4 ms·

> it was absolutely preventable, and they had the means to do so. This is not that interesting an observation by itself. For example, they could have shut down

> it was absolutely preventable, and they had the means to do so. This is not that interesting an observation by itself. For example, they could have shut down and decommissioned the reactor. What I believe you're trying to imply is that they were aware of problems and they could have fixed them in advance of the tsunami, and fixing them would not have been prohibitively expensive. But more important is the implication that the reactor was known to be particularly vulnerable in comparison to all the other nuclear power stations in Japan, and the cost of fixing all issues of comparable a priori severity was within the capacity of the funding of the system. The last bit is critical here; and it's why I disregard the narrative of Fukushima Daichi being "preventable". There are no current nuclear designs that are naturally fail-safe; all of them rely on active measures to ensure safety in the event of system failures. When the next disaster strikes, it will hopefully has as small an impact as Fukushima, but what I can guarantee is that afterwards we will have a laundry list of all the reasons why it was completely preventable based on the knowledge we had at the time.

SyzygistSix 4y ago
Even datacenters know the backup generator has to be above where it can flood. Fukushima was very much preventable. Elevating the backup generator would have prevented the meltdown.
- andrewla 4y ago
  Hindsight is 20/20 and all. How many other problems were identified at Fukushima and other reactors throughout Japan that were considered to be equally or more risky? If you cannot answer this question then how can you conclude that this was "preventable" in any useful sense?
panick21_ 4y ago
> There are no current nuclear designs that are naturally fail-safe; That is fundamentally false. There are lots of designs and many tested reactors. Passive fail safety without any active effects are very much a common thing and mostly a standard feature in pretty much all modern designs. Unless you have a different definition of 'active' then me. In the Molten Salt Reactor Experiment they literally hit the 'off' switch and went home for the weekend, shutting down the reactor was so basic. And the MSRE isn't the only reactor that has managed to do that. There just not currently available commercially to buy of the rack. If a nation seriously decided to invest in next generation nuclear, and were willing to put real money behind these designs, they could build them within a couple of years. All the companies working on these designs have greatly limited, need to deal with very long wait times very limited facilities, need to overcome regulatory system not designed for those reactors. If a nation picked a design, and really fast tracked development and deployment with all the needed work being done in parallel, a lot could be done. To give you one example, the IMSR (Integrated Molten Salt Reactor) from Terrestrial Energy is in the very final stages of Phase 2 of pre-vendor design review. They are expected to finish Phase 2 within the next couple of months and that basically means the CNSC (Canada) has considered the design and there isn't any reason why appropriate construction license request wouldn't be approved. > The objective of a review is to verify, at a high level, the acceptability of a nuclear power plant design with respect to Canadian nuclear regulatory requirements and expectations, as well as Canadian codes and standards. These reviews also identify fundamental barriers to licensing a new design in Canada and assures that a resolution path exists for any design issues identified in the review. > A vendor who has completed a Phase 2 Pre-Licensing VDR, has committed to increased regulatory efficiencies at the time of licensing. The results of Phase 2 will be taken into account mainly for the Construction Licence Application and is likely to result in increased efficiencies of technical reviews. https://nuclearsafety.gc.ca/eng/reactors/power-plants/pre-licensing-vendor-design-review/ https://nuclearsafety.gc.ca/eng/reactors/power-plants/pre-li... So if a nation like Germany did a 180 and said, we need new reactor and we need them fast, but we only accept reactors that are passively safe, but all German industrial resources are used and this is a high priority project. They could probably build such a reactor in 3-4 years. There are others one could talk about but I do like the IMSR. > But more important is the implication that the reactor was known to be particularly vulnerable in comparison to all the other nuclear power stations in Japan, and the cost of fixing all issues of comparable a priori severity was within the capacity of the funding of the system. Actually it was known that the sea wall was a particular weakness and it wasn't done, but not because of cost reasons. It was not done because it would have given anti-nuclear people a reason to say 'see it wasn't safe'. So really like Shuttle issue, its a flawed safety culture that was the real reason. The required expense wasn't that high. And that matters just as much as cost issues. However I generally agree. Saying 'it was preventable' isn't a good defense.
- andrewla 4y ago
  There are any number of speculative designs that are naturally fail-safe, but none of them in operation (outside of test facilities). I'm not a betting man, but I would wager a significant amount of money that by the time IMSR becomes widespread and approved it will require a large number of active failure measures to be considered "safe". > If a nation seriously decided to invest in next generation nuclear, and were willing to put real money behind these designs, they could build them within a couple of years. Sure -- they could do that with existing designs as well. But these things have a great deal of natural complexity, and once again, I would practically guarantee that whatever designs they choose will have active fail-safes, even if the concept does not require it in theory.

SyzygistSix 4y ago

Even datacenters know the backup generator has to be above where it can flood. Fukushima was very much preventable. Elevating the backup generator would have prevented the meltdown.

andrewla 4y ago
Hindsight is 20/20 and all. How many other problems were identified at Fukushima and other reactors throughout Japan that were considered to be equally or more risky? If you cannot answer this question then how can you conclude that this was "preventable" in any useful sense?

panick21_ 4y ago

> There are no current nuclear designs that are naturally fail-safe; That is fundamentally false. There are lots of designs and many tested reactors. Passive fail safety without any active effects are very much a common thing and mostly a standard feature in pretty much all modern designs. Unless you have a different definition of 'active' then me. In the Molten Salt Reactor Experiment they literally hit the 'off' switch and went home for the weekend, shutting down the reactor was so basic. And the MSRE isn't the only reactor that has managed to do that. There just not currently available commercially to buy of the rack. If a nation seriously decided to invest in next generation nuclear, and were willing to put real money behind these designs, they could build them within a couple of years. All the companies working on these designs have greatly limited, need to deal with very long wait times very limited facilities, need to overcome regulatory system not designed for those reactors. If a nation picked a design, and really fast tracked development and deployment with all the needed work being done in parallel, a lot could be done. To give you one example, the IMSR (Integrated Molten Salt Reactor) from Terrestrial Energy is in the very final stages of Phase 2 of pre-vendor design review. They are expected to finish Phase 2 within the next couple of months and that basically means the CNSC (Canada) has considered the design and there isn't any reason why appropriate construction license request wouldn't be approved. > The objective of a review is to verify, at a high level, the acceptability of a nuclear power plant design with respect to Canadian nuclear regulatory requirements and expectations, as well as Canadian codes and standards. These reviews also identify fundamental barriers to licensing a new design in Canada and assures that a resolution path exists for any design issues identified in the review. > A vendor who has completed a Phase 2 Pre-Licensing VDR, has committed to increased regulatory efficiencies at the time of licensing. The results of Phase 2 will be taken into account mainly for the Construction Licence Application and is likely to result in increased efficiencies of technical reviews. https://nuclearsafety.gc.ca/eng/reactors/power-plants/pre-licensing-vendor-design-review/ https://nuclearsafety.gc.ca/eng/reactors/power-plants/pre-li... So if a nation like Germany did a 180 and said, we need new reactor and we need them fast, but we only accept reactors that are passively safe, but all German industrial resources are used and this is a high priority project. They could probably build such a reactor in 3-4 years. There are others one could talk about but I do like the IMSR. > But more important is the implication that the reactor was known to be particularly vulnerable in comparison to all the other nuclear power stations in Japan, and the cost of fixing all issues of comparable a priori severity was within the capacity of the funding of the system. Actually it was known that the sea wall was a particular weakness and it wasn't done, but not because of cost reasons. It was not done because it would have given anti-nuclear people a reason to say 'see it wasn't safe'. So really like Shuttle issue, its a flawed safety culture that was the real reason. The required expense wasn't that high. And that matters just as much as cost issues. However I generally agree. Saying 'it was preventable' isn't a good defense.

andrewla 4y ago
There are any number of speculative designs that are naturally fail-safe, but none of them in operation (outside of test facilities). I'm not a betting man, but I would wager a significant amount of money that by the time IMSR becomes widespread and approved it will require a large number of active failure measures to be considered "safe". > If a nation seriously decided to invest in next generation nuclear, and were willing to put real money behind these designs, they could build them within a couple of years. Sure -- they could do that with existing designs as well. But these things have a great deal of natural complexity, and once again, I would practically guarantee that whatever designs they choose will have active fail-safes, even if the concept does not require it in theory.