2012 April 09 Monday
The Avoidability Of Fukushima Daiichi Nuclear Failures

Just getting around to posting about a Technology Review article from March about how the Japanese reactor failures could have been avoided with fairly modest measures.

"Fukushima Daiichi ... was not just due to an inadequately sized seawall—that is the wrong way to look at it," says Edward Blandford, a professor of nuclear security at the University of New Mexico and a postdoctoral fellow at Stanford University's Center for International Security and Cooperation. "The events at Fukushima Daiichi were due to a series of failures, including failures in plant defensive actions, mitigation efforts, and emergency response. If backup equipment had been stored in waterproof vaults or higher elevations, the accident would have most likely been avoided."

How hard is to built waterproof vaults that a wave could wash over? Not very. How hard to build up an elevated platform to keep some generators above a wave? Again, not very. As the article relates, the nuclear industry has resisted some forms of safety upgrades because to admit the upgrades were necessary would call into question how safe nukes were before upgrades.

Look around and ask yourself: What rare events are we not preparing enough for? I'm thinking an electromagnetic pulse from a repeat of the 1859 solar Carrington event. We should be better prepared for the potential frying of our electric power grid.

I also wonder about the costs of stockpiling enough food to handle a large volcanic eruption, perhaps one big enough to bring on another Little Ice Age or worse. The 20th century was pretty tame as far as natural phenomena are concerned. The 19th century was much worse.

By Randall Parker    2012 April 09 09:57 PM   Entry Permalink | Comments (11)
2011 April 17 Sunday
Oil Industry Macondo Response Lesson For Nukes

In response to the BP Macondo Deepwater Horizon oil well blow-out and resulting 87 days of oil gushing into the Gulf of Mexico 2 consortiums of oil companies and formed 2 companies to develop devices that can be brought on after a blow-out to cap a run-away well within a couple of weeks of a blow-out. 10 major oil companies (e.g. Exxon and COP) that account for 70% of the oil pumped in the Gulf put up $1 billion to fund the Marine Well Containment Company (MWCC) to create what are called capping stacks. Some of these stacks are ready for deployment.

A separate group of oil companies founded Helix Well Containment Group which has developed their own capping stacks. The first Helix design works to 5600 feet, which is deeper than the 5000 feet water depth of the Macondo well. The next Helix capping stack due this summer will be good to 10000 feet. Many of proposed new drilling sites are at 10,000 feet and deeper. MWCC is on a similar path to develop greater well capping capability.

The MWCC interim well containment system is ready for deployment with the capacity to contain up to 60,000 barrels per day of fluid in up to 8,000 feet of water. Work is also under way on the expanded system for delivery in 2012 to handle up to 100,000 barrels per day of fluid in up to 10,000 feet of water.

To my mind this aspect of the oil industry's response to this disaster (develop much better tools for handling worst case scenarios) illustrates what the nuclear power industry needs to do: develop a set of portable capabilities that can be rapidly deployed to any nuclear reactor site to rapidly recover from major systems failures. These capabilities are not a substitute for improvements that reduce the odds of such failures. But industries such as nuclear power and oil extraction should admit their best laid plans (which are often not laid out all that well in the first place) can fail and fail very badly.

I would like to see the nuclear power industry explain how they can develop a number of capabilities including:

Portable reactor cooling systems for cooling system failures.
  • Reactor wall patching systems for cases when a reactor breach occurs.
  • Portable shielded reactor control centers for when normal control centers become damaged or their radiation levels go too high.
  • Everything that went wrong at Fukushima should be dealt with by consortia of nuclear power industry companies by developing technologies that can substitute rapidly for damaged systems and do more rapid repair of reactor sites.

    Update: The fact that new nuclear reactors can be designed to be less susceptible to the failure mode at Fukushima is a good thing. But it is besides the point for the already existing hundreds of nuclear reactors around the world. Unless those are going to be shut down soon (and with the possible exception of Germany that appears very unlikely) we need better ways to handle failures at reactors already in place. Of course existing reactors can be upgraded (e.g. with cooling pumps that won't get knocked out by a tsunami wave). But if one fails we need better tools to deal with the consequences.

    By Randall Parker    2011 April 17 10:16 AM   Entry Permalink | Comments (19)
    2011 March 14 Monday
    New Designs Would Avoid Japan Reactor Failures?

    In Technology Review Kevin Bullis highlights nuclear reactor design improvements that would cut the risk of cooling system failures.

    The latest nuclear reactor designs could help avoid the overheating and explosions that have occurred at the Fukushima Daiichi nuclear plant in Japan following the powerful earthquake and tsunami that struck on Friday. Newer reactor designs propose the use of passive cooling systems that would not fail after a power outage, as happened in Japan, as well as other novel approaches to managing reactor heat.

    Passive systems are key in my view. Human operators make mistakes and active systems can get damaged when you most need them. According to a Bloomberg report the Japanese reactors had back-up generators designed to withstand 6.3 meter waves but the plant was hit by 7 meter waves. So for want of an additional 0.7 meters of protection the reactors have undergone partial meltdowns. Ouch.

    On the bright side, big nuclear reactor failures are like big passenger aircraft accidents: They get heavily picked over and analyzed by large numbers of skilled engineers. We learn from failure. Unfortunately, the failure of nuclear plants in Japan is starting to look more like Chernobyl and less like Three Mile Island in terms of the scale of the disaster.

    An LA Times story points to mistakes made by Japanese engineers that have exacerbated the problems.

    Engineers had begun using fire hoses to pump seawater into the reactor — the third reactor at the Fukushima No. 1 complex to receive the last-ditch treatment — after the plant's emergency cooling system failed. Company officials said workers were not paying sufficient attention to the process, however, and let the pump run out of fuel, allowing the fuel rods to become partially exposed to the air.

    Once the pump was restarted and water flow was restored, another worker inadvertently closed a valve that was designed to vent steam from the containment vessel. As pressure built up inside the vessel, the pumps could no longer force water into it and the fuel rods were once more exposed.

    One can guess these reported mistakes are not the only mistakes that have been made so far. Under intense pressure in a crisis situation people will make mistakes. Emergency handling can not depend near perfect decision making.

    The Diablo Canyon nuclear power plant at Avila Beach California is designed to handle only a 7.5 earthquake. San Onofre is designed for a 7.0 quake and has a 25 foot high wall to protect from tsunami. A fault 5 miles offshore could let loose some day. The absence of a subduction zone off of SoCal is supposed to put us at much lower risk of a tsunami. Also, SoCal Edison claims San Onofre has more safety layers than the older reactors in Japan. Plus, it has an emergency cooling source that is gravity driven.

    But what about the soundness of the probabilities of geological risks and systems reliability that are fed into models for choosing nuclear reactor sites and designs? As Joel Achenbach wrote in a WaPo piece a few days ago, Japanese scientists were expecting the Big One to occur south of Tokyo, not north of it. Japan's preparations were oriented in the wrong direction. The coast off of Sendai hadn't had a huge quake for at least 1,000 years. What other geological surprises lay in store?

    If an earthquake followed by a large tsunami is "beyond what anyone could expect" then how can the nuclear power industry claim it can choose sites and designs that will avoid events like the ones happening at the Fukushima nuclear power site?

    Richard Meserve, a physicist and former NRC chairman from 1999 to 2003, said the Japanese reactors experienced a "one-two punch of events beyond what anyone could expect or what was conceived."

    What is so unlikely about a big earthquake followed by a big tsunami along a Pacific Rim subduction zone?

    Update: Diablo Canyon is 85 feet above the ocean. Plus, it has a gravity fed back-up water reserve. So it looks like it is at much lower risk of a tsunami.

    What the Japanese ought to do for their remaining undamaged nukes: Build back-up generator buildings that can survive a tsunami flooding over them. Be able to ride out the worst that nature can throw at you can keep on going. Earthen berms and lots of concrete would do the trick. Another idea: Build back-up generators that are well inland and run underground cables to the nukes near the coast.

    Any other spectators want to tell the nuclear plant engineers how to make their nukes safe from tsunamis?

    By Randall Parker    2011 March 14 10:59 PM   Entry Permalink | Comments (59)
    2010 December 26 Sunday
    Macondo Horizon Drilling Rig Events

    If you want to understand how the events in Transocean's Deepwater Horizon led to the disaster and lives lost read this long New York Times piece. From blowout to explosion was 9 minutes. This report tells the story from the perspectives of many participants. Worth reading.

    The result, the interviews and records show, was paralysis. For nine long minutes, as the drilling crew battled the blowout and gas alarms eventually sounded on the bridge, no warning was given to the rest of the crew. For many, the first hint of crisis came in the form of a blast wave.

    The paralysis had two main sources, the examination by The Times shows. The first was a failure to train for the worst. The Horizon was like a Gulf Coast town that regularly rehearsed for Category 1 hurricanes but never contemplated the hundred-year storm. The crew members, though expert in responding to the usual range of well problems, were unprepared for a major blowout followed by explosions, fires and a total loss of power.

    They were also frozen by the sheer complexity of the Horizon’s defenses, and by the policies that explained when they were to be deployed. One emergency system alone was controlled by 30 buttons.

    I am reminded of the comments made after the Three Mile Island nuclear plant accident about overly complex control systems. The nuclear power industry changed a great deal due to TMI. I hope the same happens with offshore drillers as a result of the Horizon accident and loss of life. So many things went wrong and so many safety systems failed that the failure speaks to something much deeper than mistakes made by a single crew or oil company. They need to learn from this accident the way the airline industry has learned from the succession of aircraft accidents over a period of decades.

    By Randall Parker    2010 December 26 02:56 PM   Entry Permalink | Comments (1)
    Site Traffic Info