Slashdot Mirror
NRC Approves New Nuclear Reactor Design
hrvatska writes "The NY Times has an article about the U.S. Nuclear Regulatory Commission approval of the design of Westinghouse's AP1000 reactor for the U.S., clearing the way for two American utilities to continue the construction of projects in South Carolina and Georgia. The last time a nuclear power plant in the U.S. entered service was 1996. The AP1000 was discussed on Slashdot a few years ago."
Ignoring the massive earthquake, tsunami and the ancient reactor design of course...
In other words, ignoring things that happen in the real world, and that even a first-world country like Japan can't get around human nature (laziness) and business imperatives to cut corners and defer upgrades.
Nuclear power would be great, if we didn't have to depend on humans to run it.
Remain calm! All is well!
Or the U.S. could just let them spend the money and take all the risks in terms of designing and testing the new reactors, then steal the designs and build the reactors themselves, forcing the Chinese firms to eat the R&D costs....
Wait, something about this sounds familiar. I sense a pot and a kettle are involved.
Check out my sci-fi/humor trilogy at PatriotsBooks.
Nuclear power would be great if humans didn't have irrational fear about things there don't bother to understand. If reactor construction had not stopped after the Chernobyl disaster, very few of these old, crappy designs would still be in use. Most of the problems in the modern nuclear industry are related to ancient systems that have had their lives extended due to the lack of replacement plant.
Funny, when they built those "ancient" systems they promised us those were safe too.
But then, concentrating material that will remain highly radioactive for longer than any empire in history has stood, and for longer than any region of the world has gone without war, could never be safe when you stop and think about it.
The real problem with nuclear power is something everyone understands -- namely, people's ability for sloth and cheapness. A properly constructed and maintained nuclear reactor can be exceedingly safe. The problem is, those that run said plants will cut corners everywhere -- construction, maintenance, etc. -- and when they do, the consequences can be huge.
+1. If a building collapses due to an earthquake, it's not a civil engineering disaster, it's a NATURAL DISASTER. But somehow, no matter what hits a nuclear plant (be it an earthquake or an asteroid), its still a nuclear disaster.
But they *have* proven *relatively* safe. It depends on the benchmark you judge "relative" to.
Fossil fuels kill people all the time. Coal miners, for example. The men on the Deep Water Horizons drilling platform. They sicken and kill people every single day through pollution. And if you believe in the scientific consensus on anthropogenic climate change, it is likely they damage ecosystems on a global scale and (statistically speaking) kill people through extreme weather events.
The problem is that the killing, sickening and destroying fossil fuels do aren't visibly tied to fossil fuel use. We know these things happen in an intellectual way, but we don't viscerally associate them with flicking on the power switch and burning a little more coal in a plant twenty miles away.
The problem with nuclear power is that its risks are at the opposite extreme. Nuclear disasters are exceedingly rare, so our assessment of risks is based on assumptions built on very little practical experience with nuclear disasters. We don't really have a good basis for judging the risks of having, say, ten times as many nuclear power plants as we do now. The nuclear economy scenario is full of situations where an error in some assumption has non-linear effects on the probability of outcome. For example if you assume the height of a once-in-a-century tsunami is six meters, but in fact it is twelve, you don't *double* the probability of an accident. You transform what is for practical purposes a statistical impossibility into a near-certain disaster.
So what's the rational thing to do? I think it is to move away from a fossil fuel economy and *toward* more diverse energy sources in which nuclear power will be a key part. But I wouldn't go on a crash course to try to solve all our problems in a decade by building as many nuclear plants as we can. The almost certain result of that will be ending up with lots of white elephant designs which proved to be more problematic than we'd hoped. A measured increase allows us to gain experience with designs, and to develop approaches to problems like decommissioning, nuclear waste and, for certain designs, nuclear proliferation. It also provides space for other technologies to take larger roles in the energy economy, spreading our risk over many sources and thus limiting our exposure to problems with any one. Getting ten percent of our energy needs from biomass might be very helpful to us as oil becomes more costly; trying to get 20% might have disastrous effects.
The real problem with nuclear power is something everyone understands -- namely, people's ability for sloth and cheapness. A properly constructed and maintained nuclear reactor can be exceedingly safe. The problem is, those that run said plants will cut corners everywhere -- construction, maintenance, etc. -- and when they do, the consequences can be huge.
I totally agree, having spent most of my 25 year US Navy career serving aboard nuclear powered submarines I have no problem living in the same ship as those 60's design reactors. The training and quality assurance programs that were required when I was on active duty insured safe operation.
JoeR
The question was of why would fukishima need active cooling when passive cooling is so "easy" to do.
That's like asking why the Ford Model-T couldn't do 200mph since a modern Ford Mustang can.
The answer is because the Fukishima Reactor wasn't designed to be passively cooled, the AP1000 is.
"Don't be a martyr -- BE THE ONE WHO GOT AWAY!"
"I dare you to name just a single nuclear accident in the last few years"
"Fukushima Daiichi?"
I wouldn't call that an accident. One must keep in mind that it was hit by an earthquake and a tsunami. What else would you expect?
If it were an error due to an operator or faulty equipment, then that would be a different story.
Almost all of the post 1970s technology in the AP1000 came directly from the nuclear division of Toshiba in Japan after merging with Westinghouse. It's technology bought off Japan instead of China but still looks like what you are worried about.
Beat me to the punch. The AP1000 is not a "new" design, it's a slightly warmed-over 1970s design that got NRC approval because it was close enough to the antiques currently in operation that no bureaucrat had to risk his pension by sticking his neck out and approving something that would be a genuine improvement (I'm lumping the Gen IIIs in with the Gen IIs here because they're mostly incremental improvements obtained from experience in running Gen IIs) . When the NRC approves anything Gen IV like a PBR or, heaven forbid, something genuinely modern like a TWR, then it's time to celebrate.
No. The decays in question occur when a neutron within fission product (the nuclei created after the U235) converts into proton together with an electron and a neutrino. Each decay releases around 1 MeV of energy (order of magnitude) as opposed the 200 MeV from the fission process. The processes reduces in intensity in time. Right after a scram the "decay heat" is 7% of the full power of the reactor. After 3 days it reduces to around 0.2% of the original power.
It was by no means easy to design an economical reactor with the kind of passive safety cooling provided by the AP1000. I can imagine why you think it was.
It is easy to criticize the event in the aftermath of three meltdowns and say that the design was flawed and that the response was mismanaged. You might even go as far as to say that the accidents weren't caused by the earthquake and tsunami, but by the failure of humans to properly design and operate the nuclear power plant (which in fact you did).
There is a point where you can't design for the most improbable events. A meteor landing in the ocean or North Korea bombing the plant aren't items you can design for. You also need to make a cutoff for earthquakes and tsunamis. An earthquake 10 times larger than any earthquake previously recorded in Japan's extensive seismic record might qualify. Add in the fact that seismologists didn't think the nature of the fault lines could even theoretically allow that powerful of an earthquake to occur.
You would have needed to build a 15m (45 ft) seawall to protect the plant from an event that experts didn't think could theoretically occur. The country that coined the term 'tsunami' didn't forget to design for it. They got hit by an extremely improbable occurrence that may not happen again for another 10,000 years. The probability of this type of massive tsunami destroying nuclear plants has not increased just because it happened recently. Nuclear plants are no less safe now than before the accidents. It is just more apparent that they have their limits, like any piece of technology.