13 ms·

Yes, cost is the issue. Everyone assumes that anyone hesitant about Nuclear power is focussed on the environmental or safety concerns, but that is wrong. The

Yes, cost is the issue. Everyone assumes that anyone hesitant about Nuclear power is focussed on the environmental or safety concerns, but that is wrong. The costs (and cost overruns) of the construction and maintenance of Nuclear reactors have been enough reason on their own to be bearish on Nuclear. I'm hopeful that new technologies, such as small nuclear reactions that can be built in a factory, will address this.

virissimo 4y ago
Our nations were far less wealthy when the existing stock of nuclear power plants were built, the technology available for building them have improved since then, and the raw resources needed as their inputs aren't a constraint, so why have costs risen so much relative to what we are willing to spend on it? Environmental and safety regulations are supposed to be explanations for this rise in cost, not rival explanations to "cost is the issue" (since they take that as given).
- KptMarchewa 4y ago
  >the technology available for building them have improved since then Broadly, construction is the singular industry where productivity hasn't improved with modern technology. https://constructionphysics.substack.com/p/sketch-of-a-theory-of-construction https://constructionphysics.substack.com/p/sketch-of-a-theor...
  - MichaelZuo 4y ago
    In fact, the opposite, there’s been a noticeable productivity decline in the US in most of the construction industry.
  - towaway15463 4y ago
    Concrete infrastructure projects in particular seem to have gotten more expensive and time consuming.
    - lotsofpulp 4y ago
      I imagine moving all that mass and the energy intensive process to form and break concrete directly scale with rising oil prices. Plus price of labor going up.
- LatteLazy 4y ago
  They built them only for non economic reasons. Either because they wanted material for bombs, or as national vanity projects or because they convinced themselves it would be cheap. If you want those reasons, go for it. Iran for instance is pursuing at least one of these goals when it works towards nuclear "power". Just don't expect affordable electricity at the end...
- crote 4y ago
  - Nuclear power plants are rare enough that each project is basically unique. - Nuclear power plants are very complex. They are more "engineered" than "constructed", with hundreds of kilometers of tubes and wires. Pouring the concrete is the easy part. - Nuclear power plants deal with an incredibly dangerous environment: high pressures, high temperatures, radioactivity, hazardous chemicals. Everything has to be tested, re-tested, and certified. Many parts will be near impossible to replace once it enters production, and the plant is supposed to be operational for decades. - Nuclear power plants are safety-critical. Contrary to many other structures of similar complexity, things can get way worse than a big explosion. They have the potential to contaminate the site (or even the surrounding country) for decades or centuries. Failure is simply not acceptable. - Safety regulations have been getting stricter over the years, because supposedly "100% safe" plants keep having accidents and the population is not very happy about that. The existing stock of nuclear power plants is an offshoot of several military programs. The most common design is pretty much a submarine power plant on steroids, which turned out to be less than ideal. Things like "safety" and "profit" were almost seen as a suggestion more so than a requirement. Newly-built power plants are required to incorporate over 50 years of innovation, but it turns out almost nobody has the skills to actually build them. They have simply gotten too complex to construct!
  - ncmncm 4y ago
    All of the above is true enough, but it is not the full story on why US nukes cost like they do. The ones attempted recently in South Carolina and Georgia cost so much because people in a position to block building them want it that way. Once a nuke project starts, a torrent of money starts flowing. Everybody with a stake expects a share, and none want it ever to end. Finishing a plant would plug that flow. If it looks like the flow will dry up regardless, they might deliver something at the end. We see the same process with urban tunnels and big military procurements. It is all perfectly legal. It is probably not fixable. Thus far, wind and solar projects have mostly avoided it, maybe in part because their legitimate costs are easy to estimate. Existing nukes also cost more than renewables to operate, because they actually have operating costs. Big steam turbines, in particular, need regular overhauls. So, it is hard for anything with a steam turbine to compete with something that has none. Reactors need also operation, inspections, refueling, security, compliance monitoring... the list goes on. Solar panels might need to be dusted off, or replaced. The worst that can happen to a wind turbine is to catch fire.
- nine_k 4y ago
  * Technology makes machines cheaper, and makes human work more expensive in comparison. Nuclear projects involve a lot of highly qualified human work. * Technology makes repetitive operations cheaper. You build a factory for hundred million dollars, make a billion gadgets on it, and every gadget costs you ten cents to make. Nuclear projects lack the economy of scale: even the Navy is going to order reactors by a dozen, and civil power plant reactors may see even fewer installations per model. Thus the huge costs of the R&D and the factory are amortized over but a few reactors, making each of them very expensive. * Due to small production scale, various custom materials needed for nuclear reactors, like special steels, are much more expensive than more widely used materials. Nuclear is now in a position similar to solar cells 15 years ago: a promising technology which is too expensive due to small scale and bespoke nature of their production. It took a decade of betting on them, pouring money into them, and giving various discounts to the customers to get where we are now, with solar panels which are efficient, affordable, and available. I suppose nuclear tech would need the same to become cost-efficient. France did / does something along these lines; the US does not.
  - pfdietz 4y ago
    Solar experience relentless experience effects, with cost dropping by about 20% for each doubling of cumulative production. Nuclear has not shown good experience effects.
    - Schroedingersat 4y ago
      Nuclear's learning rate is about the same magnitude! ...In the opposite direction, but about the same magnitude.
      - dylan604 4y ago
        hmmm, I've never thought about magnitude being signed or unsigned. Is it little-endian or big-endian I'm now wondering
  - cycomanic 4y ago
    > Nuclear is now in a position similar to solar cells 15 years ago: a promising technology which is too expensive due to small scale and bespoke nature of their production. It took a decade of betting on them, pouring money into them, and giving various discounts to the customers to get where we are now, with solar panels which are efficient, affordable, and available. I suppose nuclear tech would need the same to become cost-efficient. France did / does something along these lines; the US does not. Unlike solar 15 years ago, nuclear has been around 60 years already and received huge amounts of subsidies (much more than solar, excluding all the extra military subsidies). So what is fundamentally different now that would change the essentially linear scaling to an exponential scaling and why should we not invest into the tech where we see ongoing exponential scaling of cost already with no indication of slowing (solar and wind)?
- kragen 4y ago
  My best guess is what Freeman Dyson said in his autobiography Disturbing the Universe: he stopped working on nuclear reactors after TRIGA (01958) and Project Orion (01957 to 01961) because it stopped being fun. It became a question of bureaucracy and national security and empire-building and whatnot. So people like Dyson, who learned calculus by spending his Christmas vacation working his way through a textbok for fun, were no longer available to design reactors. So progress on, for example, thorium reactors ended in the US in 01966 (thorium BWR is from 01960, thorium LWBR is from 01962, thorium MSR is from 01964, thorium HTGR is from 01966). In other countries it took a few more years. Teller, who worked on TRIGA with Dyson, seems to have stopped working on reactor design in 01960. The elves left Middle Earth. So hospitals still use TRIGA today for nuclear medicine. I don't see the key issue here as being that Dyson and Teller were brilliant, though they were brilliant. Rather, it's that Dyson and Teller were curious and playful. I think curious and playful people exploring the possibilities of nuclear reactors will come up with many improvements, even if they are the ordinary kind of stupid people. It might take them five times as long as it would have taken people like Dyson, and there might be more accidental deaths along the way, but they will get there. But today the occasional curious and playful person who tries to investigate nuclear reactors is likely to get arrested, even if they pose less risk to their neighbors than Marie Curie: https://en.wikipedia.org/wiki/Richard_Handl https://en.wikipedia.org/wiki/Richard_Handl (or assassinated by the Mossad: https://en.wikipedia.org/wiki/Assassination_of_Iranian_nuclear_scientists https://en.wikipedia.org/wiki/Assassination_of_Iranian_nucle...) and so the Navy is still using reactors very similar to the ones they used 60 years ago. I see questions like "the technology available has improved, so why have costs risen?" as symptomatic of the worldview that anything can be bought. Teller, Dyson, and Handl were not optimizing their life decisions to maximize their earning power; they were curious about the world and wanted to preserve liberal democracy. You can't buy that. If you announce that you are going to spend a lot of money on nuclear reactor development, it will attract people who optimize their life decisions to maximize their earning power, not people who are curious about the world. You maximize your earning power by owning things, not by learning things or figuring things out. People who optimize their life decisions to maximize their earning power will be no good at ferreting out possible improvements that can be made to nuclear reactors, because you don't get paid any more for making breakthroughs than you do for just plodding along. Less, in fact, because nine tenths of the time when you're doing the kind of things that lead to breakthroughs, the things you try don't work, so you aren't delivering anything of value to anybody. The other issue is that "Why can't we build nuclear power plants?" is a subset of "Why can't we build?" and I think the answer to that is basically that people aren't free to build. But nuclear power plants are probably the kind of building that people are least free to build.
- cycomanic 4y ago
  That argument should apply to houses as well, but building houses has become more expensive in the last 50 years not less. Generally construction does not significantly benifit from economies of scale, while fabrication does. Now we could build lots of small scale reactors, but while construction costs might come down (so far completely unproven) running costs most definitely would go up. You don't want to run even a small scale nuclear reactor without qualified staff, security etc..
- imtringued 4y ago
  As I said, the problem with nuclear power are all these ragtag powerplants that were hastily built at a time people didn't understand the potential for catastrophic failure. If you get rid of these ancient power plants, the risk of failure goes down dramatically and the need for further regulation disappears. Meanwhile everywhere you keep hearing how e.g. Germany should keep obsolete power plants even though this will ruin the long term reputation of nuclear further.
jjk166 4y ago
Nuclear is not fundamentally expensive[0], and most nuclear power plants currently in operation were cheap when they were built. The high costs of building new powerplants is a result of going decades without building any, leading to a loss of experience. Even still, nuclear remains competitive with other traditional power generation technologies like coal. Only with recent advancements in fracking did natural gas supplant coal as the go to fuel, and Wind and solar have become extremely inexpensive, but only in the past 10 years, in large part due to a concerted effort to scale up those industries. In other parts of the world where nuclear power did not have a decades long hiatus it remains extremely competitive with even these renewables, for example in South Korea where LCOE for nuclear is half that of solar. Of course you may have only became bearish on nuclear when the costs got high, but there are a large number of people who were bearish before then, which is what made costs high. [0] https://www.sciencedirect.com/science/article/pii/S0301421516300106 https://www.sciencedirect.com/science/article/pii/S030142151...
- BonoboIO 4y ago
  Building it maybe can get cheaper ... destruction and disposal will be even more expensive. Look at Germany. They are the first to really destruct the first power plant and it will take 20 to 25 years. Everything has to be checked.
  - jjk166 4y ago
    The cost of decommissioning is included
    - bigfudge 4y ago
      Let us know how that’s gone when you’ve actually decommissioned one and dealt with all the storage issues.
    - BonoboIO 4y ago
      Sorry, but nope. Nowhere near.
  - Galaxeblaffer 4y ago
stjohnswarts 4y ago
We'll all be bearish when it's 110 outside on the regular in the NE USA
DelightOne 4y ago
If that is true: why do people care about cost when they don't pay the price themselves?
tuatoru 4y ago
Cost is indeed the issue for commercial power generation. So let's look at costs. Especially, let's look at cost trends. My sources are: Lazard's 2022 levelized cost of energy estimates; NREL's Spring 2022 Solar Industry Update: the Global Wind Energy Council's Global Wind Report 2022 highlights; and especially the 2022 report in Cell (Cell's "Joule" sub-publication) called "Empirically Grounded Energy Forecasts and the Energy Transition". On the basis of the trends in costs shown there, I conclude that 1) coal, nuclear, and large-scale hydro are obsolete for power generation; 2) combined-cycle gas, peaker gas, and on-shore wind generation are obsolescent. (Biomass thermal, concentrated solar thermal, and geothermal are essentially lab experiments. A few hundred MW or maybe 2 GW total globally each. Oil is of course obsolete too.) Once current project pipelines are emptied (2035 maybe?), there won't be much of any of these built for commercial power generation, because they won't be able to compete except in obscure niches. Onshore wind comes closest to being competitive, but its costs are not dropping fast enough. The costs of the others are rising. (According to NREL and the GWEC respectively, cumulative global solar PV installed capacity is 939 GWdc, most of it installed in 2018-2021, and cumulative global wind capacity installed is 837 GW, most installed 2012-2019. Onshore wind, while it has not yet peaked in terms of yearly increment, has peaked as a share of total generation, while PV is bottlenecked by permitting queues alone.)
- pclmulqdq 4y ago
  So what is supposed to replace all of these obsolete technologies that currently account for the vast majority of energy generation? Lithium batteries, solar power, and offshore wind power? There are so many problems with this scheme, not least of which is that you have pretty much no base load and batteries can't be manufactured at anywhere near enough scale to successfully do this "transition" for at least the next 20 years.
  - onlyrealcuzzo 4y ago
    In the US, we already have about ~25% of power coming from nuclear and Hydro. That's a huge chunk of our base load. Add in ~7% offshore wind, ~4% for onshore, and ~4% storage for pumped hydro, and we would be just fine with the rest coming from solar & a hodge podge of other renewables. And we'll have natural gas power plants in commission for another 30 years. Why shut them down? We can use them in the Winter. 30 years is more than enough time to transition to near 100% renewables. If mini nuclear reactors pan out and can replace even a fraction of the coal plants out there - our base load will be covered. At night, in the winter, we use about half as much electricity as during the day in the summer - when Solar works best.
moonchrome 4y ago
I think the cost argument is disingenuous - solar camp likes to tout how it took a decade of subsides before economies of scale kicked in - no reason to believe nuclear wouldn't come down with same kind of subsides - on a longer scale due to longer iteration cycle.
- pfdietz 4y ago
  Nuclear already had those subsidies, in spades. Costs didn't come down.
- dragonwriter 4y ago
  > solar camp likes to tout how it took a decade of subsides before economies of scale kicked in - no reason to believe nuclear wouldn’t come down with same kind of subsides - on a longer scale due to longer iteration cycle. There is: nuclear has been subsidized on a much longer scale than solar already.
  - Nimitz14 4y ago
    Please share more information. What subsidies existed for nuclear?
    - MaKn 4y ago
      Well, since nuclear plants cannot be insured these costs as well as the waste cost are mostly carried by the state. There are various calculations on these costs online. Not sure about other costs, but I'm sure there are more (e.g. security for transportation)
- adrianN 4y ago
  The problem is that we don't have another forty years to wait for nuclear costs to come down if we want to prevent catastrophic warming.
asdff 4y ago
The costs are only seen as high relative to other investments. If you tax polluting industries for the damage the cause, for example, maybe those cost overruns for nuclear reactors would be just as easily stomached as the countless highway projects that also see cost and time overruns and are still paid for nonetheless, relative to having to pay massive taxes on dirty energy.
philwelch 4y ago
Yes, the OP is the last in an entire trilogy of articles discussing this issue in detail.
jillesvangurp 4y ago
Nuclear is not just a little bit more expensive but more like an order of magnitude. And as cost for renewables continues to drop, that might turn into two orders of magnitude. Small nuclear reactors might reduce the cost a little; but a little is not nearly enough. Cost needs to come down by at least an order of magnitude for it to at least somewhat keep up. Anything else is just not going to be cost competitive. And that's a problem when renewables are still dropping rapidly in price.
imtringued 4y ago
If cost is the issue why is Japan spending so much e.g. estimates of 600 billion USD on cleaning up Fukushima? Rebuilding the entire nuclear power plant fleet of Japan would have been cheaper than a single meltdown. Oh right, cost is an issue because obsolete nuclear power plants keep giving nuclear a bad name and end up increasing regulatory requirements on modern designs even if they don't have failure modes from 50 years ago. In other words, the nuclear power industry really likes shooting themselves in the foot.
olejorgenb 4y ago
Here is a free book on this topic: http://www.xylenepower.com/gordian_w8.pdf http://www.xylenepower.com/gordian_w8.pdf I have not read it, but I've read an article [1] which cites it. Apparently a lot of the current regulation is created based on a faulty model (LNT, ALARA) of how harmful radiation is. This model predicts that the harm is linear in dose, with no time component. Ie.: that receiving a dose of 1/n for n days is as bad as a single high dose of n for one day. According to the article and book, this is model has been proven wrong (a long time ago). [1] (in Norwegian) https://www.minervanett.no/energi-kjernekraft-klima/hvordan-kjernekraft-floppet/412426 https://www.minervanett.no/energi-kjernekraft-klima/hvordan-...
guidedlight 4y ago
NuScale looks to address this, by creating small, modular, self-managing nuclear reactors. https://arstechnica.com/science/2022/07/us-regulators-will-certify-first-small-nuclear-reactor-design/ https://arstechnica.com/science/2022/07/us-regulators-will-c...
md2020 4y ago
Recently on a podcast John Carmack said somebody should look into building small fission fragment direct energy conversion reactors, since despite their lower efficiency, they’re apparently much cheaper to build. Since Carmack is a software guy (albeit a very good one) I’m somewhat skeptical, since it seems like if it’s not that complicated a design somebody would have tried it already and found good reasons why it doesn’t work. The only resources I could find on such a thing was a DoE paper about some prototypes Sandia and General Atomics built. Does anyone know more about the merits of this idea?
- lambdatronics 4y ago
  Good question. I think historical designs supported the fuel elements on material structures that ended up absorbing a large fraction of the fragments, leading to inefficiency & overheating the fuel. Here's an outside-of-the-box idea, using a dusty plasma cloud. [0] In this case, I'd be concerned about a criticality accident if the fuel dust containment failed & all the dust landed in a pile. [0] www.rbsp.info/rbs/PDF/aiaa05.pdf

virissimo 4y ago

Our nations were far less wealthy when the existing stock of nuclear power plants were built, the technology available for building them have improved since then, and the raw resources needed as their inputs aren't a constraint, so why have costs risen so much relative to what we are willing to spend on it? Environmental and safety regulations are supposed to be explanations for this rise in cost, not rival explanations to "cost is the issue" (since they take that as given).

KptMarchewa 4y ago
>the technology available for building them have improved since then Broadly, construction is the singular industry where productivity hasn't improved with modern technology. https://constructionphysics.substack.com/p/sketch-of-a-theory-of-construction https://constructionphysics.substack.com/p/sketch-of-a-theor...
- MichaelZuo 4y ago
  In fact, the opposite, there’s been a noticeable productivity decline in the US in most of the construction industry.
- towaway15463 4y ago
  Concrete infrastructure projects in particular seem to have gotten more expensive and time consuming.
  - lotsofpulp 4y ago
    I imagine moving all that mass and the energy intensive process to form and break concrete directly scale with rising oil prices. Plus price of labor going up.
LatteLazy 4y ago
They built them only for non economic reasons. Either because they wanted material for bombs, or as national vanity projects or because they convinced themselves it would be cheap. If you want those reasons, go for it. Iran for instance is pursuing at least one of these goals when it works towards nuclear "power". Just don't expect affordable electricity at the end...
crote 4y ago
- Nuclear power plants are rare enough that each project is basically unique. - Nuclear power plants are very complex. They are more "engineered" than "constructed", with hundreds of kilometers of tubes and wires. Pouring the concrete is the easy part. - Nuclear power plants deal with an incredibly dangerous environment: high pressures, high temperatures, radioactivity, hazardous chemicals. Everything has to be tested, re-tested, and certified. Many parts will be near impossible to replace once it enters production, and the plant is supposed to be operational for decades. - Nuclear power plants are safety-critical. Contrary to many other structures of similar complexity, things can get way worse than a big explosion. They have the potential to contaminate the site (or even the surrounding country) for decades or centuries. Failure is simply not acceptable. - Safety regulations have been getting stricter over the years, because supposedly "100% safe" plants keep having accidents and the population is not very happy about that. The existing stock of nuclear power plants is an offshoot of several military programs. The most common design is pretty much a submarine power plant on steroids, which turned out to be less than ideal. Things like "safety" and "profit" were almost seen as a suggestion more so than a requirement. Newly-built power plants are required to incorporate over 50 years of innovation, but it turns out almost nobody has the skills to actually build them. They have simply gotten too complex to construct!
- ncmncm 4y ago
  All of the above is true enough, but it is not the full story on why US nukes cost like they do. The ones attempted recently in South Carolina and Georgia cost so much because people in a position to block building them want it that way. Once a nuke project starts, a torrent of money starts flowing. Everybody with a stake expects a share, and none want it ever to end. Finishing a plant would plug that flow. If it looks like the flow will dry up regardless, they might deliver something at the end. We see the same process with urban tunnels and big military procurements. It is all perfectly legal. It is probably not fixable. Thus far, wind and solar projects have mostly avoided it, maybe in part because their legitimate costs are easy to estimate. Existing nukes also cost more than renewables to operate, because they actually have operating costs. Big steam turbines, in particular, need regular overhauls. So, it is hard for anything with a steam turbine to compete with something that has none. Reactors need also operation, inspections, refueling, security, compliance monitoring... the list goes on. Solar panels might need to be dusted off, or replaced. The worst that can happen to a wind turbine is to catch fire.
nine_k 4y ago
* Technology makes machines cheaper, and makes human work more expensive in comparison. Nuclear projects involve a lot of highly qualified human work. * Technology makes repetitive operations cheaper. You build a factory for hundred million dollars, make a billion gadgets on it, and every gadget costs you ten cents to make. Nuclear projects lack the economy of scale: even the Navy is going to order reactors by a dozen, and civil power plant reactors may see even fewer installations per model. Thus the huge costs of the R&D and the factory are amortized over but a few reactors, making each of them very expensive. * Due to small production scale, various custom materials needed for nuclear reactors, like special steels, are much more expensive than more widely used materials. Nuclear is now in a position similar to solar cells 15 years ago: a promising technology which is too expensive due to small scale and bespoke nature of their production. It took a decade of betting on them, pouring money into them, and giving various discounts to the customers to get where we are now, with solar panels which are efficient, affordable, and available. I suppose nuclear tech would need the same to become cost-efficient. France did / does something along these lines; the US does not.
- pfdietz 4y ago
  Solar experience relentless experience effects, with cost dropping by about 20% for each doubling of cumulative production. Nuclear has not shown good experience effects.
  - Schroedingersat 4y ago
    Nuclear's learning rate is about the same magnitude! ...In the opposite direction, but about the same magnitude.
    - dylan604 4y ago
      hmmm, I've never thought about magnitude being signed or unsigned. Is it little-endian or big-endian I'm now wondering
- cycomanic 4y ago
  > Nuclear is now in a position similar to solar cells 15 years ago: a promising technology which is too expensive due to small scale and bespoke nature of their production. It took a decade of betting on them, pouring money into them, and giving various discounts to the customers to get where we are now, with solar panels which are efficient, affordable, and available. I suppose nuclear tech would need the same to become cost-efficient. France did / does something along these lines; the US does not. Unlike solar 15 years ago, nuclear has been around 60 years already and received huge amounts of subsidies (much more than solar, excluding all the extra military subsidies). So what is fundamentally different now that would change the essentially linear scaling to an exponential scaling and why should we not invest into the tech where we see ongoing exponential scaling of cost already with no indication of slowing (solar and wind)?
kragen 4y ago
My best guess is what Freeman Dyson said in his autobiography Disturbing the Universe: he stopped working on nuclear reactors after TRIGA (01958) and Project Orion (01957 to 01961) because it stopped being fun. It became a question of bureaucracy and national security and empire-building and whatnot. So people like Dyson, who learned calculus by spending his Christmas vacation working his way through a textbok for fun, were no longer available to design reactors. So progress on, for example, thorium reactors ended in the US in 01966 (thorium BWR is from 01960, thorium LWBR is from 01962, thorium MSR is from 01964, thorium HTGR is from 01966). In other countries it took a few more years. Teller, who worked on TRIGA with Dyson, seems to have stopped working on reactor design in 01960. The elves left Middle Earth. So hospitals still use TRIGA today for nuclear medicine. I don't see the key issue here as being that Dyson and Teller were brilliant, though they were brilliant. Rather, it's that Dyson and Teller were curious and playful. I think curious and playful people exploring the possibilities of nuclear reactors will come up with many improvements, even if they are the ordinary kind of stupid people. It might take them five times as long as it would have taken people like Dyson, and there might be more accidental deaths along the way, but they will get there. But today the occasional curious and playful person who tries to investigate nuclear reactors is likely to get arrested, even if they pose less risk to their neighbors than Marie Curie: https://en.wikipedia.org/wiki/Richard_Handl https://en.wikipedia.org/wiki/Richard_Handl (or assassinated by the Mossad: https://en.wikipedia.org/wiki/Assassination_of_Iranian_nuclear_scientists https://en.wikipedia.org/wiki/Assassination_of_Iranian_nucle...) and so the Navy is still using reactors very similar to the ones they used 60 years ago. I see questions like "the technology available has improved, so why have costs risen?" as symptomatic of the worldview that anything can be bought. Teller, Dyson, and Handl were not optimizing their life decisions to maximize their earning power; they were curious about the world and wanted to preserve liberal democracy. You can't buy that. If you announce that you are going to spend a lot of money on nuclear reactor development, it will attract people who optimize their life decisions to maximize their earning power, not people who are curious about the world. You maximize your earning power by owning things, not by learning things or figuring things out. People who optimize their life decisions to maximize their earning power will be no good at ferreting out possible improvements that can be made to nuclear reactors, because you don't get paid any more for making breakthroughs than you do for just plodding along. Less, in fact, because nine tenths of the time when you're doing the kind of things that lead to breakthroughs, the things you try don't work, so you aren't delivering anything of value to anybody. The other issue is that "Why can't we build nuclear power plants?" is a subset of "Why can't we build?" and I think the answer to that is basically that people aren't free to build. But nuclear power plants are probably the kind of building that people are least free to build.
cycomanic 4y ago
That argument should apply to houses as well, but building houses has become more expensive in the last 50 years not less. Generally construction does not significantly benifit from economies of scale, while fabrication does. Now we could build lots of small scale reactors, but while construction costs might come down (so far completely unproven) running costs most definitely would go up. You don't want to run even a small scale nuclear reactor without qualified staff, security etc..
imtringued 4y ago
As I said, the problem with nuclear power are all these ragtag powerplants that were hastily built at a time people didn't understand the potential for catastrophic failure. If you get rid of these ancient power plants, the risk of failure goes down dramatically and the need for further regulation disappears. Meanwhile everywhere you keep hearing how e.g. Germany should keep obsolete power plants even though this will ruin the long term reputation of nuclear further.

jjk166 4y ago

Nuclear is not fundamentally expensive[0], and most nuclear power plants currently in operation were cheap when they were built. The high costs of building new powerplants is a result of going decades without building any, leading to a loss of experience. Even still, nuclear remains competitive with other traditional power generation technologies like coal. Only with recent advancements in fracking did natural gas supplant coal as the go to fuel, and Wind and solar have become extremely inexpensive, but only in the past 10 years, in large part due to a concerted effort to scale up those industries. In other parts of the world where nuclear power did not have a decades long hiatus it remains extremely competitive with even these renewables, for example in South Korea where LCOE for nuclear is half that of solar. Of course you may have only became bearish on nuclear when the costs got high, but there are a large number of people who were bearish before then, which is what made costs high. [0] https://www.sciencedirect.com/science/article/pii/S0301421516300106 https://www.sciencedirect.com/science/article/pii/S030142151...

BonoboIO 4y ago
Building it maybe can get cheaper ... destruction and disposal will be even more expensive. Look at Germany. They are the first to really destruct the first power plant and it will take 20 to 25 years. Everything has to be checked.
- jjk166 4y ago
  The cost of decommissioning is included
  - bigfudge 4y ago
    Let us know how that’s gone when you’ve actually decommissioned one and dealt with all the storage issues.
  - BonoboIO 4y ago
    Sorry, but nope. Nowhere near.
- Galaxeblaffer 4y ago

stjohnswarts 4y ago

We'll all be bearish when it's 110 outside on the regular in the NE USA

DelightOne 4y ago

If that is true: why do people care about cost when they don't pay the price themselves?

tuatoru 4y ago

Cost is indeed the issue for commercial power generation. So let's look at costs. Especially, let's look at cost trends. My sources are: Lazard's 2022 levelized cost of energy estimates; NREL's Spring 2022 Solar Industry Update: the Global Wind Energy Council's Global Wind Report 2022 highlights; and especially the 2022 report in Cell (Cell's "Joule" sub-publication) called "Empirically Grounded Energy Forecasts and the Energy Transition". On the basis of the trends in costs shown there, I conclude that 1) coal, nuclear, and large-scale hydro are obsolete for power generation; 2) combined-cycle gas, peaker gas, and on-shore wind generation are obsolescent. (Biomass thermal, concentrated solar thermal, and geothermal are essentially lab experiments. A few hundred MW or maybe 2 GW total globally each. Oil is of course obsolete too.) Once current project pipelines are emptied (2035 maybe?), there won't be much of any of these built for commercial power generation, because they won't be able to compete except in obscure niches. Onshore wind comes closest to being competitive, but its costs are not dropping fast enough. The costs of the others are rising. (According to NREL and the GWEC respectively, cumulative global solar PV installed capacity is 939 GWdc, most of it installed in 2018-2021, and cumulative global wind capacity installed is 837 GW, most installed 2012-2019. Onshore wind, while it has not yet peaked in terms of yearly increment, has peaked as a share of total generation, while PV is bottlenecked by permitting queues alone.)

pclmulqdq 4y ago
So what is supposed to replace all of these obsolete technologies that currently account for the vast majority of energy generation? Lithium batteries, solar power, and offshore wind power? There are so many problems with this scheme, not least of which is that you have pretty much no base load and batteries can't be manufactured at anywhere near enough scale to successfully do this "transition" for at least the next 20 years.
- onlyrealcuzzo 4y ago
  In the US, we already have about ~25% of power coming from nuclear and Hydro. That's a huge chunk of our base load. Add in ~7% offshore wind, ~4% for onshore, and ~4% storage for pumped hydro, and we would be just fine with the rest coming from solar & a hodge podge of other renewables. And we'll have natural gas power plants in commission for another 30 years. Why shut them down? We can use them in the Winter. 30 years is more than enough time to transition to near 100% renewables. If mini nuclear reactors pan out and can replace even a fraction of the coal plants out there - our base load will be covered. At night, in the winter, we use about half as much electricity as during the day in the summer - when Solar works best.

moonchrome 4y ago

I think the cost argument is disingenuous - solar camp likes to tout how it took a decade of subsides before economies of scale kicked in - no reason to believe nuclear wouldn't come down with same kind of subsides - on a longer scale due to longer iteration cycle.

pfdietz 4y ago
Nuclear already had those subsidies, in spades. Costs didn't come down.
dragonwriter 4y ago
> solar camp likes to tout how it took a decade of subsides before economies of scale kicked in - no reason to believe nuclear wouldn’t come down with same kind of subsides - on a longer scale due to longer iteration cycle. There is: nuclear has been subsidized on a much longer scale than solar already.
- Nimitz14 4y ago
  Please share more information. What subsidies existed for nuclear?
  - MaKn 4y ago
    Well, since nuclear plants cannot be insured these costs as well as the waste cost are mostly carried by the state. There are various calculations on these costs online. Not sure about other costs, but I'm sure there are more (e.g. security for transportation)
adrianN 4y ago
The problem is that we don't have another forty years to wait for nuclear costs to come down if we want to prevent catastrophic warming.

asdff 4y ago

The costs are only seen as high relative to other investments. If you tax polluting industries for the damage the cause, for example, maybe those cost overruns for nuclear reactors would be just as easily stomached as the countless highway projects that also see cost and time overruns and are still paid for nonetheless, relative to having to pay massive taxes on dirty energy.

philwelch 4y ago

Yes, the OP is the last in an entire trilogy of articles discussing this issue in detail.

jillesvangurp 4y ago

Nuclear is not just a little bit more expensive but more like an order of magnitude. And as cost for renewables continues to drop, that might turn into two orders of magnitude. Small nuclear reactors might reduce the cost a little; but a little is not nearly enough. Cost needs to come down by at least an order of magnitude for it to at least somewhat keep up. Anything else is just not going to be cost competitive. And that's a problem when renewables are still dropping rapidly in price.

imtringued 4y ago

If cost is the issue why is Japan spending so much e.g. estimates of 600 billion USD on cleaning up Fukushima? Rebuilding the entire nuclear power plant fleet of Japan would have been cheaper than a single meltdown. Oh right, cost is an issue because obsolete nuclear power plants keep giving nuclear a bad name and end up increasing regulatory requirements on modern designs even if they don't have failure modes from 50 years ago. In other words, the nuclear power industry really likes shooting themselves in the foot.

olejorgenb 4y ago

Here is a free book on this topic: http://www.xylenepower.com/gordian_w8.pdf http://www.xylenepower.com/gordian_w8.pdf I have not read it, but I've read an article [1] which cites it. Apparently a lot of the current regulation is created based on a faulty model (LNT, ALARA) of how harmful radiation is. This model predicts that the harm is linear in dose, with no time component. Ie.: that receiving a dose of 1/n for n days is as bad as a single high dose of n for one day. According to the article and book, this is model has been proven wrong (a long time ago). [1] (in Norwegian) https://www.minervanett.no/energi-kjernekraft-klima/hvordan-kjernekraft-floppet/412426 https://www.minervanett.no/energi-kjernekraft-klima/hvordan-...

guidedlight 4y ago

NuScale looks to address this, by creating small, modular, self-managing nuclear reactors. https://arstechnica.com/science/2022/07/us-regulators-will-certify-first-small-nuclear-reactor-design/ https://arstechnica.com/science/2022/07/us-regulators-will-c...

md2020 4y ago

Recently on a podcast John Carmack said somebody should look into building small fission fragment direct energy conversion reactors, since despite their lower efficiency, they’re apparently much cheaper to build. Since Carmack is a software guy (albeit a very good one) I’m somewhat skeptical, since it seems like if it’s not that complicated a design somebody would have tried it already and found good reasons why it doesn’t work. The only resources I could find on such a thing was a DoE paper about some prototypes Sandia and General Atomics built. Does anyone know more about the merits of this idea?

lambdatronics 4y ago
Good question. I think historical designs supported the fuel elements on material structures that ended up absorbing a large fraction of the fragments, leading to inefficiency & overheating the fuel. Here's an outside-of-the-box idea, using a dusty plasma cloud. [0] In this case, I'd be concerned about a criticality accident if the fuel dust containment failed & all the dust landed in a pile. [0] www.rbsp.info/rbs/PDF/aiaa05.pdf