On February 6, 2013, Dan Stout, who is the Tennessee Valley Authority’s senior manager for its Small Modular Reactor project, gave a colloquium at the University of Tennessee in Knoxville. Much of the talk was just nuclear-boosterism. For instance, he claimed that “nuclear power was tested hard in 2011. It remains safe reliable and affordable.”

There was no mention of the fact that post-Fukushima, about 60 of the world’s light water reactors were closed for one reason or another, mainly in Japan and Germany, taking them from 24/7 power generation to 0/365. He said nothing of the enormous social, economic, and environmental dislocation and loss that has afflicted the Fukushima region and beyond since that time. And there was nothing of the Nuclear Regulatory Commission’s Task Force report of July 2011 that found US regulations seriously lacking in a number of respects (http://pbadupws.nrc.gov/docs/ML1118/ML111861807.pdf). But there was some refreshing candor about Small Modular Reactors (SMRs) mixed with this sales talk. His talk is archived at http://160.36.161.128/UTK/Viewer/?peid=fa73ded60b7b46698e9adc0732101a76

SMRs are supposed to overcome the loss of economies of scale by using assembly mass-manufacturing techniques such as are used for cars and passenger aircraft. The site set up would be standardized and the set up and commissioning on site would be relatively quick (36 months). So “economics of replication” would replace the “economies of scale” (one of the principal reasons that reactors got larger as time went on).

But there is a chicken and egg problem here, to quote a cliché. You’ve got to have a lot of orders before you can set up your assembly line and produce cheap reactors, but you have to have demonstrated your reactors are certified before you get the nuclear masses lining up to order them, given the risks involved. There are no such orders yet; no assembly line is in sight.

So for now, SMRs would be “cobbled together” in existing facilities. “Does the federal government want to help?” he asked rhetorically. “I don’t know,” he went on. “We’re going to find out. I am not going to make your electric bills all triple because of this project. That’s just …TVA won’t do that.” (Italics added.)

So for the foreseeable future without massive subsidies, nuclear power using SMRs will be the same as nuclear power with the present Atomic Leviathans – expensive and in need of government subsidies. But you have to hand it to Mr. Stout for one thing. Unlike Georgia Power and South Carolina Electric and Gas, two utilities building the new large behemoth variety (the AP1000), he is not willing to saddle TVA ratepayers with the cost of yet another nuclear gamble. TVA has been there and done that, and is still doing it with large reactors. A large part of TVA’s indebtedness is from 1970s and early 1980s was due to the cancellation mid-stream of costly and unneeded reactors. No, prudently for TVA, Mr. Stout wants the taxpayers to take the deal.

And the federal government has obliged by committing up to half of the $452 million proposed for SMRs to Babcock & Wilcox, designer of the mPower, 180 megawatt reactor, and the TVA to advance design and certification. ^[1] B&W had spent over $200 million on the project as of the middle of last year. Where will it lead other than the “cobbled together” machine? Specifically, what will it take to get an assembly line? Here is how Mr. Stout explained it:

So the concept is that you gotta to have an assembly line cranking out repeatable parts, achieving a standardized vision of lots of mPower reactors. That creates the nth of a kind plant that has the efficiency in cost. I’m building Unit One. I don’t want to pay for B&W’s factory with automation to crank out repeatable parts. So that creates a contracting challenge… So as you scratch your head and puzzle how does work, remember the math won’t work on one unit. In fact our unit is most likely going to be, I’m going to use the word “cobbled together”, it’s going to be manufactured within existing facilities. But if B&W can get an order backlog of a hundred SMRs and they are going to start delivering them in China and India etc. then they’ll be able to go get the financing and build the building and have new stuff put in place to crank out these parts in a more automated manner. So as long as the design is the same this should all work. The devil is going to be in the details and in the oversight and in the inspections. [italics added]

A hundred reactors, each costing almost $450 million, would amount to an order book of $45 billion. That does not include construction costs, which would double the figure to $90 billion, leaving aside for now the industry’s record of huge cost escalations (see B&W 2012 presentation at http://www.efcog.org/library/council_meeting/12SAECMtg/presentations/GS_Meeting/Day-1/B&W%20mPower%20Overview%20-%20EFCOG%202012-Ferrara.pdf for total estimated cost figure of $5000 per kW). This would make the SMR launch something like creating a new commercial airliner, say like Dreamliner or the Airbus 350. There were a total of 350 orders for the A350 in 2007, when it was seriously launched as a rival to Boeing’s Dreamliner. The list price of the A350 is between about $250 million and $400 million (rounded — http://www.airbus.com/presscentre/corporate-information/key-documents/?eID=dam_frontend_push&docID=14849), which would make the initial order total the same order of magnitude in cost as 100 completed mPower SMRs.

The end of this decade is the target for the commissioning of the first mPower reactor (the most advanced SMR in planning and design and the only one with substantial federal government support so far). It would take some years after that – well into the 2020s – to fully prove the design and, if needed, debug it. It stretches credulity that China and India, which along with Russia, are the main centers of nuclear power construction today, would put in orders totaling a hundred reactors much before 2020. Indeed, if they were that attracted to SMRs, why would they not pay the license fees and set up the assembly line themselves? Most notably, China, where 28 reactors are under construction (http://www.world-nuclear.org/info/inf63.html), already has a much better supply chain than the United States. So the government subsidy to B&W and TVA would likely pave the way for an assembly line in China! Unless…

The federal government orders a hundred reactors or entices utilities to do so with massive subsidies. We are talking putting tens of billions of taxpayer dollars at risk – at a time when the air is full of talk of furloughs for teachers, air traffic controllers and civilian Pentagon employees, among others, and cutbacks in training of military personnel.

What happens if a common design or manufacturing problem is discovered as it was with the Dreamliner batteries? How is a mass manufacturer of reactors, whose internals become radioactive after commissioning, going to recall them or their major components? No word from Mr. Stout, or that I am aware, from any of the SMR enthusiasts, about this. For safety issues see the testimony of Dr. Ed Lyman of the Union of Concerned Scientists, at http://www.ucsusa.org/assets/documents/nuclear_power/lyman-appropriations-subcom-7-14-11.pdf.

IEER and Physicians for Social Responsibility wrote an SMR fact sheet in 2010, outlining such concerns. (http://ieer.org/resource/factsheets/small-modular-reactors-solution/) They have only deepened with time. SMRs are highly unlikely to provide the industry the nuclear renaissance that has so far eluded it. Indeed, by the time the first ones are being commissioned, solar photovoltaics, smart grids, and a host of other disturbed electricity technologies will be more economical. Wind-generated electricity already is. No one has a perfect crystal ball in the energy business, but mine tells me that SMRs will not solve our CO2 woes. They are likely to be economically obsolete before this decade is done – and the end of this decade is the target date for the first mPower reactor to come on line. So why go there, to a time when there are going to be only costs and little prospect of benefits?