How likely is a Fukushima-style cooling-failure scenario after a huge tornado outbreak in the US?

I am about as qualified to answer this as Homer Simpson is, so dont take my word for it. I suspect that wind alone would not be enough to cause problems, even some flying debris should be no problem. Japan was hit by one of the largest earthquakes in recorded history, and a tsunami. I have a lot of respect for the power of natural forces, I’m not trying to belittle what a tornado can do by any means, but there are disasters and then there are disasters.

p.s. If i am wrong, and a tornado could cause a power plant to have a meltdown….. ummm, fix it? anyone for fix it? We have concrete, we have a tape measure, make the walls thicker or something.

A meltdown is caused by a loss of cooling. The earthquake in Japan did not damage the reactor core casing and neither did the tsunami. What needed to be fixed is cooling capability. Take a look at this:

“Storms in the US caused a loss of external power at three nuclear reactors at the Browns Ferry nuclear power plant in Alabama which were in automatic shutdown after external power was cut to the plant. The powerful storms and tornadoes that swept the region yesterday, killing more than 250 people, downed much of the local transmission network, causing large blackouts. According to an incident report submitted to the US Nuclear Regulatory Commission last night, an emergency was declared at the plant following a loss of external power at 16:35 CDT. Backup diesel generators kicked into action to keep the reactors’ cooling systems operational, and although some power has been restored to the plant since, the backup generators will continue to operate until full power is restored, according to updates on the TVA website. The reactors at Browns Ferry are of a similar design to those at Fukushima-Daichii.

Nuclear power plants have multiple incoming power lines, so that if one or more should fail external power remains available. A total loss of external power at a nuclear power plant, as seems to have happened at the Browns Ferry nuclear power plant, is “not very common”, says George Felgate, managing director of the World Association of Nuclear Operators (WANO).”

http://blogs.nature.com/news/thegreatbeyond/2011/04/storms_in_us_cause_loss_of_ext.html

So what if the backup generators don’t work? How could strong tornadoes influence the operation of these backups?

The saving grace in your scenario, @mattbrowne, would be that tornadoes, as bad as they can be (and have been) are generally “local” events. Even massive tornadoes that combine and cover relatively large areas (for tornadoes) don’t wipe out the entire infrastructure of roads.

So even if your assumption proved true (which is a stretch in itself, but let’s suppose “it could happen here”) I cannot envision a realistic scenario where a backup power supply can’t be trucked or even airlifted into place within a day.

As for emergency shutdowns and 100% blackout power plants, that’s a not-at-all-uncommon occurrence. Every power plant will have events of one kind or another that lead to complete loss of power for varying lengths of time, and that’s accounted for, not only in “disaster” and “emergency” planning, but in day-to-day operational planning. It’s really a non-event until things start to cascade as they did in Japan.

Frankly, I can’t imagine the non-LOCA (Loss of Coolant Accident) event that would disturb what was happening in the containment vessel regardless of what happened outside. The buildings themselves – the concrete containment and the steel-frame auxiliary building which holds non-nuclear processing equipment as well as the turbine-generator set, are designed for very high wind speeds. The skins of the buildings could be ripped off in a tornado; that’s likely enough, and that would look pretty ugly, but the buildings themselves would survive just fine.

@WasCy – Suppose there’s a lot of debris around the plant. How would airlifted new backup power supply be connected to the system?

Personally, I think nuclear power is way too dangerous for us to be playing with. We are going to wind up blowing up the planet. And, yes, of course natural disasters such as tornados or earthquakes could damage a nuclear reactor and cause catastrophic failure.

@mattbrowne: Temporary cabling, same as it would be hooked up in any case.

@Skaggfacemutt: “Nuclear power” isn’t going to “blow up the planet”. If it’s not contained right, then it can cause radioactive contamination over large or small areas, but “blowing up” isn’t part of the scenario for nuclear power. (Other than the potential for pressurized steam accidents that can happen with any kind of thermal-based steam generated power production.)

A lot of plants in the U.S. have dead man switches. I’m not exactly sure how they work, though, and whether or not they require power.

I like how one writer whose blog I follow put it: “Comparing keeping nuclear weapons to running nuclear power plants, in terms of safety, is like comparing keeping a gun locked safely in your house and heating said house with ammunition.”

@incendiary_dan

The deadman systems in US pressurized water reactors involves boron control rods suspended in the reactor head. (Boron absorbs the excess neutrons that would continue a fission reaction, effectively killing the fission process.) The suspension is accomplished by means of “always-on” solenoids. Kill power to the solenoids, and the control rods drop into the reactor automatically, and the fission process is stopped.

The other thing that’s somewhat equivalent to a deadman switch (though I confess I no longer recall how it’s activated) is the Safety Injection Tanks suspended around the top of the containment vessel. In a LOCA, the Safety Injection Pumps (generally running four different pipe lines on two separate and independent electrical buses) dispense their contents into the bottom of the containment vessel (assuming the worst, that the reactor pool itself is also breached), keeping the fuel bundles covered with borated water to cool the bundles and kill the fission process.

This is one of those things where I think that ignorance induces fear yet people would rather be afraid of nuclear power than learn a little science and engineering.

Suffice it to say that US plants operate a bit differently than those elsewhere. If nothing else, TMI made it politically impossible for there not to be at least a dozen levels of redundancy and more backup/contingency plans than you would believe. As I said elsewhere before, the basic tenet for a nuclear reactor is to think about what can go wrong, know what to do about that, have a plan in place in case that fails, have a plan in place in case that fails, have a plan in place in case that also fails, have a plan in place in case that fails too, have a plan in place in case even that fails, and then make a backup plan just in case things go really wrong.

Now, @WasCy is correct as far as he goes, but that is just the first step. We tend to be a bit more paranoid than other places, and in this instance, paranoia is actually a good thing.

@Skaggfacemutt How much do you actually know about reactors?

Well, it’s possible to have too many layers of redundancy. From a practical point of view, if you have layers of redundancy, and that redundancy is engineered and built into the system and factored into safety margins and procedures, then the system requires that all of those layers be normally functional. So if routine checking and maintenance of a minor part of your fourth or fifth layer of backup is keeping it from being operational, then your plant is down, which is a huge economic loss. When your plant is down for a non-critical reason (or the critical reasons pile up on top of each other), then the losses become unsustainable. So there has to be a sense of practicality: ‘what is likely to go wrong, and how can that be backed up?’ (That’s after you solve for: ‘what would kill the plant if it failed, and how can that be prevented?’)

Experience teaches us the things that are likely to go wrong over time. It’s one of the ways that engineering develops: discover a new problem and then solve for that, while retaining solutions for all previously known problems. That’s why cars are built differently now, for example, than they were just 10 years ago, and why nuclear power will be safer tomorrow than it is today. It’s an iterative process. Unfortunately, disasters occur on major and minor scale to give us the “experience”, since we can’t know everything and we can’t very well predict things (or the full effects of things) that have never happened – such as a tsunami wiping out the support systems for a nuclear power plant. (In hindsight, one wonders why engineers at the most seismically active place on Earth, and at the place where they invented the word tsunami, did not do a better job of predicting those things, but that’s in the past now.)

@WasCy I was thinking about a hydrogen-bomb kind of boom. Glad to hear that at least that wouldn’t happen – but releasing radioactivity into the air is pretty scary, too.