7 ms·

Every SMR startup is failing. The more they progress, the more they revise their costs upwards. SMR make as much sense as space datacenters. You can gaslight i

by Arodex 4mo ago

Every SMR startup is failing. The more they progress, the more they revise their costs upwards. SMR make as much sense as space datacenters. You can gaslight investors, you can gaslight HN, you can gaslight a national parliament full of lobbyists, but you can't gaslight thermodynamics.

RealityVoid 4mo ago
How are SMR's "gaslighting themodynamics"? I mean, sure, I can accept that they're not economical with current tech, but it's not a frigging' perpetuum mobile, it's feasible technology.
- Arodex 4mo ago
  Thermodynamics are the reason why SMR aren't, and will never, be economical. A bigger nuclear reactor will always undercut your price per watt.
  - holoduke 4mo ago
    The mindset that makes people stuck in time. Sorry but SMRs are potentially very cheap. Not at this point. ,but when operated on scale they will be. You need to start
    - dalyons 4mo ago
      This is just faith without evidence
      - RealityVoid 4mo ago
        I mean... you've got to have faith in your theory before testing it out. I am not having any opinion here about this, but the cycle in my mind is theory, belief, test, updated model of reality. I can imagine similar things said before we managed to have powered flight, "Pshhhsst! We'll never fly! The law of gravity forbids it." You do need to have unreasonable goals for things once in a while. In this case, I don't have a particularly fixed stance about SMR's, but the claim that "Thermodynamics are the reason why SMR aren't, and will never, be economical" feels a bit stronger than it is warranted. Never is a damn' looooong time. And I can easily imagine things being less efficient, but having other advantages that make them more economical. Claiming it's impossible is just stretching.
  - chickenbig 4mo ago
    > Thermodynamics are the reason why SMR aren't, and will never, be economical. And the link between thermodynamics and the price of electricity is what?
    - Arodex 4mo ago
      Your small nuclear reactor is going to need almost as much engineering , plumbing, safety mechanism, personnel, maintenance, etc... as your big nuclear reactor.
      - 0-_-0 4mo ago
        Not if it's mass produced
        palata 4mo ago
        Let's distribute the risk everywhere, what could go wrong?
        chickenbig 4mo ago
        Siting of an SMR is somewhat different (albeit related) to the SMR concept itself. You might cluster them together (like the plans for 3 RR SMR at Wylfa and 3 at Ringhals in Sweden).
        crote 4mo ago
        Mass-produced reactors don't need operators or guards?
      - chickenbig 4mo ago
        > almost as much engineering , plumbing, safety mechanism, personnel, maintenance, etc Sure, that is economics, not thermodynamics. I don't necessarily agree with the SMR manifesto, but it is conceivable that improved financing, construction, operation and oversight could make an SMR cheaper than a larger reactor.
  - Manuel_D 4mo ago
    The value propositions of SMRs are logistics and re-use of existing infrastructure. The idea is that you could have easily transportable reactors that you can plop down in an existing coal plant, and then reuse the turbine, dynamo, etc. that are already in place. The fact that we haven't seen more widespread use of SMRs suggests that you're right. But it's important to point out that there are cost saving opportunities that could potentially reduce the net price per watt despite worse thermodynamic efficiency.
    - crote 4mo ago
      But they'll never be small enough to be truly portable. It'll be closer to another Akademik Lomonosov [0] than a truck-sized diesel generator, which severely limits its deployment options. The additional per-site engineering required to reuse things like turbines and dynamos is almost certainly going to kill any savings it would have. If you're already shipping a building-sized reactor, what's one more turbine? Realistically the main reusable component is the grid hookup itself - but that would incentivize building a large-scale reactor on the site. As would reusing the turbines, for that matter: you can't exactly power the turbines of a 100MW coal plant with a 10MW reactor, and shipping ten inefficient 10MW reactors to the site just so you can reuse the existing ancient turbine isn't exactly an attractive option either. [0]: https://en.wikipedia.org/wiki/Akademik_Lomonosov https://en.wikipedia.org/wiki/Akademik_Lomonosov
2muchcoffeeman 4mo ago
Doesn’t China have SMRs?
- hangonhn 4mo ago
  Just one so far and it's not particularly small compared to a more conventional reactor. Russia actually does have a smallish SMR but it wasn't terribly cheap to build nor operate. IIRC it is in the form of a ship and used to power a city somewhere in the north. SMR has a place for sure but no one has demonstrated the unit costs savings of making a lot of them yet. You can actually get some, if not most, of the economy of scale by doing a fleet build of one specific design. The US seems to be working on that and picked the Westinghouse AP-1000. I think that initiative has a decent chance of succeeding. The first few will be slow and expensive to build (even China has had delays with their nuclear roll out) but the subsequent ones will get cheaper and faster to build. This is how some countries did it during the first nuclear power expansion era.
john_strinlai 4mo ago
>SMR make as much sense as space datacenters. you are in this thread a lot, so i am guessing you must be very familiar with the industry. maybe you can help me understand: is the wikipedia on SMRs incorrect/lying when they say that there are commercially operating SMRs since 2020? and how have so many smart people and companies been duped into seriously considering SMR technology if SMRs apparently break the laws of thermodynamics?
- Arodex 4mo ago
  >commercially operating And struggling, propped up by taylor-made laws and public money. >how have so many smart people and companies been duped into seriously considering SMR technology if SMRs apparently break the laws of thermodynamics? Never said they break the laws of thermodynamics. They are just inefficient and will never be more efficient than alternatives such as... Bigger nuclear reactors. Or solar. And how long have you been out there? Have you never seen investors dumping and wasting billions in dead-ends? Never seen a mania before? Nuclear attracts clever people, but it isn't smart nor wise.
  - jkman 4mo ago
    >'taylor-made' Says it all, doesn't it
  - vablings 4mo ago
    I think you have a misunderstanding of what a SMR is supposed to be. Nuclear power plants are eye watering levels of expensive. The require massive scale and cost with lengthy approvals and requirements, the fundamental idea of SMRs is to move that cost and approvals into a smaller scale so that multiple standard units can be produced and deployed in a turnkey situation, they still will be expensive but the time to deploy and cost will be significantly reduced. We also know SMRs work very well, considering the majority of the US Navy is powered entirely with SMRs and have been for a very long time. Off the top of my head ship power has been exported to local areas for disaster relief Solar is absolutely fantastic and your average person should not be hawking at solar for your home to offset your power bill. The problem with solar is that you need power 24/7 and solar will not make power in the night. I don't think the likes of Westinghouse, Siemens, Rolls Royce and GE are duped. They are trying to solve a very hard problem!
bakies 4mo ago
What are the challenges they face?
- redsocksfan45 4mo ago
  [dead]
- Manuel_D 4mo ago
  One is regulatory. At least in the US, every nuclear reactor that produces at least 100 MW needs to carry a 375 million dollar insurance policy at minimum. Under 100 MW there is an alternate schedule that ranges from 5 million to 75 million scaling based on output. But the net result is that it's still more profitable to built a single large reactor, since a 1 GW reactor is less to insure than 10 100 MW reactors. This is written into law, it would require Congress to change it. Second is that nuclear reactor efficiency tends to improve with size. The ratio of thermal watts to electric watts tends to be better with large reactors. I'm not super well versed on the engineering tradeoffs here by my rough understanding is that waste heat scales with surface area while useful energy extraction scales with volume.
  - projektfu 4mo ago
    In the grand scheme of things, that doesn't seem to be too expensive. According to the NRC, insurance is per site, and additional reactors on a site don't increase the insurance nearly as much as the first. The example they give is $1.1 million annual premium for a $500 million policy, with multi-reactor sites going up to $1.5 million. They also mention property insurance ($1.06 billion policy) and they didn't discuss the premiums on that, but as it's not liability insurance it's probably cheaper. The big costs are still going to be the cost of siting and building the reactor, the fuel, and the ongoing cost of running it. They pay off over a very long time horizon, so it's also the opportunity cost. https://www.nrc.gov/reading-rm/doc-collections/fact-sheets/nuclear-insurance https://www.nrc.gov/reading-rm/doc-collections/fact-sheets/n...
mpweiher 4mo ago
> Every SMR startup is failing. https://news.ycombinator.com/item?id=48586648 https://news.ycombinator.com/item?id=48586648
- Arodex 4mo ago
  Have you read the comments?
  - mpweiher 4mo ago
    Yes. Have you? Are you of the opinion that the comments on that article are conclusive evidence for all SMR startups failing? If so, how? Please try to be specific. I don't see evidence of even one startup failing there, never mind all of them.
    - p2detar 4mo ago
      Not only that, but the landscape of SMRs research and development is becoming very rich [0]. I think we are going to see a renaissance of reactor technology in the coming decade, and it will be well deserved. 0 - https://world-nuclear.org/information-library/nuclear-power-reactors/small-modular-reactors/small-modular-reactor-smr-design-database https://world-nuclear.org/information-library/nuclear-power-...
    - Arodex 4mo ago
      [flagged]
      - mpweiher 4mo ago
        Fitting description of what you are peddling. I am not buying. And neither is anyone else these days.
    - Arodex 4mo ago
      >Please try to be specific. https://www.rechargenews.com/analysis/smrs-v-renewables-mini-nuclears-costs-judgement-day-is-coming/2-1-1983485 https://www.rechargenews.com/analysis/smrs-v-renewables-mini... "What about the costs, both in terms of Capex to build them and the levelised cost of the power they produce? Gadomski said: “We are very suspect of projections of costs from companies because they're inclined to be more favourable, as opposed to being more realistic. I think that getting a good cost estimate is something that is not possible right now because we haven't built any.”" >I don't see evidence of even one startup failing there "SMR developers will all want to avoid the fate of NuScale, the sector trailblazer that saw a pioneering US power deal cancelled when its estimated LCOE soared to $89/MWh from a previous $58/MWh." Just a 100% cost mistake. Oops. And even if they could have reached their 58$/MWh target, let's see what is the cost of competition... https://www.pv-magazine.com/2026/05/06/firm-solar-and-storage-costs-fall-to-54-mwh-says-irena/ https://www.pv-magazine.com/2026/05/06/firm-solar-and-storag... "Firm wind-plus-storage costs in 2025 ranged from around $59/MWh in Inner Mongolia to $88/MWh to $94/MWh across Brazil, Germany, and Australia, with costs projected to fall to roughly $49/MWh to $75/MWh across those markets by 2030." Oops oops. "Construction timelines are also shortening, with projects typically built within one to two years of securing permits and grid connection." Oops oops oops. Renewable is running circles around nuclear. Every renewable technology is beating forecast. Every nuclear technology is breaking costs predictions and deadlines.
lightedman 4mo ago
"SMR make as much sense as space datacenters." So a whole lot of sense given the entire US Navy uses them and I already have one datacenter operating up in space (small test unit that over 3 months has provided ZERO issues) and a bigger one heading up into orbit next year when it's done being made. "but you can't gaslight thermodynamics" No but you can certainly conflate them like you're doing right now.
- RealityVoid 4mo ago
  Ok, this is interesting. I am skeptic about DC's in space, but I do appreciate people actually doing stuff. What is it computing up there. How did you get it up? How does one usually talk with their satellite. I guess you don't merely have a dish since it's probably not geostationary.
  - lightedman 4mo ago
    "What is it computing up there." Hyperspectral satellite imagery - think ASTER/LANDSAT/MODIS but more modern, for surficial minerals study. "How did you get it up?" How else? Paid a rocket company to launch it into orbit after proving various flightworthiness tests and getting various certifications and permissions from relevant gov't authorities. "How does one usually talk with their satellite. I guess you don't merely have a dish since it's probably not geostationary." K-band. Don't need tons of power, just a good LoS from ground on your target. And yea, not Geostat, I'm in LEO.
    - crote 4mo ago
      Right, so a regular satellite. That's indeed as relevant to the multi-100kW-scale "space datacenter" idiots like Musk are proposing as naval SMRs are to commercial power plants.
      - lightedman 4mo ago
        All satellites are regular satellites, there is literally nothing special about anything in orbit outside of what it carries - it's still just a falling body in space. 100kw is literally nothing to generate in space. At typical silicon efficiencies that's football field in size, and about 70% that if you jump to more expensive multi-junction cells. I can make a folding panel the size of a compact car that'd unfurl out to cover that. That's maybe 4 hours in NX just retooling my current design. The only limitation is the capabilities of the launch vehicle. I've already got one small (single 4U) datacenter in orbit. It works. It works GREAT. It can scale up to constellation quantity. And I don't have to waste any water for cooling or constantly pollute the air for power generation or throw extra waste heat into our atmosphere, as a side bonus. It makes plenty of sense to those with the education. What's hilarious is I'm doing all this on a GED.

by Arodex 4mo ago

RealityVoid 4mo ago

How are SMR's "gaslighting themodynamics"? I mean, sure, I can accept that they're not economical with current tech, but it's not a frigging' perpetuum mobile, it's feasible technology.

Arodex 4mo ago
Thermodynamics are the reason why SMR aren't, and will never, be economical. A bigger nuclear reactor will always undercut your price per watt.
- holoduke 4mo ago
  The mindset that makes people stuck in time. Sorry but SMRs are potentially very cheap. Not at this point. ,but when operated on scale they will be. You need to start
  - dalyons 4mo ago
    This is just faith without evidence
    - RealityVoid 4mo ago
      I mean... you've got to have faith in your theory before testing it out. I am not having any opinion here about this, but the cycle in my mind is theory, belief, test, updated model of reality. I can imagine similar things said before we managed to have powered flight, "Pshhhsst! We'll never fly! The law of gravity forbids it." You do need to have unreasonable goals for things once in a while. In this case, I don't have a particularly fixed stance about SMR's, but the claim that "Thermodynamics are the reason why SMR aren't, and will never, be economical" feels a bit stronger than it is warranted. Never is a damn' looooong time. And I can easily imagine things being less efficient, but having other advantages that make them more economical. Claiming it's impossible is just stretching.
- chickenbig 4mo ago
  > Thermodynamics are the reason why SMR aren't, and will never, be economical. And the link between thermodynamics and the price of electricity is what?
  - Arodex 4mo ago
    Your small nuclear reactor is going to need almost as much engineering , plumbing, safety mechanism, personnel, maintenance, etc... as your big nuclear reactor.
    - 0-_-0 4mo ago
      Not if it's mass produced
      - palata 4mo ago
        Let's distribute the risk everywhere, what could go wrong?
        chickenbig 4mo ago
        Siting of an SMR is somewhat different (albeit related) to the SMR concept itself. You might cluster them together (like the plans for 3 RR SMR at Wylfa and 3 at Ringhals in Sweden).
      - crote 4mo ago
        Mass-produced reactors don't need operators or guards?
    - chickenbig 4mo ago
      > almost as much engineering , plumbing, safety mechanism, personnel, maintenance, etc Sure, that is economics, not thermodynamics. I don't necessarily agree with the SMR manifesto, but it is conceivable that improved financing, construction, operation and oversight could make an SMR cheaper than a larger reactor.
- Manuel_D 4mo ago
  The value propositions of SMRs are logistics and re-use of existing infrastructure. The idea is that you could have easily transportable reactors that you can plop down in an existing coal plant, and then reuse the turbine, dynamo, etc. that are already in place. The fact that we haven't seen more widespread use of SMRs suggests that you're right. But it's important to point out that there are cost saving opportunities that could potentially reduce the net price per watt despite worse thermodynamic efficiency.
  - crote 4mo ago
    But they'll never be small enough to be truly portable. It'll be closer to another Akademik Lomonosov [0] than a truck-sized diesel generator, which severely limits its deployment options. The additional per-site engineering required to reuse things like turbines and dynamos is almost certainly going to kill any savings it would have. If you're already shipping a building-sized reactor, what's one more turbine? Realistically the main reusable component is the grid hookup itself - but that would incentivize building a large-scale reactor on the site. As would reusing the turbines, for that matter: you can't exactly power the turbines of a 100MW coal plant with a 10MW reactor, and shipping ten inefficient 10MW reactors to the site just so you can reuse the existing ancient turbine isn't exactly an attractive option either. [0]: https://en.wikipedia.org/wiki/Akademik_Lomonosov https://en.wikipedia.org/wiki/Akademik_Lomonosov

2muchcoffeeman 4mo ago

Doesn’t China have SMRs?

hangonhn 4mo ago
Just one so far and it's not particularly small compared to a more conventional reactor. Russia actually does have a smallish SMR but it wasn't terribly cheap to build nor operate. IIRC it is in the form of a ship and used to power a city somewhere in the north. SMR has a place for sure but no one has demonstrated the unit costs savings of making a lot of them yet. You can actually get some, if not most, of the economy of scale by doing a fleet build of one specific design. The US seems to be working on that and picked the Westinghouse AP-1000. I think that initiative has a decent chance of succeeding. The first few will be slow and expensive to build (even China has had delays with their nuclear roll out) but the subsequent ones will get cheaper and faster to build. This is how some countries did it during the first nuclear power expansion era.

john_strinlai 4mo ago

>SMR make as much sense as space datacenters. you are in this thread a lot, so i am guessing you must be very familiar with the industry. maybe you can help me understand: is the wikipedia on SMRs incorrect/lying when they say that there are commercially operating SMRs since 2020? and how have so many smart people and companies been duped into seriously considering SMR technology if SMRs apparently break the laws of thermodynamics?

Arodex 4mo ago
>commercially operating And struggling, propped up by taylor-made laws and public money. >how have so many smart people and companies been duped into seriously considering SMR technology if SMRs apparently break the laws of thermodynamics? Never said they break the laws of thermodynamics. They are just inefficient and will never be more efficient than alternatives such as... Bigger nuclear reactors. Or solar. And how long have you been out there? Have you never seen investors dumping and wasting billions in dead-ends? Never seen a mania before? Nuclear attracts clever people, but it isn't smart nor wise.
- jkman 4mo ago
  >'taylor-made' Says it all, doesn't it
- vablings 4mo ago
  I think you have a misunderstanding of what a SMR is supposed to be. Nuclear power plants are eye watering levels of expensive. The require massive scale and cost with lengthy approvals and requirements, the fundamental idea of SMRs is to move that cost and approvals into a smaller scale so that multiple standard units can be produced and deployed in a turnkey situation, they still will be expensive but the time to deploy and cost will be significantly reduced. We also know SMRs work very well, considering the majority of the US Navy is powered entirely with SMRs and have been for a very long time. Off the top of my head ship power has been exported to local areas for disaster relief Solar is absolutely fantastic and your average person should not be hawking at solar for your home to offset your power bill. The problem with solar is that you need power 24/7 and solar will not make power in the night. I don't think the likes of Westinghouse, Siemens, Rolls Royce and GE are duped. They are trying to solve a very hard problem!

bakies 4mo ago

What are the challenges they face?

redsocksfan45 4mo ago
[dead]
Manuel_D 4mo ago
One is regulatory. At least in the US, every nuclear reactor that produces at least 100 MW needs to carry a 375 million dollar insurance policy at minimum. Under 100 MW there is an alternate schedule that ranges from 5 million to 75 million scaling based on output. But the net result is that it's still more profitable to built a single large reactor, since a 1 GW reactor is less to insure than 10 100 MW reactors. This is written into law, it would require Congress to change it. Second is that nuclear reactor efficiency tends to improve with size. The ratio of thermal watts to electric watts tends to be better with large reactors. I'm not super well versed on the engineering tradeoffs here by my rough understanding is that waste heat scales with surface area while useful energy extraction scales with volume.
- projektfu 4mo ago
  In the grand scheme of things, that doesn't seem to be too expensive. According to the NRC, insurance is per site, and additional reactors on a site don't increase the insurance nearly as much as the first. The example they give is $1.1 million annual premium for a $500 million policy, with multi-reactor sites going up to $1.5 million. They also mention property insurance ($1.06 billion policy) and they didn't discuss the premiums on that, but as it's not liability insurance it's probably cheaper. The big costs are still going to be the cost of siting and building the reactor, the fuel, and the ongoing cost of running it. They pay off over a very long time horizon, so it's also the opportunity cost. https://www.nrc.gov/reading-rm/doc-collections/fact-sheets/nuclear-insurance https://www.nrc.gov/reading-rm/doc-collections/fact-sheets/n...

mpweiher 4mo ago

> Every SMR startup is failing. https://news.ycombinator.com/item?id=48586648 https://news.ycombinator.com/item?id=48586648

Arodex 4mo ago
Have you read the comments?
- mpweiher 4mo ago
  Yes. Have you? Are you of the opinion that the comments on that article are conclusive evidence for all SMR startups failing? If so, how? Please try to be specific. I don't see evidence of even one startup failing there, never mind all of them.
  - p2detar 4mo ago
    Not only that, but the landscape of SMRs research and development is becoming very rich [0]. I think we are going to see a renaissance of reactor technology in the coming decade, and it will be well deserved. 0 - https://world-nuclear.org/information-library/nuclear-power-reactors/small-modular-reactors/small-modular-reactor-smr-design-database https://world-nuclear.org/information-library/nuclear-power-...
  - Arodex 4mo ago
    [flagged]
    - mpweiher 4mo ago
      Fitting description of what you are peddling. I am not buying. And neither is anyone else these days.
  - Arodex 4mo ago
    >Please try to be specific. https://www.rechargenews.com/analysis/smrs-v-renewables-mini-nuclears-costs-judgement-day-is-coming/2-1-1983485 https://www.rechargenews.com/analysis/smrs-v-renewables-mini... "What about the costs, both in terms of Capex to build them and the levelised cost of the power they produce? Gadomski said: “We are very suspect of projections of costs from companies because they're inclined to be more favourable, as opposed to being more realistic. I think that getting a good cost estimate is something that is not possible right now because we haven't built any.”" >I don't see evidence of even one startup failing there "SMR developers will all want to avoid the fate of NuScale, the sector trailblazer that saw a pioneering US power deal cancelled when its estimated LCOE soared to $89/MWh from a previous $58/MWh." Just a 100% cost mistake. Oops. And even if they could have reached their 58$/MWh target, let's see what is the cost of competition... https://www.pv-magazine.com/2026/05/06/firm-solar-and-storage-costs-fall-to-54-mwh-says-irena/ https://www.pv-magazine.com/2026/05/06/firm-solar-and-storag... "Firm wind-plus-storage costs in 2025 ranged from around $59/MWh in Inner Mongolia to $88/MWh to $94/MWh across Brazil, Germany, and Australia, with costs projected to fall to roughly $49/MWh to $75/MWh across those markets by 2030." Oops oops. "Construction timelines are also shortening, with projects typically built within one to two years of securing permits and grid connection." Oops oops oops. Renewable is running circles around nuclear. Every renewable technology is beating forecast. Every nuclear technology is breaking costs predictions and deadlines.

lightedman 4mo ago

"SMR make as much sense as space datacenters." So a whole lot of sense given the entire US Navy uses them and I already have one datacenter operating up in space (small test unit that over 3 months has provided ZERO issues) and a bigger one heading up into orbit next year when it's done being made. "but you can't gaslight thermodynamics" No but you can certainly conflate them like you're doing right now.

RealityVoid 4mo ago
Ok, this is interesting. I am skeptic about DC's in space, but I do appreciate people actually doing stuff. What is it computing up there. How did you get it up? How does one usually talk with their satellite. I guess you don't merely have a dish since it's probably not geostationary.
- lightedman 4mo ago
  "What is it computing up there." Hyperspectral satellite imagery - think ASTER/LANDSAT/MODIS but more modern, for surficial minerals study. "How did you get it up?" How else? Paid a rocket company to launch it into orbit after proving various flightworthiness tests and getting various certifications and permissions from relevant gov't authorities. "How does one usually talk with their satellite. I guess you don't merely have a dish since it's probably not geostationary." K-band. Don't need tons of power, just a good LoS from ground on your target. And yea, not Geostat, I'm in LEO.
  - crote 4mo ago
    Right, so a regular satellite. That's indeed as relevant to the multi-100kW-scale "space datacenter" idiots like Musk are proposing as naval SMRs are to commercial power plants.
    - lightedman 4mo ago
      All satellites are regular satellites, there is literally nothing special about anything in orbit outside of what it carries - it's still just a falling body in space. 100kw is literally nothing to generate in space. At typical silicon efficiencies that's football field in size, and about 70% that if you jump to more expensive multi-junction cells. I can make a folding panel the size of a compact car that'd unfurl out to cover that. That's maybe 4 hours in NX just retooling my current design. The only limitation is the capabilities of the launch vehicle. I've already got one small (single 4U) datacenter in orbit. It works. It works GREAT. It can scale up to constellation quantity. And I don't have to waste any water for cooling or constantly pollute the air for power generation or throw extra waste heat into our atmosphere, as a side bonus. It makes plenty of sense to those with the education. What's hilarious is I'm doing all this on a GED.