It happened at Chernobyl in Russia, it happened at Fukushima in Japan, and it has already nearly happened many times at various nuclear power plants in America. We've been lucky, and luck runs out.
The Nuclear Regulatory Commission's rules for making rules are the problem, because of the layer upon layer of roadblocks to logical conclusions. The inevitable result will be a meltdown in America, because the rules are getting softer, not tougher, just as the industry is in need of more and more loopholes and shortcuts to make money and stay mildly competitive. In some states, including California, they are helped by the states' Public Utilities Commissions, which insures the utility a profit, in many cases by blocking cheaper (for the ratepayer) renewable energy options.
The NRC has many rules, and many of them change all the time. Some get weaker, if things don't go wrong, while others get stronger, if things do. (The American nuclear industry is still bucking even the relatively minor changes (which would cost about two dozen plants in America a few tens of millions of dollars) following the Fukushima accident three years ago.)
Among these rules, however, are a few very special rules designed to prevent duplication of effort, wasted expense, wasted time, and uneven regulatory enforcement.
These rules make it very difficult to revisit an issue which has been previously decided in the nuclear industry's favor. If a new plan for a replacement part is submitted to the regulators and rejected for some reason, the industry can always make a change and resubmit the application. But once the plan has been accepted, there is little that can be done to withdraw or alter that acceptance. Furthermore, the NRC assumes that fabrication of parts is always done correctly, and that every operation they fail to inspect was properly done. None of these assumptions are based on reality.
Over time, a bias is introduced in favor of anything that passed acceptance, rightly or wrongly. Over time, the effect of that bias grows, with one inevitable result: A meltdown.
For example, computer-controlled equipment abounds at nuclear power plants. It's more efficient. Efficiency means a lot when you're trying to make money by boiling water, to turn into steam, to turn into a spinning object, to turn into a magnetic field, to turn into electrons flowing back and forth in wires, to convert through many different amperages and phases along miles of wires every inch of which introduces additional losses, in order to deliver electricity to homes and businesses which are flooded with it from wind and sun daily.
The overall inefficiency is staggering, so of course they've computerized everything.
Then along came computer viruses. So the industry bought anti-virus software.
Then along came Stuxnet, which attacked the motor controller boards at a nuclear facility in Iran. Many features of it were very easy to copy, and many varieties quickly followed.
Along with these software threats came counterfeit parts inside the "mil-spec" hardware components that are used to make the motherboards that control the machines.
The military has a huge problem with counterfeit parts. Every industry does. The nuclear industry is both a major user of enormous quantities of computer equipment, and a major target of industrial, political and terrorist espionage attempts.
A perfect storm.
Late to the party, the NRC is (finally) holding special hearings (mostly or entirely closed-door, as far as I know) with other agencies about hackers, viruses, encryption and other software security issues. (I'm not sure the hardware dangers are going to be considered at this time.)
Nuclear power plants are notoriously complicated machines. Try as they might, no one has been able to simplify them. In fact, it's only gotten worse with added computerization in the control room and in the machinery throughout the plant.
The nuclear reaction itself is a complex balance of the density of neutron moderators against the available quantity of various isotopes for fissioning and for neutron absorption. To use the energy that is released by the reaction, water, which also acts as a neutron moderator, is heated under extremely high pressure (about 1200 PSI for Boiling Water Reactors (BWRs), and about 2200 PSI for Pressurized Water Reactors (PWRs). Change the density of the water just a little bit (by changing its temperature), and the water's ability to moderate neutrons (slow them down) can change dramatically. Stop circulating the water, and the metal overheats, the zirconium catches fire, the fission products are released, and finally, the ceramic pellets of uranium and plutonium oxides melt down through the steel reactor pressure vessel in a radioactive blob known as corium.
There are three blobs of corium in Japan and one in Russia. No American reactor accident has resulted in an unapproachable (even with proper protection) blob of corium, although SL1, Santa Susana, Fermi 1** and Three Mile Island all suffered partial or complete meltdowns. (Two were relatively small "research" reactors.)
Even without a meltdown, no one can go near the reactor while it is operating for months at a time. Fixing problems usually requires shutting the reactor down, which is not risk-free. Utilities are very limited in how often they are allowed to shut a reactor down. Half a dozen times in a year would be very alarming, for instance, to the regulators. That is, until relicensing time, when all that will be forgotten as it is expected that licenses will soon routinely be extended to 80 years.
Accidents can be so severe that at least one of the reactor's backup systems "MUST" function. At least, that's the claim -- that at least one backup system will work and prevent a catastrophic release of radiation to the public.
To prevent that, reactors are designed to withstand the worst earthquake postulated for the region, the worst tornado, the worst tsunami, the worst flood, the worst fire, the worst power outage, the worst attack by a small group of lightly armed non-suicidal terrorists ... but of course, once it's been decided that something is the worst threat in a particular realm (worst earthquake expected, worst tsunami, etc.) the NRC is very reluctant to revisit that issue, even as new science comes along. In fact, especially as new science comes along that could result in getting a reactor shut down permanently. And even if a new, more strict ruling is made, it can take decades to get that rule implemented: Reactor companies ask for extensions and exemptions all the time -- and they are almost invariably granted.
All the while, each operating nuclear reactor is creating radioactive waste at the rate of about 10 pounds per hour per reactor (10 tons per day for the U.S. as a whole) which must be guarded essentially forever.
Legally, for at least a million years.
The NRC affirmed the million-year figure recently, by concluding that the proposed (and cancelled and reborn) Yucca Mountain repository would protect humanity for that long. The timing of the release of the NRC's conclusions appears to be a political move prior to the November midterm elections. The NRC did not provide backup documentation for their decision -- that's still to be published. The NRC has a habit of withdrawing documents, then rereleasing watered-down or altered versions prior to releasing any backup documentation, if it ever comes out at all. And the backup documentation may not even back up their conclusions -- this author has seen that happen many times, as well.
The State of Nevada is fighting the legal battles one might expect them to be concerned about: They don't want Yucca Mountain to happen at all and have numerous good reasons why it's inadequate: Water intrusion, earthquake issues, volcanic issues, transportation issues (they don't want that waste anywhere near Las Vegas, but that's exactly how some of the routes go), human fallibility issues (will it be constructed according to specifications and who's gonna check? And who's gonna check the checkers?).
However, the State of Nevada doesn't want to appear anti-nuclear or -- egads -- unAmerican, so in their official submissions opposing Yucca Mountain, to the NRC or to a judge, they always state that on-site storage of spent fuel at the reactor sites is safe and will remain so until an alternative site is selected.
However, onsite storage is NOT safe, and every other potential permanent repository location has already been eliminated for one reason or another, time and again. Nobody in their right mind wants the waste anywhere near them. Especially not a wised-up American.
Having no other place on earth for the highly radioactive used reactor cores puts residents of southern California at grave risk of a catastrophic spent fuel accident, with no solution in sight. Even though San Onofre is closed, the spent fuel is still a problem, and so is Diablo Canyon: Los Angeles is far closer to Diablo Canyon than Tokyo is to Fukushima. And Tokyo might have had to be abandoned if Fukushima's "corium" blobs had exploded violently (or might still have to be, if the corium explodes some time in the future...). So Los Angeles is certainly not yet safe from San Onofre, let alone, from Diablo Canyon.
Much of the risk these reactors have imposed on Californians has been the inevitable result of the NRC being a "captured" regulatory agency. And for what? For energy we don't need: Losing SanO did not cause blackouts, and there is already more than enough electricity available in California to make up for Diablo Canyon's output to the grid, and more renewable energy is coming online every day. And with a new Public Utilities Commissioner, perhaps we will see even more renewable energy (hard to believe we could do worse than Michael Peevey has been).
San Onofre's steam generator replacement project failure has become legendary within engineering circles, and somebody has to take the blame. SCE deserves a lot of the blame, but the NRC chose to absolve them of it all and instead, take the blame for the guilty-as-sin utility's engineers and executives, who the NRC protected from criminal indictments at every step. Probably to protect their own skin.
For example, after citizens demanded a "thorough" investigation into what failed in San Onofre Unit 3 on January 31, 2012, and what -- if anything -- was different about Unit 2's virtually identical replacement steam generators, NRC formed an Augmented Inspection Team (AIT) and experts were consulted. But the experts couldn't agree on the cause, and the committee concluded it might be one thing or another, but then again it might be something else, and that was the end of the investigation. The net result? It could happen again, at another reactor site. The "root cause" was never found.
The question came up (like a freight train rumbling through a small village) of whether or not Southern California Edison (SCE) should have applied for a license amendment before replacing the steam generators, -- a license amendment which SCE specifically tried to avoid. After the AIT investigation into what happened that went nowhere, NRC concluded that SCE had satisfied all the requirements for submitting data about the replacement steam generators to the NRC. However, the NRC could not produce the documentation they supposedly used to support their conclusion.
After the leak that shut down the plant, the 18-month delay (from January 31, 2012 to June 7, 2013) before deciding to permanently close San Onofre cost ratepayers in California hundreds of millions of dollars. That money could have gone directly into solar and other renewable projects.
Prior to SCE's abandoning the reactor the NRC, to their credit, delivered several dozen technical questions to San Onofre's owners for their engineers to answer. But realistically, the NRC could have and should have told Southern California Edison there was no way they were EVER going to approve any restart of EITHER reactor -- even if the steam generators are replaced again -- because SCE was no better at keeping records of what went wrong than the NRC was, and the documentation for who screwed up the calculations appears to be missing -- so who would know what else might be bogus in that hunk of junk by the sea?
NRC should never have allowed the steam generator replacement project to move forward anywhere, but especially at San Onofre. In making their decision, the NRC apparently never considered what else might have been going wrong at the plant (such as: not going on fire watches, worker intimidation, not test-starting or inspecting backup generators for years at a time, not properly setting the welding speeds on automatic welding equipment, etc. etc. etc. ).
NRC should have shut San Onofre down -- and all the others -- because of the waste problem. It's not just unsolved. It's unsolvable.
This Monday (October 27th, 2014) the NRC will hold a hearing (in Carlsbad, where I live) regarding how to proceed with decommissioning San Onofre, after SCE finally decided it couldn't squeeze the public to pay for a thousand useless high-paying jobs anymore, and so they cut the workforce to a "skeleton" size of about 350 people, and submitted a decommissioning plan (called a PSDAR) which calls for completing the project as quickly as possible -- except for the waste, which will just sit there.
The NRC meeting is to hear from the public regarding SCE's PSDAR, which is filled with utterly fantastic predictions: That the nuclear waste problem will be solved by 2024 when a national repository will open; that San Onofre's waste will be removed from the site by 2049; that giving workers a few hundred REM of radiation is okay; and that the public has been properly engaged and knows what's going on at the plant.
When the decommissioning process starts, hundreds more people will be hired. SCE is keen to see that happen. SCE already has (our) money to pay them, and the money has to just sit there, hopefully earning enough interest so that whenever decommissioning does start, there will still be enough money to pay for it. But if not, they'll just get it from the ratepayers.
Local unions want decommissioning to start -- they see jobs, and if it's dangerous -- well, that pays even better. They can't see radiation. They're not afraid of it because construction work -- and deconstruction work -- always carries a risk. That's what people do to build skyscrapers and so forth -- they take a risk that the cables and crossbeams will hold: That the design is correct. That the materials are good.
NRC has calculated the overall radiation doses the workers will receive. If deconstruction is delayed for a few generations of workers instead of started immediately as SCE is proposing, the cumulative doses those workers who finally do the deconstruction will get will be about 1/20th of the radiation dose that workers will get if the deconstruction work is started as soon as possible (10s of REM cumulatively, versus several hundred REM cumulatively).
Figures were not available for doses to the public, but those would be reduced as well -- although those are already promised to be extremely low.
The Office of Inspector General's (OIG's) report on the NRC's handling of the San Onofre steam generator replacement project was very critical of the NRC -- but still didn't go far enough. As if in response, almost immediately after the OIG's report was released, this ne'er-do-well, corrupted, captured, hook-line-and-sinker pro-nuclear regulatory agency published a ruling that is supposed to stand for a million years or more!
And the NRC's chairperson resigned -- poof, off to a cushy university.
When she was appointed, Allison Macfarlane had been lauded by Obama for her work on the Blue Ribbon Commission -- which accomplished nothing except to say that we've got to stop allowing citizens to stop a nuclear waste dump, just because they live near a small group of tribal Indians or other "sovereign" land owners who are willing to take money for taking nuclear waste.
This BRC "solution" is called such Orwellian terms as "community choice" for "interim storage solutions." (In the case of Yucca Mountain, that site is within the Nevada Test Site, which is on land already taken from the Shoshone people, and they don't want it further desecrated.)
And speaking of Orwellian terms, 70+ years and as many billions of dollars later, the unsolved nuclear waste problem is now called the "waste confidence" problem, as if half-inch thick stainless steel canisters with thousands of pounds of nuclear waste in each one is somehow safe, and as if such flimsy canisters are a defense-in-depth (as was promised by the nuclear industry and the NRC) method to store nuclear waste for hundreds of years, come what may: Fires, floods, wars... larger-than-design-basis earthquakes... the crumbling sands of time... tsunamis...
The dry casks are only tested, built for, assumed to be able to withstand 40 feet of submersion, while the waves that washed over Fukushima were nearly double that, and underwater landslides among the submerged canyons off the coast of California can cause mountainous tsunamis far taller than that. A new study was promised for decades but of course, never materialized. And even Allison Macfarlane has admitted, you can't predict when these things will happen.
The NRC says the canisters will last at least 100 years right here on our coast, but anyone who lives here knows metals don't last very long along the coast. They become embrittled, they rust, they flake, they crack... and the NRC's own studies* confirm that we'll be lucky to get 30 years from the dry casks (the ones in Diablo Canyon are showing signs of stress corrosion cracking after only about two years). If we can't figure out a good way to thoroughly inspect these canisters periodically, we won't know if they've even started to crack until it's a through-wall crack and the canisters have already started to leak.
And yet, the NRC is America's last line of defense, which is why, despite it all, I encourage all citizens to come to their meetings and complain.
Ace Hoffman
Carlsbad, CA
* Actually, the NRC didn't do any cask degradation studies themselves, they commissioned Sandia National Labs to do some, because the NRC doesn't have the expertise or resources to do the studies, and the industry doesn't have the interest in doing them.
** Fermi 1 was added after the following letter was received from Kay C.:
Hi Ace,
Just a reminder that one of Michigan's reactors, Fermi 1, also had a partial meltdown in 1966. John Fuller wrote a book about it: "We Almost Lost Detroit." From my understanding, it was a lot like Santa Susanna. -----------------------------------------------------------------------------------------------
-----------------------------------------
-----------------------------------------
Ace Hoffman
Author, The Code Killers:
An Expose of the Nuclear Industry
Free download: acehoffman.org
Blog: acehoffman.blogspot.com
YouTube: youtube.com/user/AceHoffman
Carlsbad, CA
Email: ace [at] acehoffman.org
----------------------------------------
Note: This communication may have been intercepted in secret, without permission, and in violation of our right to privacy by the National Security Agency or some other agency or private contractor.
-----------------------------------------