Slashdot Mirror
Alabama Nuclear Reactor Gets 'F' Grade
GatorSnake writes "The US federal government issued a rare red finding against an Alabama nuclear power plant after an emergency cooling system failure. 'In an emergency, the failure of the valve could have meant that one of the plant's emergency cooling systems would not have worked as designed (PDF).' Does this further erode the argument that Fukushima was just an isolated incident in the 'modern' nuclear power age?"
that adds another zero to the zero deaths from nuclear this year. thats zero up from last year. gonna need some big design changes to catch up with fossil fuels.
That is exactly the problem. This is no different from the tragedies frequently encountered in coal mines. They cut corners and costs in the name of greater profits. And then when bad things happen, they say "whoops! This is an isolated incident. And we will fire someone for doing what we encouraged and even told them to do!"
The nuclear industry in the US has amazingly fearsome oversight. It happens that I word for a nuclear technology company and I can tell you first hand that "NRC" is mentioned in seemingly every business conversation with numerous and frequent meetings that involve NRC. So if the NRC didn't find this sooner, I have to wonder why. Has the government been cutting back on the NRC? I hope not and if they have, they need to reverse it and fast.
Nuclear energy is the best we have right now. But it also needs to be regulated and monitored closely. No one questions that fact.
Modern nuclear age? What?
The Browns Ferry Nuclear Plant began construction in 1966 (Fukushima Dai-ichi dates from 1971). Furthermore, both use General Electric boiling water reactors. The major difference seems to be that Browns Ferry is/was expected to continue to operate until 2033.
Similarly designed technology dating from a similar time has similar flaws. In most areas engineers learn from their mistakes and upgrade regularly for precisely this reason. Then we actually would be in the 'modern nuclear age', and discovering a new flaw would be disturbing news as opposed to being a wholly predictable consequence of expecting to keep dodgy, ancient crap running for well over half a century.
All nuclear plants are not created equal. The far far bigger problem is continuing to use early reactor designs past their end of life! It's like a 30 year old car that has not spent those years in a garage. It needs considerable work to stay usable, often to the point of requiring it to be rebuilt. Well the same thing holds true to nuclear plants, but we just don't spend that sort of money renovating the old ones. So they start to fail. How much effort is actually required to have severe problems is rather interesting, but I for on do not expect them to simply keep working.
We should have continued building and updating designs over the last 30 or 40 years, but anti-nuclear nuts have left us all pretty damn screwed.
we are all invisible unless we choose otherwise
I tend to agree in many ways. It's not entirely an engineering problem.
The real risks come as a result of our system, which is squarely rooted in human greed and fallibility. We're risk-takers by nature, and the risk/reward equation is skewed toward danger.
For example:
If I'm a CEO and build a reactor, cutting costs by attenuating the safety systems specified by the engineers (e.g. using cheap materials for failsafes, or not installing them at all), my profit goes up. I saved a lot of money during construction, didn't I!
However, if something goes wrong and my poorly implemented safety mechanisms fail, my personal risk is actually quite low. I probably won't notice an impact on my earnings, I certainly won't go to jail, and once the media is done feeding on the corpse of my disaster, it's back to "business as usual."
This is a far cry from the careful designs of the engineer, and the scenario gets played out all the time, in various disciplines (see also: BP oil spill, mortgage-backed securities, etc).
Maybe the solution is to let the engineers control the nuclear industry, soup-to-nuts, and send the MBA's packing?
"Beware of bugs in the above code; I have only proved it correct, not tried it." -- Donald Knuth
There are no modern nuclear reactors running commercially in the United States.
And that's the problem - the United States is not part of any "modern nuclear age.". We're stuck in the 1950s and 1960s, design-wise - retrofits really don't substitute.
"It is our blasphemy which has made us great, and will sustain us, and which the gods secretly admire in us." - Zelazny
Solar doesn't require batteries. It can feed directly into the grid via an inverter. Solar panels are near 100% recyclable and most manufactures have free recycling schemes. The carbon payback from manufacturing is as low as 1 year.
You also need to stop thinking of solar as a domestic production source - that's just perverse. Solar on industrial scales is already approaching parity with coal power stations and was cheaper than nuclear last year.
And yes, yes, it doesn't produce power at night. Maybe you've heard of power storage, which is already used in many places to help balance grid loads.
There are plenty of challenges, but so many geeks have blinkers on when it comes to solar.
lemonade was a popular drink and it still is
It isn't just the "nuclear nuts", though they probably haven't improve the R&D supply. Properly decommissioning a plant, especially one that really deserves it, is not inexpensive, and turns a reasonably profitable(once the construction/startup expenses have been amortized or written off) baseline unit into a big cost center. There is, thus, a strong built in incentive to keep patching and running as long as possible. Best case, you can continue to use the plant as a generating asset. Worst case, if you've had to make a number of repairs that compromise capacity, it may well still be cheaper to keep the lights on and the plant "operating" than it is to tear it down.
To put it another way, at any instant the fusion reaction vessel contains about 1 second's fuel, whereas a fission reactor contains more than two years fuel. Extrapolating to the limit (which is not reasonable, but informative), in the worst accident possible by the laws of physics, the fusion reactor will blast of one second's output from the plant and then be inert, which the fission reactor will blast off an unknown fraction of that two years output and keep the rest in a dangerously grumbling state.
Consciousness is an illusion caused by an excess of self consciousness.
Much like for a teacher who only gives out A's being a phoney, having a review hand out a failing grade give me more confidence in the system. It shows that the USG is not glossing over problems.
Democracy is a sheep and two wolves deciding what to have for lunch. Freedom is a well armed sheep contesting the issue
No, it merely underscores that we do not *have* a "modern" nuclear age.
People, please remember that the vast majority of nuclear reactors in use were built in the 50's and 60's. They were built based on early reactor designs. Reactor designs have improved considerably in the last 20 years but because the public basically has a "no nukes" position, very few new design reactors have actually been built. We are still basically running old reactor designs, many of which are long past their design lifetimes. Until we replace them with modern, safer reactor designs or forms of renewable energy, there will be a danger of another Fukushima/Chernobyl type of catastrophe.
Following the Fukushima accident I've asked several times about the Davis-Besse near miss. What happened there was that boric acid had beed leaking undetected from a crack onto the reactor chamber for more than ten year. When it was finally discovered, it had eaten through the 20 cm of the pressure vessel's steel (the so-called "first containment chamber"); the remaining barrier containing the reactor's material was the 1 cm (or 5 mm, not clear) internal stainless cladding of the vessel, bearing alone the 170 bars of internal pressure. The cladding had bulged but did not break - by mere luck one would say.
Had it eventually given, then the high-pressure reactor coolant would have escaped in a jet; due to the location of the leak, it could have jammed the adjacent control rod mechanism, preventing insertion of the rods. So the Davis-Besse plant was literally at that time half-an-inch away from a total loss of coolant accident with a core on full power and no way to stop it. Right in Ohio, in the middle of the US. What would have happened then? I've asked several times but the only response I got was basically Nothing to see here, move along.
Not that I like to dwelve in shaden-freude but really this kind of answer, coming from people who pride themselves so much of being smart and rational, looks disturbing. Shouldn't we try to assess the reality of the situation rather than build a fantasy world that suits our desires, conveniently ignoring uncomfortable facts?
We should have continued building and updating designs over the last 30 or 40 years, but anti-nuclear nuts have left us all pretty damn screwed.
Blaming anti-nuclear people for the lack of upgrades/maintenance of existing nuclear plants is wrong.
The real problem is that energy companies don't allocate enough money to that matter. As long as it works and produces energy, they keep maintenance to a minimum level to maximize profits.
Aye, I am all for it. Especially remove the arbitrary regulations regarding liability and let the power companies fully insure their reactors themselves. Wait, what? No insurance company would be willing to do that? Score one for the free market!
Ubi solitudinem faciunt, pacem appellant.
That is exactly the problem. This is no different from the tragedies frequently encountered in coal mines. They cut corners and costs in the name of greater profits. And then when bad things happen, they say "whoops! This is an isolated incident. And we will fire someone for doing what we encouraged and even told them to do!"
The problem is in most cases nobody is explicitly told to cut safety. What they are told is "Here is your budget, do everything". Most mid level managers don't have the balls to reply "Sorry can't do everything with that budget", and instead bounce it downstairs to where it finally gets to the team responsible for execution. They're given tasks which take 36 hours per day, and when they don't get done in the timeframe alotted everybody shrugs and says "we will get to it next week".
When bad things happen everybody starts pointing fingers.....guys upstairs saying "I told them to do it", guys downstairs saying "didn't have enough time/people/resources", and the lawyers saying "isolated incident".
Something this dangerous should not be in the hands of profit making corporations...the budgets are always set so the profit margin is there. As the plants age the budgets for maintenance need to increase eroding overall profit. Today nobody worries about profit in 30 years.
The precise degree of regulatory capture at any given time is going to be a politically determined matter; but you really can't expect any other stance: Nuclear plants are very expensive to build, and very expensive to decommission; but the cost of fuel is low, and the cost of temporary-turning-into-permanent-on-an-installment-plan 'disposal' of fuel is also fairly low. Thus, unless the maintenance situation is so bad that you have a crack squad of Godzilla slayers on staff, the economics are basically never in favor of replacement if you can keep the sucker running. Even if you can't, decommissioning costs are likely t dwarf the costs of putting it on some sort of "standby" and leaving it until you can retire away from the problem.
It's very much unlike, say, gas units, which are pretty cheap to put up and tear down; but burn fairly expensive fuel(and, worst case, just sort of explode a little bit, spreading not-very-scary natural gas combustion products), where the economic incentives to take down old plants and put up more efficient ones work out comparatively well.
The NRC, on the other hand, is pretty much in the business of delivering bad news in order to head off low-probability, but very bad, potential accidents. People that unpopular need institutional cultures of iron to avoid subversion.
I don't think you can absolve the "invisible hand of the Free Market" from blame in this regard.
"Cost-cutting" has seemed to be an on-going theme in nuclear disasters.
You are welcome on my lawn.
It seems every time there's a problem with a nuclear power plant, some people trot out the excuse "Oh, it was an old design", like that's supposed to make things better.
The fact remains, we keep nuclear power plants running for decades. Just like all power plants of that generating capacity, nuclear plants are hugely expensive to build, so you need to keep them running for decades to make them cost effective. If we're going to declare nuclear power designs obsolete and unsafe so soon after they are built, then there is no way they will ever be cost justified.
You can't handwave the problem away by saying "they're old".
dragonhawk@iname.microsoft.com
I do not like Microsoft. Remove them from my email address.
Does this further erode the argument that Fukushima was just an isolated incident in the 'modern' nuclear power age?"
The principles of reliable and robust engineering and risk management do not change no matter how "modern" the device. Fukushima was fundamentally not a failure of technology but one of risk assessment and mitigation. They knew that an earthquake and tsunami combination was a virtual inevitability but they failed to build the seawall protections and backup generator system to withstand the most severe events that could reasonably occur. 9.0 earthquakes occur fairly regularly along the Pacific rim. It was absolutely possible for engineers to build adequate protections but for various reasons (cost undoubtedly among them) they chose not to. Despite the design being an older design the problems at Fukushima still could have been prevented with adequate backup systems and/or improved seawalls.
When auditing risks you evaluate three things: Frequency, Severity, and Detectability. When talking about nuclear plants severe events are fairly rare but the potential severity is extremely high. That's potentially ok if the risk is detectible but as Fukushima illustrates, sometimes flaws are only obvious to the people looking after the fact. Complexity typically increases frequency of problems and decreases their detectability. Nuclear plants are unquestionably complex and some parts of them are difficult to evaluate for problems.
The problem with the analysis is that it's still possible to underestimate or even completely miss a failure mode. The engineers at Fukushima clearly understood the severity part of the equation but they seem to have underestimated the frequency or likelihood of a 15 meter high tsunami and then failed to develop adequate mitigation plans. Sadly this sort of mistake is all too common in every human endeavor.
These are old reactors and due to "environmentalist" blocking of building new (safe) ones they are kept functioning. Is it strange they start to rot?
There is no such thing as a 100% safe nuclear (fission) plant. These plants are designed by people and even the best intentioned people make mistakes. We might decide the risks are acceptable but there will be risks. Newer designs have the potential to be safer (safer not safe) but without adequate risk analysis and maintenance, they can be every bit as dangerous as older designs.