4 ms·

This is a solid write-up. I particularly like this part: > Most fusion concepts are just a more complicated way to heat water. It's kind of wild that our mean

by jmyeet 4y ago

This is a solid write-up. I particularly like this part: > Most fusion concepts are just a more complicated way to heat water. It's kind of wild that our means of power production do ultimately mostly boil down (pun intended) to turning a turbine with steam and, as the article notes, this is an inherent cost problem whereas solar is a direct form of energy. I hadn't really thought about it in those terms but it's true. It's also kind of wild to consider that we want to heat up hydrogen to a 100 million degrees... to boil water and turn a turbine. I'll also highlight this: > As we've seen, traditional LWRs have a cost problem. That is why the PR ignores costs or focuses only on operating costs. My own view: 1. There are several hundred nuclear power plants in the world. Not one of them has been built without government assistance. This goes to the capital cost issue; 2. Nuclear power plants take too long to plan, build and bring online. IIRC it's at least 11 years; 3. We still don't have a good long-term plan for dealing with processing waste at scale; 4. We still don't have a good long-term plan for dealing with fuel waste at scale; and 5. I just don't trust humans, particularly within the corporate structure, to build and operate nuclear power plants safely. The temptation is simply too high to increase profits by cutting costs.

SemanticStrengh 4y ago
> Nuclear power plants take too long to plan, build and bring online. IIRC it's at least 11 years It's crazy what people come to believe. To generalize the pathetic bureaucratic failures like the EPR to all model couldn't be farther from reality. Russian commercialized VVER 1200 models take 57 month to build. Their successor, the VVER-TOI take 40 month, so 3.5 year to build. It output 1300MWe Besides the russian competitive advantage, China build plants in 6 years.
- polski-g 4y ago
  Yeah these things don't take 11 years to build. Government makes them take that long.
  - nrp 4y ago
    I would be interested in understanding what drives this. Even in Korea, where large scale infrastructure projects like new subway lines get built very quickly, new nuclear reactors seem to take >10 years (both the ones installed domestically in Korea and the export ones in places like UAE).
    - epistasis 4y ago
      Nuclear technology requires ridiculous complexity, with miles upon miles of piping that need precision, accurate welds to last 40 years. It requires high performance concrete. It requires huge logistics to get all the concrete pours and rebar and everything in place. These things are massive projects, utterly massive. There's no reason to think they should be cheap.
  - epistasis 4y ago
    That's absolutely false, check out what has happened in Georgia and South Carolina and you will not be able to find any government interference or regulation slowing things down. It's just rank incompetence, bad management, and three-party contracts where each party is preparing for failure and a gigantic lawsuit in the end. Meanwhile, the latest excuse for the recent US failure in building was that they commenced building before design was complete, and if we just try again now that designs are complete, everything will proceed smoothly. Meanwhile others in this thread are complaining that the NRC requires complete plans before building. Those who complain about imaginary regulations can neither state the regulations that slow things down, nor describe suitable regulations, nor apparently even agree on what the current regulations are. I've been asking "which regulations" for years, nobody has ever, literally ever pointed me to which. They just have supreme confident that it's the answer, despite not knowing specifics.
    - SemanticStrengh 4y ago
      > I've been asking "which regulations" for years, nobody has ever, literally ever pointed me to which. Well it looks like you have been asking in the wrong places: https://whatisnuclear.com/economics.html#public-concern-grows-and-so-do-regulatory-guides-and-costs https://whatisnuclear.com/economics.html#public-concern-grow...
- jltsiren 4y ago
  Finland was trying to get one VVER 1200 built, with an option to build another later. The political approval for the project was given in 2010 and the contract with Rosatom was signed in 2014. In the following 8 years, Rosatom failed to produce the blueprints and other documents required for the actual construction permit. Meanwhile, Russia bombed some factories in Ukraine that were supposed to build key parts for the power plant, and it's looking increasingly unlikely that the construction will ever start. In contrast, the EPR reactor built by Areva only needed 3 years from the political approval to the construction permit. While Rosatom is probably more competent in the actual construction, they seem to have major issues in operating in countries where the regulatory authorities are strict (but still pro-nuclear).
- epistasis 4y ago
  > It's crazy what people come to believe. It's not crazy to believe the reality of what is actually happening in the US and Europe Assuming that the US and Europe can build the way that Russia or China can might be crazy though.
nine_k 4y ago
There is a known way to handle "nuclear waste": burn it in fast-neutron reactors [1]. The waste is mostly still uranium. The rest can still be burned, turning it into much less active isotopes and producing energy. Yes, these machines produce plutonium suitable for nuclear weapons. But plutonium is a perfectly good nuclear fuel, too, and the US already has nuclear weapons, longer than any other country. [1]: https://whatisnuclear.com/fast-reactor.html https://whatisnuclear.com/fast-reactor.html
- pfdietz 4y ago
  That only handles (some) actinides. It doesn't handle fission products, seven of which have very long half lives.
  - nine_k 4y ago
    Still, much less to bury, and the resulting waste is much less active.
  - acidburnNSA 4y ago
    You can partition the LLFPs into special irradiation targets and transmute them to short-lived things if you really want to, also in fast reactors. I'm still a bigger fan of stable geologic disposal. https://www.iaea.org/publications/7112/implications-of-partitioning-and-transmutation-in-radioactive-waste-management https://www.iaea.org/publications/7112/implications-of-parti...
    - pfdietz 4y ago
      No one has done that. It's not clear it can be done reasonably or economically. You'd have to isotopically separate the LLFPs from stable isotopes of the same elements. I will say that if those LLFP can be separated, it's probably simpler and cheaper to just shoot them into space for disposal.
- Manuel_D 4y ago
  Or, bury it underground in impermeable rock. Short of a meteor strike right on top of the repository, it will not make its way out to the surface.
pfdietz 4y ago
Among the fusion concepts that does not heat water is Helion's. For that reason (and some others) I consider it the least dubious of the various commercial fusion approaches.
ClumsyPilot 4y ago
> 3. We still don't have a good long-term plan for dealing with processing waste at scale; Who is 'we' that should come up with the plan? Government? Most westerners don't like government planning - your own post considers government involvement as a downside. We expect the market to manage most things, but markets don't come up with long term plans, that's not their function. If you come with a subject at random and ask, 'do we have a long term plan for X', generally the answer is no. We don't have a good term long-term plan for dealing with global warming, the oil running out, falling birthrates for some nations, rising house prices, or for anything really.
- webmaven 4y ago
  > Who is 'we' that should come up with the plan? Government? Most westerners don't like government planning - your own post considers government involvement as a downside. I think an even broader we was implied, that of human civilization as a whole.
  - ClumsyPilot 4y ago
    Okay, is it possible to answer my question - what aspect of life does Human Civilisation as a whole have a plan for? Because if there is no area of life that we can indetify a plan for, then the whole criticism doesn't make any sence.
    - webmaven 4y ago
      Right now, collectively we are optimizing for return on capital on a sliding scale, those who have the most capital get the highest returns (or if not a higher return, then lower risk for the same return). Over longer time frames we see the pattern failing periodically when inequality gets too severe resulting in societal breakdown in the form of wars or revolutions that end up redistributing assets through various means, including less direct ones like increased labor costs, no matter who "wins".
Chris2048 4y ago
> 3. We still don't have a good long-term plan for dealing with processing waste at scale; > 4. We still don't have a good long-term plan for dealing with fuel waste at scale; and Because there isn't enough waste produced to make such a thing economical versus current storage method i.e. there isn't enough economy of scale. > 5. I just don't trust humans, particularly within the corporate structure, to build and operate nuclear power plants safely Fukushima was built despite studies warning of the dangers. At some higher level, there wasn't enough caution in the planning stage.

by jmyeet 4y ago

SemanticStrengh 4y ago

> Nuclear power plants take too long to plan, build and bring online. IIRC it's at least 11 years It's crazy what people come to believe. To generalize the pathetic bureaucratic failures like the EPR to all model couldn't be farther from reality. Russian commercialized VVER 1200 models take 57 month to build. Their successor, the VVER-TOI take 40 month, so 3.5 year to build. It output 1300MWe Besides the russian competitive advantage, China build plants in 6 years.

polski-g 4y ago
Yeah these things don't take 11 years to build. Government makes them take that long.
- nrp 4y ago
  I would be interested in understanding what drives this. Even in Korea, where large scale infrastructure projects like new subway lines get built very quickly, new nuclear reactors seem to take >10 years (both the ones installed domestically in Korea and the export ones in places like UAE).
  - epistasis 4y ago
    Nuclear technology requires ridiculous complexity, with miles upon miles of piping that need precision, accurate welds to last 40 years. It requires high performance concrete. It requires huge logistics to get all the concrete pours and rebar and everything in place. These things are massive projects, utterly massive. There's no reason to think they should be cheap.
- epistasis 4y ago
  That's absolutely false, check out what has happened in Georgia and South Carolina and you will not be able to find any government interference or regulation slowing things down. It's just rank incompetence, bad management, and three-party contracts where each party is preparing for failure and a gigantic lawsuit in the end. Meanwhile, the latest excuse for the recent US failure in building was that they commenced building before design was complete, and if we just try again now that designs are complete, everything will proceed smoothly. Meanwhile others in this thread are complaining that the NRC requires complete plans before building. Those who complain about imaginary regulations can neither state the regulations that slow things down, nor describe suitable regulations, nor apparently even agree on what the current regulations are. I've been asking "which regulations" for years, nobody has ever, literally ever pointed me to which. They just have supreme confident that it's the answer, despite not knowing specifics.
  - SemanticStrengh 4y ago
    > I've been asking "which regulations" for years, nobody has ever, literally ever pointed me to which. Well it looks like you have been asking in the wrong places: https://whatisnuclear.com/economics.html#public-concern-grows-and-so-do-regulatory-guides-and-costs https://whatisnuclear.com/economics.html#public-concern-grow...
jltsiren 4y ago
Finland was trying to get one VVER 1200 built, with an option to build another later. The political approval for the project was given in 2010 and the contract with Rosatom was signed in 2014. In the following 8 years, Rosatom failed to produce the blueprints and other documents required for the actual construction permit. Meanwhile, Russia bombed some factories in Ukraine that were supposed to build key parts for the power plant, and it's looking increasingly unlikely that the construction will ever start. In contrast, the EPR reactor built by Areva only needed 3 years from the political approval to the construction permit. While Rosatom is probably more competent in the actual construction, they seem to have major issues in operating in countries where the regulatory authorities are strict (but still pro-nuclear).
epistasis 4y ago
> It's crazy what people come to believe. It's not crazy to believe the reality of what is actually happening in the US and Europe Assuming that the US and Europe can build the way that Russia or China can might be crazy though.

nine_k 4y ago

There is a known way to handle "nuclear waste": burn it in fast-neutron reactors [1]. The waste is mostly still uranium. The rest can still be burned, turning it into much less active isotopes and producing energy. Yes, these machines produce plutonium suitable for nuclear weapons. But plutonium is a perfectly good nuclear fuel, too, and the US already has nuclear weapons, longer than any other country. [1]: https://whatisnuclear.com/fast-reactor.html https://whatisnuclear.com/fast-reactor.html

pfdietz 4y ago
That only handles (some) actinides. It doesn't handle fission products, seven of which have very long half lives.
- nine_k 4y ago
  Still, much less to bury, and the resulting waste is much less active.
- acidburnNSA 4y ago
  You can partition the LLFPs into special irradiation targets and transmute them to short-lived things if you really want to, also in fast reactors. I'm still a bigger fan of stable geologic disposal. https://www.iaea.org/publications/7112/implications-of-partitioning-and-transmutation-in-radioactive-waste-management https://www.iaea.org/publications/7112/implications-of-parti...
  - pfdietz 4y ago
    No one has done that. It's not clear it can be done reasonably or economically. You'd have to isotopically separate the LLFPs from stable isotopes of the same elements. I will say that if those LLFP can be separated, it's probably simpler and cheaper to just shoot them into space for disposal.
Manuel_D 4y ago
Or, bury it underground in impermeable rock. Short of a meteor strike right on top of the repository, it will not make its way out to the surface.

pfdietz 4y ago

Among the fusion concepts that does not heat water is Helion's. For that reason (and some others) I consider it the least dubious of the various commercial fusion approaches.

ClumsyPilot 4y ago

> 3. We still don't have a good long-term plan for dealing with processing waste at scale; Who is 'we' that should come up with the plan? Government? Most westerners don't like government planning - your own post considers government involvement as a downside. We expect the market to manage most things, but markets don't come up with long term plans, that's not their function. If you come with a subject at random and ask, 'do we have a long term plan for X', generally the answer is no. We don't have a good term long-term plan for dealing with global warming, the oil running out, falling birthrates for some nations, rising house prices, or for anything really.

webmaven 4y ago
> Who is 'we' that should come up with the plan? Government? Most westerners don't like government planning - your own post considers government involvement as a downside. I think an even broader we was implied, that of human civilization as a whole.
- ClumsyPilot 4y ago
  Okay, is it possible to answer my question - what aspect of life does Human Civilisation as a whole have a plan for? Because if there is no area of life that we can indetify a plan for, then the whole criticism doesn't make any sence.
  - webmaven 4y ago
    Right now, collectively we are optimizing for return on capital on a sliding scale, those who have the most capital get the highest returns (or if not a higher return, then lower risk for the same return). Over longer time frames we see the pattern failing periodically when inequality gets too severe resulting in societal breakdown in the form of wars or revolutions that end up redistributing assets through various means, including less direct ones like increased labor costs, no matter who "wins".

Chris2048 4y ago

> 3. We still don't have a good long-term plan for dealing with processing waste at scale; > 4. We still don't have a good long-term plan for dealing with fuel waste at scale; and Because there isn't enough waste produced to make such a thing economical versus current storage method i.e. there isn't enough economy of scale. > 5. I just don't trust humans, particularly within the corporate structure, to build and operate nuclear power plants safely Fukushima was built despite studies warning of the dangers. At some higher level, there wasn't enough caution in the planning stage.