4 ms·

India plans to triple its nuclear capacity by 2024 to wean Asia's third-largest economy off polluting fossil fuels. That's a really short timeline. India's new

India plans to triple its nuclear capacity by 2024 to wean Asia's third-largest economy off polluting fossil fuels. That's a really short timeline. India's newest reactor Kudankulam 2 took 14 years from construction start to first criticality: https://pris.iaea.org/PRIS/CountryStatistics/ReactorDetails.aspx?current=853 https://pris.iaea.org/PRIS/CountryStatistics/ReactorDetails.... The only American reactors currently under construction, Vogtle 3 and 4, were originally supposed to be built in less than 5 years but are now projected to take 8 years. But maybe the American reactors aren't intended to be part of the generation added by 2024. India does have 7 reactors currently under construction.

roenxi 8y ago
2024 is an eyewatering timeframe, but the actual construction might not be the hardest part. The Chinese have figured out how to do prefabbed skyscrapers in 19 days [0] and the prefabbing is probably highly parallel. If a nuclear plant is 100 times harder than a skyscraper to build that is still only about 5 years as a lower bound. If they've already done the scoping, fundraising, location picking and had a construction firm who optimised for speed & quality but not cost then I can see ~5 years as achievable, although they probably won't manage it. This is one of nuclear many advantages over more usual nonrenewable, things can happen orders of magnitude more quickly because the physical footprint is so small. [0] https://www.theguardian.com/world/2015/apr/30/chinese-construction-firm-erects-57-storey-skyscraper-in-19-days https://www.theguardian.com/world/2015/apr/30/chinese-constr...
- jabl 8y ago
  AFAIK one limitation is forging of the large pressure vessels, mainly the reactor itself and the steam generators. Those take a long time. Of course you can do multiple ones in parallel, but then the limit is that there's about 25 heavy forge presses on the planet, of which only a subset are doing nuclear work. So in total the capacity at the moment is maybe 50 large pressure vessels per year worldwide. If one plant requires one reactor and 4 steam generators that would then be 10 plants per year. Of course it's possible to increase the worldwide heavy forging capacity, but again, something that takes time. Then there are many reactor designs that don't use water as a coolant or moderator; those are obviously not restricted by heavy forge capacity. But those reactors are mostly on the drawing board and not available off the shelf.
  - ohyes 8y ago
    The quality of the steel is the hard part of this. It is incredibly important to not have any defects in the pressure systems & turbine of a nuclear plant for obvious reasons.
  - mikemotherwell 8y ago
    If Nuclear was committed to, would the number of these heavy forge presses likely go up, and costs come down? I have no idea, but it does seem that if there are bottlenecks in the process, demand for nuclear could bring costs come down dramatically. Am I missing something here?
    - jbay808 8y ago
      Those heavy presses themselves seem to be nation-state level engineering projects, so it's unlikely that anyone would build more of them than they absolutely need. https://youtu.be/hpgK51w6uhk https://youtu.be/hpgK51w6uhk
    - jabl 8y ago
      Oh, absolutely! Or well, maybe costs wouldn't come down that much, since AFAIK press forging is a pretty mature technology. I'm just saying a heavy press forge is a major infrastructure project on its own, and it won't just happen by snapping our fingers. It will take time. Then there's the question of whether it makes sense to embark on a major buildout of current style large light water reactors? If the market shifts towards smaller modular reactors (SMR's), then for light water SMR's smaller forges would likely do. Or further in the future non-water low pressure designs. My opinion is that given the urgency of addressing climate change, coal pollution etc., yes we should embark on a major buildout of currently available reactor types, while also massively increasing advanced reactor R&D&commercialization funding.
  - nnq 8y ago
    Why would forging be a requirement? Hasn't material science reached a point where we can built alloys that can be just cast into the right shape? Can't the problem of them being slightly weaker be solved by just making them thicker (not a problem for nuclear I'd guess bc you don't build that many pieces and you don't need to move them much), or by just having stuff like turbines replaced more often? I know that mixing "nuclear" + "cheaper manufacturing" sounds scary, but maybe it's a better idea... why would you need "best of the best" materials for pieces that (1) stay in one place (they're not airplanes!) and (2) can be replaced/serviced/inspected on fixed schedules following fixed procedures on fixed budgets etc. Modern forges were perfected by the germans in the war because they wanted to build airplane parts (!!) and they wanter to build them fast... I'd imagine nuclear power technology has none of this requirements and can be re-designed to work with commodity low-quality parts and have the movable pieces easily replace-able... (I'd imagine that if people would get over the "uuh, nuclear, scarry, scarry" mindset nuclear technology could become a dirt cheap commodity and be build with a mixture of ceramics + cheap alloys just fine...)
mikeyouse 8y ago
Vogtle is even worse than that... Ground works started in 2009/2010 -- Best case scenario right now is first power by 2022/2023 and if they stay on budget from here, cost will be 100% more than first estimated. So much good money being thrown after bad. https://www.ajc.com/business/southern-says-vogtle-costs-exceed-25b/mvP3M8H3Db3Kw4RNMB2n0O/ https://www.ajc.com/business/southern-says-vogtle-costs-exce...

roenxi 8y ago

2024 is an eyewatering timeframe, but the actual construction might not be the hardest part. The Chinese have figured out how to do prefabbed skyscrapers in 19 days [0] and the prefabbing is probably highly parallel. If a nuclear plant is 100 times harder than a skyscraper to build that is still only about 5 years as a lower bound. If they've already done the scoping, fundraising, location picking and had a construction firm who optimised for speed & quality but not cost then I can see ~5 years as achievable, although they probably won't manage it. This is one of nuclear many advantages over more usual nonrenewable, things can happen orders of magnitude more quickly because the physical footprint is so small. [0] https://www.theguardian.com/world/2015/apr/30/chinese-construction-firm-erects-57-storey-skyscraper-in-19-days https://www.theguardian.com/world/2015/apr/30/chinese-constr...

jabl 8y ago
AFAIK one limitation is forging of the large pressure vessels, mainly the reactor itself and the steam generators. Those take a long time. Of course you can do multiple ones in parallel, but then the limit is that there's about 25 heavy forge presses on the planet, of which only a subset are doing nuclear work. So in total the capacity at the moment is maybe 50 large pressure vessels per year worldwide. If one plant requires one reactor and 4 steam generators that would then be 10 plants per year. Of course it's possible to increase the worldwide heavy forging capacity, but again, something that takes time. Then there are many reactor designs that don't use water as a coolant or moderator; those are obviously not restricted by heavy forge capacity. But those reactors are mostly on the drawing board and not available off the shelf.
- ohyes 8y ago
  The quality of the steel is the hard part of this. It is incredibly important to not have any defects in the pressure systems & turbine of a nuclear plant for obvious reasons.
- mikemotherwell 8y ago
  If Nuclear was committed to, would the number of these heavy forge presses likely go up, and costs come down? I have no idea, but it does seem that if there are bottlenecks in the process, demand for nuclear could bring costs come down dramatically. Am I missing something here?
  - jbay808 8y ago
    Those heavy presses themselves seem to be nation-state level engineering projects, so it's unlikely that anyone would build more of them than they absolutely need. https://youtu.be/hpgK51w6uhk https://youtu.be/hpgK51w6uhk
  - jabl 8y ago
    Oh, absolutely! Or well, maybe costs wouldn't come down that much, since AFAIK press forging is a pretty mature technology. I'm just saying a heavy press forge is a major infrastructure project on its own, and it won't just happen by snapping our fingers. It will take time. Then there's the question of whether it makes sense to embark on a major buildout of current style large light water reactors? If the market shifts towards smaller modular reactors (SMR's), then for light water SMR's smaller forges would likely do. Or further in the future non-water low pressure designs. My opinion is that given the urgency of addressing climate change, coal pollution etc., yes we should embark on a major buildout of currently available reactor types, while also massively increasing advanced reactor R&D&commercialization funding.
- nnq 8y ago
  Why would forging be a requirement? Hasn't material science reached a point where we can built alloys that can be just cast into the right shape? Can't the problem of them being slightly weaker be solved by just making them thicker (not a problem for nuclear I'd guess bc you don't build that many pieces and you don't need to move them much), or by just having stuff like turbines replaced more often? I know that mixing "nuclear" + "cheaper manufacturing" sounds scary, but maybe it's a better idea... why would you need "best of the best" materials for pieces that (1) stay in one place (they're not airplanes!) and (2) can be replaced/serviced/inspected on fixed schedules following fixed procedures on fixed budgets etc. Modern forges were perfected by the germans in the war because they wanted to build airplane parts (!!) and they wanter to build them fast... I'd imagine nuclear power technology has none of this requirements and can be re-designed to work with commodity low-quality parts and have the movable pieces easily replace-able... (I'd imagine that if people would get over the "uuh, nuclear, scarry, scarry" mindset nuclear technology could become a dirt cheap commodity and be build with a mixture of ceramics + cheap alloys just fine...)

mikeyouse 8y ago

Vogtle is even worse than that... Ground works started in 2009/2010 -- Best case scenario right now is first power by 2022/2023 and if they stay on budget from here, cost will be 100% more than first estimated. So much good money being thrown after bad. https://www.ajc.com/business/southern-says-vogtle-costs-exceed-25b/mvP3M8H3Db3Kw4RNMB2n0O/ https://www.ajc.com/business/southern-says-vogtle-costs-exce...