors and leads the world with only 62.6% from nuclear.

Yet, France consistently produces half of Germany's CO2 per capita: https://data.worldbank.org/indicator/EN.ATM.CO2E.PC?location...

I would consider that a giant success of nuclear energy.

> A comparable number to Denmark that has almost no hydro power.

Denmark is not relevant - they import giant majority of their energy. Right now they barely produce at all: https://i.imgur.com/69SI5J9.png

> Costs aren't the only problem, new nuclear reactors simply cannot be build fast enough to counteract the climate crisis.

They could be build fast enough if we did it _seriously_, and not as vanity projects.

https://www.scmp.com/news/asia/article/2027347/south-korea-s... Is the suggestion to just import the energy?

Are you talking about consumption or production?

In term of production a country can easily go above 100% renewable by selling a lot of it during periods of optimal conditions. Naturally, a country can not above 100% in terms of consumption. Denmark for example is a massive exporter in terms of production, but also a massive importer in terms of consumption and has a very large dependency on imports. They are not self sufficient despite producing more energy that they themselves consume.

Iceland both produces and consumes 100% electricity from renewables for instance.

You have the issue you talk about with all generation that isn’t load-following, including nuclear. France solves this by exporting subsidized electricity in time of low demand while importing in time of high demand, which is winter. In the surrounding countries like Italy and Germany you have gas-plants that jump in when needed.

Nuclear and renewables/batteries actually have the same biggest problem: the vast majority of the cost of future energy is up front capital costs. That’s what makes natural gas so attractive: the vast majority of costs are fuel amortized over the lifetime of power generation.

That’s the thing we need to fix regardless of the power source.

> or risk

That really casts doubt on your assertions. Nuclear carries significantly fewer risks than coal (which operates in the nuclear failure state all the time).

Nobody's arguing in favor of building more coal plants. The fact that nuclear is better than coal is irrelevant to the discussion of what new plants to build.

> Nuclear is the only hope we have to do to make up for the gaps in generation for solar and wind.

Nuclear would be entirely unsuited for this task. Nuclear provides baseload, it doesn't fill in gaps. If you try to run the reactor intermittently to counterbalance an intermittent source the cost of its output increases massively.

That is true when the market price is fairly static and doesn't change much. This is how it used to be when fossil fueled power dominated the grid, since the biggest cost is the fuel and that could scale based on demand.

For EU this is no longer the case. The difference in market price between low and high can be above 100x. In theory a power plant could earn as much in 4 days as an other plant earn in a year worth of power generation. This is why all those nations started to bailout the power bills of businesses and citizens last winter. A single month for some people costed more than a years worth of power. For companies with contract obligations, paying what ever the market demanded was the lesser evil.

The US is not in the same situation, but the energy grid there is still a market based one. There is also other technologies that could in theory compete in such volatile market.

Those high points are better addressed by such things as hydrogen-burning turbines. Hydrogen produced from excess renewables during the price troughs, stored underground, then burned at the peaks. Because the capacity factor of these turbines would be low the cost of fuel would be acceptable, and their capital cost would be an order of magnitude below what a nuclear plant would cost, per unit of power output.

Europe has enough salt formations to store many petawatt hours of hydrogen, far far more than would be needed.

People are naturally allowed to invest in what ever technology they think will fit the role best, such as green hydrogen. No one is investing in that, there exist no hydrogen-burning turbines that burn green hydrogen, but someone could become the first and do that.

Producers of green hydrogen are currently more interested in delivering green steel, which pays much better than hydrogen-burning turbines. The general idea is that this will in the future reduce prices down to energy grid levels, and a researcher here in Sweden working on such project estimated prices to drop to those levels around ~2060-2080.

This could happen much earlier if prices continue to increase as they do, but who knows. It would make for a good A/B testing to produce both and see which one was the cheaper option, and if the green hydrogen power plant fail they can always just produce more hydrogen for steel production.

Plenty of investment is going into green hydrogen, to mature the remaining technologies (such as electrolysers and hydrogen combustors) and drive their costs down learning curves.

Note, however, that as long as your grid is still burning natural gas for power, it doesn't make much sense to burn green hydrogen on it too. Eventually the natural gas will get very expensive (CO2 charges if nothing else), but for now price spikes are short term because LNG can be brought in (this is what Europe has done after Russia shut off the gas.)

Green hydrogen is going to have to be something that is produced, because the world uses about 100 million tonnes of hydrogen a year. Ammonia is an essential commodity chemical made from hydrogen. Some 6% of current world natural gas consumption goes to making hydrogen. These markets can and will be served by green hydrogen even before hydrogen is used for grid generation, and this will serve as dispatchable demand to help smooth renewable intermittency even without hydrogen being burned for power generation. And once there are large stockpiles of green hydrogen, it will be a small step to divert some of it for backup generation.

The ACES project for green hydrogen and electricity production is currently under construction in Utah:

https://www.deseret.com/2022/5/9/23046910/green-hydrogen-ene...

The project will take that excess solar and wind capacity and through a process called alkaline electrolysis it will separate oxygen and hydrogen from water through 220 megawatt electrolyzers, producing up to 110 tons of hydrogen a day.

It is called green hydrogen because it is derived from renewable power sources.

...

The other “wow” factor of the project is the salt cavern storage reservoirs.

“Those salt caverns will be the largest single storage site for hydrogen, globally,” Ducker said.

He pointed out that the battery storage capacity across the United States sits at two gigawatt hours via lithium ion batteries. The Utah project will have storage for 300 gigawatt hours of energy.

...

The salt domes for storing the hydrogen will be 3,500 feet underground and will be as deep as the Empire State building is tall — about 1,500 feet.

Ducker said the caverns will enable long duration storage of energy and prevent monthly curtailments of solar and wind energy.

Mitsubishi delivered the gas turbine generators for the project a few days ago:

https://power.mhi.com/regions/amer/news/20230727

And note that these storage caverns are only a small fraction of the potential of that salt formation.

Whats the energy->hydrogen conversion efficiency these days of processes that don't dump CO2? (aka ones that aren't using Natural gas as the feed-stock).

This is such an incredible misrepresentation. The only thing nuclear can't do is respond to sudden demand surges. But that's what batteries can do (it's also the only thing they can do well).

This[1] is the Australian energy market operator dashboard. Note the demand curve. It is not sudden in anyway - it is highly, highly predictable. Nuclear reactors can handle that sort of curve just fine - you roll the control rods in when it's low, pull them out when it's high.

The "inability" of nuclear reactors to handle variable loads is to do with the thermal mass of the reactor pile which can't be changed rapidly, but electrical load generally doesn't change rapidly - it changes very, very predictably at large scale.

Nuclear reactors can handle normal electrical demand flows just fine.

[1] https://aemo.com.au/en/energy-systems/electricity/national-e...

Your counterargument there betrays a complete misunderstanding of the point I was making.

I wasn't claiming that nuclear power plants couldn't technically ramp up and down. I will happily stipulate that they could. I was arguing it was economically ludicrous to do so. That's because most of the costs of nuclear are fixed: capital cost, financing costs, fixed manpower costs. If you operate the power plant at low capacity factor, the cost per unit of energy produced increases inversely, just because these fixed costs are being spread over less output.

Nuclear either makes sense for baseload or it doesn't make sense at all. Trying to retreat to an application for which it isn't suited, like covering for intermittent renewables, is a losing game. There are any number of alternatives that would be much cheaper.

Take a real close look at the scale on that dashboard. While peak power is about 9Gw, sometimes up to 12GW, the lowest power output overnight is 6GW and that's the absolute bottom.

So in a standard 24 hour day, there is at minimum 6GW of capacity which is always available and always demanded. And for most of year that's 2/3rds of the total demand which ever applies.

The question you are not answering is whether the cost of building a proportionally larger nuclear plant - i.e. one which can meet the upper ends of this scale - is substantially more expensive then a smaller one.

The answer is pretty obviously no: nuclear plants are front-loaded in capital and construction costs, but their relative size has very little impact on the cost of building them, or their maintenance needs, or even fueling costs.

The reality is renewables haven't got anything on that except from an electricity market which mostly doesn't care about them. Start asking renewable generators to well you guaranteed kilowatt-hours throughout the year and watch the "cheap" power skyrocket in cost.

If your argument there was correct, people would be building ginormous nuclear plants and not focusing on SMRs.

But many components of a nuclear plant have size that scales with power. There are some economies of scale, but they're fairly marginal. It's mostly to amortize fixed operating costs (like personnel) over more output.

you can nuclear in load following mode (France does)

but it hurts the economic efficiency massively, as costs for nuclear are almost entirely fixed at construction time

Until we’re willing to let Iran, Pakistan, the Talibani Afghanistan build nuclear power plants, nuclear is no hope of anything other than an energy apartheid.

And that’s the optimistic scenario where it actually works well, scales, can be built out rapidly, and is not extremely expensive.

The US Navy seems to be able to do it

The navy uses nuclear reactors for tactical reasons, not economic reasons. Nuclear-powered ships are significantly more expensive.

The US Navy builds many reactors for the price of this power plant.

A Virginia class submarine in 2023 costs $4.3 billion (that's for the whole submarine), it has a 210MW S9G reactor.

The Vogtle Unit 3 was $17 billion over budget, for a $30 billion total budget, for 1,100 MW.

$204MM / MW for the US Navy vs $273MM / MW for Georgia Power et. al.

This project makes nuclear submarines look like a bargain!

Sources:

* https://en.wikipedia.org/wiki/Virginia-class_submarine

* https://apnews.com/article/georgia-power-co-southern-climate...

Following the references from the Wikipedia article, I found that the S9G reactor is rated for 210 megawatts of thermal output. The AP1000 reactor that this article is about has a 3415 megawatt thermal output:

https://www.westinghousenuclear.com/energy-systems/ap1000-pw...

The naval reactor is more expensive per unit of output.

How can costs be reduced? You can't skimp on over-engineering nuclear reactors because they have to be designed and built to deal with rare 'black swan' events, such as jetliners crashing into the reactor core. E.g.

https://www.nytimes.com/2009/02/18/us/18nuke.html

> "The rule, approved by the commission in a 4-to-0 vote, requires that new reactors be designed so their containment structure would remain intact after a plane crash, cooling systems would continue to operate and spent fuel pools would be protected."

You can't risk a failure in the primary cooling system, and since reactors need active cooling in the event of a regional grid power failure just to avoid core meltdown, you need onsight power generation capable of running the cooling loop 24-7 (failure in this system led to the Fukushima explosions). These systems (from cooling loops to steam generators) are under constant stress and have relatively high maintenance costs (a major factor in the closure of California's San Onofre reactor).

Then you have to add in the cost of the uranium fuel rods, which is a complex supply chain issue in many countries (the recent coup in Niger has shut down 1/3 of France's uranium ore supply chain for their reactors, say news reports). Uranium supplies are limited and historically uranium prices get volatile when it seems a reactor boom is coming (look at right before Fukushima). Then you have the long-term costs of spent fuel treatment and secure storage, and eventual reactor decommissioning.

I really don't see anyway to reduce these costs such that nuclear will be anywhere near cost-competitive with today's solar/wind/storage complexes, that are entirely capable of producing reliable 24/7 grid power at costs well below that of a comparable nuclear power plant in most locations.

We've effectively forgotten how to build (specifically engineer) nuclear reactors. This is very similar to problems faced by the space program somewhat recently. During the 3 decade pause people retired and memories faded, there were no apprentices to carry the wisdom forward.

> How can costs be reduced?

Build more reactors, and re-learn how to build them. Note that this doesn't touch on economies of scale, which will likely never really apply to nuclear power. Nuclear is likely to always have immense up-front costs, but it shouldn't cost this much.

Wrong answer. > How can costs be reduced?

Stop building nuclear reactors.

But you're delusional if you think wind/solar are a serious alternative to fossil fuels. Nuclear energy actually is.

No, he's not delusional, but you are deeply ignorant of the current state of knowledge.

https://ieeexplore.ieee.org/document/9837910

"The main conclusion of most of these studies is that 100% renewables is feasible worldwide at low cost."

"Even former critics must admit that adding e-fuels through PtX makes 100% RE possible at costs similar to fossil fuels."

Large infrastructure projects (and nuclear is certainly one) don't benefit from economies of scale. Half of a nuclear power plant is essentially the same as a coal plant and they have not gone down in price either.

The reality is even assuming that we can not overcome shortages in solar and wind by overprovisioning and storage (and studies say otherwise), it does not make any sense to build nuclear instead of solar/wind as long as we are still running coal. We get much bigger CO2 reduction bang for our buck with solar and wind. Building nuclear would therefore effectively increase our CO2 over alternatives. This is especially true as nuclear plants have a relatively long ROI (in terms of CO2.

If half is the same as a coal plant could existing coal plants be converted to nuclear? They would be in the right places for power generation and it would be a win by reducing coal and not having to build a whole new plant.

Thats already the plan for some coal sites. TerraPower (Bill Gates is the Chairman) wants to convert the first one in Wyoming.

https://www.energy.gov/ne/articles/doe-report-finds-hundreds...

Large infrastructure projects (and nuclear is certainly one) don't benefit from economies of scale.

If you do them regularly, then sure they do. You already have trained workers, an intact supply chain and contact with the regulators.

“seven years late and $17 billion over budget. At its full output of 1,100 megawatts of electricity”

Ignoring interest for those 7 years and all other costs just the overage is insane. 17 billion / ( 1,100,000 kW * 90% capacity factor * 24 hour * 365 days * 50 years) = 4 cents per kWh. Add interest etc and someone lost an incredible amount of money on this project.

> someone lost an incredible amount of money on this project.

All roads lead to Ratepayers (electrical customers).

> Georgia Power’s residential customers are projected to pay more than $926 apiece as part of an ongoing finance charge and elected public service commissioners have approved a rate increase. Residential customers will pay $4 more per month as soon as the third unit begins generating power. That could hit bills in August, two months after residential customers saw a $16-a-month increase to pay for higher fuel costs.

> The high construction costs have wiped out any future benefit from low nuclear fuel costs in the future, experts have repeatedly testified before commissioners.

> “The cost increases and schedule delays have completely eliminated any benefit on a life-cycle cost basis,” Tom Newsome, director of utility finance for the commission, testified Thursday in a Georgia Public Service Commission hearing examining spending.

> The utility will face a fight from longtime opponents of the plant, many of whom note that power generated from solar and wind would be cheaper. They say letting Georgia Power make ratepayers pay for mistakes will unfairly bolster the utility’s profits.

> “While capital-intensive and expensive projects may benefit Georgia Power’s shareholders who have enjoyed record profits throughout Vogtle’s beleaguered construction, they are not the least-cost option for Georgians who are feeling the sting of repeated bill increases,” Southern Environmental Law Center staff attorney Bob Sherrier said in a statement.

This will likely be the last commercial nuclear generator ever reaching criticality for the first time on US soil. Consider the current interest rate environment and the appetite for backstopping a multi decade construction project.

https://www.lazard.com/media/2ozoovyg/lazards-lcoeplus-april... [pdf, start at page 4]

It's nice to have a captive customer base. Georgia Power serves more than a million additional people now than it did when this thing broke ground. And somehow all the pre-existing customers, plus the million+ new residents, all have to pay $4 extra for the power generated here. This, in addition to the fees they pay for electricity.

Vogtle isn't going to be a good sales pitch for expanding nuclear.

It's no wonder Georgia Power doesn't provide a calculator to let people know how Vogtle coming online affects the per-unit cost to consumers.

And that's why I left the nuclear industry. Nuclear power is the safest form of power production we've ever produced (more people are killed per MWh installing and repairing wind turbines than in the nuclear industry) but it has never been anywhere near cost effective and no nuclear project has ever been completed anywhere near on-schedule. Every time I hear someone talk about how the world needs tons of new nuclear plants and solar and wind can't possibly meet the demand quickly enough, that person seems to imagine that none of the next wave of nuclear plants magically will not have any of the problems that every previous nuclear plant had. Meanwhile solar and wind are beating estimates year after year. I still love nuclear physics (and I am excited about several up-and-coming fusion projects) but I just don't believe in the nuclear (fission) power industry anymore.

The sad part is, this is a solvable problem. If the nation really wanted to build a large number of safe, effective nuclear power plant, we could, and probably in a time frame of months, not years. But entrenched interest from natural gas and coal producers combined with anti-nuclear sentiment practically guarantee that will never happen.

Nuclear power is held to too high of a standard to really be a viable source of power, especially in a democracy.

> Nuclear power is held to too high of a standard to really be a viable source of power, especially in a democracy

Don’t tell that to France.

France can't do it these days either. Look at the fiasco of the EPR; it drove their nuclear corporation to insolvency.

None of that is the root problem. Nuclear plants are too big, and we can't build big things cheaply in this country.

The idle corruption of middle managers is what kills these kinds of projects.

It's not corruption; people in power don't want nuclear, and so the industry is regulated into its current quagmire. No other industry faces moving goalposts like nuclear does.

The USA can move mountains if it really wants to. But as it stands, the country is doing everything it can to keep that mountain right where it is.

> The USA can move mountains if it really wants to.

not cheaply, no.

>people in power don't want nuclear, and so the industry is regulated into its current quagmire. No other industry faces moving goalposts like nuclear does.

I think it's the other way around. We only have nuclear today because people in power wanted it for strategic reasons. Getting lots of energy now and paying for it later seemed like a good idea at the time, so the goalposts were artificially placed to make it look financially viable. "We'll figure out the waste thing later, let's go!"

Politicians are tempted by the same thing today. Voters worry about electricity prices, so give them power today and let their grandchildren pay for it!

France dictated the price of nuclear to a point that bankrupted the operator, even though the same operator basically neglected maintenance to a point where the plants have to be taken offline for - I kid you not - rust.

The problem is that now we are the people having to pay, and nuclear has the unfortunate property that you have no choice but to pay, no matter how wrong the estimates have turned out to be you can't just abandon the project.

Any other power plant can simply be abandoned if it turns out it was a bad idea, and the costs will stop.

Doubtful.

Nuclear is very specific tech. I don't think there's a whole bunch of companies out there sitting on their hands just waiting for a bunch of contracts to show up.

> If the nation really wanted to build a large number of safe, effective nuclear power plant, we could, and probably in a time frame of months, not years. But entrenched interest from natural gas and coal producers combined with anti-nuclear sentiment practically guarantee that will never happen.

meanwhile the Chinese have 22 power stations under construction and 70 planned

Interesting to hear an inside perspective! While they definitely aren't cheap, do you feel they're not worth the price, given the urgency and rapidly rising costs of climate change? It seems like they'd be competitive with, or even cheaper than, fossil fuel plants if we priced in their externalities, and nuclear can shore up the areas where renewables have difficult to solve gaps. Any thoughts?

It’s hard to see projects that take ~14 years to produce the first kWh urgent solutions to anything.

Spend the same money starting in the same year and the first solar project makes enough money to fund a second which then also comes online before your nuclear power plant is ready.

Which doesn’t make nuclear useless, but it’s really disappointing if you’re in the industry.

I feel like the root of the pro/against nuclear debate comes down to your outlook on storage :) I'm skeptical that storage can meet our needs (would love to be wrong! haven't seen a convincing analysis, yet), so I think we need a reliable form of continuous power. The only option currently on the table for that is nuclear, so, I boost nuclear. I dislike the nuclear nay-saying, because it further delays an already slow process. When (IMO) the optimistic plan for storage doesn't pan out, we're really going to regret not having started building those nukes ~14 years ago... so why not hedge our bets and get 'em started right now? In the grand scheme of climate change costs, they are absolutely dirt cheap.

What exactly do you find unconvincing? Just comparing grid storage needs vs EV’s suggests battery production capacity isn’t going to be an issue so it simplifies to cost across say 20+ years.

There’s a lot of very simplistic analysis that makes things seem vastly worse but you don’t need crazy assumptions for things to look reasonable. Basically the more sources of energy you can add to the mix the cheaper things become as you only include them when you save money. Thus if you ignore Wind in an estimate adding wind can only lower costs, ditto for nuclear, geothermal, etc.

Hydro is reliable and better than continuous power because it’s flexible enough you can fill in gaps based on what you forecast production will be across the next week. US only gets 6.2% of its power from hydro so the rest of the world actually looks better than these numbers.

How much Solar to build is part of a complex optimization problem, simplify it 2c/kWh of capacity but 1/2 of all solar power is wasted over a year ~= 4c/kWh. Further, let’s assume 1/2 of all solar is used directly with the rest available to charge daily batteries again pessimistic with that kind of oversupply and HVDC lines etc but whatever. 50% - 5% (hydro) = 45% of total demand from batteries. (Wind drops this by a lot.)

Projecting LCOE for batteries and Battery degradation is closely correlated with use so 120$/MWh is again high long term.

Now our worst case is something like 4c/kWh (solar/hydro) + 12c/kWh (batteries) * 45% = 9.4c/kWh for 24/7/365 power with zero fossil fuels and zero nuclear. On the surface that might seem terrible or awesome depending on what your local rates look like, but this was a very pessimistic estimate.

Our pro nuclear investor needs to be pessimistic in the other direction. Remember demand is shifted to cheaper nighttime rates, if daytime electricity is cheaper then nighttime use drops further. Baseline numbers of say 4c/kWh production + 80$/MWh batteries * 30% = 6.4c/kWh 24/7/365. If you’re considering building nuclear and projecting 60+ years out those kind of numbers represent a real threat without any dramatic breakthroughs.

PS: You can plug your own assumptions in here, but remember people are trying to minimize costs and we already have existing infrastructure that isn’t being replaced that quickly. I doubt we’ll average under 20% natural gas within 20 years due to flexibility, which dramatically reduces the need for batteries and excess solar. That sounds bad, but mix in significant EV adoption and it still becomes a vast reduction in CO2 emissions, while even lower costs discourage Nuclear.

> What exactly do you find unconvincing?

Most of the estimates I've seen for storage requirements for current electricity usage are not optimistic[1]. When you add in the huge amount of energy used for heating, which is currently not electric, it gets even worse.

I'd like to see an analysis of how much generation and storage is required to handle Minnesota's natural gas usage[2], with a reasonable guarantee that power will not be lost for more than an hour or two during the entire yearly six month cold weather span. Are those generation & storage estimates reasonable? How much will they cost, how reliable are they, how much room will they take, how many natural resources will they consume? Keep in mind Minnesota will not be able to bogart the entire battery manufacturing capacity of the globe, and remember that this analysis is in addition to the amount required to meet current electricity usage needs. Then, extrapolate that analysis to all other cold weather areas, such as other northern US states and most of Canada. Is renewable+storage really feasible to meet that need?

I've never seen this analysis done, and what I've seen from other analyses makes me think it's not feasible without depending on unproven future tech. But I'd love to be wrong!

[1] For example, https://news.ycombinator.com/item?id=36949165

[2] Approximately 500 billion cubic feet of natural gas per year. Note also that the usage is not evenly distributed through the year, we use a lot more in February than we do in July. https://www.eia.gov/dnav/ng/NG_CONS_SUM_DCU_SMN_A.htm

The person you are quoting in [1] is a chronic bad faith actor. Examine his arguments with great skepticism. The argument he is making there is obviously wrong -- he's implying (without justification) that battery production cannot be greatly expanded, and ignores non-battery storage.

I don't agree with that characterization, but regardless, that was just a handy link. There's plenty of well-justified skepticism regarding storage out there. Either way, I'd love to see the analysis I suggested. If it's that clear-cut, surely it's not too hard for someone better informed than me to put together the numbers.

I think if you really dig into the arguments he is making, you will spot the evasions and non sequiturs, and conclude he's not arguing in good faith.

There is so much peer reviewed work saying the opposite of what he claims that you should default to skeptical about him. It's the same way one should treat a creationist.

It's because of the urgency that nuclear is not the solution. You need more than a dozen of years to build one. By that time it will be too late for the climate.

>>Nuclear power is the safest form of power production we've ever produced.

This is a common claim but it abuses statistics. By looking only at the one metric (deaths) it ignores the enormous enterprise required to keep nuclear as safe as it has been. No other power source needs anything like it.

It also ignores the statistical value of lives. From a policy point of view, a life saved is worth about $9 M in the US. The putative lives saved by nuclear (vs. renewables) cost much more than this. If one accepts that this was worthwhile, it also implies the NRC is not regulating nuclear sufficiently, since additional improvements (like filter containment venting systems) would pay off in expected lives saved.

I would argue that ratepayers would lose significantly more, in the form of

iq decreases, cancer, early death from air pollution, and ongoing, worsening weather damage from climate change

were this plant not constructed and the grid powered from natgas or coal instead.

I hate these discussions about power, because in every case the quoted price of alternatives does not factor in the huge and obvious externalities.

https://www.ewg.org/news-insights/news-release/2023/01/2024-... (By 2024, one-fourth of U.S. electricity will come from renewables: EIA)

https://www.reuters.com/sustainability/us-power-regulator-we... (US moves to link more wind and solar projects to electric grid) ("Today there is more than 2,000 gigawatts of renewable power waiting to be connected to the grid -- nearly double the amount of current U.S. generation capacity, Federal Energy Regulatory Commission Acting (FERC) Chairman Willie Phillips said at a press conference following the unanimous vote.")

https://www.energy-storage.news/us-utility-scale-battery-sto... (US utility-scale battery storage industry deployed 4GW/12GWh in record-breaking 2022)

https://cleantechnica.com/2023/03/31/turning-america-into-a-... (Turning America Into A Solar Manufacturing Powerhouse)

Low carbon power from Vogtle is welcome, because it's here and what is done is done, but there is no point in throwing good money after bad on commercial fission.

I would add to the list of externalities if we do not build more nuclear power plants: CHAOS

chaos that will ensue when we will reach peak oil/gas/coal, as we extract and consume faster than earth produce them. When the price to extract oil/gas/coal would be too high, chaos will result of our economies addiction to fossil fuel.

The sooner we use alternatives the better.

Housing: So, yes, it is annoying to have new house built with only electricity (i.e. without a gas line) but unless energy becomes super abundant and cheap so we can create gas (out of CO2 and water), electricity is the best bet long run.

Transportation: Same, the sooner we can electrify transportation the better starting with trucks and trains. Boats and planes might get the last drop of oil.

Industry: Should get incentives to move to electricity.

And all electricity should be eventually non-fossil (nuclear + solar + wind + hydro).

Those are non-intermittent kilowatt hours, though (and the maintenance that does need to happen is known in advance). One energy demand at peak production is saturated, intermittent sources become a lot more expensive since storage needs to be provisioned. Some markets are fast approaching this scenario: https://reneweconomy.com.au/california-duck-curve-now-a-cany...

Wouldn’t solar, as an intermittent source, be well correlated with demand in Georgia?

Sun’s out = heavy AC load?

Doing any grid 100% solar needs a lot of storage, but there’s gotta be a perfect point that mostly shaves peaks really nicely.

(Yeah yeah, it can be cloudy and hot+humid, but still)

Demand typically peaks in the evening, around 9pm. This is right as the sun sets, where solar is tapering off. It's not "shaving peaks" about half your electricity consumption (including peak consumption) is going to happen during periods of non-production.

And yet their on-peak times in the summer ends at 7pm:

> Shift some of your summer energy usage away from the on-peak time periods (2-7 p.m., Monday–Friday, June–September, excluding holidays).

Demand shifting is just a long winded way of saying, "we can't generate enough electricity, you just need to stop using power at certain times of day." And it's easier said than done. Residential demand shifting, and some forms of industry (e.g. arc furnaces) can shift demand. But others cannot: the pumps powering your sewage system needs to work all day around. Same with data centers, and plenty of other industries.

Right, but we don't need to shift all demand for it to have a big impact on the amount of storage required. Any demand shifted is energy that doesn't need to be stored.

Many smaller intermittent sources with non-correlated outages is better for availability than a single dominant source even with preplanned maintenance intermittency. What is the emergency and disaster planning impact if the nuclear plant goes down in the middle of a summer heatwave?

Smaller intermittent sources don't really help since they're still subject to the same weather patterns and earth rotation. Distributing generation across continents would involve massive costs in energy transmission. People like to tout that intermittent sources are more distributed and less centralized. But that's negative facet of renewables, since demand is centralized in population centers, and distributed production means more transmission costs.

The same criticism can be applied to any electricity source: What happens wind stops during a heatwave? Or when the sun goes down? You might say nuclear is less reliable, but the reality is nuclear power has the highest capacity factor of all generation sources: https://www.energy.gov/ne/articles/what-generation-capacity#....

Instead of proposing cross-continent distribution lines, just build more small regional renewable sources... unlike nuclear, their costs keep going down. Smaller renewable solar and wind sources with battery storage are perfectly fine, and the correlation in heat with sun works to reduce the need for storage capacity.

Capacity factor does not address reliability or emergency planning...

> Smaller renewable solar and wind sources with battery storage are perfectly fine, and the correlation in heat with sun works to reduce the need for storage capacity.

Again, peak electricity demand does not happen at noon. It happens at around 9pm, when the sun either has set or is about to set. Unfortunately, solar's production does not match demand patterns.

> Instead of proposing cross-continent distribution lines, just build more small regional renewable sources...

Again, all the renewable sources in the same region are subject to the same weather patterns and and day/night cycles. Sure multiple solar farms gives you redundancy against some sort of mechanical failure that causes one specific solar farm to go down. But the most common failure mode in intermittent sources is cloud cover or lower-than-expected wind speeds. A backup solar plant a mile away isn't going to give you redundancy against weather patterns. This is why a lot of plans for renewable grids are contingent on thousands of miles of HVDC lines to move energy across continents.

> Capacity factor does not address reliability or emergency planning...

Capacity factor describes the uptime of a power plant. Nuclear power has the highest uptime. The point is it's less likely to go down than all alternatives.

Cloud cover doesn't stop all production, it just reduces it. Just build more solar and more storage and it will still be a fraction of the cost of nuclear. I could answer point by point but all of your "failures" are artificially constrained.

Cloud cover reduces solar production by 10-25%. If you need 10x overproduction, plus storage it's not cheaper than nuclear. That's not even counting factors like axial tilt (during winter the Earth is rotated away form the sun, reducing incoming light per M^2 of land), and the transmission infrastructure required to accommodate the decentralized nature of solar and wind [1] (which nuclear conveniently avoids since it's much more energy dense).

These aren't artificially constrained issues. These are real practical barriers. Why haven't countries the world over completely to renewables if it's cheaper? Because corporations want to screw up the environment because... they're moustache-twirling evil people or something? But the reality is that intermittent sources still have significant barriers to implementation that won't go away without massive, orders-of-magnitude improvements in storage performance. Renewables are good to deploy in an opportunistic fashion, supplementing dispatchable sources during periods of production and then turning the gas back on when they're not producing. But actually producing a primarily renewable grid becomes vastly more challenging due to the intermittency.

1. https://www.vox.com/videos/22685707/climate-change-clean-ene...

Fossil fuel companies are those mustache-twirling profit optimization villains, and there is demonstrated evidence that they have run a decades long campaign against any form of energy that is not fossil fuel...

10-25% short fall doesn't require 10x over capacity, it requires a mix of overcapacity + storage + energy conservation measures.

The energy storage at relevant scale is nowhere near feasible as other comment chains explain. "Energy conservation" is just an admission that these sources can't generate enough energy.

Other comment chains have explained the storage scale you claim is needed, isn't a great estimate.

Energy conservation is probably a better payback than either nuclear or renewables and better resilience for people infrastructure. For example, better insulation reduces energy costs in both cold winters and heat waves, and reduces the impact if there are grid/source failures by any technology. It also doesn't have a big crossover on the logistics tail of either nuclear or renewables, and so is a great thing to do in parallel to address climate change in general.

"Conservation is an admission of failure" is a purely ridiculous position.

The estimate of 12 hours of storage was one of the more optimistic ones. It's not just day/night cycles there's also weather that results in long term fluctuations in output. Some of the estimates call for weeks of storage.

Energy conservation is likely not going to be feasible as more and more transportation gets electrified, as gas heating is replaced with electric, and industrial processes like smelting are decarbonized. Conservation is indeed an admission of failure, because success involves accommodating the growing demand for electricity in the future. Remember, electricity production is only about a third of total energy consumption: https://www.eia.gov/energyexplained/us-energy-facts/

Estimates for weeks of storage are made by people who don’t understand the fundamental cost optimization going on.

More production costs X which reduces the need for batteries by Y, as long as X > Y you build more production even if most of it’s output is unnecessary. Also 16% of global electricity comes from hydroelectric generation, we really don’t need that many batteries.

Actual optimized energy storage is on the order of 2 to 6 hours total globally ~30 years from now depending on how much prices drop. With larger numbers representing lower battery prices. In the shorter term it’s a tiny fraction of that.

2-6 hours is nowhere near enough to offset daily fluctuations, let alone seasonal fluctuations. Overproduction only gets you so far, unless this proposal also involves transoceanic HVDC cables. Hydropower is not exactly the same thing as storage: you can stop releasing water and build up the reservoir, but you cannot pump water up river. Even just 6 hours of storage -which would not be sufficient even with several factors of overproduction - is 15,000 GWh of storage to satisfy present global electricity use.

Overproduction means you don’t ever see daily fluctuations you have a surplus every single day of the year. Unless storage gets insanely cheap that’s by far the cheapest safe option.

Hydropower dams are already storage. When you don’t release the water it’s saved until you do.

If you have enough rain from 3 months ago to release 10 MW 24/7 over a week then you also have enough water to release 15 MW for 12 hours and 5 MW for the other 12 every day etc. You can’t save 100% of the water or the river downstream runs dry but the minimum is generally well below the average flow rate.

This isn’t unusual, most dams ramp production up and down to maximize the value of the stored water.

> Overproduction means you don’t ever see daily fluctuations you have a surplus every single day of the year.

No, this is not even remotely true. No amount of solar overproduction will let you produce energy at night. 12 hours a day (on average, depend on the season) you will not have any production regardless of overproduction. You'll have a lot of excess energy during the day, but still no energy when the sun has set. Infinity times zero is still zero. Not unless you create transoceanic HVDC cables that connect Eurasia to the Americas, literally piping electricity from one side of the world to the other. Wind also has windless days, with more extreme seasonal fluctuations. Overproduction is not a silver bullet that eliminates the need for storage. Just because you have net surplus over the course of a 24-hour period does not mean demand was satisfied at all times. This is why intermittent sources are so challenging.

> Hydropower dams are already storage. When you don’t release the water it’s saved until you do.

But you can only save energy at the rate that rainfall refills the dam. If you have a dam that refills at a rate of 10MW, you can only effectively store 10MW of excess energy. If you have 20MW of excess production, you can still only store 10MW and the other 10 MW is wasted. Dams are not batteries, they can store a lot of energy but they can only be refilled at the rate dictated by rainfall.

You also can't completely shut off a dam or the river will run dry with serious ecological consequences, and impairing downstream water supply.

Excess per day doesn’t mean you get solar power every single second just per day, that’s what batteries are for.

> But you can only sacs energy at the rate rainfall refills the dam. > can’t completely shut off

You’re completely misunderstanding how dams are used look here: https://en.wikipedia.org/wiki/Hoover_Dam

Installed capacity 2,078.8 MW, Capacity factor 23%

Minimum flow rate isn’t 0, but is plenty low enough to have a great deal of flexibility.

> Excess per day doesn’t mean you get solar power every single second just per day, that’s what batteries are for.

...which would require an enormous and infeasibly large amount of batteries. You're right: overproduction doesn't mean you get power for every second per day. But a grid does need sufficient energy at every single second per day or you have blackouts. This is why overproduction of intermittent sources is not as useful as it sounds.

Nothing in your link about dams contradicts what I wrote: their rate of recharge is limited. You can shut off much of the turbines and let the reservoir build up, but they cannot be recharged faster than the rate at which rainfall refills the reservoir. Judging by a 2 GW capacity and 23% capacity factor, one would infer that it's refilled at a rate of ~500 MW (in reality, less than that since Lake Meade is shrinking). Completely shutting off the turbines would only refill at a rate of 500 MW. If you have 2 GW of overproduction, you can't refill Lake Meade with 2 GW of potential energy. You can shut down its turbines and let the reservoir refill at a rate of 500 MW, but the other 1,500 MW can't be used to recharge Lake Meade.

A dam is sort of like a battery but its rate of recharge is much more limited than it's rate of discharge, which is a big disadvantage when trying to capture the overproduction from intermittent sources.

> ...which would require an enormous and infeasibly large amount of batteries

False, but you can’t substantiate your argument by simply saying the words you need to back it up with something such as actual calculations etc.

> Their rate of recharge is limited

That’s completely irrelevant here. Rainfall is so concentrated in short periods that they often have months of water in reserve and can decide when exactly to release it over that kind of timeframe. 95% of the time there is less water flowing out of a dam than flowing into it. That’s why we build dams.

If you want to calculations on just how far short batteries fall, I do so here: https://news.ycombinator.com/item?id=36949165

The rate of recharge is absolutely relevant, because you can't actually capture excess production from intermittent sources. If you're relying on a dam to fulfill periods of non-production, you need a way to put the excess energy during periods of overproduction back into the dam. But a dam can only shut down its turbines, it can't be recharged faster than the rate that rainfall refills it. If you need 40% of your electricity coming from dams during periods of non-production, then you need rainfall sufficient to produce that much energy. It's not like a battery where you can take excess production and store it back in the dam. That's how pumped hydro electric storage works: you run turbines backwards and refill the dam with excess energy. But pumped hydro requires a very specific set of geographic features, and is not easy to scale up.

> If you're relying on a dam to fulfill periods of non-production, you need a way to put the excess energy during periods of overproduction back into the dam. If you're relying on a dam to fulfill periods of non-production, you need a way to put the excess energy during periods of overproduction back into the dam.

I understand your point but it’s based on faulty assumptions.

Simply not using existing water means it’s still there. If you have 10,000$ in your bank account and you don’t buy something you still have the 10,000$. Dams are the same way if you have 20,000 MWh worth of water and can average 20 MWh for the next 1,000 hours then generating 10 MW for the 500 hours of those hours and 30 MW for the other 500 hours hits zero at exactly the same time.

Recharge is important long term but irrelevant in the short term. You might expect to receive water from the spring thaw, but that’s a long way away. Large dams like the hover are built to contain multiple years worth of average flow for a river. It took more than a full year just to collect enough water for them to start generating hydropower.

As to your analysis,

> 500 GWh globally

That’s an outdated estimate for last year even just EV’s broke 500 GWh. “Automotive lithium-ion (Li-ion) battery demand increased by about 65% to 550 GWh in 2022, from about 330 GWh in 2021” https://www.iea.org/reports/global-ev-outlook-2023/trends-in...

Your number was an estimate for 2022 total production made during 2022, and they got it wrong which isn’t that surprising as EV sales ended up 55% from 2021 and average battery sizes also increased. 2023 numbers are hard to estimate for similar reasons.

> Production of batteries may grow in the future

Again, the rates have been increasing by double digits per year for a long time, that’s wildly faster than the increase in electricity demand. We don’t need to talk in hypothetical terms here just current factories already wildly invalidate your calculation let alone any kind of longer term estimates when grid storage may start to pick up.

> 12 hours

As I mentioned that’s a monumental overestimate, but not particularly relevant compared to the first two issues. We can quibble about specifics here but compared to even a 50% EV world grid storage simply isn’t a major factor.

Rate of recharge is the limiting factor of the total amount of energy you can release from a dam without reducing the total amount of water in the reservoir over time. Hoover dam may be able to produce 2GW of power, but if it averages more than 500 MW it will drop in water level over time (actually less than that, since it's already reaching record low fill levels). True, rainfall is not constant throughout the year. But it doesn't really matter in the big picture: the total amount of energy you can produce with the dam without lowering fill levels is fixed, and it isn't changed by overproduction.

Overproduction doesn't help you here: if you have 3 GW of solar energy and 2 GW of electricity demand, you can't use the remaining 1 GW to refill Hoover Dam. If you produce enough solar to meet demand, you can shut off hoover dam's turbines. But if you overproduce - if you produce more energy than the grid needs - it's wasted, you can't use excess energy to refill the dam beyond just shutting off the turbines.

> That’s an outdated estimate for last year even just EV’s broke 500 GWh. "Automotive lithium-ion (Li-ion) battery demand increased by about 65% to 550 GWh in 2022"

Demand, not production. Just under 500 GWh of batteries was produced, a good chunk of demand went unfulfilled. If anything, all your source shows is that grid storage is even more infeasible because we can't even satisfy EV's storage needs.

> As I mentioned that’s a monumental overestimate

No, if anything it's an optimistic underestimate. 12 hours is just enough for diurnal storage. But you also need storage to offset seasonal fluctuations.

Non-interment kilowatt hours are actually less valuable. Demand dips at night, the weekend, spring and fall etc. So when negotiating power purchase agreements grid operators are pricing most of that production when it’s least valuable.

Battery backed solar is becoming really popular with grid operators because you get the upsides of load following without the overhead of mostly idle production. The economics get interesting as more solar comes online, but things are only getting much worse for nuclear and coal. Which is why there is basically nothing in the pipeline for either one.

Incorrect, demand peaks right as night starts and the sun goes down. See that peak [1] at 9pm? Battery backed solar is not even remotely feasible. To put this in perspective, the USA alone uses ~11.5 TWh of electricty each day. A battery to last through the night is over 5,000 GWh.

1. https://b698061.smushcdn.com/698061/wp-content/uploads/2023/...

There are several battery backed solar power plants in the US already built.

Collocated Solar with batteries increases efficiency as DC from solar is used to directly charge batteries without the cost or losses associated with DC>AC>DC you would see if these where separate. LCOE is already below 60$/MWh because you only need batteries for a fraction of total production.

https://atb.nrel.gov/electricity/2022/utility-scale_pv-plus-...

Sure, but those battery backed solar plants usually only have 2-4 hours worth of storage. They're not actually producing full output throughout the night.

The cost of batteries would skyrocket if countries actually tried to provision 12 hours of battery storage. Because the US alone uses 500 GWh of electricity each hour. This is greater than the global battery production figures in 2022 [1].

1. https://www.spglobal.com/mobility/en/research-analysis/growt...

You’re calculating that storage number as if the solar power plant would have produced power for 24 hours.

To simplify if you have 8 hours of power and you store 1/2 of it you need 4h of storage. So our hypothetical 4GW solar power plant produces 2GW for 8 hours and 1 GW for 16. That already batter fits the demand curve than steady state output.

However things get better if you instead consider real world demand and the actual production of solar panels being spread across the entire day a 4 hour of storage is close to perfect. Except, the real world isn’t a hypothetical 100% solar grid so 2h ends up working fine in most areas.

Again, the demand curve is the opposite of what you claim: peak production happens during the night, at around 9 pm, not during the day. You'd need to supply more energy during periods of non-production, not less. You can't get away with 50% output at night, unless you have some other energy source picking up the slack. And remember, during the winter you're getting less than 12 hours of sunlight, maybe even less than that if you consider that cloud cover eliminates 75-90% of solar power output.

These factors do depend on geography. Hotter climates do see more energy use during the day to power A/C. But on the flip side, colder climates see even more energy demand during the night to heat homes. And unfortunately this also coincides when solar is producing the least amount of energy due to fewer sunlight hours, axial tilt, and more cloud cover.

There are niches solar can carve out: Las Vegas and Australia have loads of empty space, clear weather, and hydroelectricity that can fill in periods of non-production. But nuclear power is a much more flexible solution. All you really need is water to cool the reactor, and 80% of the population lives within a hundred miles of the coastline (you just need water, not fresh water).

Energy use for heating can be quite easily reduced just by insulating houses (and installing more efficient heating equipment such as heat pumps).

Do we have any idea what the cause of that 9pm spike is? That seems like a bizarre time to be the peak to me (I can't work out what the primary energy users would be at that time).

Replacing gas heating with heat pumps is going to increase, not reduce, electricity use.

Replacing gas heating with solar thermal is even cheaper and doesn’t. Electric heat pumps are viable when electricity is cheap or demand is low and you want AC, the more they drive up electric costs the more alternatives become viable.

Net effect you tend to see them in areas where there’s a surplus of winter electricity which also tends to be places with minimal demand for heating like Virginia rather than Mane. 46% in South Carolina, 42% in North Carolina, 30% in Virginia etc: https://www.statista.com/statistics/1327164/share-of-househo...

Solar thermal heating is not going to produce much energy in the winter when there's fewer hours of sunlight and more hours of cloud coverage - which just so happens to coincide with places that would need heating during the winter. Solar thermal performance is worse with diffuse light (read: overcast days) than PV solar, in fact that's a big part of why it was largely abandoned in the 2000s.

The main advantage of solar thermal is it’s really cheap so you can just scale collection to cover winter heating. That’s why adoption is so high in China even relatively far north. As to the US, the cost savings just isn’t that appealing for most people. It’s frankly ugly, complicated, and has minimal payoff at current prices.* Passive solar design can have significant benefits while looking much nicer. But if you’re increasing the cost of electricity or gas then suddenly it’s looking significantly more attractive.

* Except of course when people want to heat a pool, that takes crazy amounts of energy.

“Israel became the world leader in the use of solar energy per capita with 85% of households using solar thermal systems (3% of the primary national energy consumption)” https://en.wikipedia.org/wiki/Solar_water_heating

As per your link, solar thermal water heating is very variable depending on geography:

> The amount of heat delivered by a solar water heating system depends primarily on the amount of heat delivered by the sun at a particular place (insolation). In the tropics insolation can be relatively high, e.g. 7 kWh/m2 per day, versus e.g., 3.2 kWh/m2 per day in temperate areas.

Because of this, many solar thermal systems are supplemented by convention heaters.

> In winter, the percentage of your hot water heated by the sun drops to as low as 10-20%—as you might expect with short days and weak sun in December. That’s why practically every solar water installed in the US will be connected to a backup conventional water heater to ensure that your hot water needs continue to be met even in January.

https://www.aspessolarproducts.com/solar-water-heater/how-do...

Again batteries smooth over instantaneous demand.

If you really want to model things you need to consider the vast drop off between midnight and 5 am in which that steady state I talked about would be wasted. Batteries don’t really care about when the peak is just the total demand minus power production. And again steady state like nuclear looks terrible by comparison.

Running summer numbers you see an extension in the number of hours the panels are producing which then reduced the gap you need to fill with batteries. Winter sees the reverse as average output drops but conversely batteries are storing a higher percentage of total energy produced. Sizing is further adjusted based on expectations for daily output not hypothetical maximums.

Which of those is more critical depends on local conditions but the grid is always sized for absolute worst case not the average one. There’s lots of micro optimization such as aiming panels East or West to shift production, local hydropower and wind resources etc etc.

None of which is that important compared to the overall effect I was describing where batteries are sized to the daily output not the peak output. The important number is the cost which is running 60$/MWh vs the daily demand not whatsoever the nameplate numbers people love to talk about to make projects seem more important.

And yet again: batteries are not available at a scale where they can even remotely smooth over demand. Even just provisioning 1 hour of storage for the USA requires 500 GWh of storage. A steady state like nuclear works just fine for the grid: it's only a problem for intermittent sources that are made redundant by a source of decarbonized energy that works 24/7. The disparity between peak and trough energy demand is not that great: it's typically 20-30%.

Pointing to one specific project is not relevant: we can't replicate this project to the point where demand is fulfilled because that would require an order of magnitude more batteries than are produced.

We don’t need batteries on that scale any time soon. Let’s project out 25 years and we might need 3-5 hours of 100% grid electricity storage nationwide depending on your assumptions but that’s about it. It’s not like hydroelectricity is going away anytime soon and that’s extremely flexible not just in terms of time of day but also days of the week, and wind doesn’t care when the sun is shining, and solar can shift east to west to cover more of peak evening demand on the east coast and peak etc etc. We can quibble about the numbers but it’s just not important.

As to storage we’re talking 4h * 500 GWh +/- whatever that’s only 2,000 GWh ish, but the specifics aren’t important. By comparison we’re talking a nearly 100% EV cars so 75 kWh per car * 282 million, that’s ~ 21,150 GWh. Those numbers are so huge we could bump things to 16 hours of electricity and it’s still fairly trivial by comparison. Build enough generation and storage could actually be even lower, it’s just a cost tradeoff and renewables are really cheap. (Edit: There’s been a lot of research into it and there’s plenty of raw materials for a 100% EV transition and on this timescale ramping battery manufacturing isn’t an issue.)

Also, that 30% number is wildly incorrect. Even just over a week the CA demand is expected to go from a low of 24,867 to a high of 42,007 in a single day, but you can’t design a grid for a single day. https://www.caiso.com/TodaysOutlook/Pages/default.aspx#secti... Things look similar nationally, we don’t have enough interconnects to really smooth things out and even if you did the daily variation is quite significant.

However, the actual disparity between peak annual demand and minimal annual is vastly higher and you can’t smooth it out. That delta is why idling a grid of mostly nuclear power would be so wildly expensive. Without France style exporting and importing from non nuclear countries cost go insane. The only way it’s even vaguely viable in the US is with similarly vast amounts of storage. But at that point adding just a little cheap solar + wind saves money, and that keeps being true as you use less and less nuclear.

TLDR; Cost effective mostly nuclear grid needs even more storage due to capacity factor issues when you can’t follow France’s model of importing and exporting a large fraction of your power to non nuclear countries.

And on the flip side, some energy markets like New England see a minimal variation in energy demand throughout the day, less than 10%.

Provision enough generation to satisfy peak demand and spend the excess power doing things like carbon dioxide sequestration or desalination in places with limited drinking water. Too much energy is a vastly easier problem to solve than too little. And the great thing about nuclear is that it's no more expensive to run at 100% capacity than 50%.

That sounds great until you consider the opposite.

The issue with nuclear is it costs twice as much per kWh at 50% capacity than it does 100% capacity. Nuclear is expensive at 90% capacity factor, it’s insane at 45% capacity factor.

Further, daily demand isn’t that problematic. It’s really seasonal demand that kills people’s dreams of scaling nuclear power. Over 24 hours going from a 100% high down to 50% low might average say 70-80% depending on specifics. But if 1/2 the year you never get over 80% capacity, and you also need to curtail on nights and weekends year round, and take the plant offline for weeks to refuel etc, then things look much worse.

The temptation is to say batteries to the rescue, but then you’re directly competing with solar on price per kWh. It’s really hard to make things work without natural gas to pick up the slack.

Again, you exaggerate the fluctuations in energy demand. This is a good primer on the hourly and seasonal fluctuations of electricity demand: https://www.eia.gov/todayinenergy/detail.php?id=42915

Even in the summer in Texas, where we experience the largest demand fluctuations, minimum demand is around 60% that of peak demand. And this is indeed a good use case for rooftop solar, especially since it can be used to power A/C in the same building on which its mounted - I never said we should have zero solar power, just that it's infeasible to use for the primary source of electricity. Rooftop solar is indeed a good way to mitigate A/C power draw. But outside of summer peak and minimum power draw is only about 20% difference, and again the peak power is at night.

I don't doubt that nuclear power is expensive. But at least is feasible to build. "Feasible but expensive" is much better than "not feasible regardless of cost". Intermittent sources are only feasible in a grid backed by fossil fuel and hydroelectric sources that can flexibly respond to solar and wind's variation. Once you enter the realm of a predominantly renewable grid, this changes drastically. Storage at a scale anywhere close to what's required to smooth the intermittency of renewables is not even feasible regardless of cost. No amount of money is going to provision the amount of storage required to smooth out the daily fluctuations of solar, let among the seasonal variations of both solar and wind. Overproduction and HVDC connections only take you so far unless you're going to cross the Atlantic (which is also of dubious feasibility). Even just a few hours of storage is well outside of reach. Don't be misled by a handful of battery stations in the MWh scale: actually trying to providing 2 TWh of storage would cause prices to skyrocket because production cannot remotely fulfill that level of demand. Not to mention it'd kill electric vehicle adoption.

This is why most proposals for a primarily renewable grid involve a novel storage system. Will hydrogen, or compressed air, or giant flywheels deliver the required scale? Maybe, in the same vein that maybe deep-drilled geothermal power will deliver cheap decarbonized energy. Those are unknown factors in that one cannot assume will work out. Would you consider it a reasonable plan to assume deep-drilled geothermal will solve decarbonization? After, Iceland has used geothermal for decades, and we have lots of experience drilling for oil? This is how we ought to regard things like hydrogen electricity storage or ammonia: they're in the realm of possibility not feasibility.

Again, it's not a question of cost it's a question of what's even possible without an engineering breakthrough. Maybe we'll figure out a way to build storage at grid scale. And maybe we'll figure out geographically independent geothermal power. But both of those are things not presently within our technical capabilities.

> ...and take the plant offline for weeks to refuel etc, then things look much worse.

Nuclear has historically had the highest capacity factor of all generation sources [2]. This hasn't been an issue with existing generation sources with even more downtime, so why would it be an issue with nuclear?

1. https://www.deegesolar.co.uk/do_solar_panels_work_in_the_win...!

2. https://www.eia.gov/tools/glossary/index.php?id=Capacity_fac....

This is the Australia energy market price and demand dashboard[1].

Note the two peaks: 1 happens at 7am. There's very little solar at 7am even though the sun is up, but demand is at a maximum. The other happens at around 6pm - when people get home. There's also very little solar at 6pm (sun sets at 5pm in Winter, during the day it's longer).

Also note the demand fall off in the evening: that's a very lopsided peak, because people stay at home (whereas in the morning they turn on the kettle, then leave). But also note the absolute magnitudes: at the minimum, which is about 4am, demand last night bottomed at about 6,800MW. The peak was 9,800MW. So even the "low demand" was 2/3rds of peak demand, and it took 8 hours to get there. Most of the night, consumption was a lot higher.

But it gets worse: before you're going to recover any real capacity from your solar, the largest demand of the day is about to happen in 3 hours, with the sun just barely over the horizon.

[1] https://aemo.com.au/en/energy-systems/electricity/national-e...

The better you model this the better the number look. First demand falls off a cliff from midnight to 5 am, a steady state nighttime rate will be extremely wasteful but batteries can shift demand to these peaks just fine. Remember 100% of their output is offsetting battery demand which is significant even just at 9AM or 6PM.

Also, your not actually getting 100% solar output over the day. Batteries that could store 50% of your hypothetical maximum are well over half of your expected average output especially in winter when these peaks are most pronounced.

That’s for March 31st, presumably in the northern hemisphere. I assume the curve looks dramatically different for July 31st.

The curve also looks dramatically different in January 31st. March is about the middle between summer and winter.

That's 'net load', so the graph doesn't show rooftop or grid solar or wind's contribution.

You can see the difference here for today (choose a previous day if it's near the start of the day in California):

https://www.caiso.com/TodaysOutlook/Pages/default.aspx#secti...

The non-grid PV and wind peak is lowered, and shifted a couple of hours later.

And that graph still doesn't show consumer and behind the meter commercial PV impact. Historically there was a double peak, one of which has been wiped out entirely.

The peak still occurs in the evening around 8-9pm. The chart I linked shows a drop in net demand around midday, due to solar's contribution. But my point is the mismatch between peak electricity demand and solar production, which is the same regardless.

That's all from the grids perspective, they only count their own solar and wind.

Here's an animated diagram that explicitly calls out BTM (behind-the-meter) solar:

https://isonewswire.com/wp-content/uploads/2023/02/btm_forec...

Right, and you see where maximum demand is at 8pm (even before any solar was installed)? That peak isn't going down by provisioning more solar. Thanks for the gif, it explains exactly what I'm talking about.

A Tesla megapack holds 3MWh and is warranted for 15years. If you use the entire capacity daily for duck curve shifting and it dies the day after its warranty runs out, (3MWh * 365 * 15), that's 16 TWh. A megapack costs $1.8M, for a cost of 11 cents per kWh. That's cheaper than the fully loaded cost of Vogtle.

But of course you don't have store every kWh used. Peak demand in the summer is caused by A/C and corresponds closely with peak solar generation and can be used directly. New solar farms have a cost of about 0.5cents per kWh, but we'll use 1cent per kWh to be generous. So if you use 3/4 of your power directly and shift 1/4 you end up with an average cost of 0.751c + 0.25(11+1) = 2.75 cents.

And those costs aren't theoretical. For a concrete example, the 8minute energy Eland project proves 24h battery+solar energy for under 4 cents per kWh.

3 MWh is a trivial amount of storage. That's less than 10 seconds of this plant's output. To put this in perspective, the USA uses 11.5 TWh of electricity each day. That's just under 500 GWh per hour. You'd need a lot of megapacks to provision 8 hours of storage. The Eland project you mention has 4 hours of storage, it's not a 24 hour production system.

The reality is that renewables are currently only viable to supplement a grid primary backed by a dispatchable source of energy. If you have loads of hydroelectricity, that's fine, but the regions that don't have hydroelectric potential are going to be stuck burning fossil fuels until a massive storage breakthrough is found.

It's a trivial amount of storage for a trivial amount of money. Do the math, don't hand wave.

The Eland project provides 24 hour power with only 4hr of storage. That's the demand curve in action.

> It's a trivial amount of storage for a trivial amount of money. Do the math, don't hand wave.

Sure thing! Right now we have an annual battery production rate of 500 GWh globally [1]. If we're going to use global battery production figures, we need to use global electricity consumption, which is about 70 TWh per day [2]. How much storage we'll need varies, depending on the mix of solar and wind. Estimates I can find say 12 hours on the low end, 3 weeks on the high end [3].

So even with the optimistic estimates of 12 hours, that means we'd need 35,000 GWh of storage. This is 70 years of global production at our current rate, for the optimistic storage estimates. And of course we can't dedicate all battery production to grid storage - we need them for electric vehicles, and electrical devices

Production of batteries may grow in the future, but then again so will electricity demand as countries develop and transportation becomes more electrified. Furthermore, we're not counting the fact that batteries have limited lifetimes. It depends on depth of discharge, but we're usually looking at 1,500 to 3,000 cycles before they're substantially degraded.

As your can see, the scale of battery production and the scale of energy storage required to make intermittent sources variable are totally mismatched. The reality is there is no amount of money that will provision the battery storage required, because if countries across the world start trying to buy terawatt hours of batteries when only 500GWh of batteries are produced then the cost of batteries will skyrocket. Cathode material already

> The Eland project provides 24 hour power with only 4hr of storage. That's the demand curve in action.

The "demand curve" means Eland doesn't provide 24 hours of power at its rated output. It provides a fraction of its rated power at night and tells customers not to use as much electricity. This may work for some consumers, but not others. The pumps powering your sewage system can't demand shift if you want to flush your toilet at night. The reality is that peak energy demand happens at night [4], when storage isn't producing electricity. Eland can do this demand shift because other producers are picking up the slack.

When you read about storage projects you need to be on the lookout for weasel-words like this. Demand curve means they produce a fraction of the rated power output during periods of non-production. If I have a plant that produces 1,000 MW during the date and 100=MW at night, that's technically 24 hours of production. But clearly this is not the same thing as a nuclear plant that produces 1n000 MW at all hours.

1. https://www.spglobal.com/mobility/en/research-analysis/growt...

2. https://www.statista.com/statistics/280704/world-power-consu...

3. https://pv-magazine-usa.com/2018/03/01/12-hours-energy-stora...

4. https://reneweconomy.com.au/california-duck-curve-now-a-cany...

The US alone has 800GWh of battery plants in the pipeline, to come online before 2026.[1] China has multiple TWh's worth. We can build 35TWh or even 350TWh of batteries a lot faster and than we can build the multiple TW of nuclear plants that would be necessary to decarbonize electricity without storage.

1: https://publications.anl.gov/anlpubs/2022/11/178584.pdf

> The reality is that peak energy demand happens at night

Peak net energy demand happens at night. Peak gross demand is during the day.

> Eland can do this demand shift because other producers are picking up the slack.

Eland is producing at a rate identical to the California demand curve, it's in their contract. It's the solar producers who don't have solar along with consumer rooftop solar that's causing the duck curve daytime demand drop.

But the former costs 1/10th of the latter, so you build 10 of them to get 1000MW at night and 10000MW during the day.

The "pipeline" you're referring to is a measure of battery manufacturing capacity. This is not nearly the same thing as actual production figures. Capacity utilization in 2022 was under 35%. In other words, 100 GWh of capacity only translated into 35 GWh of battery production. This is because the majority of cost of lithium ion batteries is in raw materials, namely cathode material [1]. A huge amount of capacity is useless if you don't have the input materials to feed your factories.

You can't store energy in a battery factory, you store energy in batteries. Cite the actual production figures, not the stated capacity figures (spoiler alert: it was just under 500 GWh last year.).

And to reiterate, the vast majority of this production is not going to grid storage, it's going to EVs and electronics. Even if battery production matches the predicted growth, it's still vastly insufficient to provision grid storage without heavily crippling EV rollout.

> Eland is producing at a rate identical to the California demand curve, it's in their contract. It's the solar producers who don't have solar along with consumer rooftop solar that's causing the duck curve daytime demand drop

Again, electricity demand at night is still high: https://www.caiso.com/TodaysOutlook/Pages/default.aspx

Demand remains high well into midnight. I'm not sure why you think matching the demand curve is somehow going to mean you're going to get away with less storage. Unless Eland is going to be producing much less than its nameplate capacity at all times of day, 4 hours of storage is nowhere near enough for it to match the demand curve. And remember, solar is also subject to cloud cover. I'm sure Eland has clauses exempting it during periods of cloud coverage otherwise they'd need weeks of storage not hours.

> If I have a plant that produces 1,000 MW during the date and 100=MW at night, that's technically 24 hours of production. But clearly this is not the same thing as a nuclear plant that produces 1000 MW at all hours. But the former costs 1/10th of the latter, so you build 10 of them to get 1000MW at night and 10000MW during the day.

The former doesn't have a price tag, because no amount of money in the world will buy you that much lithium ion batteries. Again, the world uses 70,000 GWh of electricty per day, most of that being consumed when solar is not producing electricity. No amount of money can fulfill that amount of storage.

1. https://www.visualcapitalist.com/breaking-down-the-cost-of-a...

Excellent comment.

> If we're going to use global battery production figures, we need to use global electricity consumption, which is about 70 TWh per day

Don't forget about all the energy usage that isn't currently electricity, but will need to be! Especially heating & transportation.

Your link 3 also contains this line: "The solar heavy network wouldn’t need energy storage with an HVDC network."

IOW, the US could build a 100% solar+wind+hydro grid WITHOUT ANY STORAGE. The wind is always blowing somewhere in the US.

Of course that much HVDC and overbuild would be ridiculously expensive, but some HVDC and some batteries are a lot cheaper than only HVDC or only batteries.

> Production of batteries may grow in the future,

To me it seems reasonable to assume that not only will battery consumption grow, it will grow exponentially (over the medium term - say the next 10 - 50 years). Rationale:

- There is high demand

- The production process is well established technology that can be easily replicated

- There are no obvious limits to growth in the medium term (In the short-term there are resource constraints as mines are opened).

And even ignoring grid storage, battery production must increase dramatically for electric vehicles.

The problem isn't intractable. It just requires time, effort, and capital.

https://www.tesla.com/ns_videos/Tesla-Master-Plan-Part-3.pdf

The paper also assumes we'll make have feasible CO2 energy capture and hydrogen electrolysis and storage. It required time, effort, capital, and and multiple engineering breakthroughs. If your plan is contingent on technologies that aren't available... it's just a long winded way of saying you don't have a plan.

Actually, we just need improvements in fusion power and then we don't need solar, wind, or battery storage!

The paper assumes CO2 energy capture at $200/ton. That's "scale up existing technologies" pricing, not "engineering breakthrough" pricing.

It also assumes a vast hydrogen electric grid storage network - this is only theoretical, nobody has actually deployed a facility that converts electricity to hydrogen, and back to electricity. At best this is, "scale up heretofore unproven technologies", not "existing technologies".

In the same sense that commercial breeder reactors are theoretical (which is conservative, as you can't actually buy a breeder reactor). But nuclear advocates don't talk about those much, even though they'd be needed for a nuclear powered world.

But we don't need commerical breeder reactors. The USA just needs to build 4 PWRs for each reactor it presently has to generate electricity entirely from hydro and nuclear. This is vastly more feasible than provisioning terawatt hours of energy storage.

"Just scale up" is a good summary of how to decarbonize through nuclear power. We have decades of experience building nuclear plants. Not so with hydrogen based electricity storage. Plans for renewable grids call for the use of novel electricity storage systems because none of the existing storage mechanisms are feasible. Until we've built the storage mechanisms proposed, any plan involving it's use is effectively hand-waving a big part of its implementation.

Powering the world (all energy, not just the grid, and not just the grid in the US) with PWRs won't work, because we run out of cheap uranium too quickly. Breeding is needed for sustainable use of nuclear.

Incorrect, we have plenty of uranium both terrestrially and through seawater extraction: https://www.forbes.com/sites/jamesconca/2016/07/01/uranium-s...

Sure seawater extraction is more expensive. But procuring raw uranium is a tiny fraction of nuclear's cost. You'll see statistics saying nuclear fuel is a significant cost, but most of that expense is from enrichment not extracting raw uranium.

Seawater extraction is even less plausible than breeders. The polyamidoxime fibers that article discusses do not have sufficiently long service lives. Also, if the uranium were to be burned in PWRs, the effective power per ocean area of uranium collectors would be much less than the effective power/area from photovoltaics.

Uranium is a small fraction of nuclear's cost at current uranium prices. But eventually that runs out, and the price of uranium would increase dramatically. This has always been the motivation for breeders.

Again, the energy density of uranium is such that even if seawater extraction remains expensive, it has negligible impact on nuclear power costs: https://en.m.wikipedia.org/wiki/Economics_of_nuclear_power_p...

> The cost of raw uranium contributes about $0.0015/kWh

Even if this increases by an order of magnitude, this is not significantly impacting the cost of nuclear power. Heck, even two orders of magnitude still amounts to ~1% increase in net cost per KWh.

RE your edit after I commented:

> Uranium is a small fraction of nuclear's cost at current uranium prices. But eventually that runs out, and the price of uranium would increase dramatically. This has always been the motivation for breeders.

Again, the cost of raw uranium extraction amounts to $0.0015/kWh in nuclear generation. A 100x cost increase will not amount to even a quarter of a cent per KWh. This is the power of fissile energy density: it's so energy dense that the cost of extraction is largely decoupled from the net cost of nuclear power.

By comparison, how would the price of lithium ion batteries be affected if the price of lithium carbonate increases by 100x? Half of a battery's cost comes from the cost of cathode material: https://www.visualcapitalist.com/breaking-down-the-cost-of-a...

If $0.0015/kWh is the current fuel cost, a 100x increase would bring it up to $0.15/kWh. That's 15 cents per kilowatt hour, not a fraction of a cent. It would make nuclear power among the most expensive of electricity sources from fuel cost alone.

If lithium increased in price by 100x, we'd switch over to one of the many other options for energy storage. If uranium increases in price by 100x, burner reactors are screwed (well, even more screwed than they already are.)

"many other options" like what? You can't just say we'd use alternatives and then neglect to specify what those alternatives are. All storage options available to us fall short. Hydropower is geographically limited. Batteries are in too short supply, and are mostly being directed to other applications. Plans for a mostly intermittent grid invariably call for hydrogen, compressed air, giant flywheels, or something else to solve the storage problem. We have no practical experience building electric storage with these systems, so it's effectively a giant hand-wave.

A plan that's dependent on something like hydrogen electric storage is like a plan calling for widespread deep-drilled geothermal power: We have plenty of experience with drilling, and steam turbines. Iceland has plenty of geothermal power - but it sits right on a fault line. That's no guarantee we'll actually be able to build geographically-independent geothermal power. Would you view a plan that involves widespread installation of geothermal power as feasible?

There are many chemistries for batteries. We're even seeing some of them pushed to commercialization. Chemistries based on common elements like sodium or iron would evade concerns about material availability.

There are thermal storage technologies. An example is pumped thermal storage. This involves (1) adiabatically compressing argon, (2) transferring heat from the compressed argon to a hot store (say, molten "solar salt", a potassium/sodium nitrate salt mix) by a countercurrent heat exchanger, (3) expanding the cooled argon back to the initial pressure, (4) using that now cold argon to extract heat from a "cold store", say liquid hexane, cooling it to -100 C. To discharge, reverse this process. Round trip efficiencies similar to pumped hydro could be achieved. The high temperature side of this process is within the creep range of ordinary steel, so no exotic materials are required.

Resistively heated thermal stores would not be quite as efficient (maybe in the low 50s%) and involve higher temperature (~1200 C), but could work with existing gas turbines. Babcock and Wilcox are commercializing this now, using their very nifty direct contact sand/gas fluidized bed heat exchanger. The storage medium here would be ordinary sand, of which there is an unlimited supply.

This last approach also allows an external heat source, such as hydrogen combustion, to act as a backup heat source. So if your thermal stores run out, you can keep running them by burning hydrogen (or some other e-fuel). The marginal capital cost of this capability would be very low, just that of adding a fluidized bed hydrogen combustor to heat the sand.

Thermal storage has only been used for district heating. There is no commercial electric thermal storage project in existence. Babcock and Wilcox have not broken ground on a prototype thermal electric storage plant, let alone a commercial one. They signed an intellectual property agreement [1], this is not even remotely the same thing as commercialization.

Hydrogen electric storage has issues producing hydrogen without emitting fossil fuels: almost all hydrogen produced today is through steam reformation which emits carbon dioxide. Electrolysis has issues with corroding electrodes, in particular. We've known about electrolysis for decades (centuries?) but its disadvantages have not been solved. Likewise, how long have sodium and iron batteries been on the verge of commercialization? How long did lithium ion batteries take to reach the scale sufficient for EVs? Sources say that they're projection sodium ion batteries to be produced at 20 GWh per year by 2030 [2]. Even if that level of optimism pans out, this is nowhere near a scale sufficient for grid storage.

People still hope for lithium ion batteries to deliver, because it's the best (or least-bad) option and none of the competitors are set to unseat it. And remember, almost all of this battery production is going to EVs and electronics, only a fraction of it is going to grid storage.

1. https://www.babcock.com/home/about/corporate/news/babcock-wi...

2. https://cen.acs.org/business/inorganic-chemicals/Sodium-come...

Peak summer demand from A/C can be shifted by making ice, or even by precooling building structure during the low part of the duck curve. All that's needed is to expose consumers to the actual price signals.

Once cheap intermittent sources have in the moment saturated the market, nuclear cannot charge what it needs to pay off the very large capital cost. Nuclear is then dependent on making that money during brief spikes, but storage and demand dispatch will be flattening those spikes. Nuclear will have nowhere to hide.

Once cheap intermittent sources have saturated the market during peak production, further construction of intermittent sources will do nothing to displace the other 50% of demand that will continue to be fulfilled by fossil fuels. And then we either keep burning fossil fuels and wait until a storage breakthrough happens, or we start building non-intermittent sources of decarbonized energy.

Or we just raise taxes on the fossil fuels until they stop being burned. This will have to be done anyway (along with CO2 tariffs to isolate countries that refuse to do so.)

After that, storage gets rolled out. Nuclear never makes sense, and eventually the existing NPPs go away.

We raise taxes on fossil fuels, and then what? More investment in intermittent sources works until peak production is saturated. But now thanks to fuel taxes, lithium becomes a lot more expensive to mine since the heavy machinery is more expensive to fuel. Taxes on fossil fuels would reduce emissions, mostly by contracting the economy. It doesn't translate into more storage build out, and it certainly doesn't solve the orders-of-magnitude mismatch between the amount of storage available and the amount of storage required to smooth out intermittent sources.

Raw lithium is a small fraction of the cost of battery, using recent costs your looking at ~4% of total battery price from low purity lithium. People focus on it because of the name rather than nickel-cobalt-manganese or whatever the specific chemistry actually used. They also confuse mining costs with manufacturing costs when looking at ultra pure lithium ready for use in batteries but that’s like conflating the cost of 99.999% pure monocrystalline silicon with the cost of silicon in glass.

Batteries don’t need much Lithium and the stuff is more abundant than lead, tin, iodine, mercury, etc.

Cathode material is half the cost of a lithium battery. Manufacturing is only 24%.

https://www.visualcapitalist.com/breaking-down-the-cost-of-a...

“The cathode material determines the capacity and power of a battery, typically composed of lithium and other battery metals.”

Also, the cost of a battery cell isn’t the cost of a battery pack, let alone the cost of all the associated equipment, land, installation, etc needed for grid scale batteries which is what matters here.

Wind + solar + hydro can go well past 50% without any batteries. Batteries are getting ever less expensive and production just keeps ramping up even faster.

We really don’t need any major breakthroughs beyond economies of scale to see sub 7c/kWh rates 24/7/365. People are already deploying 50% battery backed solar because it’s cost competitive vs both base load and peaking power.

Hydroelectricity alone can produce 100% of some countries electricity, but other countries can't produce any [1]. Hydropower is geographically limited, in that you need the right combination of a river and a valley to build a dam. You can't just scale out hydroelectric power.

No country produces the majority of its electricity from wind and solar. Australia produces the 15% and 13% of its electricity from solar and wind respectively. Denmark produces 2.5% and 42% from wind and solar [1].

Unfortunately, battery prices have stopped dropping and started increasing [2].

1. https://en.wikipedia.org/wiki/List_of_countries_by_renewable...

2. https://about.bnef.com/blog/lithium-ion-battery-pack-prices-...

> No country produces the majority of its electricity from wind and solar.

Yet, but infrastructure lasts far longer than the recent dip in prices which favors renewable energy. As to battery costs that article specifically mentions expectations of near term drop in prices, these are long term trends not specific annual guarantees. Solar prices have occasionally increased due to supply shortages even in the middle of vast longer term price drops.

Also, countries electric grids aren’t independent outside of a few islands.

Globally Hydropower is ~16% of global electricity generation and quite flexible. A country with 100% hydropower generation can easily add wind and solar panels and then export flexible and cheap hydroelectric energy. In fact if you look at countries running over 50% of hydroelectric power they tend to export quite a bit.

The good news is that after it is paid for, it will cost about $20-$25/MWh. The bad news is that at least for the few decades it will cost $75, which is about the same as solar+battery LCOE today.

Those numbers are wildly optimistic, these overruns mean the plant is going to be repaying construction costs over it’s full lifespan.

I don’t know where you messed up but if you’re running the numbers include the full cost not just the overrun, interest rates, increasing operating costs during loan repayment, capacity factor, and decommissioning costs. Opening costs late in life is a real killer as older power plants mean ever more equipment needs to be replaced which then lowers capacity factors etc.

Also, LCOE battery backed solar is also under 60$/MWh https://atb.nrel.gov/electricity/2022/utility-scale_pv-plus-... Numbers may be slightly higher in Georgia, but still well below your estimate.

> 4 cents per kWh.

For comparison: I live in SF Bay Area and pay 42 cents per kWh.

For comparison: I pay 6.269¢ per kWh for energy costs.

For comparison I pay 6 cents per kWh in Quebec (and that’s Canadian money).

Just to make sure no one misunderstands (not saying you did, just that this is easy to misread):

The 4 cent/kWh figure is the additional cost it would have to charge, over the life of the plant, to make up for the budget overage, and would go on top of whatever ratepayers would have to pay for its "regular" energy output.

Further pedantry: you'd have to factor in time value, so that kind of understates it, as the entity that built it would have had to borrow that much more to finish, and pay back with interest.

Thanks. I can now see how my comment could be misunderstood.

My point is that while 4 cents per kWh for 50 years is a very big number in terms of cost overruns and the taxpayers in Georgia will have to eat it, I as a California resident, somehow pay through the nose despite abundant and cheap solar energy, especially during the daytime.

Nuclear power plants now run more than 50 years. We are pushing some to 80 years already.

https://www.energy.gov/ne/articles/whats-lifespan-nuclear-re...

A large fraction of all nuclear power plants decommissioned recently didn’t make 50 years.

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

We have our own nuclear albatrosses that we are overpaying for in California.

We are getting ripped off.

To be fair it's more like 12 cents per kWh for actual electricity and the other 30 cents for delivery.

> To be fair it's more like 12 cents per kWh for actual electricity and the other 30 cents for delivery.

It seems that in Quebec (see the other reply with 6 cents per kWh all included), moving electrons through a copper wire is much-much cheaper. Something is very wrong with the PG&E costs.

>“seven years late and $17 billion over budget. At its full output of 1,100 megawatts of electricity”

It's all projects. This just happens to be a big one.

Everyone likes to point to over-runs, over-budget, late projects.

Nobody goes back and asks "who did the original estimate that we eventually went over?".

Many projects start with /s 'everyone knowing it will go over budget, but if we give a real estimate it wont get off the ground'/s.

Then later someone has to get blamed. If everyone is (quietly)honest they kind of spread around the blame and the success. If it is a contentious project, everyone is playing tag at the end.

Every other kind of power project is built using fixed price bids. If a company bids $3B for a natural gas plant or solar plant and the project costs more than $3B to build the company is on the hook, not the ratepayer.

I didn't dig into details on specific of this project.

But any large fixed price project I've been on, it all hinged on verbose specifications. And then any change resulted in change orders. And it becomes a pissing match of who will pay. It is rarely cut and dry.

They are power contracts, not construction contracts. The power company isn't buying a power plant, they're buying power. That's what makes the contract so much simpler.

I'll have to pass. I don't know the power industry that well.

My rule of thumb thought, is that nobody looses 17 billion. That gets passed to the consumer somewhere somehow. In someway obviously or not obviously.

On the news page, there is a link til an article on India purchasing 1,200 MW facility from China for 3,5 billions. Edit: https://apnews.com/article/pakistan-china-nuclear-power-plan...

If you ever doubt that the US has a simply fucking insane advantage in primary resource extraction and consumption, consider your comment. You think $40 per MWh is expensive.

Wholesale rates in Europe hit SIX HUNDRED AND SEVENTY FIVE EUROS PER MWh last year because of gas supply issues and unfavorable weather. Spot prices have gone higher still, over one order of magnitude higher actually, those were futures contracts for a useful period of time. Because THAT'S WHAT PEOPLE WILL PAY WHEN THERE IS NO ALTERNATIVE.

Consumer rates of 30 cents per kWh are perfectly normal. 100 not unheard of.

Oh, fun fact; the largest producer of nuclear power in Europe is suing its government because it was forbidden from selling at that market rate. It had to sell at 40 cents per kWh. Not to consumers of course, to the fucking glorious private sector, aka resellers, who did sell it to consumers at market rate. The ones who hadn't gone bankrupt and fucked off earlier when market conditions were against them, that is. Although they did spend a lot arguing, successfully, they didn't need to pay producers then, either. Because the glorious efficient private sector can't have competition.

Yes, I'm bitter. Going to an industry conference and seeing no one able to run plants properly because of unreliable power, whilst neighouring Germany sets a new coal-burning record, in unnatural heat, does that.

Texas spot prices recently peaked at 9,000$/MWh briefly that’s in no way what I am talking about.

4c/KWh is the minimum additional cost across 50 years ignoring interest not the total cost of this power. In other words if inflation adjusted electricity would have been X$/kWh in 2070 it’s now at least X + 4c / kWh whatever the baseline would have been. Europes average electricity prices for the year aren’t that far above average, across 50 years it’s a tiny blip by comparison.

Tell me more about that coal burning record in Germany.

> Oh, fun fact; the largest producer of nuclear power in Europe is suing its government because it was forbidden from selling at that market rate. It had to sell at 40 cents per kWh. Not to consumers of course, to the fucking glorious private sector, aka resellers, who did sell it to consumers at market rate.

don't you all love the European Union's Single Electricity Market?

Germany's coal use is at an all time low. If you're really going to conferences, you need to be better informed. I've posted the data many times here, but if you can post so many words without bothering to do a brief check for easy to find data, then I'll spare myself the time.

Your own math points out that this is a cost effective technology. 4 cents per kWh is quite cheap electricity.

That's an additional flat charge, _in addition to_ the cost it was last month.

Even then, that's not too bad. In California, electricity is closer to 30-40 cents per kWh. They're nowhere near hitting anything that could be called "high cost."

2 units, both 1,100 MW each. So 2 cents / kWh.

The second unit isn’t online and the total overrun is more than 17 billion.

Still cheaper than commercial solar or wind.

If by cheaper you mean 5x more expensive…

That 17 billion isn’t total construction cost it’s how far they went over budget so far. They still need to pay the initial estimate + ongoing operating costs + decommissioning costs.

Awesome! Nuclear is a fantastic and needed addition to the grid. The cost was high, but still quite cheap relative to the cost of continuing to build & run fossil fuel plants. It's a real shame we let 30+ years slip by, letting our ability to build projects like this wither. I hope there's lessons learned from this experience that will help get costs down for future plants. This should spur discussions on streamlining the regulatory side of things, and there's a lot of exciting stuff going on in modular reactors.

(In case it needs saying, which it shouldn't: yes, we should be building out wind & solar, too! We need all hands on deck, wind, solar, and nuclear, right now, to kill coal. We already blew our chance to do it cheaply, so now we have to pay the price.)

Mostly agree... though I think Solar tech still needs to improve a bit. Nuclear power is definitely needed for the grid, though preferably more inland in places less likely to be affected by natural disaster. I get why the NY plant was closed.

I also hope that the build and cost timelines can be shortened. I think the push for electric cars is a bit of a miss in this, only because the grid needs to improve dramatically before such efforts can be effective and are barely keeping pace with current demands.

EV’s have a fairly trivial impact on the grid. Cars last 25+ years so even if everyone all new cars where EV’s we are talking less than 1% increase in total electricity demand per year. The grid has expanded dramatically faster at several points.

Even more importantly we can shift most of that demand to cheaper times of the day or even day of the week.

> Cars last 25+ years

This is not a truth universally acknowledged.

The average age of cars on the road in the US is 12 years. There aren't many 100 year old cars skewing the average, so that implies that most cars last over 24 years.

I'm not sure about most cars lasting 25+ years... that said, even a 25-30x increase in EV usage in the next 15-20 years is still a massive increase. And it's not even the total load, it's load to the houses, the charging stations, etc. A lot of homes are regularly facing brownouts and rolling blackouts in the summers as-is.

Average car in the US is 12.2 years old. To offset cars purchased yesterday someone needs to be driving a 24.4 year old car to average 12.2 years.

As to electric demand, “The average miles driven per year is 13,489 in 2021.” Efficiency varies, at say 4 miles per kWh that’s 3,372 kW per year, we’re talking roughly 0.38 kW bump that to 0.4kW per car * all 762,883 EV’s sold in 2022 and your up to 305 MW nationwide. Meanwhile this single reactor is putting out 1,100 MW so ~3.5 years of EV sales at the current rate.

Critically unlike watching TV people have a great deal of flexibility when can charging EV’s and therefore mostly use cheap energy Aka when other demand is low therefore minimizing risks of brownouts.

> Average car in the US is 12.2 years old. To offset cars purchased yesterday someone needs to be driving a 24.4 year old car to average 12.2 years.

12.2 was a year ago [0], 12.5 now [1], but you could also get to that with 9 at 13.9 years for each one today, instead of one at 25.

[0] https://www.spglobal.com/mobility/en/research-analysis/avera...

[1] https://fortune.com/2023/05/15/average-age-of-cars-on-u-s-ro...

The rate of new car production is fairly consistent. The only way to get 9 at 13.9 years is for 13.9 years ago cars to have been produced 9x as fast, or to randomly pairing things off which doesn’t tell us anything.

If we pair the newest with oldest car and work to the middle things don’t quite work due to antiques, the pandemic, and increasing production, but it’s close enough for 25 years to be a safe estimate.

> The rate of new car production is fairly consistent. The only way to get 9 at 13.9 years is for 13.9 years ago cars to have been produced 9x as fast.

Its not at all consistent, and the reason the average age has gone up is supply chain issues driving costs up and sales down for new cars.

I didn’t say constant, just reasonably consistent. Individual years don’t matter much here it’s the overall trend we care about and the trend shows nothing close to an 9x increase and decrease:

https://fred.stlouisfed.org/series/TOTALSA (Note the graph minimum is 7.5 not 0.)

Ehh Indian Point (the NY plant) wasn't really a concern with regards to natural disasters. We get some occasional (very minor) earthquakes and a hurricane once in a decade or so, but I doubt the latter would do much vs many tons of reinforced concrete.

The plant definitely had issues -- some due to age (construction started in 1956!), some due to mismanagement, and some due to dumb regulations [0]. My chem class went there on a field trip in high school, and the guy giving the tour definitely gave off "engineer who has been overruled by management in very dumb ways" vibes while explaining that their waste silos were almost full because they weren't allowed to transport the spent rods across state lines so that they could be recycled back into fissile materials.

As an aside, the control rooms had a real "retro-futurism" look. Lots of manual dials and brightly-colored plastics [1]. Gene Roddenberry eat your heart out, etc.

[0]: https://en.wikipedia.org/wiki/Indian_Point_Energy_Center#Saf...

[1]: https://www.lohud.com/story/opinion/2016/02/12/indian-point-...

I think you have that reversed, solar is great, and nuclear needs to improve.

Nuclear is rather old tech, and the more we learn about it the more expensive it gets. It's not a good fit for modern economies, because construction is so labor intensive and difficult to automate. Plus, the industry itself is a managerial failure and can't plan or execute on those plans effectively.

Storage is a great complement to the grid, because it lessens the need for transmission. And our original "storage" on the grid, hydro, was put there to deal with the shortcomings of nuclear, if you listen to Jigar Shah (though he is far far more bullish than me on the ability of the nuclear industry to deliver in the future).

Nuclear is not a needed addition to the grid, unless you have a compulsion to waste money.

About time!

> "This hadn’t been done in this country from start to finish in some 30-plus years," Chris Womack, CEO of Atlanta-based Southern Co. said Monday in a telephone interview.

IIRC, scientists are working on Gen 4 reactors, and there are a number of Gen 3 reactors operating in commercial capacities around the world; but the US is still stuck on Gen 2 due to regulation.

That's a major oversimplification. The reason the US is mostly (entirely) running Gen 2 reactors is because we simply lost interest in building new reactors for a long time. There were regulatory hurdles that caused this, but there were tons and tons of other factors that were (IMO) more important. Especially economic factors related to the cost and mismanagement of large nuclear projects, public opinions shifting over nuclear power, and alternatives like natural gas being super cheap.

The NRC has been approving Gen 3 designs for a while now but nobody wanted to follow through on building them.

Are costs and mismanagement directly related to regulations? Genuine question. This is impression I get from nuclear advocates like Mark Nelson and doomberg.

Yeah that's the complicated part. The regs are responsible for some of the mess, but far from all. It's a complicated subject, probably too much to get into here but you can look at the history of Vogtle as a start. Blaming the NRC is justified at parts, but it really only tells part of the story.

This project is what killed nuclear in the US. Not regulations, not the NRC, just plain old incompetence, bad planning, bad EPC, bad design.

The AP1000 was supposed to be a "modular" design, where most of the difficult welding could be done off site and delivered complete, with paperwork. And failure in this modularity is what caused the project to be such a flop, and essentially kill all new large nuclear in the US (people are trying "small" modular, but their target costs are still far too high to be economically feasible).

See, for example, this 2017 report on just one aspect of what went wrong:

https://www.enr.com/articles/43325-witness-to-the-origins-of...

> To build the first new nuclear reactors in the U.S. in three decades—South Carolina’s V.C. Summer Units 2 and 3 and Georgia’s Plant Vogtle Units 3 and 4—the design and construction team would face a steep learning curve. However, says Hartz, learning wasn’t much of a priority in the rush to start work at Lake Charles. “They were clueless” about the complex geometry of nuclear welds, the nuclear supply chain and the need for a nuclear safety culture, he notes, adding, “I wasn’t a whistle-blower. I was just a senior procurement manager who was concerned.”

> Westinghouse would issue drawings to Shaw Nuclear in Charlotte. When Shaw reviewed the drawings and asked Westinghouse to correct a detail, problems ensued. The work processes were unnecessarily complicated by the separation of the team members. Giving an example of how the process got out of hand, Hartz says that, if a design called for a 3⁄8-in.-wide, 12-in.-long fillet weld, the welder might make it 14 in. long. “Instead of having Westinghouse right there saying, ‘That’s no problem,’ ” recalls Hartz, “we had to write a nonconformance report that was processed and reviewed by Shaw and then sent to Westinghouse for disposition. It was insane. From Lake Charles to Pittsburgh to Charlotte then back to Shaw Modular before the red nonconformance tag could be taken off, saying it’s OK now.” He adds, “Each change went through the same tortuous path, taking months and months.”

In regards to the story about the weld length, that is the process working as intended. As a constructor you don't just decide that something can be a different dimension. And as a engineer you don't just stand "right there" saying "That's no problem". That is how people get killed.

I 100% agree that such changes need to be checked, but that's not the point of the anecdote. The point is that this process took months rather than being streamlined.

It is a management failure to make common necessary things, like signing off on small changes, a month long process.

This sort of bad management is endemic throughout the entire build.

Which is to say, the problem isn't the regulation requiring that things are build as designed, the problem is the management structure that makes such changes so uneconomical, combined with the frequency of such changes due to miscommunication between designers and builders.

That reactor could last until 2100.

https://www.energy.gov/ne/articles/whats-lifespan-nuclear-re...

I live in Georgia, and the only thing this plant will do is raise our rates.

It's late (of course) and HUGELY over budget (also of course). The planners always knew it would be way late and way over budget. None of that mattered, because the 'regulatory' system guarantees they get it all back (and more) from the rate payers.

It was also painfully obvious WHEN THIS WAS PLANNED that building Solar and/or Wind would get done sooner and far less expensively.

So to power every home in the USA would roughly cost 3 trillion at these prices. if they reused the same teams and designs maybe they could get it somewhat cheaper. plus we have increasing renewable plus existing hydro and nuclear. for the price of a year or two of national debt we could go full renewable in this country.

So we could decarbonize the entire US power grid for what we spent on COVID relief/stimulus?

That seems like a good deal, even without economies of scale...

Not just decarbonize, but become a major fossil fuel exporter and be all smug about like a certain Nordic country…

> become a major fossil fuel exporter

Wouldn't doing that undo a lot of the gains from this? If we export fossil fuels instead of burning them, we aren't really decarbonizing. We're just having the carbonization take place somewhere else.

It offsets significant carbon production elsewhere (especially if we only export natgas), and defunds from some of the worst governments on earth.

> It offsets significant carbon production elsewhere

Does it? Or does it simply increase the amount of fossil fuels being used elsewhere?

> defunds from some of the worst governments on earth.

That has nothing to do with decarbonization, though.

you are right, oil extraction is literally producing carbon (which is then processed and burned at later stage somewhere else)

so we need to separate: carbon production and carbon emissions.

a lot of so called decarbonized industries are decarbonized simply because they outsourced carbon intensive parts to third world, but carbon is still being produced and emitted.

I think this is why some people are supporting De-Growth movement, that just by reducing consumption of nonessential stuff, we would decrease carbon production and emissions globally

Yup. The Georgia plant seems expensive until you remember it cost half of what Twitter did.

$30B is more than half of $44B.

"Cheaper than twitter" is a canonical example of damning with faint praise.

What if climage change is a big hoax and we create a better world for nothing?[1]

[1]: 2009 - https://upload.wikimedia.org/wikipedia/en/1/1e/What_if_it%27...

The alternative isn't carbon emission, the alternative is far cheaper wind, solar, hydro, geothermal, battery storage (both lithium and many others)

The only question is if nuclear can offer something compelling to keep it in the mix. We have roughly 100 1GW of nuclear reactors in the US that are reaching end of life, and 90% of those communities would like to see replacements to keep the jobs in town. However, nobody will ever invest in a project that looks like Vogtle, due to the cost. Unfortunately, recent projects in France, Finland, and the UK all look quite similar in cost of dollars and time.

Which is to say, that nuclear will not be able to provide a climate solution at all. It's too late, we can not even build 100 rectors in the next 15 years to replace what needs to be retired!

If nuclear advocates want more nuclear, they need to focus on the details and the engineering and the management, not merely rah-rah for the tech. The devil is in the details, and projects like Vogtle have virtually guaranteed that nuclear will not be built in the US in quantity for at least a generation.

The only reason Vogtle was built was because of the fixed price contract Westinghouse signed. Just as in the first nuclear buildout in the 1960s, fixed price contracts revealed themselves to be ruinous for the NPP builder.

No utility going forward is going to sign a contract that puts them on the hook for cost escalation (which turned out to be this contract too, once Westinghouse went bankrupt.)

And EDF learned the hard way at Olkiluoto about saying they can build something at a specified price.

The only way for nuclear or be built in advanced economies is to have some government provide subsidies far beyond what renewables have ever reviewed. And at least with renewables there was the goal of making it cheaper for the next round of production. We don't that get that benefit when we throw an extra $10B at a nuclear build.

> 3 trillion

The US federal budget deficit was 2.8 trillion in 2021. One year. I can't remember any US leader of any party even mentioning it.

> seven years late and $17 billion over budget.

When does this become a national security issue? Zoinks!

Good. More, please.

What this article is missing is that if you don’t consider refueling, defouling, valve refitting, coolant issues, cooling tower furloughs, transmission plant failures, early retirement, funding issues, staffing issues, or pressure vessel refurbishment the uptime is 100%! Beat that, renewables.

> ...seven years late and $17 billion over budget...

And how much of that was a result of red tape from the NRC, DOE, and EPA?

I don’t know, but new nuclear projects in the West tend to be extremely late and over budget even with friendly regulators.

Olkiluoto 3 in Finland is a Gen 3 reactor that just came online earlier this year. Its original target date was 2010. The original budget was 3 billion euros, but the final cost was over 11 billion.

The project was of great national importance because this single unit provides around 14% of power to the country, and the Finnish nuclear power regulator was extremely motivated to make it happen. So the delays and cost overruns were not due to red tape.

It's easy to find detailed retrospectives of the multitude of failures here.

I haven't found a single one that said excess regulation was a problem, but I have found a huge number that showed project management, bad design, bad communication between engineers and EPC, etc. were all to blame.

Here's one I was reading recently from 2017, about the welding issues. Every other aspect, such as concrete, exhibited similar failures.

https://www.enr.com/articles/43325-witness-to-the-origins-of...

But if you have an idea of which regulations to change, or how to fix project management, you can pick up a half-completed pair of reactors in South Carolina on the cheap. That boondoggle often gets forgotten when examining Vogtle.

My 2c:

Why nuclear power is a dumb idea.

Countries with nuclear reactors: 32.

Countries which have had nuclear leaks or meltdowns: 15.

Number of nuclear leaks and meltdowns since 1952 (only those which resulted in loss of human life or >US$50K property damage): ~100.

About 60% of those have been in the USA, allegedly the most advanced country in the world with the bestest regulation of such things.

Note that the USA requirements for nuclear reactor waste (yes, they produce toxic waste; they are not clean), last time I checked, required the canisters to be able to survive for 300 years. The waste lasts longer than 300 years. So all you can do with the waste, at best, is leave it for someone else to handle later.

Two years ago the USA had a leak which spilled ~400,000 gallons of radioactive water into a major river system, and it was covered up for two years. You can not trust governments or nuclear power companies about this stuff.

The entire ecosystem is getting poisoned by all that waste water Japan is dumping. Almost 50% of countries with nuclear reactors have had significant leaks and meltdowns, and it only takes one significant event to screw up the entire natural environment.

Finally: If you are not willing to have a nuclear reactor right beside your house, but are willing to have one beside someone else's house, you are a coward and are not really in favour of nuclear power.

The 300 years comes from the length of time the waste remains self-protecting against amateur diversion of plutonium. After that time, the fission products have decayed so much that the gamma rays are, conservatively, too weak to inhibit such diversion.

It's not that waste canisters can't last more than 300 years, it's that something has to be done with the waste by that time so there's no point requiring they be certified to last longer than that.

The 300 years came from a specification I read, produced by the US government, for the design and construction of the canisters used to contain the waste. This was many years ago, and may have been updated.

Reply on news.ycombinator.com