Economics and similar, for the sleep-deprived
A subtle change has been made to the comments links, so they no longer pop up. Does this in any way help with the problem about comments not appearing on permalinked posts, readers?
Update: seemingly not
Update: Oh yeah!
Friday, April 01, 2011
Big things, fine tolerances
OK, probably worth going through this one in a bit of specific detail - I've mentioned it before, but it's a common source of confusion in nuke debates. George Monbiot sets out a version of the thesis:
Last week I argued about these issues with Caroline Lucas. She is one of my heroes, and the best thing to have happened to parliament since time immemorial. But this doesn't mean that she can't be wildly illogical when she chooses. When I raised the issue of the feed-in tariff, she pointed out that the difference between subsidising nuclear power and subsidising solar power is that nuclear is a mature technology and solar is not. In that case, I asked, would she support research into thorium reactors, which could provide a much safer and cheaper means of producing nuclear power? No, she told me, because thorium reactors are not a proven technology. Words fail me.
Me too, to an extent, but actually whether by accident or design, Caroline Lucas is fundamentally right here.
The point is this - more research isn't going to help you with thorium reactors. The lab research program is complete; we know how to make a thorium reactor. There are probably about a dozen thorium reactors currently reacting in the world, all of them in university labs. The problem is now one of industrialisation and commercial scale. Because while there are a dozen or more lab-size reactors, what we would need would be a proper commercial-scale one. There are two problems here:
1) is that we don't have a magic size-increasing ray that would allow you to simply increase the size of the lab models to make them big enough to run a power plant off. Making a big casting is a more expensive process than making a small casting, for a given percentage tolerance, and:
2), even if we had a magic size increasing ray, if we used it to scale up a lab model by 10 times, we would be scaling up all the little cracks and imperfections by a factor of 10 times too. And this would mean that our magically scaled-up reactor would be unusable, because the size of a gap that something nasty can get through is a fixed quantity, not a percentage of the size of the object (dollhouses don't have mice in the walls). A working commercial-scale reactor has to be manufactured to a much finer percentage tolerance than a lab model.
How these problems are usually solved with new reactor technologies is by taking a deep breath and deciding to build a power plant anyway, hoping that in the process of doing so, you will come up with enough clever engineering wheezes and process improvements to mitigate the extraordinary expense of making very big objects to very fine tolerances - and that the lessons learned in doing so will mean that your second nuclear reactor is a lot better-designed than your first and so on. It has kindasorta worked sometimes in the past.
But this makes it clear what the issue is - there's no separate research issue in solving the problem of building a commercial-scale thorium reactor. The remaining problem to be solved is just literally the problem of "building a commercial-scale thorium reactor", and the only research methodology for solving that problem is also called "building a commercial-scale thorium reactor". If you think that thorium's a commercially viable technology, commission one. At present, most people with their own (or their government's) money at stake don't - most of the people who do are perennial Mittyish nukeman optimists of exactly the sort that taxpayers and investors have learned to fear.
"More research", in other words, is basically the "no fly zone" of nuclear power plant development. It's a way of trying to be a little bit pregnant.
 Actually probably more like "commissioning a number of commercial-scale thorium reactors" - there isn't really much point in just building one.
Update: Alex's comment about the Indian thorium project prompts me to make a slight qualification here, because there is an intermediate stage between lab and commercialisation - the "prototype". But a prototype isn't really a research project - they are commissioned as step one in the development of a series of plants. They do give you the option of walking away having spent only massive amounts of money if the technology turns out to be unworkable, and that's important, but my objection to Monbiot still stands because you don't simply decide to build a prototype plant independently of your decision to build a load of real plants - that's pretty much the meaning of the word "prototype".
this item posted by the management 4/01/2011 01:08:00 AM