4 ms·

Nuclear engineer here: These plants need on-site diesel generators to provide site-power in case of loss of off-site power (grid failure). Here the operator ha

Nuclear engineer here: These plants need on-site diesel generators to provide site-power in case of loss of off-site power (grid failure). Here the operator had leaks in fuel lines that fed both the A and B generators. These are the only generators of on-site emergency power. An operating reactor needs approximately 2-5 days to cool due to build-up of radioactive decay products in the core, which continue to generate heat even after the reactor has been fully shutdown (in the case of an emergency, a sudden shutdown of the reactor is called a SCRAM). On-site power serves the primary purpose of driving core-recirculating coolant loops to control the core temperature. Chief metrics in reactor safety generally relate back to core-damage-frequency (CDF), because when one damages the fuel-casing or core, this is the most-likely precursor to releasing radioactivity. CDF determinations are almost directly linked to loss-of-coolant-accidents (LOCAs) which imply that one can no longer cool the nuclear fuel core or elements. Total station blackout, which is a result of loss of off-site and on-site power, combined with a SCRAM is essentially the scenarios that led to Fukushima and Chernobyl disasters. The NRC will not treat this lightly.

AnthonyMouse 3y ago
These reactors are all the 20th century designs that require active cooling after a SCRAM. There are newer designs that can do this with passive cooling and thereby don't require power, aren't there? How come we don't build those?
- AngryData 3y ago
  Yes there are newer designs that are suppose to be passively safe. But of course the money and government approval needed to build them, especially an as of yet untested design, are all locked behind political doors and hurdles.
  - scoobydooxp 3y ago
    The US currently has one AP1000 I believe and one under construction (see Vogtle 3&4). Other countries are building them. China for example has AP1000's and CAP1000's/CAP1400's (their modified version of the AP1000) in use and being built. The AP1000 design is a generation III+ reactor which is safer than 99% of the existing plants in the U.S.
- omgJustTest 3y ago
  New reactor designs have a serious approval burden via the NRC. Almost all of the reactor designs that are in operation are plants that were designed pre-1970. After the Atomic Energy Commission was disbanded the Nuclear Regulatory Commission was formed. The NRC has had an ultra-conservative policy against new technologies, creating a chicken-and-egg-problem for new tech: a tech must be proven in a relevant nuclear environment before it can be approved by the NRC, but it cannot be proven without installing it in a relevant environment unless it is already in a grandfathered reactor design. Additionally while there are a lot of interesting ideas, simulations and paper-reactors, "approved" real reactors face large capital costs (~$5bn / 1 GWe) as well as long build times (~10-20years). The real crux of this problem is that the utility cannot get guarantees that when it builds a reactor that it will be allowed to operate it. At the end of a build, there is a licensure process that must pass a community comment period. The political landscape 10-20years in a community's future can change dramatically, leaving a utility with a huge gamble that can result in brand-new nuclear-reactors, which are blessed by the NRC, to be shuttered.
  - dr_kiszonka 3y ago
    Thanks for sharing your expertise! The first Google snippet had me think you can build a nuclear powerplant in six years, but you are right [1,2]. Why does it take so long, say, 15 years? Is it simply the difficulty? Supply chain issues? Regulation? Changes in the political climate?Something else? 1. https://www.statista.com/statistics/1333678/nuclear-plants-construction-time/ https://www.statista.com/statistics/1333678/nuclear-plants-c... 2. https://www.world-nuclear.org/gallery/world-nuclear-performance-report-gallery/median-construction-times-for-reactors-since-1981.aspx https://www.world-nuclear.org/gallery/world-nuclear-performa...
    - omgJustTest 3y ago
      The timeline is unique to the US and the NRC. China can and has recently constructed plants in very short build timelines (https://en.wikipedia.org/wiki/Ningde_Nuclear_Power_Plant https://en.wikipedia.org/wiki/Ningde_Nuclear_Power_Plant, for example has had many installations over not that much time). Vogtle-3 and 4 are the first nuclear reactors to come into commercial market in like 20-30 years. https://en.wikipedia.org/wiki/Vogtle_Electric_Generating_Plant https://en.wikipedia.org/wiki/Vogtle_Electric_Generating_Pla..., permits started in 2008 and finally operation will happen this year. Westinghouse, the designer, went bankrupt in 2017 and the utility was given at least $12bn in _federal_ loan guarantees to help the reactors be completed. These two units cost $15bn each. A lot of operators blame construction timelines on community comment periods, though I suspect that like every other large capital investment in the US, the timelines are extended by mistakes, regulatory oversight / indecisiveness and some misaligned incentives of the government, construction companies and utilities.
- cwmma 3y ago
  Fukushima had passive cooling, but for various reasons it wasn't operating during the disaster, so it's still very important to have these even if you have passive cooling.
jona-f 3y ago
> An operating reactor needs approximately 2-5 days to cool due to build-up of radioactive decay products in the core,... Wait, aren't there cooling pools for spent rods that take years to cool down? What do you mean by "cool"? Fukushima made it rather clear that reactors don't cool down in 2-5 days. Looks to me the copium is still high in the field.

AnthonyMouse 3y ago

These reactors are all the 20th century designs that require active cooling after a SCRAM. There are newer designs that can do this with passive cooling and thereby don't require power, aren't there? How come we don't build those?

AngryData 3y ago
Yes there are newer designs that are suppose to be passively safe. But of course the money and government approval needed to build them, especially an as of yet untested design, are all locked behind political doors and hurdles.
- scoobydooxp 3y ago
  The US currently has one AP1000 I believe and one under construction (see Vogtle 3&4). Other countries are building them. China for example has AP1000's and CAP1000's/CAP1400's (their modified version of the AP1000) in use and being built. The AP1000 design is a generation III+ reactor which is safer than 99% of the existing plants in the U.S.
omgJustTest 3y ago
New reactor designs have a serious approval burden via the NRC. Almost all of the reactor designs that are in operation are plants that were designed pre-1970. After the Atomic Energy Commission was disbanded the Nuclear Regulatory Commission was formed. The NRC has had an ultra-conservative policy against new technologies, creating a chicken-and-egg-problem for new tech: a tech must be proven in a relevant nuclear environment before it can be approved by the NRC, but it cannot be proven without installing it in a relevant environment unless it is already in a grandfathered reactor design. Additionally while there are a lot of interesting ideas, simulations and paper-reactors, "approved" real reactors face large capital costs (~$5bn / 1 GWe) as well as long build times (~10-20years). The real crux of this problem is that the utility cannot get guarantees that when it builds a reactor that it will be allowed to operate it. At the end of a build, there is a licensure process that must pass a community comment period. The political landscape 10-20years in a community's future can change dramatically, leaving a utility with a huge gamble that can result in brand-new nuclear-reactors, which are blessed by the NRC, to be shuttered.
- dr_kiszonka 3y ago
  Thanks for sharing your expertise! The first Google snippet had me think you can build a nuclear powerplant in six years, but you are right [1,2]. Why does it take so long, say, 15 years? Is it simply the difficulty? Supply chain issues? Regulation? Changes in the political climate?Something else? 1. https://www.statista.com/statistics/1333678/nuclear-plants-construction-time/ https://www.statista.com/statistics/1333678/nuclear-plants-c... 2. https://www.world-nuclear.org/gallery/world-nuclear-performance-report-gallery/median-construction-times-for-reactors-since-1981.aspx https://www.world-nuclear.org/gallery/world-nuclear-performa...
  - omgJustTest 3y ago
    The timeline is unique to the US and the NRC. China can and has recently constructed plants in very short build timelines (https://en.wikipedia.org/wiki/Ningde_Nuclear_Power_Plant https://en.wikipedia.org/wiki/Ningde_Nuclear_Power_Plant, for example has had many installations over not that much time). Vogtle-3 and 4 are the first nuclear reactors to come into commercial market in like 20-30 years. https://en.wikipedia.org/wiki/Vogtle_Electric_Generating_Plant https://en.wikipedia.org/wiki/Vogtle_Electric_Generating_Pla..., permits started in 2008 and finally operation will happen this year. Westinghouse, the designer, went bankrupt in 2017 and the utility was given at least $12bn in _federal_ loan guarantees to help the reactors be completed. These two units cost $15bn each. A lot of operators blame construction timelines on community comment periods, though I suspect that like every other large capital investment in the US, the timelines are extended by mistakes, regulatory oversight / indecisiveness and some misaligned incentives of the government, construction companies and utilities.
cwmma 3y ago
Fukushima had passive cooling, but for various reasons it wasn't operating during the disaster, so it's still very important to have these even if you have passive cooling.

jona-f 3y ago

> An operating reactor needs approximately 2-5 days to cool due to build-up of radioactive decay products in the core,... Wait, aren't there cooling pools for spent rods that take years to cool down? What do you mean by "cool"? Fukushima made it rather clear that reactors don't cool down in 2-5 days. Looks to me the copium is still high in the field.