4 ms·

Thankfully, we understand the physics a lot better now. The failure states for modern reactors are safe states. We've basically planned things so that if they

Thankfully, we understand the physics a lot better now. The failure states for modern reactors are safe states. We've basically planned things so that if they do fail, they turn off.

Retric 11y ago
In theory I might agree with you. In practice a lot of 'safe' designs for a wide range of things have failed. For a single reactor in a single location the odds might be very good. But, build a few thousand and the odds get worse. What happens if there is no power for months. What happens if an earthquake causes a reactor end up on it's side? What happens if there is a mistake while refueling? What about active sabotage? What if someone messes up in construction? How about a dam breaking causing massive wave? What about a combo of all of the above?
- Etzos 11y ago
  There are risks associated with all kinds of power generation. Most of the things you list would apply to a lot of them. However, I'll attempt to address most of the ones you listed (and I'm assuming a boiling water reactor, there are alternatives that may be naturally more stable): Modern reactors have passive cooling systems[1] which will only require an outside pool to be refilled every 72 hours (easy to automate or just drop a pipe from another location) so it could be left without power for a very long time. Huge earthquakes which would shake a reactor apart would be strong enough to take down entire cities. A reactor going supercritical, while bad, is still less likely to kill. Furthermore, as long as the passive systems aren't completely destroyed they will still function. Refueling is done by shutting down the reactor, so the risk is low. Active sabotage would be difficult to accomplish because there are multiple layers of redundancy and control systems and people that would all have to be bypassed. Construction mistakes could happen to anything, although again with multiple redundancies it doesn't matter quite as much. Most of these issues are really boiling down to one-off incidents. Modern nuclear facilities have tons of redundancy and checks in place. Things can go south with any system, but modern nuclear facilities have a lot of layers to prevent catastrophic failure in all but the most extreme situations (situations which would cause plenty of death and destruction in general). I just think it's a little flippant to discount modern nuclear power generation because of the failings in the past when the field was still in its infancy. [1] https://nuclear.gepower.com/build-a-plant/products/nuclear-power-plants-overview/esbwr.html https://nuclear.gepower.com/build-a-plant/products/nuclear-p...
  - Retric 11y ago
    require an outside pool to be refilled every 72 hours In other words the design is not passively safe. As to earthquakes a huge sinkhole might not be obvious before hand. http://robertkaplinsky.com/wp-content/uploads/2013/01/sink_hole_close1_large.jpg http://robertkaplinsky.com/wp-content/uploads/2013/01/sink_h...
    - Etzos 11y ago
      No power generation facility is passively safe for an extended period of time. Solar panels will leech harmful chemicals into the environment if they sit long enough. Coal plants will degrade and leech. Dams will degrade and fall apart. So it all depends on how long you think it is necessary to define as "passively safe". It is passively safe for 72 hours and then it will require an externally housed pool to be refilled. In the event of an incident something like that will be monitored very carefully, so it really doesn't have to be completely passive for very long.
      - Retric 11y ago
        72 hours is something of a best case assuming the pool is intact. Also, dam's can generally be left alone for months without issue many are on annual inspection schedules. Similarly solar can be left alone for years without problems. 3 days is far less time in the middle of a major disaster than you might think.
        Etzos 11y ago
        In order to cause enough damage to need this kind of backup system it would have to be huge. Something of that magnitude would also cause similar (if not more immediately catastrophic) problems for a dam or even wind turbines. Making the pool itself redundant in case of its failure is not all that difficult to do, although largely unnecessary. I'm only highlighting one of a myriad of systems that are designed to deal with worst-case of worst-case scenarios. I realize you're attempting to pick out those kinds of things and highlight that something can go wrong. But for any system something can go wrong given incredible situations. Modern nuclear reactors account for a large number of incredibly unlikely scenarios and provide mitigations for them.

Thankfully, we understand the physics a lot better now. The failure states for modern reactors are safe states. We've basically planned things so that if they do fail, they turn off.

Retric 11y ago

In theory I might agree with you. In practice a lot of 'safe' designs for a wide range of things have failed. For a single reactor in a single location the odds might be very good. But, build a few thousand and the odds get worse. What happens if there is no power for months. What happens if an earthquake causes a reactor end up on it's side? What happens if there is a mistake while refueling? What about active sabotage? What if someone messes up in construction? How about a dam breaking causing massive wave? What about a combo of all of the above?

Etzos 11y ago
There are risks associated with all kinds of power generation. Most of the things you list would apply to a lot of them. However, I'll attempt to address most of the ones you listed (and I'm assuming a boiling water reactor, there are alternatives that may be naturally more stable): Modern reactors have passive cooling systems[1] which will only require an outside pool to be refilled every 72 hours (easy to automate or just drop a pipe from another location) so it could be left without power for a very long time. Huge earthquakes which would shake a reactor apart would be strong enough to take down entire cities. A reactor going supercritical, while bad, is still less likely to kill. Furthermore, as long as the passive systems aren't completely destroyed they will still function. Refueling is done by shutting down the reactor, so the risk is low. Active sabotage would be difficult to accomplish because there are multiple layers of redundancy and control systems and people that would all have to be bypassed. Construction mistakes could happen to anything, although again with multiple redundancies it doesn't matter quite as much. Most of these issues are really boiling down to one-off incidents. Modern nuclear facilities have tons of redundancy and checks in place. Things can go south with any system, but modern nuclear facilities have a lot of layers to prevent catastrophic failure in all but the most extreme situations (situations which would cause plenty of death and destruction in general). I just think it's a little flippant to discount modern nuclear power generation because of the failings in the past when the field was still in its infancy. [1] https://nuclear.gepower.com/build-a-plant/products/nuclear-power-plants-overview/esbwr.html https://nuclear.gepower.com/build-a-plant/products/nuclear-p...
- Retric 11y ago
  require an outside pool to be refilled every 72 hours In other words the design is not passively safe. As to earthquakes a huge sinkhole might not be obvious before hand. http://robertkaplinsky.com/wp-content/uploads/2013/01/sink_hole_close1_large.jpg http://robertkaplinsky.com/wp-content/uploads/2013/01/sink_h...
  - Etzos 11y ago
    No power generation facility is passively safe for an extended period of time. Solar panels will leech harmful chemicals into the environment if they sit long enough. Coal plants will degrade and leech. Dams will degrade and fall apart. So it all depends on how long you think it is necessary to define as "passively safe". It is passively safe for 72 hours and then it will require an externally housed pool to be refilled. In the event of an incident something like that will be monitored very carefully, so it really doesn't have to be completely passive for very long.
    - Retric 11y ago
      72 hours is something of a best case assuming the pool is intact. Also, dam's can generally be left alone for months without issue many are on annual inspection schedules. Similarly solar can be left alone for years without problems. 3 days is far less time in the middle of a major disaster than you might think.
      - Etzos 11y ago
        In order to cause enough damage to need this kind of backup system it would have to be huge. Something of that magnitude would also cause similar (if not more immediately catastrophic) problems for a dam or even wind turbines. Making the pool itself redundant in case of its failure is not all that difficult to do, although largely unnecessary. I'm only highlighting one of a myriad of systems that are designed to deal with worst-case of worst-case scenarios. I realize you're attempting to pick out those kinds of things and highlight that something can go wrong. But for any system something can go wrong given incredible situations. Modern nuclear reactors account for a large number of incredibly unlikely scenarios and provide mitigations for them.