Japan, and nuclear disasters


The images from Japan are shocking and depressing, and what seemed at first an example of the difference between a 1st and 3rd world earthquake has produced a 5 figure death toll. But the nerd and engineer in me has to wonder about some of the things I've seen.

Youtube disaster?

While there has been some remarkable footage, some of it in HD, I was surprised at how underdocumented things were, considering Japan's reputation as the most camera-carrying nation of the world -- and the place where all the best cameras come from. I had expected this would be the "Youtube disaster" where sites like YouTube would fill with direct observer HD videos from every town, but the most of what was uploaded there in the first few days was stuff copied from the TV (in fact, due to DRM, often camcorders pointed at TVs.) Of course, the TV networks were getting videos from private individuals, but we saw the same dramatic videos over and over again, particularly the one from destroyed village of Miyako where the water swept boats and cars over the seawall and under a bridge.

Yes, there was a lot of individual reporting, but I expected a ton, an unprecedented amount, and I expected to see it online first, not on the news first.

Cell phone shutdown

Japan is also one of the world's most connected countries, with phones for all. Not a lot has emerged about the lost of cell phone service. Some reports suggest some areas of the network were switched into texting-only mode for civilians to leave capacity for emergency workers. Other reports say that landlines were often up when cell lines were down. The world still awaits Klein Gilhousen's plan to allow cell phones to text peer to peer which I reported on in 2005.

Nuclear plant worst case

The public is now fully aware of many of the issues with nuclear reactors which require active stabilization using external electricity. A lot had to happen to get to the pump shutdown:

  • The reactors themselves were auto-shutdown after the earthquake. Wise, though in theory the subsequent problems would not have happened if one reactor had remained up and powering the plant.
  • The quake or tsunami shut off the external power. A week later it's still not up. It seems that restoring it should have been a top priority for TEPCO. Was the line so destroyed or did they not prioritize this?
  • The backup generators were damanged by the tsunami, all 19 of them. I have to admit, most people would think having 19 backup generators is a very nice amount of backup. But this teaches that if you have lots of backups, you have to think about what might affect all of them. 1 backup generator or 100, they all would have failed if unable to withstand the wave
  • The batteries supposedly lasted for 8 hours. This does not seem unreasonable. But they either did not realize that they had to get something else going in the 8 hours, or expected other power. Their procedure manuals should have had a "what to do if you have only 8 hours of battery left" contingency, but I can believe they didn't because it seemed so unlikely.
  • That said, I believe the best backup plan has a fallback that involves emergency-level external resources. In particular, I have heard of no talk of sending a ship with a few hundred meters of cable to the docks there, one of which appears to be under 100m from reactor #1 and presumably the internal power grid. Many ships have big generators onboard or can deliver them.
  • Failing that, a plan for helicopter delivery of a generator and fuel in case all other channels are out.
  • Apparently they did bring in a backup generator by truck, but it was incompatible, and they are still without power.
  • It's a hard question to consider whether they should have restarted a reactor while on batteries. There would not be enough time for a full post-quake, post-tsunami inspection of the reactor. On the other hand, they clearly didn't realize just how bad it was to lose all power, and/or probably presumed they would get power before too long.
  • Everybody has now figured out the problem with spent fuel storage without containment in a zone where the chamber might crack and drain. Had nobody worried about that before? Most reactors don't store all their spent fuel this way, but some do, and I have to presume work is underway to address this.


Japanese skill in robotics is world-leading. I've seen examples of some of that going on, but I'm surprised that they haven't moved just about every type of robot that might be useful in the nuclear situation to near the nuclear plant. If they should ever have a situation where they must evacuate the plant again, as they did on Wednesday, it could be useful to have robots there, even if only to act as remote cameras to see what is happening in the reactors or control rooms.

There are also remote manipulator robots, and I am surprised no media organization has managed to get some sort of camera robot in the plant to report. Of course, keeping the robot powered is an issue. Few robots are actually able to hook themselves up to power easily, but a number of the telepresence robots can do that.

Many of the "work in danger zone" robots have been built for military applications, and the Japanese don't have that military need so perhaps they are not so common in Japan. But they do have stair climbers, telepresence and basic manipulators. Even if the robots can do very little it would make the public feel better to know that something is there.

The Chernobyl cleanup was in part done by remote control bulldozers that the Russians made.

Future of Nuclear Power

The reactor failure is causing much public examination of nuclear power. This disaster does show just how bad the older designs are, but makes us question why companies were running them when it's been known for decades that those designs were a poor idea. Obviously investors will not be keen on saying, "Oh, we made mistakes back then, let's write off the billions."

There is also an argument that a technology can't develop without going through a phase where it is less well understood and designs are not as safe as can be. Would we have developed newer, safer designs if nobody had been able to build the older ones?

I have been seeing tons of ads on CNN by the coal, gas and oil industries about how wonderful their technology is. In spite of the fact that there have been quite a large number of deaths from these technologies, and tons of pollution, and now the fear of greenhouse gases.

According to one agency in Europe, I found a quote that the world's nuclear plants had generated 64.6 trillion kwh in the period up to 2009, or 6.4 x 10^16 watt-hours. A watt-hour of coal produces about a gram of CO2. A watt-hour from the coal and gas plants at the US average is less than that, let's call it 0.7 grams/watt-hour more than nuclear (there is some CO2 output from the full lifecycle nuclear industry.) Correcting from original where I had used euro-billion = 10^12 which can't be right.

That's about 4 * 10^16 grams of CO2 not put into the air by the nuclear industry. I'm looking for figures to see what that means, but one that I found says that the whole atmosphere of the planet has 2.7 x 10^18 grams of CO2 in it.

The number I would like to see is what difference those 10^16 grams of CO2 have made to the total PPM of CO2 in the atmosphere, which is to say, how much did those nuclear plants retard global warming according to accepted climate models. Anybody have info?

To solve the world's energy needs, while we eventually would like to develop economical solar plants, biofuels that don't use cropland, geothermal, fusion and other sources, right now it seems that there is no choice but to build lots more nuclear if we want to stop burning so much coal. Other choices are coming but are not assured yet. If this disaster scares the public away from newer reactor designs which go to a safe state without active support or human intervention, I think that would be a mistake.

I hope that Japan is able to recover as quickly as possible, and that more of the missing are found alive. Someday something like this is going to happen here in the Bay Area -- though probably not a 9.0, but possibly an 8 -- and it won't be pretty.


A little known reactor type developed in the 1960s has several features that may now get some attention. Back then it was called MSRE, Molten Salt Reactor Experiment. Todays version LFTR pronounced “lifter”, Liquid Fluoride Thorium Reactor, would not have the problems that solid fuel reactors have. With no water it could not have a loss of coolant accident. If the core overheats a frozen plug of salt in the bottom of the reactor vessel melts sending the core fluid down into a dump tank where natural convection keeps the salt from over heating. With no Zirconium metal no hydrogen gas would be produced when the core overheats. Development was stopped because with the Vietnam war going on cash was getting tight for the government, and the competing Sodium cooled fast fission reactor could produce power, and simultaneously copious quantities of weapons grade Plutonium a product highly prized at the time. If LFTRs were involved in this Japan accident, Three Mile Island, and Chernobyl there would have had no damage done. There are several other major advantages, but I will let you go to a few links to more information on LFTR: http://en.wikipedia.org/wiki/LFTR and http://www.energyfromthorium.com/forum/

Thanks Mike

"The Chernobyl cleanup was in part done by remote control bulldozers that the Russians made.". Doubt so, cause IIRC by various distincts accounts all bots broke pretty fast, and were quickly replaced by human 'liquidators'.

Soviet tech was not great. But this is 2011 and it's Japan, so I am curious as to why they don't seem to be (as far as we can tell from the news) doing a lot putting their robotic expertise to work. Even if just to see what's happening.

Soviet tech wasn't great, indeed, but low-tech approaches are often hardening the product, and iirc the Chernobyl bots failed because they were too fragile.

A Monirobot is at work at Fukushima, http://www.newscientist.com/blogs/onepercent/2011/03/japanese-send-robots-into-fuku.html ( funny quote from Mr. Tadokoro )

"To solve the world’s energy needs ((...)) right now it seems that there is no choice but to build lots more nuclear".

Nuclear only produces grid (electricity) power, which is a just part of the problem.

An all-nuclear electricity gridpower will may one day power all cars and heat all buildings, among others needs, but for this to happen we will need at least at least decades.

Nuclear plants produce now at best 6% of such human-produced energy ( http://en.wikipedia.org/wiki/File:2004_Worldwide_Energy_Sources_graph.png ), therefore replacing other current energy sources leads to approx 20 times more nuke plants and a massive retrofit program for most our energy-consuming stuff.

Check http://en.wikipedia.org/wiki/World_energy_resources_and_consumption

Given fuel availability (we now have approx 80 years, check http://en.wikipedia.org/wiki/Uranium_market#Available_supply ) a currently economically realistic approach leads to ask for huge money in order to build 20 times more plants, for which only 4 years of fuel are readily available. Financial officers considering such program may not like recent price surges ( http://en.wikipedia.org/wiki/File:MonthlyUraniumSpot.png ), which happened without any serious "push" towards nuclear energy.

If some want to invest in such a way, I have an Eiffel Tower to sell.

Some hope that new thingies (taping new uranium sources, other fuels, fusion...) may save the day, but there is no economically and field -proven approach.

There are various other show-stoppers (waste management, adequate sites, public opinion...)

Moreover nuke plants aren't carbon-clean: http://en.wikipedia.org/wiki/File:Sovacool_2008_life-cycle_study.png .

The nuclear debate is indeed complex, all the energy debates are. And all the energy sources under consideration have a host of problems related to cost, pollution, scaling and other environmental damage. While we hunt for a solution that doesn't have this, many people have concluded we must have more nukes, because if we don't, we'll end up burning more coal, plain and simple. This is not a statement that nukes solve all the problems.

There are different reactor designs with different fuels, including thorium. And dreams for reactors like the traveling wave approach. In time, I think nuclear waste will be seen as valuable rather than a problem. I would not store it the way Fukashima did, but I don't think it needs to go underground for 10K years. Very little is carbon clean, even solar, but the point is that we're comparing nuclear and coal/gas.

The nuclear debate is indeed complex, and that's the reason why, with all due respect, IMHO we need hard facts instead of "many people have concluded". Moreover we need to double-check our sources, because "many people", in this debate, may not be objective. I'm not paranoid, and don't write about any conspiracy but about efficient "public relations" :-)

All the energy sources under consideration have a host of problems, indeed, so let's analyze and select a bunch of the less dangerous ones.

To avoid burning more coal and oil we may first try to waste less energy.

Thorium reactors are not field-ready, they need at least 10 more years and probably more. That's an already 45 years-old serious research field, not so easy to tackle. The "traveling-wave" approach. is probably even further. The concept is not easy to implement (known for approx 50 years, many attempts)

Nuclear waste may be one time seen as valuable, however let's agree on the fact that such hopes are to remain... hopes, not assertions leading to any waste-neglecting decision, OK?

I don’t know if waste needs to go underground for X years, but I pretty know that, for now, this stuff is an annoyance and that all superb plans so far devised to seriously cope with it fallen down, one after the other.

Our option are not limited to nuclear and coal/gas. Let's "stop" wasting energy, conserve it, then seriously promote decentralized production (just big enough to obtain a good conversion efficiency, not far from the consumer to reduce loss). Our way to let our overcentralized "jacobinist" governements bias towards planification and huge programs isn't adequate anymore (I'm French, and IMHO Washington isn't far from Paris from this point of view).

I apologize for saying "many people" but I did so because this article was meant to be about the Japanese situation, not the nuclear vs. non-nuclear debate. Some sources to read are David MacKay, whom I have linked to before on this blog, and Saul Griffith. Neither of them imagine we would go all-nuclear, but at the same time their calculations show that unless we develop new technologies (like Griffith's own kite-based wind tech) there needs to be a lot of nuclear. Everybody also hopes there will be savings in conservation and increased efficiency, but there are harsh realities here which show these efforts have only modest success. (Conservation -- reducing to help the planet -- seems to have a hard struggle, while efficiency -- getting the same with less energy has more luck.) I am skeptical of claims that we, as a people, are likely to conserve our way out of this crisis. It does little good to say that we can conserve, that matters is demonstrations of what actually makes that happen.

Just looking at my numbers, for example, I see strong eco-advocates putting up solar panels when they would do far more with the same money applied to conservation or efficiency. There is a psychology of it that seems hard to beat.

I will admit that my intuition that the nuclear waste problem can be solved is still just an intuition. There are several promising approaches but they are not proven. At the same time, you could take all the nuclear waste and grind it up in front of giant fans and it would still cause a tiny fraction of the devastation predicted by some forecasters by global warming due to fossil fuel use.

I think it was a very dangerous mistake not to realize that in the past, reduction in nuclear was very much correlated with increases in coal. Today I have more optimism that better methods will come along, but that's still how it is right now. I would like it to be otherwise, and expect it to become otherwise, but it is not yet that way.

Any followups on this, Brad?

The news seems to have fallen off the newscycle, but actually *gotten* worse.

(I'm a LFTR fan, myself, and my best friend likes CANDU and Pebble-bed designs seem pretty well adapted safety wise too...)

Add new comment