Sowell's Law Blog: The Truth About Nuclear Power

Subtitle: San Onofre Shutdown Saga

The twin-reactor nuclear power plant at San Onofre, California, known as SONGS (San Onofre Nuclear Generating Station), is now shutdown after a project to replace worn steam generators resulted in premature tube leaks that released a small amount of radioactive steam to the atmosphere. The plant's majority owner, Southern California

SONGS Nuclear plant aerial view
Pacific Ocean at bottom, Interstate 5 at top
source: Wikipedia

Edison, SCE, chose to shut down the plant permanently rather than determine the cause of the premature failure and correct the problem. More on the details of the technical side may be found in Part Ten, see link, and in Part 16, see link.

The reasons for choosing a different design for the worn steam generators, and the economic aftermath for ratepayers and utility shareholders are the subject of this article, Part 23 in the series.

Choosing a Bad Design

The NRC has two separate paths for allowing replacement parts at a nuclear power plant: 1) like-for-like, where the new part is so nearly identical to the old part that a comprehensive design review is not required, and 2) differences between the new and old parts are substantial, and a comprehensive design review is required. SCE informed NRC that the new steam generators, two each for each reactor at SONGS, were like-for-like. That turned out not to be true. A couple of differences were 1) more tubes in the new steam generators, and 2) each tube had a slightly smaller internal diameter.

A utility company is in business for profit, and tries to make more profit where it can and is legally permitted to do so. Even at a nuclear plant, opportunities occur to increase profits. One way to increase profit is to increase the plant's output. As with most process plants, it is normally not economically attractive to replace a constraining part because the increased benefits are small while the added costs for replacing the part are large. A constraining part in a nuclear power plant could be, for example, a steam generator or steam turbine.

That economic computation changes, however, when a part is so worn that it must be replaced to stay in business. At that point, engineers can perform an incremental project analysis to determine if the part can be replaced with a somewhat larger part that produces more profit. Due to economies of scale, making the part only a few percent larger can cost very little extra. For example, making a steam generator 5 percent larger in surface area could produce perhaps 5 percent more electrical output. This is not a hypothetical, as some nuclear plants across the US have increased their generating capacity to a few percent above the design capacity. This is very likely what SCE was trying to accomplish at SONGS.

Where SCE erred was choosing the new design for the steam generators. The design turned out to have more vibration so that adjacent tubes banged into each other, rubbing metal away so that tube walls thinned and holes formed. The holes allowed the radioactive water to leak into the steam system. Safety was compromised, and the NRC was correct in stopping SCE from running the plants until the problem was resolved.

Economic Consequences

This entire episode was also described in a newspaper article which is actually not badly written. (This is high praise from me, as most newspaper articles on technical subjects get it very wrong.) see link

The utility wanted approximately $4.7 billion in compensation from the California Public Utility Commission, CPUC, for making the utility whole. The customers, or ratepayers, would pay the entire amount in the utility's request. However, ratepayer advocates negotiated a reduced amount - but not by much. The settlement agreement provides for $1.4 billion reduction, leaving $3.3 billion for ratepayers. The CPUC will make a decision on who pays how much for what. The basics of the proposed settlement can be found here -- see link.

The higher, policy argument is this: should a monopoly utility be rewarded for making stupid economic decisions? In this case with SONGS, the amount in question is only $4.7 billion. SCE is a huge utility, with a bit more than $12 billion in annual revenue, $45 billion in assets, 5 million customer accounts serving 14 million customers, transmission peak of 22,500 MW, and employing approximately 13,000 people. (source: SCE Annual Report 2013)

As the LA Times article (linked above) noted, this proposed settlement is akin to bailing out the banks in the economic crisis of 2008-2009. As ratepayer advocates stated, it is simply wrong to reward bad behavior by a utility. In my own words, somebody should give a utility reason to pause, think it over, and realize that jobs will be lost and the company will suffer for taking such a great risk. SCE could have easily ordered identical new steam generators, the same as the ones that ran perfectly well for 20-plus years. That would have been the no-risk alternative. Instead, the new design was chosen (perhaps) to increase plant output and profit.

It is the CPUC's job to look out for the ratepayer. One hopes that the settlement agreement is not accepted, and the utility is forced to bear the entire costs of making a stupid decision.

Previous Articles

The Truth About Nuclear Power emphasizes the economic and safety aspects by showing that (one) modern nuclear power plants are uneconomic to operate compared to natural gas and wind energy, (two) they produce preposterous pricing if they are the sole power source for a grid, (three) they cost far too much to construct, (four) use far more water for cooling, 4 times as much, than better alternatives, (five) nuclear fuel makes them difficult to shut down and requires very costly safeguards, (six) they are built to huge scale of 1,000 to 1,600 MWe or greater to attempt to reduce costs via economy of scale, (seven) an all-nuclear grid will lose customers to self-generation, (eight) smaller and modular nuclear plants have no benefits due to reverse economy of scale, (nine) large-scale plants have very long construction schedules even without lawsuits that delay construction, (ten) nuclear plants do not reach 50 or 60 years life because they require costly upgrades after 20 to 30 years that do not always perform as designed, (eleven) France has 85 percent of its electricity produced via nuclear power but it is subsidized, is still almost twice as expensive as prices in the US, and is only viable due to exporting power at night rather than throttling back the plants during low demand, (twelve) nuclear plants cannot provide cheap power on small islands, (thirteen) US nuclear plants are heavily subsidized but still cannot compete, (fourteen), projects are cancelled due to unfavorable economics, reactor vendors are desperate for sales, nuclear advocates tout low operating costs and ignore capital costs, nuclear utilities never ask for a rate decrease when building a new nuclear plant, and high nuclear costs are buried in a large customer base, (fifteen) safety regulations are routinely relaxed to allow the plants to continue operating without spending the funds to bring them into compliance, (sixteen) many, many near-misses occur each year in nuclear power, approximately one every 3 weeks, (seventeen) safety issues with short term, and long-term, storage of spent fuel, (eighteen) safety hazards of spent fuel reprocessing, (nineteen) health effects on people and other living things, (twenty) nuclear disaster at Chernobyl, (twenty-one) nuclear meltdown at Three Mile Island, (twenty-two) nuclear meltdowns at Fukushima, (twenty-three) near-disaster at San Onofre, (twenty-four) the looming disaster at St. Lucie, (twenty-five) the inherently unsafe characteristics of nuclear power plants required government shielding from liability, or subsidy, for the costs of a nuclear accident via the Price-Anderson Act, and (twenty-six) the serious public impacts of large-scale population evacuation and relocation after a major incident, or "extraordinary nuclear occurrence" in the language used by the Price-Anderson Act. Additional articles will include (twenty-seven) the future of nuclear fusion, (twenty-eight) future of thorium reactors, (twenty-nine) future of high-temperature gas nuclear reactors, and (thirty), a concluding chapter with a world-wide economic analysis of nuclear reactors and why countries build them. Links to each article in TANP series are included at the end of this article.

Additional articles will be linked as they are published.

Part One – Nuclear Power Plants Cannot Compete

Part Two – Preposterous Power Pricing if Nuclear Power Proponents Prevail

Part Three – Nuclear Power Plants Cost Far Too Much to Construct

Part Four – Nuclear Power Plants Use Far More Fresh Water

Part Five – Cannot Simply Turn Off a Nuclear Power Plant

Part Six – Nuclear Plants are Huge to Reduce Costs

Part Seven -- All Nuclear Grid Will Sell Less Power

Part Eight – No Benefits from Smaller Modular Nuclear Plants

Part Nine -- Nuclear Plants Require Long Construction Schedules

Part Ten -- Nuclear Plants Require Costly Upgrades After 20 to 30 Years

Part Eleven - Following France in Nuclear Is Not The Way To Go