San Francisco — Shoreham. Marble Hill. Zimmer. Nuclear power plants representing billions of dollars in investments. They are also stark reminders that low-cost electricity from the atom is easier promised than delivered. Some utilities have been successful at building and operating nuclear plants. But others are staggering under nuclear-generated debts, converting nearly completed facilities to coal, and even abandoning nuclear projects. Can a 19th-century industry learn to manage 20th-century technology?
There is a subtle, almost subliminal message in the design of a nuclear power plant's control room.
The fact that its basic layout and confusion of dials, gauges, switches, and illuminated warning signs are strikingly similar to those found in decades-old conventional power stations clearly shows the heritage of the electric utility industry.
With roots in the inventions of Thomas Edison and Nikola Tesla, this industry is more than a century old and generally set in its ways. Like its control rooms , many of its attitudes, methods, and procedures change with near-glacial slowness.
In recent years, this has had mostly minor effects on the industry's ability deliver electricity safely and at reasonable prices. In the case of nuclear power, however, it is one of the major reasons why the near-heroic engineering effort to harness the atom for peaceful purposes is foundering on the triple shoals of economics, regulation, and public perception.
This is not to say that the utility industry is the only major actor on the nuclear power stage. As the congressional Office of Technology Assessment (OTA) put it recently in its report entitled Nuclear Power in an Age of Uncertainty: ''There are at least . . . seven sides to the coin of each issue.''
Besides the utilities, these principal players include utility investors, state public utility commissions, nuclear reactor manufacturers, antinuclear critics, the Nuclear Regulatory Commission (NRC), and the public.
But ''most of nuclear energy's problems stem from the fact that we have a 19 th-century industry trying to cope with a 20th-century technology,'' argues Victor Gilinsky, an NRC commissioner and longtime industry critic.
There is growing agreement that a fundamental mismatch exists between the technology and the institutions overseeing it. This stems not only from the industry's ultraconservatism: Even today it invests only about 0.6 percent of its sales in research and development. It is also due to the industry's extreme diversity, a legacy of pitched political battles waged early in this century between the advocates of public and private power. This has left a patchwork of institutions that includes rural electric cooperatives, tiny municipal utilities , large, private companies, and federal organizations like the Tennessee Valley and Bonneville Power Authorities. Not only do these embody a wide range of management philosophies and capabilities, but a bewildering variety of regulatory arrangements.
To appreciate the problem this poses, a few facts about nuclear fission technology are necessary. Despite all the engineering effort lavished on nuclear reactors, they have proved far more demanding to build and operate than their designers anticipated. One cubic foot within a reactor core generates as much power as 1,000 cubic feet of raging flame in a coal-fired boiler. As a result, things can happen very rapidly. In the case of certain malfunctions, plant operators have but minutes to figure out what is wrong and correct it.
A nuclear reactor is also very demanding in routine operation. For instance, the chemistry of the water flowing through the reactor must be kept within strict tolerances. If it is not, the plant's piping is subject to severe corrosion and damage.
Compounding this is the technology's complexity. A good benchmark for power plant complexity is the number of valves it has. A nuclear reactor has 40,000 - 10 times the number in a coal-fired plant.
For these reasons ''a nuclear reactor requires 10 times the management intensity as a comparable coal-fired power plant. There are a lot of dedicated people. But some have not been able to muster the management intensity required, '' says Carl Walske, president of the Atomic Industrial Forum, an industry trade group.
This is not to imply that nuclear reactors cannot be operated safely and efficiently. There are currently 83 commercial reactors in operation. And, as the OTA report observes, the excellent performance record of some of these ''indicate that LWRs (light-water reactors) can be very reliable when properly managed.''
If operating a nuclear reactor is unusually demanding, building one has proved monumental. A 1,300-megawatt reactor consumes some 75 million pounds of steel, 195,000 cubic yards of concrete, 8 million pounds of piping, 1,000 miles of wire and cable, and 52,000 man-years of labor. Small armies of specially trained workers - 2,000 to 4,000 per reactor - are required. The time required to build a reactor ranges from six to 16 years.
Because of the tremendous concentration of energy within a reactor core and the inventory of potentially lethal radioactive material that builds up, quality of construction has been a primary concern within the nuclear industry from the first. It is not uncommon for field engineers on nuclear projects to work within tolerances of 1/16 or even 1/32 of an inch.
An elaborate paper audit system that rivals the one the National Aeronautics and Space Administration uses on its space vehicles has evolved. The NRC estimates that this adds 40 to 50 percent to basic engineering and construction costs.
In short, building and operating nuclear power plants has proved the most ambitious undertaking in the history of the electrical utility industry.
It wasn't something that the industry had time to grow into gracefully. The 1950s and early '60s were comfortable periods for electric utilities. Demand was growing at a steady 6 to 7 percent a year. Prices were dropping with each new plant built. The technologies for generating electricity were mature and demanded only minor refinements, so utilities did not need much technological sophistication. The major source of excitement was trying to wrest more customers from the clutches of the natural gas industry.
In the late '60s, however, the inexorable pressures of exponential growth were building. Sustaining historic growth rates meant adding more and more generating capacity annually. Projected into the future, the number and size of new power plants that would be required rapidly became staggering. Nuclear energy, strongly championed by the federal Atomic Energy Commission, looked like the best answer. As a result, utilities small and large began ordering nuclear reactors like hotcakes. One hundred forty such orders were placed between 1970 and '74. This prospect also drove reactor vendors to scale-up their designs to match expected demand and gain economies of scale. By 1968, orders were being placed for units larger than 1,000 megawatts, although there were only three plants larger than 100 megawatts in operation.
At first, it seemed that the energy crisis, touched off by the Arab oil embargo of 1973-74 would further boost nuclear power's fortunes. President Nixon announced Project Independence, which, among other things, called for thousands of new reactors by the end of the century.
Ironically, the chaotic events in following years were to prove nearly fatal to the nuclear enterprise. Energy-crisis politics and skyrocketing energy costs thrust utility executives from comfortable obscurity into the limelight. Consumers, shocked by rising energy bills, began exerting strong pressures on state public utility commissions to hold down rate increases. This began a slow squeeze on utility finances.
Next came double-digit inflation. Then, interest rates jumped dramatically. Finally, the severest blow came as recessions and aggressive energy conservation knocked the bottom out of demand.
These factors turned the flood of reactor orders into a riptide of cancellations. Since 1974, 101 projects have been canceled: a rate of one every 5 1/2 weeks. The last new order was placed in 1978.
The pro-nuclear stampede of the early 1970s had a number of side effects that contributed directly to this situation.
Utilities were used to having power plants designed to their specifications. They insisted on continuing this practice with nuclear plants, although reactor makers tried to sell the advantages of standardization. This created a shortage of nuclear power plant designers, leading to the practice of breaking ground on a new plant when its design was only 30 percent complete. This approach resulted in a large number of costly field changes. Customization alone appears to have nearly doubled the price of nuclear power.
''The industry now realizes the high price it has paid for customization,'' observes James Moore, vice-president and general manager of the water reactor division of Westinghouse.
The industry's approach to training operators also has contributed to its current problems. Essentially, utilities slightly upgraded the qualification for operators of conventional power plants. Some reactor-safety experts suggested at the time that a better model, in terms of training and pay, were airline pilots. Operator error was a major part of the Three Mile Island accident. Since then, utilities have substantially boosted their operator training and requirements.
While utilities such as Commonwealth Edison in Illinois and Duke Power in North Carolina nurtured inhouse expertise on the technology, others relied almost exclusively on reactor builders and the US government. Because the industry was fragmented, there were no effective programs to compare and analyze operating experiences at various reactors. Thus, operators of the Three Mile Island plant were not aware of events at other reactors, knowledge that might have prevented the accident that destroyed one of their reactors. Since then the industry has instituted such a program.
Meanwhile, the regulatory process was evolving, not smoothly but reactively. There were three major flurries of regulatory activity. These followed a legal decision in 1971 that required environmental-impact statements for nuclear power plants, the 1975 fire at the Brown's Ferry nuclear plant, and the accident at Three Mile Island in 1979. New regulations evolving from these events severely disrupted the schedule of plants under construction at the time, driving up costs substantially. At the same time, utilities adopted a series of changes to improve plant performance as the industry gained experience with the large reactor designs.
''Together these accounted for most of the 240 percent increase in the constant-dollar cost of nuclear reactors between the beginning and end of the 1970s,'' calculates Charles Komanoff, an independent industry analyst.
Next, as more and more plants moved from the design to the construction phase , a shortage of qualified workers developed.
The first companies to venture into nuclear power tended to be the stronger, more capable companies. Even many of these have had to struggle to manage this technology properly. But the rush to go nuclear also attracted a number of less experienced, less capable companies. Nuclear projects have spelled hard times for a number of these:
* General Public Utilities, owner of the Three Mile Island nuclear power plant, continues to skate at the brink of insolvency. After five years, the stricken TMI-2 reactor has not been totally cleaned up and the utility has not received permission to restart the undamaged reactor at the site.
* The Washington Public Power Supply System - a consortium of more than 100 public utilities in the Pacific Northwest - defaulted late last year on $2.25 billion in municipal bonds floated to finance part of an overly ambitious five-reactor construction program.
* Cincinnati Gas & Electric announced that it was going to convert its Zimmer nuclear power plant, 97 percent complete, to coal. CG&E had severe problems with quality control during construction. The NRC was forced to require that much of the plant be torn down and rebuilt, which would have almost doubled the plant's construction cost.
* Public Service of Indiana had some severe problems with the quality of the concrete work on its two-unit Marble Hill project. They managed to surmount these problems, but ran out of money. So they abandoned the project, although they had already spent $2.5 billion.
''Things will get a lot worse before they get better,'' says Mr. Komanoff. He predicts that in coming months, half a dozen more investor-owned utilities will face insolvency because of their nuclear projects.
Critics have worried about the capabilities of utilities for some years. But the concern was couched primarily in terms of public safety. But recently, this has been more than displaced by a growing appreciation that nuclear power may be economical in the hands of capable utilities but prohibitively expensive in the hands of those that are not. ''There is a high correlation between high costs and inexperience,'' Komanoff says.
The nuclear industry itself has increasingly realized that it is ''held hostage'' by its weakest members.
A reflection of this fact is the Institute of Nuclear Power Operations (INPO) , the industry group formed after the accident at Three Mile Island. INPO collects, analyzes, and redistributes reports of operating experiences within the industry. According to the OTA, the group has been very successful in getting utilities to comply voluntarily with its recommendations.
INPO also has been evaluating how well companies operate their reactors and ''has made a noticeable difference,'' says James Taylor, deputy director of NRC's division of inspection and enforcement.
Ironically, while nuclear power's problems have dominated the nation's headlines, the atom remains the fastest growing US electricity source. Since early 1980, one new reactor has been put on line every three months. For every industry problem-child, there are two or three plants that have been built with relatively few problems and little controversy.
It is clear, however, that radical changes in either the institutions, the technology, or the economy will be required before a new nuclear plant is ordered in the US.