If cost and schedule is the real hinderance to new nuclear, there are available solutions
More and more articles about nuclear energy are appearing to reach the conclusion that the technology is safe enough, reliable enough, and competitive enough except for the fact that the plants take too and cost too much. Some people in the pronuclear camp have determined that the only way to solve that problem is to invent a radically new type of nuclear plant that will be simpler, quicker and cheaper to construct than a typical light water reactor that boils water to produce steam.
I think that is the wrong path to take.
I confess. For a number of years I thought that the solution to the nuclear cost and schedule challenge was to shift to a revolutionary technology that could avoid some of the costs associated with using high pressure water and steam that are unavoidable. In 1991, I wrote a paper about the potential benefits of combining the low fuel cost and zero emissions capability of nuclear energy with the proven low capital costs of gas turbine heat engines. I got so excited by this concept, which had first been discussed in engineering literature in 1946, that I left active duty and formed Adams Atomic Engines, Inc. to pursue the idea full time. For a variety of reasons, that effort did not succeed.
However, I have always recognized that there were countless ways to improve the cost and schedule performance for light water reactors. I recognized that the first time I led an effort to resurface a 4 inch diameter flange that was leaking steam. The paperwork associated with that job was fully 3 inches thick.
No cost conscious nuclear professional would ever claim that we have, as an industry, ever done much to reduce the amount of work or time required to complete a task. I am not talking about cutting corners, but I am talking about paying attention when extra layers of requirements are added for no apparent reason. I am also talking about working hard to reduce existing requirements that reduce reliability with an increased investment and operational costs for unnecessary equipment.
Time is also a factor that can be reduced. On large construction and manufacturing projects, time really is money – every additional day adds to the cost of labor, the cost of rent on large equipment, and the amount of interest charged. It also delays the time when revenue can start flowing to begin paying off the accumulated investments and loans.
The potential improvements of series production also cannot be underestimated. Every manufacturing and construction project team that produces the same design more than once will experience a learning curve effect. Every doubling of cumulative unit volume of the same design with the same team will result in a predictable cost and schedule reduction. Please note the caveats of that general rule. It is hard for traditional site-built plants that take five to ten years to construct to take any advantage of learning curves. Recognition of that fact is a major reason that there is so much interest in smaller, more modular reactors where substantial sections can be built in a factory and delivered to the site.
Modular construction is not just for small plants. Many of the large reactors are designed from the ground up to take advantage of modularity and factory production of substantial sections of the plant that can be delivered to the site already assembled. I have been especially impressed by the efforts made by the AP1000 team that includes Westinghouse and Shaw Group to implement modular construction techniques and to recognize the cost saving benefits of learning curves and unit volume.
I like to take most people at their word – unless they have already proven to be unreliable. When people tell me that they would be in favor of nuclear if it was not so expensive and it did not take so long, my new approach will be to attempt to engage them as allies in a long struggle to address the cost and schedule issues. History shows that we have the opportunity for significant improvement. The very first nuclear plant built in the US took just 4 years. That project had to invent ways to manufacture fuel, had engineers using slide rules, and had no overhead cranes available. We have a lot of advantages today that the Shippingport engineers and builders did not have.
Here is the article that stimulate this particular line of thought this morning, but it is only one of many like it. Lifting nuclear ban may not mean much. The article discusses the recent successful steps being taken to eliminate the new nuclear plant moratorium in Minnesota, but it includes quotes like the following:
Even if the utility needed another big power plant, Xcel officials say building a new nuclear reactor poses many uncertainties.
“First of all the estimates are very high on what it will cost,” said Betsy Engelking, Xcel’s director of resource planning. “It will be cheaper to run because the fuel costs are low. It is very high in terms of building it. And there is just a lot of uncertainty about how long it will take you to get through a process to actually approve a plant.”
Nuclear plant suppliers have to recognize the importance of addressing initial costs, but they cannot allow that effort to cause them to make decisions that may increase costs or risks later. There is a balance that has yet to be reached, but if we really do want access to lower cost, cleaner energy, there are steps we can all take to reduce barriers and eliminate uncertainty.
That statement is especially true for those people who spend their time working to slow down nuclear development while at the same time complaining that the plants are too expensive to consider.
The computer aided AP1000 design allowed the engineers to see on the computer what the plant would look like at each step. One of the design processes was to have construction experts think through how each new module would be connected and wired. This process resulted in a 36 month estimate for the time between first concrete and start of testing. The first four Chinese AP1000’s are on a 42 month time schedule and the schedule is being met. The Chinese could do us another favor by increasing the construction speed especially on the 2nd, 3rd, and 4th plant at a site. Do you think that AP1000’s can be built as fast in the United States as they are in China?
I think that one of the problems in the US is that no one wants to be first and take on the associated risk. One way to prove that false is to have the manufacturer (i.e. B&W mPower) take on the risk of building and operating the first plant itself. It would show confidence in the design. When I was initally hired by Applied Materials (AMAT), all new hires were given a book written by then-CEO James Morgan, “Cracking the Japanese Market: Strategies for Success in the New Global Economy”. One of the problems related in the book in getting Japanese companies to buy from AMAT was that the Japanese companies did not believe that AMAT intended to have a longterm relationship with them. Whether or not it was true, they felt that AMAT was just trying to sell them something and would not support it afterwards. At that time, AMAT was selling their products through a 3rd party representative. AMAT’s solution was to create the subsidary Applied Materials Japan (AMJ), a physical presence in Japan with Japanese personnel and president. Only then did the Japanese semiconductor companies begin to believe that AMAT was behind their product 100%.
The analogy is not perfect, but history is full of examples where companies actually had to build and demonstrate their products before customers accepted and bought them.
I’m worried that the problems that plague large DoD acquisition programs may make there way to nuclear plant construction in the US. We actually reward schedule slips and cost overruns–no General ever earned another star by delivering a program on time and wthin budget, and defense contractors don’t earn more profit by meeting the government’s original cost and schedule goals. In other words, I’m afraid that there may hidden incentives for companies, labor groups, and government officials to find a way to lengthen construction time tables.
People need to realize that the nuclear industry is an exception to the rule of free enterprise. The kind of rights and freedoms that we take for granted in most area of our lives don’t apply to nuclear energy. So I totally agree with Rod and see this as an opportunity to say the obvious (to most readers) but it still needs repeating here: That the regulatory system needs changing or replacing or both. Then costs and schedule concerns will both improve enormously when fair practice of rules and regulations are observed.
I started a comment like this earlier but deleted it since it didn’t feel politically correct. The regulatory environment is only friendly to those who are in opposition to the industry regulated. Why else would Vermont Yankee and Pilgrim Station have License Renewal applications with the NRC for more than 5 years when similar plants in Nebraska and Iowa are approved in less than half that time? Why does the NRC have a chairman who is clearly anti-nuclear? I in no way intend to demean the many good people at the NRC especially the plant residents who are knowledgeable and clearly highly responsible, but at times the NRC appears to act like it was a science fair or was in competition for government grants. Constantly revisiting and reevaluating what at one time were settled issues. It seems, at times, that the NRC is a make work project rather than a regulating body.
Running up the costs by the use of legal gymnastics, and filibustering hearings and such, are the stock-in-trade of the antinuclear movement. Only massive changes in legislation are going to cut those avenues of attack off.
I think Martin points out a likely path. Once China really gears up for the construction of AP1000’s and their variant (CAP1400), schedules will shorten and costs will come down. Then, we can have a discussion of ‘what is different’? Even Japan (hardly a developing country with low labor costs) built reactors for relatively low costs (half of what the US is projecting) in the late 90’s.
Although it may be painful that Westinghouse has turned over 70,000 documents to the Chinese on the AP1000, at least the Chinese are letting them build the darn things. After the first four are completed (and maybe Westinghouse will get a few more contracts) they will have worked out a lot of kinks. So, when they build them in the US (Vogtle seems pretty certain), I expect it will go relatively smoothly. If regulatory / legal issues are used to slow construction, then, when people ask why it is different in China, there will be some concrete answers.
@SteveK9 – my issue with that analysis is that learning curves get reset when you either change the design OR when you try to bring in a new team in a new location. Many of the lessons learned are learned at the trade skill and apprentice level; they will not directly translate half way around the world and from Chinese to English.
We will be setting ourselves up for and enormous failure if we think that we will be able to match the Chinese 4th of a kind cost on an American FOAK unit. I have spoken to some management level folks who think that sending engineers and supervisors to work in China for a year or so at a time will provide equivalent benefits, but I do not buy it based on my own front level repair and maintenance experience. I HOPE I am wrong, but hope is not a strategy.
I’ve also read about the Vogtle folks going over to China and vice versa. You’re probably right that building our first will not be like the Chinese building their 4th, but I’m sure some pitfalls will be identified (in fact I recall one of the Westinghouse executives discussing a particular instance (I don’t remember the details though) and avoided at Vogtle.
I want to say thanks to Rod Adams for the great effort put into the Atomic Insights Blog. I follow it daily and I have no doubt that it is serving to quicken the pace of the nuclear resonance.
I am an emeritus professor of biology, so I am looking in on nuclear power industry activity from the outside. I have greatest respect for engineers. I know that engineers are responsible for a great deal of the breakthrough innovation in all fields. Ideas are often born when one is involved in hands on application.
From my position on the outside looking in, I do raise one concern. That is the tendancy to fall in love with the current technology that one lives and breathes on a daily basis. The extra cost associated with high pressure is not the only draw back to the LWR. How about the 250 tons of uranium that needs processing for a one GWyear of kWhs? Fission products amount to only one ton of the 250. Anti-nukes have a hayday with the so-called long term waste.
Breeder technology will almost certainly be the future energy source replacement for current technology. The Russians report a good experience over 30 years with the BN 600 and they will soon deploy the updated BN 800.
There are reports that China has also bought into the BN 800.We have plans for two closed cycle reactors, the IFR and the LFTR.
China recently announced plans to build a Thorium MSR after a visit to Oak Ridge. We have had 40 years to pick up this ball. I suppose that we should be glad that somebody is going to attempt its development. Two Nobel laureates, Eugene Wigner and Edward Teller of the Manhattan project endorsed the TMSR or LFTR reactor. Sadly, politics killed this promising project in the 1970s.
TerraPower, a company founded by Bill Gates for the express purpose of building a Traveling Wave Reactor (TWR) with potential to produce electricity cheaper than dirty coal plants. Gates says that high operating temperature means that it puts out a lot of energy for its size. It is designed to use the spent fuel from our current LWRs and it will be loaded with enough fuel to operate for 60 years without refueling. He is looking to partner with China. Gates says,
@John – I am not sure which thorium evangelist is providing you with the numbers that you use for light water reactor fuel use.
it takes about 8-9 kilograms of natural uranium to produce one kilogram of low enriched uranium fuel, depending on the “tails assay” concentration.
The typical LWR burn up is about 45,000 MW-days/ton. That is 4.5-5% of the amount of energy that would be released by complete fission.
By my rough numbers, about 4-5% of the used nuclear fuel is fission products. The depleted uranium is not counted as HEU – in fact it can be disposed of in low level repositories, but it better used by saving it for future breeders.
Even if you count the depleted uranium as part of the waste volume, you still only end up with about 160-180 tons per ton of fission products.
Anyway, I have no beef with breeders. I think it is a great technical direction, but you cannot jump from an 8086 to a dual core Core Duo without going through many technical iterations. I am not “satisfied” with what we can build today, but they are already a lot cheaper than coal plants to operate. Compared to coal plants that meet today’s legal requirements for new construction, I would venture to say that LWRs cost about as much to build as compliant coal plants.
Focusing on fuel use improvements violates one of my fundamental engineering design principles – “fix the big noise first”.
In the 8 – 10 cents per kilowatt hour projected by the MIT Future of Nuclear Power study, only 0.7 cents was the cost of the entire fuel cycle, even with all of its current waste.
If we were disposing of the non used fuel, I would be more worried, but we are simply storing it up for future generations.
We do agree re: the NRC Chairman.
I acknowledge that I follow both thorium and IFR evangelists. I was citing from Robert Hargraves’ Aim high power point presentation. His numbers are not that different from yours. I believe that he suggest 3.5% enrichment. Thirty five tons of uranium for fuel rods with 215 tons of depleted uranium left over. He lists spent fuel as being 34.4 tons U 238, 0.3 tons U235, 0.3 tons Pu and 1 ton of fission products. I am aware that only fuel that has been in a reactor requires isolation to protect from radio-toxicity., None the less , enrichment which produces depleted U becomes a part or the equation when estimations about the abundance of nuclear fuel for LWR technology in ore deposits. It would appear that the fuel supply for breeders is 250 times greater than for LWRs. I am really not concerned about running out of nuclear fuel. As you point out nuclear fuel is cheap owing to its energy density. Harvesting from the ocean would not be prohibitively expensive. Phil Morrison in 1943 observed that even the 4 ppm uranium in granite could be mined for use in a breeder. The world supply of uranium and thorium appears to be inexhaustible.
@John – I like Bob and like his Aim High presentation. However, he is not playing fair by comparing the performance of obsolete light water reactor fuel against the performance of molten salt reactors that exist only in either dim memory as prototypes or as future visions in the minds of forward thinking people like Kirk, Charles and Bob. I am not saying they are wrong about the potential, but so far it is unrealized potential that will need some major expenditures in engineering and technology development to achieve.
On the other hand, Bob’s numbers do not include improvements in LWR fuel that are in wide use today. The difference is not large, yet, between what he includes and what commercial reactors already achieve, but there are innovations that have already been fully tested that double the burn-up and offer a path towards additional improvements. Those innovations have not yet been employed, partly because they are not economical if natural uranium is cheap.
If the basic material is cheap enough, doing quick and wasteful processing is often the most economical way to proceed. That concept violates the sensibility of many engineers who prefer more elegant, less wasteful solutions, but businessmen often do not care much about elegance.
Mr Pronuclear, you assert that the regulatory system needs to be replaced on reformed. Many others have made the same claim. However, I can not find documents, blogs, on any information that lays out what a reformed regulation system would look like. Can you point me to a list of specifics or a 20 page position paper that has been written about changes to the regulations system. Since I am not from the nuclear industry I can not tell the difference between someone complaining about irritating paper work and someone that has an idea about changes that can be made.
A reformed regulation system? Some ideas:
*Hard time limits for every NRC action except perhaps design certifications and manufacturing licenses; if the NRC fails to finish within the time limit, the action is automatically approved. Say 6 months for early site permits, 12 months for COLs.
*Since the NRC is a highly technical agency, it obviously has much greater technical competence than courts do. Eliminating judicial review of NRC decisions except where the NRC fails to follow its own rules is thus appropriate, as the courts don’t know what they’re doing.
*Elimination of any and all environmental impact statements as part of the siting process. The environmental impact of a nuclear plant is minimal; provided the plant can safely be built on the site, there is negligible environmental impact, and statements are just useless red tape.
*Full and explicit Federal pre-emption of any and all state level regulation of nuclear power related issues beyond the decision of the state PUC/PSC to allow construction.
C’mon Rod, isn’t this what you work on all day? I’d love to hear more about the specific.
I think the rational plan to expand nuclear in the short term is to build cheaper LWRs with a more repeatable process. That said, the picture today still looks a lot better than it looked in the late 1970’s, when orders for plants were being canceled. Back then, the day-to-day operation of LWRs hadn’t quite been mastered, so potential owners would be afraid of spending a lot of money to buy a lemon.
Today, LWR operation is mature, and we’re seeing steady and gradual improvements in fuel performance over time.
As for life cycle issues, it’s true that, with the kind of assumptions most people make today, reprocessing costs somewhat more than direct disposal of spent fuel rods, but when you consider the total cost of the system, it’s not such a big deal — the decision would better be made based on long term sustainability of the industry, proliferation risks, or other issues. As is the case with the LWR, a mature reprocessing and MOX fabrication capability exists that can be replicated internationally, so we don’t need to be afraid of another West Valley (which struggled to extract plutonium) or Sellafield (which was never able to fabricate usable MOX fuel)
There’s a big literature on metal-cooled fast reactors, and people like Frank von Hippel will point with glee to estimates that LMFBRs that use water as a working fluid will have capital costs 50% or more greater than the LWR, which requires astronomically expensive uranium for them to become practical. However, there’s a more modern line of development that assumes a dry system that uses a gas as a working fluid in a Brayton cycle turbine. The total size and mass of a system could be much smaller, and some people imagine that such a reactor could be cheaper than a LWR. However, this requires extensive technology development, and it’s fair to assume that operational challenges would take some time to sort out — however, the pessimism that one could get from the French and Japanese experience could be tempered by the fact that we’re entering an age of robotics, so that maintenance doesn’t require draining all the coolant, purging the atmosphere and admitting human workers.
The story with thorium MSRs is similar, but there’s less literature. Elegant integral reprocessing may be possible, but there’s a lot of technological development to do. I know many MSR advocates envision that the MSR would make a good small reactor, but the Teller-Moir idea of building large MSR nuplexes makes more sense to me.
@Paul – You know I cannot tell you what I do all day. 🙂 That would not be fair to the people who pay my salary.
Rod, I like your idea of streamlining paperwork and IMPROVING regulatory processes.
I grew up in the IT industry. In the 1990s business process re-engineering was the technique that really improved ROI for businesses. The technique is to TOTALLY redefine all business processes, using a team of experienced, capable, dedicated managers from ALL business functions. IT analysts and consultants would conduct top-down business functional decomposition and also logical data base design of all the information a business needed. The stockmarket runup post 2000 is partly due to this. The success of SAP software is due to this sort of corporate-wide systems engineering work.
A key lesson is the idea of defining WHAT work should be done independent of WHO does it, in order to circumvent organizational politics.
This could be done at the NRC as well. With good consulting leadership, we could do top-down re-engineering of all the NRC business processes and information needs, developing information systems and documenting business processes to operate more efficiently and paperlessly. We can retain all the regulatory functions we choose to, but eliminate inefficiencies.
Of course business process re-engineering improved ROI by cutting costs — cutting labor costs — cutting jobs, and the same result would be obtained within NRC.
The best re-engineering efforts within industry take 1-2 years, and this could happen within NRC with proper management direction and funding. [2 years is one election cycle.]
Comments are closed.
Recent Comments from our Readers
The Clinton Nuclear Plant also in Illinois was shutdown essentially for almost 2 years before it was taken over by…
Good Podcast – Very informative One thing that was not discussed is how to deal with a particular fear that…
Renewables people are masters in marketing. Unreliable intermittent generators whose output is all over the place, and usually badly correlated…
Looking at their lineup, Westinghouse seems bound and determined to keep Gen IV in its “place” which is apparently the…
So they are developing a scaled down version of the AP1000, which is a scaled up version of the AP600,…