5 ms·

Would love to know about expected failure modes or this reactor type. Is there any info on this?

by roomey 4y ago

alexb_ 4y ago
You can read the safety evaluation here: https://www.nrc.gov/reactors/new-reactors/smr/licensing-activities/nuscale/ser-open-items.html https://www.nrc.gov/reactors/new-reactors/smr/licensing-acti...
cratermoon 4y ago
https://thebulletin.org/premium/2021-07/can-small-modular-reactors-help-mitigate-climate-change/ https://thebulletin.org/premium/2021-07/can-small-modular-re...
AtlasBarfed 4y ago
"https://www.youtube.com/watch?v=vU-IlqiP4sU https://www.youtube.com/watch?v=vU-IlqiP4sU" Well, it says its a LWR, video says "fuel rods", which means solid rods and meltdown risk. Eh. If it's a solid fuel rod, then if you get a runaway reaction, and if circumstances mean the safety systems go offline (see: Fukushima) then meltdown. Contrast this with something like LFTR: the liquid fuel needs to stay in a certain shape/containment/vessel to maintain criticality. If it starts overreacting/heating, the "plug" at the bottom of the containment melts and the liquid flows into a shallow distributed pool that, per nuclear physics, is impossible to maintain criticality. That type of system is inherently meltdown-proof, even if all the systems go offline, the plug will melt. You know, assuming gravity still works. A pebble bed, where the solid fuel rod is instead a bunch of solid pellets, but if they get too hot you can similarly melt a plug and the pellets fall into a shape that likewise wouldn't stay critical, might also be similarly meltdown-proof, but I haven't read nearly as much on pebble bed designs. But a solid fuel rod? Nope.
- acidburnNSA 4y ago
  Nuclear engineer here. That's not quite how it all works. Solid vs. liquid fuel is not tied directly to reactivity stability, as quantified in the power coefficient of reactivity. If it goes up in power, you want the chain reaction to naturally go down. In solid fueled reactors, this is usually accomplished via the moderator. If the moderator temperature goes up, it reduced in density, thereby reducing the overall neutron moderation in the core. Thus, fewer neutrons make it to to the energy range that causes fission. Thus, reactivity goes down and the reaction stops. This is inherently stable, just like fuel density in a fluid fuel reactor going down and reducing the overall fission rate. LFTRs are pre-melted. You melt 100% of the core and then bring it critical. That's a lot of pretty mobile fission products! As for the melt plug, that's also a false solution. Given that achieving subcriticality is trivial in modern solid and fluid fuel reactors, the challenge in an accident is afterglow heat removal. As you may know, the plants at Fukushima had all rods in and were fully subcritical an hour before the tsunami hit). But when they lost afterglow heat removal, it still melted some containment barriers. Same can happen with fluid fuel, regardless of whether or not you've moved it from one tank to another. Fluid fuel is not the panacea many people want to think it is. Passive afterglow heat removal is the thing that lets reactors of any fuel form be safer than today's typical reactors, which generally require backup power to run the cooling systems. If you use certain molten salt, liquid sodium metal, liquid lead metal, etc. cooling configurations you can achieve indefinite heat removal without any external power. That reduces core damage frequencies by about 100x from modern large LWRs. Again, regardless of fuel form. That said, 100x safer than how safe current nuclear is is kind of just playing with very small numbers. Fossil and biofuel combustion kills 8 million per year from particulate emissions, according to the WHO, and also cause climate change. So we should just be building hundreds of regular large water cooled reactors now and then switch over to fancier cooling ones later, and also breeders that are ~infinitely sustainable for the long term.
  - Symmetry 4y ago
    For people who don't know this stuff, like myself a few years ago: When a uranium atom is hit by a neutron some of the will split immediately releasing new neutrons but some of them will go into an unstable state and then split some period of time later. If the instantaneous splits are enough to keep the reaction going that's called a "prompt critical" configuration and usually seen only in atomic weapons. If the neutrons released by both the immediate reactions and the delayed reactions are enough to keep the reaction going that's only "critical." Because there are many atoms in the reactor that have been hit by neutrons and are unstable but haven't split yet a reactor continues to release a lot of heat even when it's no long critical, somewhere on the order of 10% as much power as when it was fully on.
  - pfdietz 4y ago
    > That's a lot of pretty mobile fission products! https://gain.inl.gov/SiteAssets/MoltenSaltReactor/Module2-OverviewofMSRTechnologyandConceptsML17331B114.pdf https://gain.inl.gov/SiteAssets/MoltenSaltReactor/Module2-Ov... (slide 23) "Gaseous Fission Products Inherently Evolve from Fuel Salt" "Many FPs have Xe or Kr precursors – Over 40% of FPs leave core – Large fraction of cesium, strontium and iodine end up in offgas"
  - panick21_ 4y ago
    I agree freeze plug isn't magic, passive heat removal can be much easier if you can move the fuel into a different contaiment with different geometry and different passive heat removal features. However it has to be noted that most of the companies that work on molten salt reactors dont use that method. Terrestrial Energy, Moltex Energy for example.
    - acidburnNSA 4y ago
      Meh, I dunno I think freeze plugs are really falling out of favor in general. They take a long time to melt, aren't that reliable or predictable, and can spuriously actuate, dumping the whole core. Why make two vessels when you already need one? Just add passive afterglow heat removal to the one and you're done. Moving stuff around for no reason doesn't add anything.
- convolvatron 4y ago
  I think the point about pebble-bed designs is that the density of reactive material in the pebbles is low enough that without external control it maintains a moderate temperature. low enough to not cause a meltdown under any circumstances
  - pfdietz 4y ago
    It also means the fuel takes up much more volume. This has two downsides: the fuel now becomes more expensive to make, and it becomes more expensive to dispose of.
  - VLM 4y ago
    Engineered to cool by convection and gravity, yes. Somewhat safer than reactors that melt without continuous active cooling. But that comes at a cost: the good parts of low power density AND the bad part of low power density, such as being "huge" compared to a tiny little submarine reactor of similar power level. Really any reactor "could" be engineered this way but it does make them big. And "big" competes against the natural desire of engineers to run them at high temps thus high pressures to keep efficiency high. But what if lower efficiency results in net cheaper and safer electricity; its not like they're paying silicon valley prices for the land and nuclear fuel is stunningly cheap so burning twice as much is still cheaper than coal, LOL. A lot of the "old school" reactor design was based around the nuclear navy where both weight and volume are NOT cheap, not cheap at all. I don't think you could ever have a pebble bed reactor in an aircraft carrier.
    - pfdietz 4y ago
      This is one of the things that annoys me about the hype about DT fusion reactors. "They can't melt down like fission reactors!" "Yeah, if your fission reactor had 1/40th the volumetric power density of a PWR it would be really hard to make it melt down too." Sometimes in these discussions you see some precious spirit advocating a DT fusion reactor for use in ships.
- gene-h 4y ago
  LFTRs involve onsite reprocessing which is politically difficult to get approved in the US and may introduce other risks.
p1mrx 4y ago
Here's a presentation on their approach to safety: https://www.youtube.com/watch?v=JhrxFCtCPUo&t=2360s https://www.youtube.com/watch?v=JhrxFCtCPUo&t=2360s They estimate that the risk of core damage is 3-5 orders of magnitude lower than traditional nuclear plants.

Would love to know about expected failure modes or this reactor type. Is there any info on this?

by roomey 4y ago

Would love to know about expected failure modes or this reactor type. Is there any info on this?

You can read the safety evaluation here: https://www.nrc.gov/reactors/new-reactors/smr/licensing-activities/nuscale/ser-open-items.html https://www.nrc.gov/reactors/new-reactors/smr/licensing-acti...

cratermoon 4y ago

https://thebulletin.org/premium/2021-07/can-small-modular-reactors-help-mitigate-climate-change/ https://thebulletin.org/premium/2021-07/can-small-modular-re...

AtlasBarfed 4y ago

"https://www.youtube.com/watch?v=vU-IlqiP4sU https://www.youtube.com/watch?v=vU-IlqiP4sU" Well, it says its a LWR, video says "fuel rods", which means solid rods and meltdown risk. Eh. If it's a solid fuel rod, then if you get a runaway reaction, and if circumstances mean the safety systems go offline (see: Fukushima) then meltdown. Contrast this with something like LFTR: the liquid fuel needs to stay in a certain shape/containment/vessel to maintain criticality. If it starts overreacting/heating, the "plug" at the bottom of the containment melts and the liquid flows into a shallow distributed pool that, per nuclear physics, is impossible to maintain criticality. That type of system is inherently meltdown-proof, even if all the systems go offline, the plug will melt. You know, assuming gravity still works. A pebble bed, where the solid fuel rod is instead a bunch of solid pellets, but if they get too hot you can similarly melt a plug and the pellets fall into a shape that likewise wouldn't stay critical, might also be similarly meltdown-proof, but I haven't read nearly as much on pebble bed designs. But a solid fuel rod? Nope.

acidburnNSA 4y ago
Nuclear engineer here. That's not quite how it all works. Solid vs. liquid fuel is not tied directly to reactivity stability, as quantified in the power coefficient of reactivity. If it goes up in power, you want the chain reaction to naturally go down. In solid fueled reactors, this is usually accomplished via the moderator. If the moderator temperature goes up, it reduced in density, thereby reducing the overall neutron moderation in the core. Thus, fewer neutrons make it to to the energy range that causes fission. Thus, reactivity goes down and the reaction stops. This is inherently stable, just like fuel density in a fluid fuel reactor going down and reducing the overall fission rate. LFTRs are pre-melted. You melt 100% of the core and then bring it critical. That's a lot of pretty mobile fission products! As for the melt plug, that's also a false solution. Given that achieving subcriticality is trivial in modern solid and fluid fuel reactors, the challenge in an accident is afterglow heat removal. As you may know, the plants at Fukushima had all rods in and were fully subcritical an hour before the tsunami hit). But when they lost afterglow heat removal, it still melted some containment barriers. Same can happen with fluid fuel, regardless of whether or not you've moved it from one tank to another. Fluid fuel is not the panacea many people want to think it is. Passive afterglow heat removal is the thing that lets reactors of any fuel form be safer than today's typical reactors, which generally require backup power to run the cooling systems. If you use certain molten salt, liquid sodium metal, liquid lead metal, etc. cooling configurations you can achieve indefinite heat removal without any external power. That reduces core damage frequencies by about 100x from modern large LWRs. Again, regardless of fuel form. That said, 100x safer than how safe current nuclear is is kind of just playing with very small numbers. Fossil and biofuel combustion kills 8 million per year from particulate emissions, according to the WHO, and also cause climate change. So we should just be building hundreds of regular large water cooled reactors now and then switch over to fancier cooling ones later, and also breeders that are ~infinitely sustainable for the long term.
- Symmetry 4y ago
  For people who don't know this stuff, like myself a few years ago: When a uranium atom is hit by a neutron some of the will split immediately releasing new neutrons but some of them will go into an unstable state and then split some period of time later. If the instantaneous splits are enough to keep the reaction going that's called a "prompt critical" configuration and usually seen only in atomic weapons. If the neutrons released by both the immediate reactions and the delayed reactions are enough to keep the reaction going that's only "critical." Because there are many atoms in the reactor that have been hit by neutrons and are unstable but haven't split yet a reactor continues to release a lot of heat even when it's no long critical, somewhere on the order of 10% as much power as when it was fully on.
- pfdietz 4y ago
  > That's a lot of pretty mobile fission products! https://gain.inl.gov/SiteAssets/MoltenSaltReactor/Module2-OverviewofMSRTechnologyandConceptsML17331B114.pdf https://gain.inl.gov/SiteAssets/MoltenSaltReactor/Module2-Ov... (slide 23) "Gaseous Fission Products Inherently Evolve from Fuel Salt" "Many FPs have Xe or Kr precursors – Over 40% of FPs leave core – Large fraction of cesium, strontium and iodine end up in offgas"
- panick21_ 4y ago
  I agree freeze plug isn't magic, passive heat removal can be much easier if you can move the fuel into a different contaiment with different geometry and different passive heat removal features. However it has to be noted that most of the companies that work on molten salt reactors dont use that method. Terrestrial Energy, Moltex Energy for example.
  - acidburnNSA 4y ago
    Meh, I dunno I think freeze plugs are really falling out of favor in general. They take a long time to melt, aren't that reliable or predictable, and can spuriously actuate, dumping the whole core. Why make two vessels when you already need one? Just add passive afterglow heat removal to the one and you're done. Moving stuff around for no reason doesn't add anything.
convolvatron 4y ago
I think the point about pebble-bed designs is that the density of reactive material in the pebbles is low enough that without external control it maintains a moderate temperature. low enough to not cause a meltdown under any circumstances
- pfdietz 4y ago
  It also means the fuel takes up much more volume. This has two downsides: the fuel now becomes more expensive to make, and it becomes more expensive to dispose of.
- VLM 4y ago
  Engineered to cool by convection and gravity, yes. Somewhat safer than reactors that melt without continuous active cooling. But that comes at a cost: the good parts of low power density AND the bad part of low power density, such as being "huge" compared to a tiny little submarine reactor of similar power level. Really any reactor "could" be engineered this way but it does make them big. And "big" competes against the natural desire of engineers to run them at high temps thus high pressures to keep efficiency high. But what if lower efficiency results in net cheaper and safer electricity; its not like they're paying silicon valley prices for the land and nuclear fuel is stunningly cheap so burning twice as much is still cheaper than coal, LOL. A lot of the "old school" reactor design was based around the nuclear navy where both weight and volume are NOT cheap, not cheap at all. I don't think you could ever have a pebble bed reactor in an aircraft carrier.
  - pfdietz 4y ago
    This is one of the things that annoys me about the hype about DT fusion reactors. "They can't melt down like fission reactors!" "Yeah, if your fission reactor had 1/40th the volumetric power density of a PWR it would be really hard to make it melt down too." Sometimes in these discussions you see some precious spirit advocating a DT fusion reactor for use in ships.
gene-h 4y ago
LFTRs involve onsite reprocessing which is politically difficult to get approved in the US and may introduce other risks.

p1mrx 4y ago

Here's a presentation on their approach to safety: https://www.youtube.com/watch?v=JhrxFCtCPUo&t=2360s https://www.youtube.com/watch?v=JhrxFCtCPUo&t=2360s They estimate that the risk of core damage is 3-5 orders of magnitude lower than traditional nuclear plants.