6 ms·

Nuclear Plant Accidents: Sodium Reactor Experiment

87 points by tehabe 10y ago

spditner 10y ago
I'm always surprised that the CANDU[1] reactor isn't more popular, as you don't require uranium enrichment, can use a mix of fuels, and has a number of fail-safes that work without power -- such as gravity-dropped control rods into the low pressure chamber, and requiring the moderator (deuterium in the original models) to be present to maintain a reaction. [1] https://en.m.wikipedia.org/wiki/CANDU_reactor https://en.m.wikipedia.org/wiki/CANDU_reactor
jdblair 10y ago
Santa Susana is notorious for a dangerously cavalier attitude about chemical safety. For years they burned toxic chemical waste in open pits. They filled barrels and shot them with rifles until they exploded. One person was killed in 1994. They disposed of sodium coated parts by letting them burn in the same open-air pits. https://en.wikipedia.org/wiki/Santa_Susana_Field_Laboratory#Sodium_burn_pits https://en.wikipedia.org/wiki/Santa_Susana_Field_Laboratory#... http://www.latimes.com/business/hiltzik/la-fi-hiltzik-20140613-column.html http://www.latimes.com/business/hiltzik/la-fi-hiltzik-201406...
jballanc 10y ago
I find various accounts of nuclear accidents fascinating because, as the intro of the article alludes to, they often involve complex systems undergoing complex failure scenarios. Then there's SL-1 (https://en.wikipedia.org/wiki/SL-1 https://en.wikipedia.org/wiki/SL-1). That was a really stupidly simple failure, but of all the accidents I've read about I find it the most poignant reminder of why nuclear reactors are such complex systems in the first place.
- hga 10y ago
  But it was well documented that you shouldn't lift that one rod above point X. Maybe the message didn't sink into the guy, maybe he screwed up, they even wonder if he committed suicide. Its big message was to eliminate all possible single points of failure, especially ones controlled by fallible humans, even at the cost of increased complexity.
  - cladari 10y ago
    It should be noted that reactivity control in military and commercial PWRs is fundamentally different. That said it's hard to understand how the SL1 operator managed to yank the rod so far out of the core. The procedure was to withdraw each rod 4 inches and attach it to the crdm, which in itself is a little mind blowing to me. The rod was withdrawn 26 inches and prompt criticality was achieved at 23 inches or so. The rod weighed 86 pounds. Even if it was stuck it's hard to imagine yanking something that heavy that far to unstick it.
Animats 10y ago
At least they didn't have a sodium fire. Monju (Japan) and Clinch River (US) both had sodium fires. The U.S. Navy tried sodium reactors in the early days. In addition to Nautilus, the first nuclear submarine, they also built USS Seawolf, with a sodium-cooled reactor. Seawolf worked, but was enough of a headache that its reactor was replaced with a Nautilus-type pressurized water reactor. The US Navy didn't try sodium-cooled reactors again. Rickover on sodium-cooled reactors: "expensive to build, complex to operate, susceptible to prolonged shutdown as a result of even minor malfunctions, and difficult and time-consuming to repair." The USSR tried sodium-cooled reactors, with moderate success.[1] They had leaks and fires too. One big reactor had 39 leaks over the years, none catastrophic. In the last 15 years of operation, none. Eventually they found all the construction defects. Basic truth: complexity in the radioactive parts of a nuclear reactor causes trouble. Sodium reactors have fires. Pebble-bed reactors have pebble jams. (There's one in Germany that can't be unjammed or decommissioned.) Helium-cooled reactors leak helium. The nice thing about water-cooled reactors is that the complexity is on the outside, where repairs are possible. This is why new fancy reactor designs are viewed sceptically in the industry. You can't have any problems inside the reactor vessel, because you can't fix them. [1] http://www-pub.iaea.org/mtcd/meetings/PDFplus/2009/cn176/cn176_Presentations/plenary_session_6/INV-08.Ashurko.pdf http://www-pub.iaea.org/mtcd/meetings/PDFplus/2009/cn176/cn1...
- pcl 10y ago
  I hadn't heard about pebble jams as a problem in pebble-bed reactors. Here's a link to (scant) details about that German event: https://en.wikipedia.org/wiki/Pebble-bed_reactor#1986_accident https://en.wikipedia.org/wiki/Pebble-bed_reactor#1986_accide...
  - Animats 10y ago
    More info: [1] Most of the pebbles were removed through the fueling system, but broken pebbles and dust remain, all highly radioactive. Bits of broken pebbles are jammed into crevices in broken graphite. "The reflector bottom of the core cavity is funnel shaped, with an inclination of 30°, ending in the 0.5 m wide vertical fuel discharge tube. In that geometrical setting it could simply be assumed that all pebbles must have rolled into the discharge tube and that there was no need for an inspection. The reflector bottom, however, contains many slits for coolant penetration, 3.4 cm wide and in radial orientation. A piece of broken pebble could have stuck in such a slit, and, standing out, one or more pebbles might lay behind it guided by the slit like by a rail." "An upper limit of 197 fuel pebbles, respectively fuel pebble equivalent in form of broken pieces, has been evaluated to be residual in the reactor vessels containing an upper limit amount of 98g of fissile material." "AVR has most probably the strongest β-contamination, and in the worst form, of all nuclear installations world wide". This is why complexity in the radioactive portion of the system is strongly undesirable. If anything goes wrong, dealing with it is extremely expensive, difficult, and dangerous. [1] http://www.wmsym.org/archives/2000/pdf/36/36-5.pdf http://www.wmsym.org/archives/2000/pdf/36/36-5.pdf
wildlogic 10y ago
This is a great read that sums up many criticality accidents and their causes over the years - "A Review of Criticality Accidents" https://www.orau.org/ptp/Library/accidents/la-13638.pdf https://www.orau.org/ptp/Library/accidents/la-13638.pdf
totally 10y ago
If you're into this, you might also like: Inviting Disaster: Lessons From the Edge of Technology: https://www.amazon.com/gp/product/0066620821 https://www.amazon.com/gp/product/0066620821
- pinewurst 10y ago
  Or Atomic Accidents by James Mahaffey - this is a great book https://www.amazon.com/Atomic-Accidents-Meltdowns-Disasters-Mountains/dp/1605986801/ref=sr_1_1?s=books&ie=UTF8&qid=1467769048&sr=1-1&keywords=atomic+accidents https://www.amazon.com/Atomic-Accidents-Meltdowns-Disasters-...
sbierwagen 10y ago
Kind of weirdly disingenuous that the article says "the workers" did this and "the workers" did that. What, was management completely absent for a period of months, and a bunch of yokels were just randomly flipping switches? If "a worker" was overriding a failsafe, it's because right behind him a supervisor was telling him they had to get the plant back online, or else he'd be fired.
- mikekchar 10y ago
  I think you are being just as disingenuous. Without knowing the processes, we have absolutely no idea what the decision chain looks like. I can think of many emergency situations where I would want workers who are close to the problem to make judgement calls. Even if they make the wrong call, at least the chance of success was optimised. I often notice a bias amongst some people that they think management is more susceptible to caving in to pressure for "success" (success being measured by getting a job done quickly/cheaply, etc). In reality, I think it's just a human condition. Imagine working in a shop where every move you make is scrutinised. If you get something done fast, everyone assumes that you cheated somehow or that you did an exceptionally poor job. People always doubt you and assume that you are on the take somehow. This would be pretty dysfunctional. Now imagine working in a situation where when you do work everyone assumes that it is of high quality. They don't have to check it because they know you are capable, hard working and reliable. If you get it done more quickly than expected, it's because you are a genius and should be promoted. People often try to build companies like the latter. The problem is that it's very easy to let your standards slide. If you do poor work very quickly you get congratulated and rewarded. If you do good work very slowly, nobody looks at it, nobody understands the quality and everyone complains that you took a long time. If you are in management, it's even worse. If you check up on the quality of work of your workers: 1. They are upset that you don't trust them. 2. If you find problems, it only causes you headaches and makes you look bad because nobody understands quality at all. They just think everything is wonderful and if you bring up a problem, then it's because you are the worst manager on the team! It's really easy to close your eyes and reward poor work. What I find really interesting is that on teams where there is absolutely no pressure to deliver, conversely programmers often imagine pressure from management to deliver quickly. Again, it's because nobody sees the quality of their work and the only thing they see being evaluated is the speed with which they deliver. In the end, management must take responsibility when workers make mistakes. Maybe the wrong people were hired. Maybe there was a lack of training. Maybe the social dynamics were poor. It doesn't matter. However, it is also possible that workers made the wrong choice, but that the system still allowed the choice with the highest chance of success to be made (given the information that they had). You are never guaranteed to have success. Even when workers and management do an excellent job, sometimes you fail because of something you didn't realise at the time. In that vein, I wouldn't assume that "the workers did X" is an indication that they were incompetent. Nor would I assume that management pressured them into doing something stupid.
mixmastamyk 10y ago
I have a bit of a connection with this incident, I grew up in Simi Valley, not far from Los Angeles; it is not an academic subject to me. In short, they built a defective reactor in a warehouse a few miles away from several cities and covered it up when it exploded. A number of workers and their spouses (known by my parents) died early in their 40s from cancer ostensibly due to contamination. There is a belief in increased birth defects in the area, but I can't find anything to cite right now. The shame is that we used to be proud of having rocketdyne nearby, testing rocket engines and such, which we could hear from our school yards. Little did we know how reckless they were, and had had a meltdown a few years prior.
cocoablazing 10y ago
Nuclear electric power production carries risks, which are outweighed by the expectations of lives saved by reducing air pollution and greenhouse gas production.
- lostlogin 10y ago
  Presumably you mean the risks are outweighed if better options aren't available. For example hydro is a good option if your power needs aren't too high and population quite small.
  - justinator 10y ago
    ...and you don't care about destroying the habitat of anything upriver, or anything that needs to get downriver/upriver (spawning fish), or anyone that relies on those critters. And you don't mind that it's not a long-term solution, as silt builds up behind the dam until your dam is worthless. But yeah, dam away!
    - lostlogin 10y ago
      Nuclear power warms water which is then usually dumped into rivers at a temperature higher than it started at. The waste issue is problematic and the cost (especially when factoring in cleanup and waste) is prohibitive to a small country with no existing nuclear program. Our dams gave been in a while here in New Zealand and the trade offs they require are vastly more acceptable to New Zealanders than the nuclear option.
    - Dylan16807 10y ago
      > And you don't mind that it's not a long-term solution, as silt builds up behind the dam until your dam is worthless. You can let some of the water flow take silt away, at least the silt anywhere near the dam. That might mean you don't have a huge reservoir, but you can generate a lot of power without a reservoir at all.
      - Dylan16807 10y ago
crististm 10y ago
Happens all the time. They ignored multiple and repeated warnings that they should stop and think things through. On a lower scale it's like (repeatedly) rebooting the PC when it's misbehaving.
lb1lf 10y ago
Reminds me of an old one-liner I saw in an email signature years ago - "There's one really terrifying word in nuclear physics. Oops." I'll lead myself out.

87 points by tehabe 10y ago

I'm always surprised that the CANDU[1] reactor isn't more popular, as you don't require uranium enrichment, can use a mix of fuels, and has a number of fail-safes that work without power -- such as gravity-dropped control rods into the low pressure chamber, and requiring the moderator (deuterium in the original models) to be present to maintain a reaction. [1] https://en.m.wikipedia.org/wiki/CANDU_reactor https://en.m.wikipedia.org/wiki/CANDU_reactor

jdblair 10y ago

Santa Susana is notorious for a dangerously cavalier attitude about chemical safety. For years they burned toxic chemical waste in open pits. They filled barrels and shot them with rifles until they exploded. One person was killed in 1994. They disposed of sodium coated parts by letting them burn in the same open-air pits. https://en.wikipedia.org/wiki/Santa_Susana_Field_Laboratory#Sodium_burn_pits https://en.wikipedia.org/wiki/Santa_Susana_Field_Laboratory#... http://www.latimes.com/business/hiltzik/la-fi-hiltzik-20140613-column.html http://www.latimes.com/business/hiltzik/la-fi-hiltzik-201406...

jballanc 10y ago

I find various accounts of nuclear accidents fascinating because, as the intro of the article alludes to, they often involve complex systems undergoing complex failure scenarios. Then there's SL-1 (https://en.wikipedia.org/wiki/SL-1 https://en.wikipedia.org/wiki/SL-1). That was a really stupidly simple failure, but of all the accidents I've read about I find it the most poignant reminder of why nuclear reactors are such complex systems in the first place.

hga 10y ago
But it was well documented that you shouldn't lift that one rod above point X. Maybe the message didn't sink into the guy, maybe he screwed up, they even wonder if he committed suicide. Its big message was to eliminate all possible single points of failure, especially ones controlled by fallible humans, even at the cost of increased complexity.
- cladari 10y ago
  It should be noted that reactivity control in military and commercial PWRs is fundamentally different. That said it's hard to understand how the SL1 operator managed to yank the rod so far out of the core. The procedure was to withdraw each rod 4 inches and attach it to the crdm, which in itself is a little mind blowing to me. The rod was withdrawn 26 inches and prompt criticality was achieved at 23 inches or so. The rod weighed 86 pounds. Even if it was stuck it's hard to imagine yanking something that heavy that far to unstick it.

Animats 10y ago

At least they didn't have a sodium fire. Monju (Japan) and Clinch River (US) both had sodium fires. The U.S. Navy tried sodium reactors in the early days. In addition to Nautilus, the first nuclear submarine, they also built USS Seawolf, with a sodium-cooled reactor. Seawolf worked, but was enough of a headache that its reactor was replaced with a Nautilus-type pressurized water reactor. The US Navy didn't try sodium-cooled reactors again. Rickover on sodium-cooled reactors: "expensive to build, complex to operate, susceptible to prolonged shutdown as a result of even minor malfunctions, and difficult and time-consuming to repair." The USSR tried sodium-cooled reactors, with moderate success.[1] They had leaks and fires too. One big reactor had 39 leaks over the years, none catastrophic. In the last 15 years of operation, none. Eventually they found all the construction defects. Basic truth: complexity in the radioactive parts of a nuclear reactor causes trouble. Sodium reactors have fires. Pebble-bed reactors have pebble jams. (There's one in Germany that can't be unjammed or decommissioned.) Helium-cooled reactors leak helium. The nice thing about water-cooled reactors is that the complexity is on the outside, where repairs are possible. This is why new fancy reactor designs are viewed sceptically in the industry. You can't have any problems inside the reactor vessel, because you can't fix them. [1] http://www-pub.iaea.org/mtcd/meetings/PDFplus/2009/cn176/cn176_Presentations/plenary_session_6/INV-08.Ashurko.pdf http://www-pub.iaea.org/mtcd/meetings/PDFplus/2009/cn176/cn1...

pcl 10y ago
I hadn't heard about pebble jams as a problem in pebble-bed reactors. Here's a link to (scant) details about that German event: https://en.wikipedia.org/wiki/Pebble-bed_reactor#1986_accident https://en.wikipedia.org/wiki/Pebble-bed_reactor#1986_accide...
- Animats 10y ago
  More info: [1] Most of the pebbles were removed through the fueling system, but broken pebbles and dust remain, all highly radioactive. Bits of broken pebbles are jammed into crevices in broken graphite. "The reflector bottom of the core cavity is funnel shaped, with an inclination of 30°, ending in the 0.5 m wide vertical fuel discharge tube. In that geometrical setting it could simply be assumed that all pebbles must have rolled into the discharge tube and that there was no need for an inspection. The reflector bottom, however, contains many slits for coolant penetration, 3.4 cm wide and in radial orientation. A piece of broken pebble could have stuck in such a slit, and, standing out, one or more pebbles might lay behind it guided by the slit like by a rail." "An upper limit of 197 fuel pebbles, respectively fuel pebble equivalent in form of broken pieces, has been evaluated to be residual in the reactor vessels containing an upper limit amount of 98g of fissile material." "AVR has most probably the strongest β-contamination, and in the worst form, of all nuclear installations world wide". This is why complexity in the radioactive portion of the system is strongly undesirable. If anything goes wrong, dealing with it is extremely expensive, difficult, and dangerous. [1] http://www.wmsym.org/archives/2000/pdf/36/36-5.pdf http://www.wmsym.org/archives/2000/pdf/36/36-5.pdf

wildlogic 10y ago

This is a great read that sums up many criticality accidents and their causes over the years - "A Review of Criticality Accidents" https://www.orau.org/ptp/Library/accidents/la-13638.pdf https://www.orau.org/ptp/Library/accidents/la-13638.pdf

totally 10y ago

If you're into this, you might also like: Inviting Disaster: Lessons From the Edge of Technology: https://www.amazon.com/gp/product/0066620821 https://www.amazon.com/gp/product/0066620821

pinewurst 10y ago
Or Atomic Accidents by James Mahaffey - this is a great book https://www.amazon.com/Atomic-Accidents-Meltdowns-Disasters-Mountains/dp/1605986801/ref=sr_1_1?s=books&ie=UTF8&qid=1467769048&sr=1-1&keywords=atomic+accidents https://www.amazon.com/Atomic-Accidents-Meltdowns-Disasters-...

sbierwagen 10y ago

Kind of weirdly disingenuous that the article says "the workers" did this and "the workers" did that. What, was management completely absent for a period of months, and a bunch of yokels were just randomly flipping switches? If "a worker" was overriding a failsafe, it's because right behind him a supervisor was telling him they had to get the plant back online, or else he'd be fired.

mikekchar 10y ago
I think you are being just as disingenuous. Without knowing the processes, we have absolutely no idea what the decision chain looks like. I can think of many emergency situations where I would want workers who are close to the problem to make judgement calls. Even if they make the wrong call, at least the chance of success was optimised. I often notice a bias amongst some people that they think management is more susceptible to caving in to pressure for "success" (success being measured by getting a job done quickly/cheaply, etc). In reality, I think it's just a human condition. Imagine working in a shop where every move you make is scrutinised. If you get something done fast, everyone assumes that you cheated somehow or that you did an exceptionally poor job. People always doubt you and assume that you are on the take somehow. This would be pretty dysfunctional. Now imagine working in a situation where when you do work everyone assumes that it is of high quality. They don't have to check it because they know you are capable, hard working and reliable. If you get it done more quickly than expected, it's because you are a genius and should be promoted. People often try to build companies like the latter. The problem is that it's very easy to let your standards slide. If you do poor work very quickly you get congratulated and rewarded. If you do good work very slowly, nobody looks at it, nobody understands the quality and everyone complains that you took a long time. If you are in management, it's even worse. If you check up on the quality of work of your workers: 1. They are upset that you don't trust them. 2. If you find problems, it only causes you headaches and makes you look bad because nobody understands quality at all. They just think everything is wonderful and if you bring up a problem, then it's because you are the worst manager on the team! It's really easy to close your eyes and reward poor work. What I find really interesting is that on teams where there is absolutely no pressure to deliver, conversely programmers often imagine pressure from management to deliver quickly. Again, it's because nobody sees the quality of their work and the only thing they see being evaluated is the speed with which they deliver. In the end, management must take responsibility when workers make mistakes. Maybe the wrong people were hired. Maybe there was a lack of training. Maybe the social dynamics were poor. It doesn't matter. However, it is also possible that workers made the wrong choice, but that the system still allowed the choice with the highest chance of success to be made (given the information that they had). You are never guaranteed to have success. Even when workers and management do an excellent job, sometimes you fail because of something you didn't realise at the time. In that vein, I wouldn't assume that "the workers did X" is an indication that they were incompetent. Nor would I assume that management pressured them into doing something stupid.

mixmastamyk 10y ago

I have a bit of a connection with this incident, I grew up in Simi Valley, not far from Los Angeles; it is not an academic subject to me. In short, they built a defective reactor in a warehouse a few miles away from several cities and covered it up when it exploded. A number of workers and their spouses (known by my parents) died early in their 40s from cancer ostensibly due to contamination. There is a belief in increased birth defects in the area, but I can't find anything to cite right now. The shame is that we used to be proud of having rocketdyne nearby, testing rocket engines and such, which we could hear from our school yards. Little did we know how reckless they were, and had had a meltdown a few years prior.

cocoablazing 10y ago

Nuclear electric power production carries risks, which are outweighed by the expectations of lives saved by reducing air pollution and greenhouse gas production.

lostlogin 10y ago
Presumably you mean the risks are outweighed if better options aren't available. For example hydro is a good option if your power needs aren't too high and population quite small.
- justinator 10y ago
  ...and you don't care about destroying the habitat of anything upriver, or anything that needs to get downriver/upriver (spawning fish), or anyone that relies on those critters. And you don't mind that it's not a long-term solution, as silt builds up behind the dam until your dam is worthless. But yeah, dam away!
  - lostlogin 10y ago
    Nuclear power warms water which is then usually dumped into rivers at a temperature higher than it started at. The waste issue is problematic and the cost (especially when factoring in cleanup and waste) is prohibitive to a small country with no existing nuclear program. Our dams gave been in a while here in New Zealand and the trade offs they require are vastly more acceptable to New Zealanders than the nuclear option.
  - Dylan16807 10y ago
    > And you don't mind that it's not a long-term solution, as silt builds up behind the dam until your dam is worthless. You can let some of the water flow take silt away, at least the silt anywhere near the dam. That might mean you don't have a huge reservoir, but you can generate a lot of power without a reservoir at all.
    - Dylan16807 10y ago

crististm 10y ago

Happens all the time. They ignored multiple and repeated warnings that they should stop and think things through. On a lower scale it's like (repeatedly) rebooting the PC when it's misbehaving.

lb1lf 10y ago

Reminds me of an old one-liner I saw in an email signature years ago - "There's one really terrifying word in nuclear physics. Oops." I'll lead myself out.