12 ms·

France has sort of done this - they have 56 reactors in operation all based on the same 3 basic designs[1]. It's pretty incredible how quickly the plants were d

by barake 3y ago

France has sort of done this - they have 56 reactors in operation all based on the same 3 basic designs[1]. It's pretty incredible how quickly the plants were designed, tested, and built. Over a span of 15 years they brought 56 reactors online[2] - in the US we'd be lucky to build and commission a single reactor in that time span. [1] https://www.iaea.org/newscenter/news/frances-efficiency-in-the-nuclear-fuel-cycle-what-can-oui-learn https://www.iaea.org/newscenter/news/frances-efficiency-in-t... [2] https://www.pbs.org/wgbh/pages/frontline/shows/reaction/readings/french.html https://www.pbs.org/wgbh/pages/frontline/shows/reaction/read...

i_am_proteus 3y ago
Friendly reminder that the US does have such a program (the Navy's nuclear propulsion program), and several reactors are commissioned every year.
- MrBodangles 3y ago
  One of the core problems with nuclear is the size and scale of risk posed by cost cutting and neglect over its lifetime, especially the waste. People neither trust for-profit enterprises or present/future governments with that sort of responsibility (and with empirically justifiable reasons). I’m aware newer models are stated to be significantly safer, but assurances mean little when these institutions have repeatedly lied and failed in the past. Either way, US carriers are probably one of the safest places for nuclear, as they’re mission critical for the life of the carrier and most likely to receive the utmost care… Plus the US has a long history of rubber stamping virtually unlimited funds to solve any military problem, whether the people approve or not. The handling of the waste is still a major concern, but what about the consequences of a torpedo compromising the reactor in warfare?
Ringz 3y ago
Many claim that France’s 1974 Messmer plan resulted in the building of its 58 reactors in 15 years. This is not true. The planning for several of these nuclear reactors began long before. For example, the Fessenheim reactor obtained its construction permit in 1967 and was planned starting years before. In addition, 10 of the reactors were completed between 1991-2000. As such, the whole planning-to-operation time for these reactors was at least 32 years, not 15. That of any individual reactor was 10 to 19 years.
godelski 3y ago
> in the US we'd be lucky to build and commission a single reactor in that time span Most countries, including the US and France, did a build out in the 70's/80's and then basically stopped. France a bit later than the US, but both essentially did the same thing. Checking the wiki list[0] and sorting by operation year you can see 4 things. 1) the vast majority of reactors were built in the 70's, 2) the newest reactor was built in the 90's (operational 2001), 3) the most recent reactors took longer to go into operation (including a few at 16 years, where the 70's build out was typically 6-7 years), 4) almost all 70s/80's reactors are of the same type and same power level (CP1, CP2, P4 REP 1300). We actually see the exact same story in the US (see Watts Bar, ouch). On the other hand, South Korea didn't do their build out till the mid 80's and continued into the 90's. Then we see the wall hit in the 2000's with the APR 1400. Japan did a bit better and strangely looks like the big success story, especially considering how many reactors such a small country built. Interestingly only Mitsubishi reactors are still operational... Canada is also a good success story but also hasn't built anything since the late 80's (but last reactor was still <10yrs). Countries like Sweden, started their build out but then there was a hard stop. Sweden had nothing past '85. Germany isn't too far off, but it is also a different story. Ditto for UK. I intentionally left out China and Russia because different economic structures and because the stories are a bit different even though might appear similar to what I'm discussing at face value (note that my comments are vastly oversimplified, with some things only being alluded to), but it is worth paying attention to the above patterns and think about how the economic structure might reinforce some of those aspects, then think about the western countries different styles during their build out phases (how it actually worked). The nuclear story is long and complicated. Even this wall of text is oversimplified. This is part of the problem: we like our simple talking points but as speakers are often unwilling to admit that these are only part of the stories or as listeners rebut the speaker as if they are only considering a single factor. It makes real conversation almost impossible and both play a role and build over time. Which is not too dissimilar to a few problems that happened in the nuclear industry. [0] https://en.wikipedia.org/wiki/List_of_commercial_nuclear_reactors#France https://en.wikipedia.org/wiki/List_of_commercial_nuclear_rea...
Gwypaas 3y ago
Lets add Flamanville 3 to the "experience" graph in the article. The only reason it gets pushed through is for France to have an industrial base enabling nuclear submarines, carriers and weapons. https://imgur.com/6G2RBa0 https://imgur.com/6G2RBa0 https://en.wikipedia.org/wiki/Flamanville_Nuclear_Power_Plant https://en.wikipedia.org/wiki/Flamanville_Nuclear_Power_Plan...
- adev_ 3y ago
  > Lets add Flamanville 3 to the "experience" graph in the article. The only reason it gets pushed through is for France to have an industrial base enabling nuclear submarines, carriers and weapons. That's garbage. - The reason Flamanville 3 takes so much time is precisely because it is a prototype on a new design that never have been produces in series, nor even tested. That supports 100% what is said here: If you want to reduce cost, mass produce. - Submarines and carrier nuclear reactors are completely different beast that have nothing to do with either Flammanville 3 or the existing nuclear park.
  - natmaka 3y ago
    > Flamanville 3 takes so much time is precisely because it is a prototype on a new design No, this reactor type (EPR) is a mere evolution. Proof: https://www.irsn.fr/savoir-comprendre/surete/presentation-historique-projet-epr https://www.irsn.fr/savoir-comprendre/surete/presentation-hi...
    - jjk166 3y ago
      It doesn't matter if it's an evolution or a revolution - a change of any sort means new tooling to produce the components, new procedures to assemble them, new analysis to certify that they will all work together, and new training for everyone involved to do all the above.
      - natmaka 3y ago
        All this was known before starting the building process, and the builder planned to deliver the reactor in 2012. It is late (right now: not delivered) mainly because project management and quality insurance were abysmal: https://en.wikipedia.org/wiki/Flamanville_Nuclear_Power_Plant#Unit_3 https://en.wikipedia.org/wiki/Flamanville_Nuclear_Power_Plan...
screye 3y ago
Genuine question. What is the stopping the US from paying every experienced French Nuclear Engineer x 2 x Usual US cost Adjustment --> and letting them build here in the US ? France and US are allies, and likely aligned on climate goals. France has similar climate as many of the planned regions and has fairly high building standards.
- Zigurd 3y ago
  But not the ones who underestimated decommissioning costs by a factor of 4 vs. Germany's estimates, and about a factor of 10 vs. UK. France is not the paragon of nuclear efficiency it is commonly portrayed as being. It's not just that NPPs are expensive to build, and unpredictably priced in ways that make the price of power generated uncompetitive. They are also a large and hard to predict liability after they stop generating power and the income from selling that power. There is no example of "this is how to do it." New designs have to emerge and be proven before it is possible to build new NPPs with as much cost certainty as other kinds of power generation.
- fransje26 3y ago
  Probably the same problems that are making (rail) infrastructure building 10x more expensive in the US than in Europe. It's not that the knowledge is inaccessible, the problem is that the not-invented-here syndrome compounded by administrative red-tape, powerful counter lobbies and greedy actors make those projects prohibitively expensive.
- roomey 3y ago
  Well... They only know metric, would be a recipe for disaster!
- godelski 3y ago
  France still hasn't built a reactor in several decades. Latest started in 91 and finished in 02. The problem is that the parent missed some parts of the story.
- AtlasBarfed 3y ago
  Look, the elephant in the room is that if a power company wants to invest a billion dollars in power generation, they are going to do it with solar/wind. The project scales out, the cost is dropping constantly, and they get power generation early in the project. A nuclear power plant is: - a 10 year investment delay (with no return until completed) - could be cancelled at any moment (high risk) - with a very uncertain price target (solar/wind + grid storage will probably be half the cost or less of what it is today) - can't be expanded - very likely to balloon in cost and be a total financial quagmire Solar/wind can be scalably purchased, installed, and expanded as needed. The costs will drop continuously, replacement and maintenance is easy, there's no nuclear waste to get rid of, and can very reliably be specced in terms of cost for generation.
pyrale 3y ago
It was more like 30 years. '70-'85 was only the CP0-1-2 part, which was about 34 reactors. P and N models were built from '78 to 2000. Also that doesn't include design. French design was done in the 60's, and resulted in UNGG prototypes which were abandoned in favor of buying a Westinghouse PWR license. All french reactors are based on that license. Still an amazing feat, considering it's what provides power to France to this day.
eastbound 3y ago
Chronological scale of the construction dates of French reactors: https://fr.quora.com/En-combien-de-temps-construit-t-on-une-centrale-nucl%C3%A9aire https://fr.quora.com/En-combien-de-temps-construit-t-on-une-... We did 6 batches of 6-to-20 reactors.

by barake 3y ago

i_am_proteus 3y ago

Friendly reminder that the US does have such a program (the Navy's nuclear propulsion program), and several reactors are commissioned every year.

MrBodangles 3y ago
One of the core problems with nuclear is the size and scale of risk posed by cost cutting and neglect over its lifetime, especially the waste. People neither trust for-profit enterprises or present/future governments with that sort of responsibility (and with empirically justifiable reasons). I’m aware newer models are stated to be significantly safer, but assurances mean little when these institutions have repeatedly lied and failed in the past. Either way, US carriers are probably one of the safest places for nuclear, as they’re mission critical for the life of the carrier and most likely to receive the utmost care… Plus the US has a long history of rubber stamping virtually unlimited funds to solve any military problem, whether the people approve or not. The handling of the waste is still a major concern, but what about the consequences of a torpedo compromising the reactor in warfare?

Ringz 3y ago

Many claim that France’s 1974 Messmer plan resulted in the building of its 58 reactors in 15 years. This is not true. The planning for several of these nuclear reactors began long before. For example, the Fessenheim reactor obtained its construction permit in 1967 and was planned starting years before. In addition, 10 of the reactors were completed between 1991-2000. As such, the whole planning-to-operation time for these reactors was at least 32 years, not 15. That of any individual reactor was 10 to 19 years.

godelski 3y ago

> in the US we'd be lucky to build and commission a single reactor in that time span Most countries, including the US and France, did a build out in the 70's/80's and then basically stopped. France a bit later than the US, but both essentially did the same thing. Checking the wiki list[0] and sorting by operation year you can see 4 things. 1) the vast majority of reactors were built in the 70's, 2) the newest reactor was built in the 90's (operational 2001), 3) the most recent reactors took longer to go into operation (including a few at 16 years, where the 70's build out was typically 6-7 years), 4) almost all 70s/80's reactors are of the same type and same power level (CP1, CP2, P4 REP 1300). We actually see the exact same story in the US (see Watts Bar, ouch). On the other hand, South Korea didn't do their build out till the mid 80's and continued into the 90's. Then we see the wall hit in the 2000's with the APR 1400. Japan did a bit better and strangely looks like the big success story, especially considering how many reactors such a small country built. Interestingly only Mitsubishi reactors are still operational... Canada is also a good success story but also hasn't built anything since the late 80's (but last reactor was still <10yrs). Countries like Sweden, started their build out but then there was a hard stop. Sweden had nothing past '85. Germany isn't too far off, but it is also a different story. Ditto for UK. I intentionally left out China and Russia because different economic structures and because the stories are a bit different even though might appear similar to what I'm discussing at face value (note that my comments are vastly oversimplified, with some things only being alluded to), but it is worth paying attention to the above patterns and think about how the economic structure might reinforce some of those aspects, then think about the western countries different styles during their build out phases (how it actually worked). The nuclear story is long and complicated. Even this wall of text is oversimplified. This is part of the problem: we like our simple talking points but as speakers are often unwilling to admit that these are only part of the stories or as listeners rebut the speaker as if they are only considering a single factor. It makes real conversation almost impossible and both play a role and build over time. Which is not too dissimilar to a few problems that happened in the nuclear industry. [0] https://en.wikipedia.org/wiki/List_of_commercial_nuclear_reactors#France https://en.wikipedia.org/wiki/List_of_commercial_nuclear_rea...

Gwypaas 3y ago

Lets add Flamanville 3 to the "experience" graph in the article. The only reason it gets pushed through is for France to have an industrial base enabling nuclear submarines, carriers and weapons. https://imgur.com/6G2RBa0 https://imgur.com/6G2RBa0 https://en.wikipedia.org/wiki/Flamanville_Nuclear_Power_Plant https://en.wikipedia.org/wiki/Flamanville_Nuclear_Power_Plan...

adev_ 3y ago
> Lets add Flamanville 3 to the "experience" graph in the article. The only reason it gets pushed through is for France to have an industrial base enabling nuclear submarines, carriers and weapons. That's garbage. - The reason Flamanville 3 takes so much time is precisely because it is a prototype on a new design that never have been produces in series, nor even tested. That supports 100% what is said here: If you want to reduce cost, mass produce. - Submarines and carrier nuclear reactors are completely different beast that have nothing to do with either Flammanville 3 or the existing nuclear park.
- natmaka 3y ago
  > Flamanville 3 takes so much time is precisely because it is a prototype on a new design No, this reactor type (EPR) is a mere evolution. Proof: https://www.irsn.fr/savoir-comprendre/surete/presentation-historique-projet-epr https://www.irsn.fr/savoir-comprendre/surete/presentation-hi...
  - jjk166 3y ago
    It doesn't matter if it's an evolution or a revolution - a change of any sort means new tooling to produce the components, new procedures to assemble them, new analysis to certify that they will all work together, and new training for everyone involved to do all the above.
    - natmaka 3y ago
      All this was known before starting the building process, and the builder planned to deliver the reactor in 2012. It is late (right now: not delivered) mainly because project management and quality insurance were abysmal: https://en.wikipedia.org/wiki/Flamanville_Nuclear_Power_Plant#Unit_3 https://en.wikipedia.org/wiki/Flamanville_Nuclear_Power_Plan...

screye 3y ago

Genuine question. What is the stopping the US from paying every experienced French Nuclear Engineer x 2 x Usual US cost Adjustment --> and letting them build here in the US ? France and US are allies, and likely aligned on climate goals. France has similar climate as many of the planned regions and has fairly high building standards.

Zigurd 3y ago
But not the ones who underestimated decommissioning costs by a factor of 4 vs. Germany's estimates, and about a factor of 10 vs. UK. France is not the paragon of nuclear efficiency it is commonly portrayed as being. It's not just that NPPs are expensive to build, and unpredictably priced in ways that make the price of power generated uncompetitive. They are also a large and hard to predict liability after they stop generating power and the income from selling that power. There is no example of "this is how to do it." New designs have to emerge and be proven before it is possible to build new NPPs with as much cost certainty as other kinds of power generation.
fransje26 3y ago
Probably the same problems that are making (rail) infrastructure building 10x more expensive in the US than in Europe. It's not that the knowledge is inaccessible, the problem is that the not-invented-here syndrome compounded by administrative red-tape, powerful counter lobbies and greedy actors make those projects prohibitively expensive.
roomey 3y ago
Well... They only know metric, would be a recipe for disaster!
godelski 3y ago
France still hasn't built a reactor in several decades. Latest started in 91 and finished in 02. The problem is that the parent missed some parts of the story.
AtlasBarfed 3y ago
Look, the elephant in the room is that if a power company wants to invest a billion dollars in power generation, they are going to do it with solar/wind. The project scales out, the cost is dropping constantly, and they get power generation early in the project. A nuclear power plant is: - a 10 year investment delay (with no return until completed) - could be cancelled at any moment (high risk) - with a very uncertain price target (solar/wind + grid storage will probably be half the cost or less of what it is today) - can't be expanded - very likely to balloon in cost and be a total financial quagmire Solar/wind can be scalably purchased, installed, and expanded as needed. The costs will drop continuously, replacement and maintenance is easy, there's no nuclear waste to get rid of, and can very reliably be specced in terms of cost for generation.

pyrale 3y ago

It was more like 30 years. '70-'85 was only the CP0-1-2 part, which was about 34 reactors. P and N models were built from '78 to 2000. Also that doesn't include design. French design was done in the 60's, and resulted in UNGG prototypes which were abandoned in favor of buying a Westinghouse PWR license. All french reactors are based on that license. Still an amazing feat, considering it's what provides power to France to this day.

eastbound 3y ago

Chronological scale of the construction dates of French reactors: https://fr.quora.com/En-combien-de-temps-construit-t-on-une-centrale-nucl%C3%A9aire https://fr.quora.com/En-combien-de-temps-construit-t-on-une-... We did 6 batches of 6-to-20 reactors.