Hacker News Reader

Comment by ogre_codes | Hacker News Reader

The anti-nuke environmentalists spend massive amounts of time and energy building public distrust of nuclear. They managed to shut down existing nuclear reactors with safe track records. This makes me pretty skeptical we'll be able to get new ones on line regardless of how much safer they are.

Getting the NRC to approve them is one thing, getting a city to allow a nuclear power plant of any kind in their back yard is a whole other ball-game.

View on HN

The PR battle has nothing to do with the lack of investment. That all comes down to well-supported reactors not even being buildable.

Even EDF in France, which has massive support, can't build nuclear these days, and France is often cited as nuclear's greatest success stories.

It's time to stop blaming boogie men for nuclear's failures, because that does nothing to help it succeed. Anti-nuke "greens" blame does allow for building a culture war, which is why I think it survives in the public imagination. However, in utility boardrooms, and at the NRC, none of that has stopped the world's attempts at building the next generation nuclear fleet. What has stopped that is the cost of construction.

It doesn't matter whether there is outsider opposition to the tech if it can't even be built economically.

The West has not been able to build any sort of infrastructure for a generation [1]. It's not that we can't build nuclear power plants on time and budget, it's that we can't build anything larger than a sports stadium period and we can only get over that by drowning it in money and hoping for the best.

Green energy is great in this regard because solar panels are just roofing and wind turbines are just really skinny 5 story buildings.

Hell you're on a website that exists only because no one is willing to do the type of R&D that build the tools we use in the first place (unix, arpanet, computers) so we instead outsourced it to people in garages and hope for something that works, or at least looks like it works and makes a ton of money before it crashes and burns (hi twitter!).

https://www.spglobal.com/en/research-insights/featured/infra...

I guess the next question is: Why is the construction and operation so expensive? Did the cost of steel and concrete skyrocket? Are new reactors excessively complex? Is there insufficient competition in the industry to keep costs under control? Excessive regulatory compliance costs compared to other forms of power generation? Is it the cost of paying a team lawyers for decades to keep fighting all of the lawsuits?

I believe the cost of construction labor has outperformed inflation & materials have tracked inflation. But the cost of energy has underperformed inflation.

If the trend continues for much longer, energy sources which require less labor will become more competitive (solar).

If you're a greenie - this is a good thing. Because oil has almost nothing going for it beside how good a source of energy it is for jet/rocket fuel.

I am 99% sur that solar requires more manpower than nuclear, for an equivalent controllable power output.

The Desert Sunlight solar farm generated 1,344,841 MWh in 2018 [1] while employing 15 people full time [2]. That is an annualized output of 10.2 megawatts per full time employee.

The Diablo Canyon nuclear power plant generated 18,213,519 MWh in 2018 with 1400 full time employees [3]. That is an annualized output of of 1.5 megawatts per FTE.

During operation, the solar farm generates more than 6 times as much energy per FTE.

Of course you need storage too for solar to supply power after sundown. I looked at some recent utility scale battery project news, like the Moss Landing project currently under construction, but I couldn't find information about these projects adding any number of permanent jobs after construction. They may be too low maintenance.

Finally, you need to amortize the project construction labor over the life of the project to get an all-in labor intensity analysis. Diablo Canyon was unfortunately built long enough ago that contemporary reporting was pre-digital. From recent nuclear projects in the United States, it looks like nuclear projects don't have lower lifetime labor intensity (inclusive of construction) than solar + storage either.

The low labor intensity of large scale renewables is an under-appreciated point. Most of the labor is front-loaded into the construction phase, and doesn't amount to a lot per lifetime MWh. When you see breathless reports about how solar power is employing a boatload of people, they're talking about short term construction work. If solar power were at steady-state, as mature as coal or nuclear power, it wouldn't provide many jobs. The only reason its job numbers are so much higher than coal/nuclear is because coal/nuclear are stagnant/declining while solar is still in the rapid growth/construction phase.

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

[2] https://www.energy.gov/lpo/desert-sunlight

[3] https://en.wikipedia.org/wiki/Diablo_Canyon_Power_Plant

Yeah, but it's more granular and you can deregulate it and still be safe, so you can leave it to the market to handle pretty easily.

I read a study a while ago arguing that reactor cost was as expensive as it is because nuclear plants are so large, complex, and infrequently built that each project is essentially bespoke.

Was that study by chance suggesting a large number of small mass produced reactors?

It's a workable idea unless you think the problem is that every reactor, no matter how large, has to deal with red tape and lawsuits from NIMBYs.

Fossil fueled power plants are cheap, quick to build, and has plenty of private companies that compete for the contracts. The cost per MWh has almost been cut in half for natural gas in the last 10 years, while nuclear is struggling to reduce costs (if at all).

The cheapest possible energy gird currently available is to use wind as much as the weather allows it since it has currently the lowest cost per produced MWh, and natural gas for everything else. Solar is also starting to reach similar cost per MWh as natural gas when weather allows it.

We might see a renewal in nuclear if natural gas (and other fossil fuels) were taxed to the point of being economical unattractive. Without that I don't expect something that cost twice as much to compete in a market with a fungible commodity.

I think this overlooks the ways that opposition can increase costs of building nuclear and the extreme safety burden imposed on nuclear plants but not other energy plants & sources

> The strict safety regulations imposed on the nuclear industry are another reason why capital costs are so high. [1]

This is actually covered a bit in a recent 80,000 hours episode: https://80000hours.org/podcast/episodes/mark-lynas-climate-c...

[1] https://thebulletin.org/2019/06/why-nuclear-power-plants-cos...

If safety costs were not high, then suddenly the boogie men might be correct about the dangers?

The fatality rates of other source of electricity are much higher, and they seem to be accepted:

* https://en.wikipedia.org/wiki/Energy_accidents

* https://ourworldindata.org/safest-sources-of-energy

So why is there an insistence on potentially over-engineering nuclear? Why the insistence on lower rates of nuclear compared to other type of generation?

If we don't want to reduce nuclear safety—which is completely reasonable—then to be consistent, other forms of generation should be made safer—which may drive their costs up.

Pick a rate that is acceptable, and have design goals around that. Either by making things less onerous for nuclear or more onerous for others.

There are all sorts of gas generators coming online (because of fracking), and while they may be better than coal, but that's hardly a useful way to eliminate emissions. Pushker Kharecha and James Hansen:

> Our findings also have important implications for large-scale "fuel switching" to natural gas from coal or from nuclear. Although natural gas burning emits less fatal pollutants and GHGs than coal burning, it is far deadlier than nuclear power, causing about 40 times more deaths per unit electric energy produced (ref. 2).

* https://www.giss.nasa.gov/research/briefs/kharecha_02/

The right balance would be to let nuclear meet the same safety standards as other power sources, instead of far higher ones.

The way I see it is you can measure how economical these things are in a country where they don't listen to the people. So I look to China and see if they think it's worth it. And Chinese nuclear power is spiking. They're building loads of reactors and loads of renewables. It looks like their renewables growth is more than the nuke growth, but they're doing the nuke thing too, which makes it seem like the economics aren't the thing.

As I see it China is just sampling a few of everything, and then observes how it performs. Can be seen by their massive building of high speed rail, which employs (almost?) every type of high speed train type from all over the world. The same seems true for their nuclear plan. AFAIK they have or build at least one of everything that's available.

Gaining knowledge that way.

And then, after maybe 20 to 30 years producing the best combination of all that, with the gained knowledge.

That is what really is happening at the moment. The real results will be visible in decaces only.

One big field test.

That's an interesting idea, for sure.

> So I look to China and see if they think it's worth it. And Chinese nuclear power is spiking.

Spiking? They approved two reactors this year and most years since Fukushima they didn't approve any. https://www.scmp.com/economy/china-economy/article/3100097/c...

...after the 2011 Fukushima nuclear plant disaster in Japan, China halted its approval of new reactors for four years. These resumed in 2015, largely driven by Beijing’s pledge to cut carbon emissions. Then, from 2016 to 2018, the country again stopped granting permission for the construction of new nuclear power plants. After three new plants were approved in June 2019, government approvals again ground to a halt until this week.

You don't need that many approvals a year. They're doing just fine. Check out the construction pipeline. They have 10.6 GW under construction right now. That's a 25% increase when it's built.

Tianwan-5 was just hooked up to the grid last month.

For comparison, China added ~30GW of solar in 2019. To make the GW of solar more comparable, one must adjust for capacity factors, which round to 1 for nuclear but is ~0.25 for new solar.

So China added the equivalent of 7 nuclear reactors worth of solar TWh last year.

They are also great at adding HVDC, which the US desperately needs.

Sheer incompetence has much more to do with our inability to build nuclear than the anti-nuke greens or onerous regulations do. Read about the comedy of errors that was Westinghouse's plans to build reactors in the US and their descent into bankruptcy.

“The cause of this latest problem seems to have nothing to do with nuclear power and everything to do with incompetent business practices, particularly Toshiba’s construction contractor,” Conca wrote.

https://www.triplepundit.com/story/2017/westinghouse-bankrup...

They can absolutely be built economically, but they are hobbled by excessive regulation due to public fear.

This argument is akin to when congress accuses the US Postal Service of losing money and being uneconomic, when in reality congress itself mandated that it must be run that way.

I'm not sure that I've ever seen somebody in the field suggest a reform that would reduce that cost, except for the small modular style construction.

That is to say, nobody involved with the process has a suggestion on what regulations are unnecessarily increasing costs. So I don't think the USPS is comparable, as there are very clear and easy fixes that would solve their problems (ie increase stamp prices).

When you look at what happened with Westinghouse and the AP-1000 in Georgia and South Carolina, it seems that parts of the problem is that the designs weren't even buildable as they were laid out, and when it came time to build the construction side just went ahead with whatever they could hobble together. So it's been suggested that if the NRC checked the plans not only for safety, but also basic construct ability, then maybe these massive projects would succeed. But because the engineers did not do their job, and construction teams did jobs they shouldn't have, and the business arrangements were such that all this went sideways instead of problems being identified and fixed, there was massive failure.

Personally, I think more regulation has a better chance of fixing the industry rather than less regulation. The executives are clearly incompetent and have no clue or long term planning ability.

>I'm not sure that I've ever seen somebody in the field suggest a reform that would reduce that cost,

You don't wind up in the nuclear field by being dumb and smart people don't say things that can easily be construed by the opposition as advocating for reduced safety.

That doesn't sound like a smart strategy to me, because it leads to the conclusion that every regulation is necessary for safety and noting can be cut.

Meanwhile at the same time lots of people are saying "too much regulation drives up costs," which means that the implication is still that nuclear advocates are asking for less safety fit cost savings.

The cost is upfront for nuclear and that's what doesn't align with society's immediate goals. France is no exception to that (and in fact if you know more about the state of research in France for example you can see that long-term investments aren't valued highly enough either).

That's why despite the knowledge being there (and exported, just check out what french companies do in China) the main obstacle is political and societal: people are still scared, and politicians oscillate constantly on the matter of nuclear energy (Fessenheim, EPR etc) and therefore cannot actually develop a consistent position over the development phase. France actually has a ton of benefits from using nuclear (check out maps pollution-related deaths, check out hybrid car CO2 emissions per km, etc) but it takes a long-term vision to make big commitments that can pay off handsomely, and most of the world is fairly myopic (and my country is no exception).

I could point you towards [1] the perception of nuclear risk (nuclear waste first, nuclear reactors second) in France being much higher than the perception of floods/fire/medical risks/terrorism, and much higher than the real numbers[2]. There is a HUGE disconnect between what people THINK nuclear power plants and nuclear waste are in terms of risk and what actually is problematic, and this oscillation has been very strong in public opinion (after all, it was Hollande 4 years ago who committed to closing a perfectly fine Fessenheim following public opinion more than anything else).

So no, it's not about being buildable or not. Nuclear power plants are buildable. They're just not giving the private sector a good return on investment in today's market, and therefore it's on governments to get moving but for them to enact change their voters need to be consistently wanting it too. It's not even just a nuclear problem: just look at how public transportation can often be a constant fight between parties which delays advancements by decades (see: Toronto).

[1] (in French) perceived moderate and high risks for various items in social awareness: https://i.imgur.com/iPpS6rM.jpg

[2] (in French) actual amounts of deaths related to many of the items of the previous image: https://i.imgur.com/fOwKjka.jpg

There is a false dichotomy being presented here. A piece of infrastructure can be both technically difficult and politically challenging. I think that applies to traditional nuclear power stations. Nuclear engineering is difficult, and that makes it more difficult to build nuclear power plants! And past success and knowledge is not a licence to success in the future.

The person I replied to said they're not buildable. This is logically false because we've been able to build them. There's no dichotomy here, just an invalid argument.

And science isn't just a game of randoms. We know full well how to build nuclear power plants, let's not play mysticism about it..

I'm thinking about "buildable" in terms of how an agent that would construct it thinks: if I spend X amount of money will I have anything to show for it?

In the US, the percentage of build failures has been too high for any utility or merchant generators to stomach the risk.

So the existence of new reactors is not sufficient to define "buildable." The existence frequency of failed builds is what scares off new projects, and what makes them "unbuildable."

Which also points to another difficulty with build ability: even if Rosatom or China or South Korea have figured out a way to build with a high success rate, that doesn't mean that it will translates to a different build crew and culture. We would have to see a string of probable 5 successes in a row in the US before a utility exec would consider risking their entire company on a new nuclear power plant. Too many gigantic companies have been destroyed in the attempt. There's a hilarious video of a SCANA (I think?) exec delivering blisteringly sarcastic remarks about the possibility of somebody else being able to pick up the pieces of their failed AP-1000 build. Nuclear is a career killer and a corporation killer in the US. It's likely that many execs in South Carolina will end up in jail for the stuff they did to try to keep the build going.

And this is from people that have none of the safety or environmental concerns about nuclear. It's purely economic and "can I ever build this thing and will it destroy the company I lead."

So you're thinking "buildable" in terms of short-term market when nuclear power is the kind of thing built to operate on the scale of a century (and really gives financial returns only when you look over many decades), ESPECIALLY because the initial capital invested is the only real cost and most of the energy income from nuclear power plants has to go back towards investors at a typical 10% (if governments don't step up to do it themselves at much lower interests), exactly because the market is expecting returns NOW.

This is a problem of framing you've got there, because by definition pretty much all the actions taken on markets have short-sightedness in mind, and at the very least nowhere near a century-long vision.

The existence of reactors AND the fact that they are being built RIGHT NOW is enough to define buildable. That you don't WANT to build them is an entirely different problem, and again it is motivated by the fact that incorrect information is constantly being thrown at nuclear (even in countries where it once was a major thing to build new reactors, i.e. France) and that this information trickles down into the way people drive their representatives (politicians) towards short-sightedness.

That last paragraph must be a brilliant excuse to use.

  * Boss - So why did the reactor vessel produce cracks?  
  * Engineer - Oh well, long standing misinformation being thrown at the industry.
  * Boss - what has that got to do with my reactor vessel.
  * Engineer - Err well, the craft was lost due to a lack of ongoing projects.
  * Boss - But I though this was just science?
  * Engineer - When I said that "we" know how to do this I was talking about humanity in general.

You can make any problem the fault of politics if you try hard enough. But it is hopelessly reductive.

Engineers don't have a say in whether things get built or not. Politics do, because politics are expressed through representatives of the people and people do.

Yes you can make any problem a problem of politics because that's EXACTLY how things get solved: people legislate, and things are made legal, illegal, decriminalized, mandated, budgets are voted, etc.

I understand that some people can misinterpret the data about nuclear and have a perception of danger, but to say that nuclear can't be built because it is advanced engineering is as absurd as saying that a tower can't be built because it is advanced engineering. You've been proven wrong literally hundreds of times, and you still cling on to the idea that the thing can't be built?!

To say that it can't be built because people don't want to build them at that cost and that it's "just it", is a very very convenient way to dilute the lack of will and foresight into this abstract beast called "the market", a beast that I will repeat once again is absolutely short-sighted (you definitely give no shit about getting a ROI on the order of a century unless perhaps you're Gates or Buffett) and ultimately it is a MUCH bigger cop-out than to say that this is about political will.

Because at least political will is about all of us learning about the issues and expressing our will, not about some magical elves deciding to invest hundreds of millions into long-term schemes they won't directly benefit from.

>Getting the NRC to approve them is one thing, getting a city to allow a nuclear power plant of any kind in their back yard is a whole other ball-game.

Some rust belt city that doesn't have the luxury of making decisions in a way other than "strictly by the numbers" in a state that isn't inclined to meddle in local matters will go for it if those numbers make sense.

There is a cost to long distance transmission which distorts those numbers, presumably.

Rust Belt cities use electricity, too.

I'm always in awe of how out of touch a lot of SV people are. The Rust Belt has cities. They have people. The rest of the country has people. The San Francisco-San Jose axis and NYC/LA are not the only parts of the country that matter. In Detroit alone, 6 million people live in the surrounding metro area. Cleveland? 2 million. Pittsburgh? 2.5.

Nuclear would be a great way to replace coal and create new jobs as well. And that part of the country is prime for innovation in the energy sector having been dominated by coal for the last century and a half.

Michigan produces ~25% of its power via nuclear. Ohio has 5 nuclear plants including several near Cleveland. Pennsylvania produces the second most nuclear power of any state and ~35% of its total power that way. Illinois has the most nuclear capacity in the country and generates over half its power that way.

The Rust Belt has tons of nuclear penetration already. It's the west coast specifically that is lacking nuclear power. Either because hydro/solar are better options or regulatory reasons, I don't know.

Perhaps even more than expected — aluminum refining uses up a huge percentage (20%?) of total US GWHs produced.

Existing nuclear reactors may have safe track records in terms of their operation, but we still have no long term solution for high-level nuclear waste disposal. At all.

The decommissioned plant outside of San Onofre, for example, is now storing nuclear waste in steel canisters on site. That site is on an earthquake fault line, right next to the pacific ocean [0].

Unless we come up with a complete end-to-end model for safely handling the waste, I don't consider nuclear to be a viable option. As it stands, it's a massive liability to the health of our future environment.

https://www.latimes.com/opinion/op-ed/la-oe-chapple-san-onof...

We could've if there was enough political will though. https://en.wikipedia.org/wiki/Yucca_Mountain_nuclear_waste_r...

Do you want trains full of other cities' nuclear waste passing through your city?

"Trains full"? The US only produces 2,000 tonnes of spent fuel a year, and has only produced 83,000 tonnes in the entire history of its nuclear generation program[1]. That entire annual production, even accounting for the weight of shielding and other safety measures, would fit on a single train the size of a typical coal or oil train.

[1] https://www.energy.gov/ne/articles/5-fast-facts-about-spent-...

Is this like....a rhetorical question? Of course no one actively wants them, but it's just a thing to transport like any other. I also don't want sewage or slaughterhouse waste transported next to my house(very smelly) but it has to be done at some point.

Let me ask you a question then - have you actually seen those railroad casks for nuclear waste? They are incredibly resilient, I'd be far less worried about having those travel through my city than living next to a 70 year old chemical plant(which I do).

Sure, nuclear waste isn’t sent for long term storage until it’s spent many years in a cooling pond radiating away the fastest-decaying and thus most dangerous components. After that it’s much less dangerous and frankly no more worrisome than plenty of industrial chemicals that are shipped around in traincars.

You way overestimate how much nuclear waste is being produced.

Wouldn't care, its hardly dangerous.

In terms of the waste, right now nuclear waste can and should be recycled which would reduce the amount of waste: https://en.wikipedia.org/wiki/Radioactive_waste

Soon it will be possible to use most of the waste as fuel:

"...Fast reactors can "burn" long lasting nuclear transuranic waste (TRU) waste components (actinides: reactor-grade plutonium and minor actinides), turning liabilities into assets. Another major waste component, fission products (FP), would stabilize at a lower level of radioactivity than the original natural uranium ore it was attained from in two to four centuries, rather than tens of thousands of years"

http://en.wikipedia.org/wiki/Integral_fast_reactor

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

While there are issues with nuclear power, the worry people have about nuclear waste is greatly overblown to say the least. The amounts generated are manageable and in a relatively short amount of time we can use most of this "waste" to generate electricity.

Sounds good in theory. But in reality the fuel rods need to be mechanically separated from their cladding, which is brittle, irradiated and whatnot else. Then chemical reprocessing of the fuel itself, and reassambly into whichever form is necessary.

It's not like you could take a used fuel-rod and throw it into some breeder to have energy. Not to speak of the old cladding.

I mean, I'm not against it, but it's not that simple.

>...It's not like you could take a used fuel-rod and throw it into some breeder to have energy. Not to speak of the old cladding. I mean, I'm not against it, but it's not that simple.

You are moving the goal posts.

Your original comment was "...but we still have no long term solution for high-level nuclear waste disposal. At all. ...Unless we come up with a complete end-to-end model for safely handling the waste, I don't consider nuclear to be a viable option. As it stands, it's a massive liability to the health of our future environment."

There is an end-to-end model for safely handling the waste. (For example, the IFR design would actually have reprocessed the fuel on-site.) Nuclear waste will not be a massive liability to our future environment. The amount waste is very manageable (the Netherlands actually stores their waste in an art museum) and in a relatively short amount of time we can use most of this "waste" to generate electricity.

That is probably because I don't care about ball sports at all :)

Yes we do have a long term solution for high-level nuclear waste. Right now.

Recycle it. You get far more energy out than the first pass as well. This technology is over half a century old.

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

Please stop saying we have no solution for nuclear waste. We absolutely do. It's just cheaper to keep storing waste than to build new plants due to the political barriers.

It would be good if there were a technology neutral expectation that all waste from energy production must be handled and addressed in both short term and long term.

France recycles yearly 1700 tons of nuclear waste, since 1966. Over half a century, that's what I would call safe enough. That's about the size of the US waste production, and represents the waste produced in France plus the Netherlands and Germany and and. So that end-to-end model exists already, it's just many people don't know some things are happening also outside US borders.

Vamos a la Playa!

> Unless we come up with a complete end-to-end model for safely handling the waste, I don't consider nuclear to be a viable option.

We only need to store the stuff for 2-3 decades, by which time it should be feasible to just yeet the waste into space.

And we can thank the anti-nuclear green activists for the waste disposal problem too. There are numerous ways to minimise waste (reprocessing, breeder plants) which have been lobbied against. And there have been good waste management options that failed only because of lobbying.

So they install one at Disney World. The Buena Vista district is privately run there. I think they even negotiated for nukes, when it was established. There are quite a few behemoth organizations that could pull it off, if the economics look ok.

It's not just them though. No one wants to handle the waste. If I recall, the middle of nowhere Nevada was designated to take on the waste from quite a few places and they fought it and won. Additionally, no one wants that waste passing through their town on the way.

I support nuclear but acknowledge the worst case scenarios of it.

Well, the bombs are way more dangerous than power applications. How do we get rid of these bomb cores?

Throw them into the reactors! Someone’s gonna have to do it,

Canada already blends ex weapons plutonium into their CANDU reactors.

http://ccnr.org/mox_griffith_1.html

Yeah all those terrible anti-nuke environmentalists making nuke expansive as hell and never done on time! They are the worst!

PS. nobody stops you from building a reactor. Get it approved, come up with the money, build it and sell the power but stop begging for tax money. It's over in the civilised western world. Get over it.

https://phys.org/news/2018-11-france-nuclear-reactors-macron...

https://www.diw.de/documents/publikationen/73/diw_01.c.74261...

Reply on news.ycombinator.com