Comments: Cool It, Rover!

Wall Street agress with you as to the safety and the liability of these nuke plants. They have no intention of even loaning money to the Nuclear Power Plant people.

Therefore, the funding for the next phase of America's renewed committment to more nuclear power comes from the President himself. He is fully intent on honoring his promise to give at least 45 billions of dollars (and possibly fifteen billion more than that) to his good friends at GE.

I mean, what could possibly go wrong!

Posted by Elise Mattu at August 25, 2011 10:20 PM

I would have walked away with the soup, ate it and left.

Posted by Susan at August 25, 2011 11:50 PM

Would that mean the cooling system is essentially dissipating a NERVA rocket's worth of heat while the plant is in operation?

Posted by Kent at August 26, 2011 04:01 AM

That trifling little quake--which I didn't even notice within a short drive from aaron's surfing adventure (though it did knock down the shelf of one ikea bookcase in my youngest son's room)--that trifling little quake almost reached the design threshold for a nuclear plant?

I used to think the invisible hand was in our pocket, but now I think it's on our throat.

Posted by N E at August 26, 2011 08:15 AM

i had the pleasure of experiencing this earthquake approximately twenty miles northwest of aaron's location. i would not describe it as "triffling" - it was, in fact the loudest and shakiest quake i have experienced in all my life, but i have no doubt been fortunate compared to some - nothing in my home was damaged, although my cat and i had our equilibrium disturbed

i agree with aaron's conclusion, It's madness to believe that this technology can be harnessed safely.



by the way, how 'bout them fukushima exclusion zones?

Posted by mistah charley, ph.d. at August 26, 2011 09:14 AM

@Kent - that's about right, yes. A 1GWe (GigaWatt-electric) plant must generate between 2-3GW of thermal power (I'm not sure of the exact number). 1GW goes out on the electric lines, 1-2GW gets dumped as waste heat into the thermal reservoir (Lake Anna, the Missouri River, the Susquehanna River, etc.)

Posted by Aaron Datesman at August 26, 2011 10:51 AM

Now we will see how it does with Irene.

Posted by Dredd at August 26, 2011 11:24 AM

Thanks for the reply (and for all your articles on this issue).

This post brings together a few things that have been bugging me for quite some time (going somewhat off topic to global warming) about the interaction between nuclear and OHC/Ice model under-prediction of warming impacts. Your answer simply enhances my concern. The idea was inspired by memories of the warmth of Lake Anna.

I've been kicking it around for a while and gotten to the point I'd almost call it a Hypothesis (admittedly, as a software engineer/programmer, these are topics I am not really qualified to research). I'm in the early stages of writing it up and bringing together some relevant studies ... and it's pretty slow going with the learning curve and all. It has occurred to me this is an awful lot of work trying to flesh out an idea that may end up appearing obviously stupid to any 1st year physics major.

You seem to have a pretty strong grasp on the basic particle physics/thermodynamics formula applications that I simply don't ... and seem game to play technical thought experiments. Any chance you could take a look and see if my logic can be trivially dismissed?

(a link just in case the answer is yes)
http://kentbutler.com/index.php?option=com_content&view=article&id=5&Itemid=2

Posted by Kent at August 26, 2011 01:05 PM

@Kent - a quick envelope-back calculation shows about 0.01 W/m^2 forcing due to waste heat dumped into water. My recollection is ~0.3 W/m^2 forcing due to excess CO2, which places this in the ballpark of "worth looking at". But please understand that I am not an expert.

Why don't you contact me offline at datesman@gmail.com? I would be interested to discuss. In return, perhaps you could teach me how to run a Perl script. I have been meaning to do some simulations relating to posts I wrote here in April.

Posted by Aaron Datesman at August 26, 2011 03:34 PM

So even a snow storm could potentially knock out the power? Lovely.

Posted by Amanda Rex at August 26, 2011 03:58 PM

"Before dawn, a piece of heavy equipment nicked an eight-foot-high, 2,000-foot-long temporary rubber berm, and it deflated. Water also began to approach electrical equipment, which prompted operators to cut themselves off from the grid and start up diesel generators."

Really? And we call ourselves intelligent beings? I'm thinking it should have been Epimetheus instead of Prometheus erected in front of the Chernobyl plant.

Posted by Amanda Rex at August 26, 2011 04:04 PM

P.S. My environ. studies professor said something about the next projected quake potentially happening in Tennessee. Get ready for round two.

Posted by Amanda Rex at August 26, 2011 04:16 PM

I have, for years, been of the opinion that nuclear power was far too dangerous to use. Now, with climate change accelerating and nothing being done about it, I am starting to have doubts. Here's a link to an explanation that really shook my convictions:

http://www.straightdope.com/columns/read/2988/is-nuclear-power-safe

I was wondering if you guys have any thoughts on this problem. Could it be that the risks presented by nuclear power are far outweighed by the impending disasters wrought by climate change?

Posted by Liebchen at August 26, 2011 04:40 PM

@Aaron Datesman:

You further write that “In terms of heat generation, a commercial nuclear power plant in the OFF position basically has the equivalent of a NERVA rocket sitting inside it…set to launch the minute the cooling system fails.” The insinuation is that a reactor in shut-down has the ability to suddenly ramp up to the 1100 MW heat output of a nuclear rocket engine. This is gossly false. A reactor with one gigawatt electrical output will have a shade less than three times as much thermal output, so let’s say 3 GW thermal output. When the reactor is placed in shut-down-“in the OFF position”—the chain reaction stops and so does almost all the heat production. The chain reaction cannot be started up again unless the operators do so deliberately; even if the reactor melts down, the fuel will be in a non-critical configuration incapable of supporting a chain reaction. What’s left after the shutdown is the residual heat production from decaying fission products in the fuel. At the instant of shutdown, this is less than 7% of the heat produced during normal operation and within one hour it drops to 1.5 % of normal thermal output, or about 45 MW--roughly 4% of the 1100 MW you suggested in your analogy. (http://en.wikipedia.org/wiki/Decay_heat) From there, the thermal power output continues decaying exponentially. This residual thermal energy is enough to cause the sort of damage we’ve seen at Fukushima—chemical reactions that produce explosive hydrogen, along with melt-down and spew—but to liken the destructive potential to that of an 1100 MW nuclear rocket engine blasting off is fanciful.

Posted by Will Boisvert at August 26, 2011 04:44 PM

It appears from witness accounts that the earthquake at Fukushima caused numerous pipes in the cooling system to break and taking the cooling system out of service before the power went off. The back-up generators are really moot at that point as the pumps are now squirting the cooling water all over the place. Nobody is mentioning the condition of the cooling system itself, and I have to wonder if pipes were cracked if not broken by the tremors.

Posted by Pragmatic Realist at August 26, 2011 05:10 PM

positive void coefficient of reactivity:

http://tinyurl.com/3pmfg7bsig2=imwKb112hP_7_1VJB-afyw&cad=rja

Posted by Amanda Rex at August 26, 2011 05:17 PM

@Aaron Datesman:
I believe you drastically over-estimated how much electricity is needed to cool a reactor. You write: “Of the electrical power generated by the nuclear plant, one-third is directly consumed in the operation of the plant itself.” That’s an extraordinary claim that doesn’t jibe with published information on the gross and net electrical capacity of the North Anna plant. (http://en.wikipedia.org/wiki/North_Anna_Nuclear_Generating_Station)(The difference between the gross electrical capacity of the turbines and their net electrical capacity that can be sold to the grid is the electrical power devoted to plant operations.) North Anna has two reactors. North Anna I has a gross electrical capacity of 973 MW and a net capacity of 903 MW, for a difference of 70 MW for plant operations. North Anna 2 has a gross capacity of 994 MW and a net capacity of 972 MW, for a difference of 22 MW. These data indicate that the plant as a whole requires 92 MW for its own operations, out of a total electrical capacity of 1,967 MW—that is, 4.6%% of output, not the 33% you suggested. These 92 MW would run about 6,000 homes, assuming an average household electricity usage of 15 KW. The electricity needed to run reactor cooling systems would be some smaller portion of that power. So it seems that the power required to cool a reactor after loss of outside power and shutdown is modest and well within the capacity of the plant’s backup diesel generators, provided these have not been destroyed by a tsunami.

But perhaps I’ve overlooked something. It would help readers assess this issue if you could provide checkable (preferably online) references with quantitative figures that substantiate your claims that 1) 33% of North Anna’s electrical output is required for plant operations; and 2) most of that power is required specifically for reactor cooling systems.

Posted by Will Boisvert at August 26, 2011 05:28 PM

(I realize NA doesn't have the same reactors as Chernobyl-my link is just a vague response to a vague assumption)

Posted by Amanda Rex at August 26, 2011 05:47 PM

@ Amanda Rex

Your link on "positive void coefficient of reactivity" led me to an error message.

In any case, light-water reactors like those in Japan and the United States have a "negative void coefficient of reactivity." That means that when they lose the light-water coolant, the chain reaction shuts down. That happens because the light-water coolant is also the "moderator"--the substance that slows down the neutrons (by collision with hydrogen nuclei) so that they can be efficiently absorbed by uranium nuclei to cause a fission event. If the light-water coolant/moderator leaks or boils away or is greatly lessened in density by steam bubbles, the chain reaction automatically shuts down without any human or mechanical intervention. So all the danger in a loss-of-coolant accident comes from residual decay heat from fission products, not from heat caused by the chain reaction, which stops immediately.

Posted by Will Boisvert at August 26, 2011 05:50 PM

mistah charley

I didn't mean to suggest the quake was trifling subjectively, but nobody in the Office Depot I was in (within a half hour's drive of Aaron and mistah charley) even noticed that there was a quake (I heard about it shortly after that on the radio).

In contrast, my wife was near the epicenter of the Whitter quake in California in 1987 in a car, and her car bounced up and down in the air hard for about 30 seconds, and the highway swayed up and down rising above the windshield and then smashing back down to the ground hard over and over for the same period, and the street lights bent at a 45 degree angle several times as she watched during those 30 seconds, all of which got her attention, especially because she was stuck in traffic under an 8-land overpass built by an engineer to specs she hadn't studied. And you have to wonder in those circumstances if it's supposed to bend like that.

Trifling compared to that, I meant, and even that Whittier quake in 1987 was an itty bitty 5.9 quake, not something at 8 or 8.5, which does happen here and there now and again. It seems to me that if it happens anywhere the plants are built to withstand a mere 6, the price of iodine will skyrocket.

Posted by N E at August 26, 2011 06:08 PM

Lazy version: http://www.google.com/url?sa=t&source=web&cd=1&ved=0CBsQFjAA&url=https%3A%2F%2Fnetfiles.uiuc.edu%2Fmragheb%2Fwww%2FNPRE%2520402%2520ME%2520405%2520Nuclear%2520Power%2520Engineering%2FTitle-Preface.pdf&rct=j&q=Dr.%20Magdi%20Ragheb%2BChernobyl%20Accident.pdf&ei=zxhYTpH_DYWDgAes972FDA&usg=AFQjCNH9ASz0188htetm4a-JvYBxV7Ih3A&sig2=3sJUwHS8-1mJBaJTZGCwwg

I won't attempt to have a technical argument because 1. there's a 95% chance every person on this thread knows more than me, and 2. I just don't care that much for arguing over trivial details when the actual issue is so much larger. My link is to a paper on the Chernobyl accident, and is entirely a response to a very general assertion that reactors (of any kind) cannot cause a large explosion.

Posted by Amanda Rex at August 26, 2011 06:21 PM

And damn if I didn't link the wrong thing! If you still want to even see it after this many failed attempts: http://www.google.com/url?sa=t&source=web&cd=10&ved=0CFoQFjAJ&url=https%3A%2F%2Fnetfiles.uiuc.edu%2Fmragheb%2Fwww%2FNPRE%2520402%2520ME%2520405%2520Nuclear%2520Power%2520Engineering%2FChernobyl%2520Accident.pdf&rct=j&q=RBMK%20reactors%20can%20develop%20what%20is%20known%20as%20a%20positive%20void%20coefficient%20of%20reactivity.&ei=wQlYToiiLtPngQewwIC6DA&usg=AFQjCNH8XbyuGjEI-WRjownDX18gVF3olg&sig2=imwKb112hP_7_1VJB-afyw&cad=rja

Posted by Amanda Rex at August 26, 2011 06:27 PM

@ Liebchen:

Your question is an excellent one. In my opinion the benefits of nuclear power greatly outweigh its risks, which are drastically overstated by anti-nuke alarmists. When you crunch the numbers properly, nuclear power, Chernobyls and Fukushimas included, is the safest form of energy bar none—especially when compared to the ravages of coal power, which kills hundreds of thousands of people every year. Switching to a completely nuclearized power system would reduce the number of lives lost to energy production by 90% to 99% (See my essay at www.slackwire.blogspot.com, http://slackwire.blogspot.com/2011/07/trying-to-be-for-nuclear-power-when-it.html for a thorough analysis, including estimates on the likely casualties from the Fukushima disaster.) Nuclear is also the only carbon-free energy source that can shoulder the burden of powering a modern economy—wind and solar are simply too feeble, fickle, expensive and land-devouring.

It pains me to say that the postings on nuclear power that I’ve read on this blog—alas, too late to make the comment deadline—have been extraordinarily misleading, and do a grave disservice both to readers and to the cause of sustainable energy.

Posted by Will Boisvert at August 26, 2011 06:28 PM

Aaron

I didn't realize that you have been misleading me with all your knowledge and information, and I am very disappointed in myself for being taken in by it. Stop doing a grave disservice to readers by expressing your extraordinarily misleading opinions, and most of all, start allowing late comments that would demonstrate how safe nuclear energy is compared to coal. It only seems fair to give the proponents of nuclear power a chance to express themselves.

Posted by N E at August 26, 2011 09:43 PM

@Will -

Thanks for chiming in. Please accept my apologies for not moving forward with the blog-match we corresponded about earlier. My life intervened. Perhaps we can move forward with it next week; I will reach out to you off-line.

Regarding your first post, I'm willing to admit that the last line is a bit poetical. You are correct that, if everything goes as it should and the reactor scrams as it is designed to, the prompt neutron flux should disappear quite quickly. Of course, as an engineer, if complicated machines routinely performed as they were designed, you would think I would have observed it at least once by now.

The point I wished to illustrate is the quantity of heat involved. We're sold this benign model that a nuclear reactor is pretty much a friendly old tea kettle, just serving to boil water. The reality, in my opinion, is that a nuclear reactor is much more like a NERVA rocket. Within certain limits, we are able to operate this rocket continuously at low power by imposing a control system upon it. (I mean this in the technical, engineering sense.) But what it really wants to do is shake off our control, and take off (or melt).

So, while I agree that the prompt neutron flux will disappear quickly, it's equally important to note that the reactor possesses tremendous thermal inertia. Just describing the falloff of the prompt neutron flux when the reactor scrams or trips is not by any means a full transient analysis. The initial condition matters too.

I do not agree that fission chain reactions will necessarily stop if the reactor melts down. For instance, there were indications of continuing fission at Fukushima long after the cores had melted down. This contention is not supported by basic physics, either. Specifically: a) even in the absence of prompt neutrons, there is always a flux of delayed neutrons, and b) criticality is a function of (in addition to other factors) geometry. Melting the core certainly changes its geometry. Therefore, it seems to me to be possible to observe a self-sustained fission chain reaction in a molten blob which used to be a reactor.

I do agree that it probably won't go on forever, and that it probably won't generate 3GW of thermal power. Somehow this doesn't console me much.

Regarding the 1/3rd number, this is what I was told by one of the plant managers at North Anna when I toured the facility in 1996. I remember it perfectly clearly, as I thought it was amazing. I will dig around a bit and see whether I can verify it. Certainly somebody in DOE must know the answer.....I hope.

Posted by Aaron Datesman at August 26, 2011 10:36 PM

Loss of on site backup power was the primary cause of the Fukushima disaster, when they were flooded, but we have never been told if the shock of the quakes or quakes themselves did damage which contributed to the problems. Damage to piping and electrical systems probably did occur there as well as structural damage which caused leaks and water loss in the fuel pools.

When they speak of what the design is built to withstand in quakes those are the things they refer to. As it turns out backup power is the most important safety feature of all. Backup of backup power should become mandatory. That means either redundant on site ideally, or the maintenance of regional portable backup generators, heavy lift helicopters to deliver and a standardized design for hooking the power up. Neither or which will happen because of the cost of course.

In Japan with so many plants and the small area of the country it was almost bizarre to me that it took as I recall a couple of weeks to even begin to get portable generation equipment there. Then they had the problem of tying it in the systems. I assume there was way to do that built in so they had to cobble it in.

Posted by rapier at August 27, 2011 08:11 AM

@Will B -

In this link, IAEA states the following:

- Severe disturbance, such as grid power failure. The NPP may be required to cut off all external ties, to run back to supply its own house load, and to be ready to accept load again up to nominal power soon after the disturbance has been cleared. During house-load operation, only about 5% of nominal power is generated by the generator while the reactor power is maintained at a higher level. (p.29)

This about agrees with your numbers (~7% for decay heat, about 4% for house load power if the reactor is shut down), although I have not thought through whether I agree with how you arrived at the second answer.

This second document, also from the IAEA, says the same thing:

During this "house-load" operating mode, the reactor operates at a reduced power level that is still sufficient to assure enough electricity for its own needs, typically 5% of full power. (p.9)

Apparently it's IAEA's opinion that, for a 1000MW commercial reactor, it requires 50MW of house-load power to cool 70MW of decay heat. Do you agree with this?

The first reference, on the same page, also says this:

These operating features enable the NPP to be independent from the grid and to keep the main pumps running (normally, the emergency electric power system with diesel set does not enable the start-up of the large primary coolant pumps and the boiler feedwater pumps.)

In my opinion, this is a very clear statement that the NPP will enter a perilous state (possibly leading to meltdown) if the plant loses its utility connection and the reactor fails to scram as designed. A good engineer asks what happens if this valve sticks or that motor burns out or the foundation shifts. It's an enormously complicated piece of machinery.

The operating power exceeds the initial decay heat by a factor of about 14x; if we scale the house load requirement by the same factor, we get >700MW to run the really big pumps required during power generation. I would expect sublinear scaling, bigger equipment being more efficient, but in view of these numbers from IAEA what I was told in 1996 (33% of the electricity generation goes to operate the NPP) seems very plausible.

I am interested to learn what the true network of power flows looks like in an operating reactor; I can not recall ever seeing one. If I can find that information, I will share it.

Best, AD

Posted by Aaron Datesman at August 27, 2011 05:05 PM

@ Rapier,

You’re right; one of the main lessons of Fukushima is that nuclear plants need to pay much more attention to back-up power. They actually do have backups of backups already—many redundant diesel generators, and electric batteries—but these need to be placed in hardened structures, some of them offsite, with adequate heavy lift capacity, and interoperability, and realistic training drills to avoid the Keystone Cops scenes we saw at Fuku. People who want to improve the safety of nuclear power instead of just abolishing it should harp on this issue. I think your claim that these measure are so expensive that they would never be implemented is unduly pessimistic; have you seen any relevant cost data?

Posted by Will Boisvert at August 27, 2011 10:47 PM

@Aaron Datesman,

On the question of how much electricity a nuclear power plant uses to run its cooling pumps, I believe you have misread the IAEA documents.

The “house load” they refer to is a common term for the amount of electricity that a power plant requires when it is operating at full power. That’s how it’s used in this source, and others. (http://en.wikipedia.org/wiki/Sizewell_nuclear_power_stations)

You seem to be under the impression that the 5% house load is the amount of electricity necessary simply to dissipate residual decay heat after reactor shut-down, and that this load must scale linearly to dissipate the much greater heat generated by the reactor at full power. That’s not true; house load is all you need to run the plant at full power, no scaling necessary.

The scenario discussed in the IAEA documents you cite is for loss of grid power with the reactor still online, not shut down. That is, the reactor is functioning and prepared to resume grid connections and ramp up to full power on short notice to shoulder grid load. In this state of readiness, the reactor is still powering all of the plant’s systems as normal, including “the really big pumps required during power generation” that transport coolant at full power. In fact, the documents clearly state that the reactor power, though reduced, is assumed to be substantially above house load, which means that the reactor is channeling just a fraction of its output to the turbines and dumping the rest as heat. That makes sense since the scenario is for a standby mode.

The documents assume this house load to be 5% of full power. This figure is in good agreement with the sources I cited, which put the difference between the North Anna generators’ gross capacity and net capacity—which is another conventional way of stating a power plant’s internal electricity usage (http://en.wikipedia.org/wiki/Net_generation)—at 4.6 % of full power.

I’m afraid you really are mistaken in asserting that nuclear plants use 33% of their power for internal operations. The correct figure is about 5%, (and the amount needed for emergency cooling operations is some smaller fraction of that). All the evidence and sources demonstrate this conclusively. I think you are simply misremembering something a tour guide told you 15 years ago.

Posted by Will Boisvert at August 27, 2011 10:48 PM

@ Aaron Datesman,

Regarding heat generation in a shut-down reactor:

--Upon shutdown, prompt neutron flux dissipates within seconds, delayed neutrons, which are less than 1% of total neutron flux, within a few minutes. After that, all the thermal energy generated by the fuel comes from residual decay heat of fission products; none comes from the chain reaction, which is extinguished. So, within seconds, the thermal output of the reactor drops to under 7% of full power, and then to 1.5 % of full power within an hour, and so on. (Granted, there’s still a lot of heat in the hot water, but that’s also a small fraction of the reactor’s thermal output at full power.)

--Your post suggested that a reactor in shut-down is in perpetual grave danger of ramping back up to 1140 MW thermal output in an instant. You’ve now conceded that the thermal power output of a reactor in shut-down is indeed a small fraction of that 1140 MW figure. So now your argument rests on the possibility that the chain reaction could resume in a reactor that’s already been shut down. Again, that’s simply not possible unless the operators deliberately restart the reactor.

--A light-water reactor like North Anna in shut-down can’t restart its own chain reaction through coolant loss and meltdown. and here’s why. First, loss of coolant means loss of the light-water moderator; without a moderator, the neutrons are travelling too fast to be efficiently absorbed by uranium and the chain reaction cannot sustain itself—the laws of physics forbid it. Secondly, a meltdown destroys the very specific, delicate fuel geometry necessary to sustain a CR. Without the right geometry, too many neutrons leak out through the perimeter of the fuel before fissioning another uranium nucleus, and a chain reaction cannot be sustained. A molten, shapeless drip of corium is most emphatically not a critical geometry, and cannot sustain a CR.

--I believe the rumor about renewed criticality in the Fuku melt-down was started by Arnie Gundersen, a paid anti-nuke propagandist. I haven’t seen any authoritative evidence for it. Have you?

--Your example of the Nerva nuclear rocket was an egregiously misleading way to illustrate the quantity of heat involved in cooling a shut-down nuclear reactor. It overstates that quantity by a factor of 25 at the one-hour mark, and more as time goes by. Your larger point that a reactor is a volatile creature champing at the bit to burst into criticality is wrong-headed. It is actually rather difficult to get a reactor to go critical; most of the art of nuclear engineering is devoted to trying to coax to life a lump of materials that would prefer to just sit there inertly.

Posted by Will Boisvert at August 27, 2011 10:50 PM

@ Aaron Datesman,

Let me try to summarize the issues at stake in your post.

Certainly, if a NPP shuts down its reactor and is cut off from grid power, that’s a serious situation. Back-up power is required immediately to cool the reactor and prevent it from melting down and possibly spewing, as at Fukushima. The question your post asks, in the context of North Anna, is this: how difficult a task is it for back-up power to cool that reactor? And you address that question by 1) estimating how much electricity is required by the cooling systems ; and 2) estimating how much heat needs to be dissipated from the reactor by those cooling systems.

Unfortunately, both of your estimates are wildly exaggerated. You insinuated that the electricity required to run the cooling systems is close to 33% of full plant output, where in fact it is under 5%. You then insinuated that the amount of thermal energy that needs to be dissipated by the cooling systems is in the nieghborhood of 1140 MW, where in fact it is a tiny fraction of that figure—under 4% at one hour and dropping. (Your replies on the comment thread concede both of these points.)

By exaggerating these figures, your post conveys a false—pardon me, “poetical”—impression that what was actually a routine, small-scale task for North Anna was a gargantuan existential crisis. This impression is utterly unwarranted by the facts. Here’s what happened at North Anna: a nuclear plant encountered an earthquake and weathered it with flying colors. That’s actually a very banal happenstance that occurs dozens of times a year all over the world. Nuclear reactors are not fragile, fraught or perpetually straining to explode; they are robust, reliable, and safer than other technologies that we take for granted. North Anna just proved it—again.

Posted by Will Boisvert at August 27, 2011 10:51 PM

Aaron--Admitting mistakes is an admirable habit to have.

Posted by N E at August 29, 2011 06:16 AM

Only if ALL the SCIENTISTS were like you Aaron!!!

Posted by Rupa Shah at August 29, 2011 10:58 AM

Getting shit wrong is my number one hobby!

And regardless, you're still the smartest scientist posting to this blog.

Posted by Amanda Rex at August 29, 2011 01:07 PM

"Admitting mistakes is an admirable habit to have.

Unfortunately, that habit is as uncommon as "commonsense"......just as 'Honesty' and 'Integrity' have become rare
Thus, Aaron, your posting of correction is a breath of fresh air.

Posted by Rupa Shah at August 29, 2011 03:11 PM

@ Aaron Datesman:

On backup diesel generators at North Anna.

--In your correction you write: “Had a piece of switching equipment (or a second backup generator) inside the plant failed, it’s likely that a serious incident would have occurred at North Anna last week.”

It’s my understanding that 1 of 5 backup diesel generators (20%) failed at North Anna after the earthquake. Your correction suggests that if a second had failed there would have been a serious accident, presumably an inability to cool the shut-down reactor. But it is also my understanding that each one of the 5 backup diesel generators is capable of powering emergency cooling systems all on its own. The Wikipedia article on nuclear safety systems (I know, not always reliable) agrees (http://en.wikipedia.org/wiki/Nuclear_safety_systems): “[emergency diesel generators] are sized such that a single one can provide all the required power for a facility to shutdown during an emergency situation which allows facilities to have multiple generators for redundancy.” This suggests that all five backup diesel generators would have had to fail for the plant to lose emergency systems, not just two, as you suggest. Do you know of any references substantiating your claim that the failure of just two diesel generators would have caused a serious incident at North Anna?

Posted by Will Boisvert at August 29, 2011 08:33 PM

@ Aaron Datesman:

On emergency reactor shut-down risks:

You write: “What worries me is the assumption that the reactor will, in fact, turn off when the operators instruct it to do so. For instance, what if some corrosion inside the reactor causes a sub-assembly of control rods to stick in their channel halfway down?”

That’s certainly something to worry about. But it’s my understanding that light-water reactors have a second independent backup system to shut down the reactor if the control-rod mechanisms fail, which involves injecting boron, a neutron absorber, directly into the reactor coolant (http://en.wikipedia.org/wiki/Nuclear_safety_systems). I thought such backup systems were mandatory; are you saying that North Anna doesn’t have one? Telling readers about the backup system would help them assess the risk of the reactor failing to shut down when ordered. The odds of two independent reactor shut-down systems failing simultaneously are, of course, drastically lower than the already tiny odds of either one failing by itself.

Posted by Will Boisvert at August 29, 2011 08:34 PM

@ Aaron Datesman:

On the general risks of nuclear power:

You write: “But a low probability of disaster is not the same as zero. Would you care to test the odds again? How about sixteen times, all at once? During the 2003 Northeast blackout, 16 nuclear power plants in the US and Canada lost utility service and went offline. The concerns I’ve outlined about North Anna after the earthquake applied to every one of those plants at that time, too.”

So, 16 nuclear plants were cut off from the grid during a blackout. The result was that the safety systems at all those plant worked fine; no damage was incurred and no one was harmed. Well, yes, I do like those odds. Anti-nukes seem to have the strange impression that whenever a nuclear plant survives a disruption unscathed, it proves that nuclear power is fragile. Logically, it proves the opposite. If 100 plants had survived the blackout, that would have proved nuclear power even safer. The more bullets nukes dodge, the better at dodging bullets they must be. Right? That’s why the conclusion we should draw from North Anna is that nukes are just a bit safer than we thought they were before the plant shrugged off that earthquake.

You’re right that the risk of a nuclear disaster is not zero. No human undertaking is absolutely safe. But the right way to assess nuclear risks is not to dream up disaster scenarios; instead, we should look at the empirical data on the safety performance of nukes in the real world—and then compare that to other energy sources. In the United States, there’s been not a single civilian fatality caused by a nuclear power plant during over fifty years of operation, Three Mile Island included. By comparison, the air pollution cause by coal-fired power plants kills 13,000 people every single year in the United States alone (www.catf.us/resources/publications/files/The_Toll_from_Coal.pdf), and hundreds of thousands of people worldwide. Chernobyl, according to the Union of Concerned Scientists’ Lisbeth Gronlund, will have killed a total of 27,000 people by the time all the radiation decays. (http://allthingsnuclear.org/post/4704112149/how-many-cancers-did-chernobyl-really-cause-updated?34df10c0) (Note, Aaron, Gronlund’s dismissal of the Yablokov study at the end.) That means that Chernobyl caused worldwide, over many decades, barely two years’ worth of the fatalities that coal-power causes in the US alone. Fukushima, a much smaller spew, will cause just a small fraction of yearly US coal-power fatalities. (Coal-fired power plants generate about 42% of world electricity, compared with nuclear’s 14%, but even after normalizing for that discrepancy, coal is still colossally more dangerous.)

By every rational measure, nuclear power is safer—and by orders of magnitude--than the fossil-fueled energy system we have today. Anti-nukes should rethink their opposition to it.

Posted by Will Boisvert at August 29, 2011 08:38 PM

Will: You make good points but I take issue with this:

>> But the right way to assess nuclear risks is not to dream up disaster scenarios

Yes it is. Don't get me wrong. I buy your argument that, on average, nuclear power is considerably safer than coal. But the worst-case nightmare scenario is what freaks people out and therefore needs to be addressed. For coal, we know what it is and it's not too bad (because the worst case is not that different from the average). But suppose that a nuclear accident that happens once in 100 years were to render half of Japan uninhabitable. Being so rare and nukes being so new, this would not have happened yet, so compiling stats (as you do) can't address this issue. But I believe that is the key issue that turns people away from nukes. So to make your case you need to forget about Fukushima, Chernobyl, etc., for a moment, and argue why the annihilation of Japan as we know it is as unlikely as the moon turning into green cheese or Obama caring for the poor. I am not saying the case can't be made. I am saying it must be made.

Posted by bobs at August 29, 2011 09:10 PM

WE'll ALL be in the nuclear physics business for the next 50,000 to 14,000,000 years, no if ands or buts. That genie isn't going back in the bottle. That being said, perhaps We must first look at the true root cause of OUR problem---GE. Corrupt practices lend to poor design and shoddy equipment eventually leading to ruin even the nifftiest science projects.

Posted by Mike Meyer at August 30, 2011 05:05 AM

@ bobs:

You’ve got a point, and I’m going to answer it. But first, a warning about “the worst-case nightmare scenario” that “freaks people out.” If febrile imaginings are to be the final arbiter of energy policy, then no progress at all can be made. They will stymie every initiative to move beyond the status quo, and that includes renewables; already, nonsense about bat holocausts and strobing-induced seizures is motivating NIMBYs to sue and vote against wind farms. Crazier stuff by far will be said—and believed—and if you don’t think so, then I’d like to introduce you to the Republican Party. The only way we can move away from an energy system dominated by fossil fuels is to shift the debate to realistic projections based on scientific expertise and the evidence of experience.

But you’re right: rare catastrophic risks loom larger in people’s minds than statistically worse but routine risks. So here’s my stab at the worst-case scenario. You told me to forget Chernobyl and Fukushima, but I can’t, because the fact is that Chernobyl really is the worst-case nightmare scenario; it just can’t get any worse. Here’s why. Chernobyl was a giant explosion that blew the reactor wide open to the elements and started a huge fire with plenty of graphite to fuel it. (And with nothing but the proverbially sluggish and incompetent Soviet state to stop it.) That fire raged for over a week, lofting tons of radioactive gas and soot way into the sky to spread far and wide. Remember, a nuclear disaster is simply any process that distributes the contents of a reactor outside the plant. (It is physically impossible for a nuclear power reactor to detonate the way a nuclear bomb does.) There is no process that can do that as efficiently, effectively and widely as a big, hot, persistent fire in a reactor that’s open to the sky. So we’ve been there and done that, at Chernobyl.

Posted by Will Boisvert at August 30, 2011 07:24 AM

@ bobs

Similarly, I’m going to argue that Fukushima is as bad as things can possibly get with the light-water reactors used in Japan and most everywhere else. Chernobyl was the result of the uniquely bad Soviet RBMK reactor design. The RBMK had a positive void coefficient, which means the chain reaction accelerated wildly when there was a loss of coolant (hence the enormous explosion) and a graphite moderator (the fuel for the fire.) Light-water reactors have a negative void coefficient, which means that the chain reaction instantly shuts down when there is a loss of coolant, even if there is no human or mechanical intervention. They have no graphite in the core, which means there is nothing to fuel the fire. And they have an immensely strong containment building—Chernobyl had none. So with a light-water reactor, you can’t get the truly worst-case Chernobyl scenario—a huge explosion that blows the reactor wide open to the sky and ignites a raging, persistent fire in it, which is the most efficient way possible to broadcast a reactor’s contents.

Of course, you can still get the Fukushima scenario, which is also quite bad: a meltdown caused by residual decay heat, which generates hydrogen, which causes a lesser explosion that creates partial breaches in containment, through which smaller quantities of radioactive core material make it outside. But Fukushima emitted only about one seventh as much radioactive material as Chernobyl did, so the difference in reactor designs was crucial.

Posted by Will Boisvert at August 30, 2011 07:25 AM

@ bobs

You posit a nuclear accident that renders half of Japan uninhabitable. Again, I think that’s just not possible with any reactor, let alone a light-water reactor. But also remember that the concept of uninhabitability is highly subjective and colored by an irrational radio-phobia. The fact that radiation is not as dangerous as people think it is—you’re soaking in it!—is something else that hopefully can inform public discussion. On the blog www.slackwire.blogspot.com I calculate the likely cancer risk for people hypothetically living inside the 20 km Fukushima evacuation zone. (www.slackwire.blogspot.com, http://slackwire.blogspot.com/2011/07/trying-to-be-for-nuclear-power-when-it.html) I find that, while they would face some increased cancer risk from radiation, it will be on average quite a bit less than a smoker would incur. (Yeah, not great, but not apocalyptic either.) Some places inside the EZ are actually less radioactive than Denver, Colorado. Moving to the EZ is probably less dangerous for your health than moving to Beijing, with its famously lethal air pollution. There are still peasants living in the Chernobyl zone of alienation—some of them are in their eighties.

Look, we’ve had nuclear power plants for 50 years, hundreds of them, operating all over the world in widely varying circumstances. They are not unknown quantities. We do know how often they spew: about once a generation. And we know how bad the spews can get—Chernobyl. It’s not a perfect track record, but it’s improving; the rate of spews per reactor-year is dropping and Fukushima casualties will be nowhere near as bad as Chernobyl’s. Nuclear risks will never vanish entirely—there will be more Fukushimas. But at this point, we really do have a good handle on what the risks are, and we can confidently say that they fall within a range that society can live with—because we already tolerate risks in our energy system that are objectively much worse.

I feel it’s time to drag this debate out of the realm of nightmare conjecture and into the light of empirical evidence.

Posted by Will Boisvert at August 30, 2011 07:28 AM

Thanks, Will. I am in no position to assess the technical merits of your analysis (I hope those who agree or disagree with you will chime in) but I appreciate the fact that you addressed, directly and thoroughly, the question that I asked.

Posted by bobs at August 30, 2011 10:55 AM

Will Boisvert: Nice sales pitch, a little high pressure, but then, that's the industry. OUR technological level with repect to nuclear power and fossil fuels as well doesn't fit well with OUR environment. WE do have off world uses and the ability move the nuclear industry off world. WE simply lack the will at this point. Chemical fires are NATURAL to this world where as intense nuclear energy use is not. WE are not ready for that yet, The builders still use CHEAP land in unstable areas just to satisfy the stockholders and the bottom line. Those two conciderations alone guarantee disaster. Coal not burning or oil exposed out in the open are NOT the health issue to large groups of population that cesium or plutonium pose and I see in this blog that YOU nicely sidestep and gloss over those facts repeatedly. Plutonium does NOT occur naturally on this planet whereas oil and coal do.

Posted by Mike Meyer at August 30, 2011 01:11 PM

I feel it’s time to drag this debate out of the realm of nightmare conjecture and into the light of empirical evidence.

Funny, I think that's where we started. Way back in fucking April.

Posted by Amanda Rex at August 30, 2011 05:54 PM

It’s not a perfect track record, but it’s improving; the rate of spews per reactor-year is dropping and Fukushima casualties will be nowhere near as bad as Chernobyl’s. So, when you said once every generation then you must have be referring to the lifespan of some simpler organism? Regardless of how dandy the improvements are, nuclear plant officials have lied in the past about amounts of radiation released into the atmosphere, and that gives me reason to trust they might not have my best interests at heart. With each improvement to reactor designs, resulting accidents ensure that people still develop cancer, human and animal DNA is still irreparably mutated, the environment is still devastated, and dead zones still result.

I will assume that you have a family, or that you are least familiar with the concept of family, and encourage you to consider how positively you might view our minimally regulated nuclear future in the case that your family were to be living within the exclusion zone of the next disaster. I think the glaring point here, which you seem to minimize with each comment despite your very insightful data, is that: technology fails, human error is rampant, people in positions of power often lie, and failure to honestly address the worst case scenario dooms us to a future of repeating our past mistakes.

It is honorable to speak up when you feel someone has misrepresented something. It is distasteful to continually shit on someone's intentions after they've made corrections to the information you originally disputed. I say this because I am grateful that someone has the courage to even consider disagreeing with the majority of happy-go-lucky nuclear supporters, and because I am familiar with some who have suffered greatly as a consequence of selfish, short-sighted decisions on the part of nuclear officials. I'm sure you're familiar enough with research science to know that profitable solutions to costly problems receive funding almost exclusively. An absence of data does not necessarily reflect an absence of fact.

Posted by Amanda Rex at August 31, 2011 01:10 PM

Correction:

On rereading press accounts, it's clear that North Anna did lose power in one of four diesel generators--as Aaron Datesman reported--not one of five, as I suggested above. Still not clear how many generators it would have had to lose before emergency cooling systems would have been knocked out. Apologies for the error.

Posted by Will Boisvert at August 31, 2011 01:11 PM

@Will, you sure do have a lot of free time. I used mine to bake cookies Monday night. They are delicious!

Regarding the number of backup generators, yes, it appears to be four - two per reactor. Therefore, if one fails, there is only one backup immediately configured to replace it.

Regarding boron injection, yes, I suppose that's true. But at every step the possibility of catastrophic failure edges closer. I might also mention that a) how do you inject fluid into a highly pressurized vessel without an operating pump, and b) what if there is a gas bubble or cavitation disturbance in one of the flow channels?

But then you and I would just go on to design a nuclear reactor, which does not interest me.

Regarding the 8:38 comment, the training I received from the DEPARTMENT OF ENERGY does not allow me to conclude that one among a set of technological choices is preferable because it kills fewer people.

Among the other points you raise, I disagree strongly about at least this subset:

a. It's ludicrous to claim that nuclear plant emissions are confined to once per generation. Every plant routinely dumps large amounts of radioactive material into the environment - some permitted, some not. You can even find tables and charts detailing permitted emissions among NRC and IAEA documents.

b. I think about 700,000 Chernobyl liquidators might disagree with your assessment that the Soviet response to Chernobyl was "sluggish".

c. The claim that Chernobyl and Fukushima are about as bad as it can get for those specific reactor configurations is basically correct - except that the radioactivity in the spent fuel pools at Fukushima also was lost to the environment. This should not be overlooked.

I lack the time to engage on other specific points, since I'm busy helping DOE to deploy wind energy. These are my basic views:

a. Modern commercial nuclear plants are a triumph of engineering. But this doesn't mean they are safe - and it doesn't mean that the engineers' opinions should be trusted. I base this opinion upon knowledge I gained in the fields of materials science and fluid dynamics while working at Bettis Atomic Power Lab.

(It was also the stated position of the senior scientist running the JIMO project on which I worked - he said straight up, in front of an auditorium of 750+, that reactor accidents will happen and can't be avoided.)

There are a tremendous number of buried assumptions in the design of a nuclear reactor. Learn about neutron embrittlement or how the Navier-Stokes equations for fluid flow are only approximations if you don't know about these things already.

b. The means we employ to determine the probable harm from radiation exposure utilize models which are based on only the flimsiest science, when they are not basically made up from whole cloth. If you calculate harm using the accepted dose models, you are already beginning in the middle of the story, and building upon a foundation which has no real scientific basis.

Even EPA says very straightforwardly that the estimates of harm from radiation are based upon extrapolations from high dose effects. Science is based upon observation, not extrapolation.

In general, it's this topic of the health effects where my greatest concerns lie. If I were able to accept the health physics models, as you do, then I would reach the same conclusion you do. However, I paid at least a little attention in statistical mechanics class, and I know that the models are ludicrously wrong. Therefore I reach the opposite conclusion from yours.

Appreciate the comments. It would help the rest of us judge their value if you would share your bio with us. I requested it from you before, but you did not supply it. I have been up front with the readers here about who I am and what my background is.

Posted by Aaron Datesman at August 31, 2011 02:22 PM

I find this thread somewhat surrealistic. Aaron is trying to convince us that nuclear power can be very dangerous. But I think we all know that. Will says that, yes, it can be bad, but never as bad as coal. I bring up the fact that worst-scenarios are what it's all about, and Will tells me (with Aaron agreeing) that Chernobyl/Fuku are the worst we can imagine. Since both agree, I'll go with that. Now, even in the best case, renewables won't replace current energy sources in many many decades, so the question is then what to do in the interim -- even assuming we go green all the way. Do we nix nukes (and not just for the cool alliteration) and rely on fossil fuels in the meantime, or do we keep the status quo or even build more nuke plants just to tide us over.

That question doesn't strike me as entirely obvious. Coal is very bad for global warming and it kills thousands every year. Nukes can be potentially catastrophic. I think a real debate would seek to resolve this question honestly by examining the tradeoffs. The position that there is no tradeoff, that it's just a question of building a gazillion wind farms by the end of the week strikes me as a cop out. Or perhaps it's possible to decommission all nuclear plants while not increasing greenhouse gas emissions. If that's possible, then I'd love to hear how to do it. And if it's not, then people like Aaron need to explain why greenhouse gas is better than nukes. Because such a choice will arise out of necessity.

Posted by bobs at August 31, 2011 08:12 PM

bobs

I think Aaron just agreed with Will that Fukushima/Chernobyl were as bad as it could get "for those reactor configurations," which is a lot different from saying those catastrophes are the worst he can imagine. Scientists almost ways have those little hanging clauses that can radically change the meaning of their statements. I doubt many of the rest of us, like me, know anything about reactor configurations and so just have to ignore qualifications like that when we read.

It's not possible for more than a tiny handful of people to understand all the technological tradeoffs, if anyone can do it, and those who understand the most are often the most biased and least admirable in other ways. So hell, we're going to have to settle for improvement and hope we survive. If we could eliminate the corrupting influence of money, that strikes me as likely to help a lot, though even that is pretty utopian given that we admire money so much in our society.

Posted by N E at August 31, 2011 09:06 PM

On that note: Why have alarms not been sounded about radiation exposure in the US? Nuclear operator Exelon Corporation has been among Barack Obama's biggest campaign donors, and is one of the largest employers in Illinois where Obama was senator. Exelon has donated more than $269,000 to his political campaigns, thus far. Obama also appointed Exelon CEO John Rowe to his Blue Ribbon Commission on America's Nuclear Future.

http://tinyurl.com/3t9t77a

Posted by Amanda Rex at August 31, 2011 09:18 PM

NE: Maybe Aaron, between two batches of cookies, will clarify that. Re. experts, that's a real problem. I don't trust nonexperts on technical matters (why should I?) and I don't trust experts either, because most of them are on the take. Politicians will tell me whatever the lobbies tell them to tell me (as Amanda documents).

And so I still don't know if greenhouse gas emissions are worse for the future than a few nukes. Whoever knows the answer, please let us know.

Posted by bobs at August 31, 2011 09:37 PM

Here's a start: http://www.epa.gov/cleanenergy/energy-and-you/affect/nuclear.html

Posted by Amanda Rex at September 1, 2011 01:02 AM

Thanks for the link, Amanda. But it does not even begin to answer my question. I know nuclear power is dangerous, generates waste that's hard to get rid of, etc. And I know that greenhouse gas emissions are terrible, too. But which is worse? That's the question one must answer. We're told AGW is the no.1 scare. Aaron seems to say nukes are. My point is that both can't be right and a choice must be made. Or if a choice need not be made, and indeed we can nix nukes while decreasing greenhouse emissions in the next 50 years then we need to be told how. Your link doesn't do that.

Posted by bobs at September 1, 2011 01:29 AM

Or perhaps it's possible to decommission all nuclear plants while not increasing greenhouse gas emissions. If that's possible, then I'd love to hear how to do it. And if it's not, then people like Aaron need to explain why greenhouse gas is better than nukes.

I was attempting to politely infer that the tools for developing an informed opinion are readily available. When I don't understand the politics/impact of something, I start with a clear definition so I can form an overview of an issue, and then seek to determine point for point where each side of the debate stands. I apologize if I came off as condescending. Someone like Aaron could share with you what they know, but you've mentioned that you don't necessarily trust the opinions of experts and non-experts alike-which I completely relate to, and is why I think the internet is an invaluable tool for seeking information on both sides of the debate. Especially when it comes to profit-driven matters, because when people live in a bubble of comfort they often fail to realize how much of an asshole they appear to be from the perspective of someone outside that bubble.

Posted by Amanda Rex at September 1, 2011 09:17 AM