nliouse Effect ESEARCH IN EPRI JOURNAL is published nine times each year (January/February, March, April/May, June, JulyIAugust, September, October, November, and December) by the Electric Power Research Institute, EPRI was founded in 1972 by the nation's electric utilities to develop and manage a technology program for improving electric power production, distribution, and utilization. EPRI JOURNAL Staff and Contributors Brent Barker, Editor in Chief David Dietrich, Managing Editor Ralph Whitaker, Feature Editor Taylor Moore, Senior Feature Writer Michael Shepard, Feature Writer Pauline Burnett, Technical Editor Mary Ann Garneau-Hoxsie, Production Editor Jim Norris, Illustrator Jean Smith, Program Secretary Christine Lawrence (Washington) Kathy Kaufman (Technology Transfer) Richard G. Claeys, Director Communications Division Graphics Consultant: Frank A Rodriquez © 1986 by Electric Power Research Institute, Inc. Permission to reprint is granted by EPRI, provided credit to the EPRI JOURNAL is given. Information on bulk reprints available on request Electric Power Research Institute, EPRI, and EPRI JOURNAL are registered service marks or trade­ marks of Electric Power Research Institute, Inc. Address correspondence to: Editor in Chief EPRI JOURNAL Electric Power Research Institute P.O. Box 10412 Palo Alto, California 94303 Cover: The buildup of CO2 and certain trace gases in the atmosphere is blocking the natural escape of much of the earth's heat from our environment. Experts believe that the resultant warming trend could have profound implications for the earth's climate in as few as 50 years. EPRIJOURNAL Volume 11, Number 4 June 1986 Editorial 2 Features 4 16 22 28 The Scientific Unknowns of CO2 The Greenhouse Effect: Earth's Climate in Transition Global warming as a result of the greenhouse effect may turn out to be the greatest environmental problem of modern times. Evolution in Combustion Turbines Increasingly efficient and reliable, combustion turbines have become an attractive option for utilities seeking small increments of new capacity. Remote Scanning of Low-Level Waste Two new methods improve the accuracy and lower the risk of waste assays by reading radiation levels and determining isotopic content from outside the waste container. Herbert Woodson: Encouraging Exploration in Research An electrical engineering professor for 32 years, Woodson keeps his hand in with utilities as a consultant, committee member, and member of EPRl's Advisory Council. Departments 3 34 35 36 57 58 60 Authors and Articles At the Institute: Board Approves 1986 Budget Revision Calendar Technology Transfer News New Contracts New Technical Reports New Computer Software R&D Status Reports 38 42 46 50 54 Advanced Power Systems Division Coal Combustion Systems Division Electrical Systems Division Energy Analysis and Environment Division Nuclear Power Division The Scientific Unknowns of CO2 Males Coming to appropriate decisions on complex environmental issues is extremely difficult, a fact that is often ignored by advocates of one or another policy prescription. Usually, it is only with hindsight that an action appears to be undeniably the preferred policy option. Only 50 years ago the heavy particulate and gaseous loadings that make up urban smog were deemed a sign of industrial progress and a necessary concomitant cost of economic growth; in the United States, these levels of urban pollution are now perceived to be unacceptable. Although opinions may differ on the appropriate degree and method of restriction, few would argue that the emissions limitations imposed by state and federal laws and regulations are undesirable. Unfortunately, the benefit of hindsight is not available for environmental issues we are studying today. The so-called greenhouse effect-the buildup of CO2 in our atmosphere and the consequent possible effect on climate-is a classic example of the difficulty of dealing with complex scientific unknowns. This issue involves tough basic questions: what happens to the CO2 that is emitted from our industrial combustion processes? what are the other sources of CO2? what are the sinks? Even when we become relatively knowledgeable about these questions, a whole series of others loom ominously: what will be the effect of various levels of CO2 in the atmosphere? how do other greenhouse gases affect the climate? what are the possible feedback loops from initial warming effects? As with all scientific issues, research will never bring absolute certainty, but well-designed research can advance the understanding and reduce the degree of uncertainty for each of these phenomena. The eventual decision on whether to limit emissions of CO2 and the other greenhouse gases and how to effect such limits will be a political one-a decision that balances risks against costs, that takes into account public values. Yet such a decision will be easier to make and will be better designed if we know more about the science of the issue. EPRI has the role of developing such scientific and technical understanding for the electric utility industry. Fortunately, DOE has a large ongoing research program addressing many of these scientific questions. EPRI can share in these results and build its much more modest research agenda as a complement to DOE's research. Similarly, other federal agencies are looking at issues related to CO2. By coordinating with these agencies, EPRI can help utilities keep up to date on a field in which they may be called upon to take policy positions. Rene H. Males, Vice President Energy Analysis and Environment Division 2 EPRI JOURNAL June 1986 Authors and Articles Hakkarinen T Hansen Spencer Dolbee he Greenhouse Effect: Earth's Cli­ mate in Transition (page 4) exam­ ines the myriad mechanisms by which carbon is exchanged between plants, oceans, and the atmosphere, and the potential outcomes. Written by Michael Shepard, Journal feature writer, who re­ ceived principal technical guidance from three EPRI research managers. Charles Hakkarinen is technical man­ ager for environmental data analysis in the Energy Analysis and Environment Division. Formerly an assistant to the di­ vision director, he has been with EPRI since 1974, when he was still completing work on his PhD in environmental sci­ ence and engineering at the University of California at Los Angeles. Alan Hansen, also in the EAE Divi­ sion, is a project manager in the Envi­ ronmental Chemistry and Physics Pro­ gram. He came to EPRI in March 1985 after eight years with Environmental Re­ search & Technology, Inc., where he was manager of environmental chemistry. He was previously an environmental analyst and a research chemist with the Environ­ mental Protection Agency, the University of California (Riverside) Air Pollution Re­ search Center, and SRI International. Hansen earned a PhD in chemistry at the University of California (Irvine). Dwain Spencer is an EPRI vice presi­ dent and director of the Advanced Power Systems Division. Before he joined the Institute in 1974, he was with the Jet Pro­ pulsion Laboratory of the California In­ stitute of Technology for 16 years, part of that time in studies of planetary atmo­ spheres in which space vehicles would Dohner Gluckman travel. Spencer has a BS in chemical en­ gineering from the University of Notre Dame and an MS in engineering from Purdue University. • E volution in Combustion Turbines (page 16) reviews the thinking and the R&D that are adapting an already­ modular prime mover for intermediate and baseload use. Written by Michael Shepard, Journal feature writer, aided by members of EPRI's Advanced Power Systems Division. Albert Dolbee has managed his divi­ sion's research in power machinery since 1978. He came to EPRI a year earlier after nearly 26 years with General Electric Co., where he became manager of design en­ gineering for gas turbine and combined­ cycle plant controls. He is an electrical engineering graduate of Manhattan Col­ lege. Clark Dohner, a project manager in Dolbec's program, has been with EPRI since 1982, guiding R&D in gas turbines. He formerly was with General Electric for 20 years, the last 10 in the gas turbine division. Dohner has BS and MS degrees in mechanical engineering from Johns Hopkins University and the University of Pittsburgh, respectively. Michael Gluckman heads the Engi­ neering and Economics Evaluation Pro­ gram. He was named to the position in 1980 after five years as project manager in gasification system research. Between 1971 and 1975 he was an associate pro­ fessor of chemical engineering at City College of the City University of New York, where he earned his PhD. • Welty Robinson R emote Scanning of Low-Level Waste (page 22) presents two new, fast, safe, and thorough ways to measure cer­ tain radioisotopic contents of waste ma­ terial. Written by Taylor Moore, the Jour­ nal's senior feature writer, with the coop­ eration of two research managers in the Nuclear Power Division. Charles Welty, a project manager in low-level waste and coolant technology research, has also managed EPRI re­ search for the utility Steam Generator Owners Group. He came to EPRI in 1978 after three years as production manager and manager of the Dillingham Corp. shipyard in Honolulu. A graduate of the U.S. Naval Academy, Welty was an offi­ cer in the navy's nuclear submarine force for nine years. Patricia Robinson, also a project man­ ager for low-level waste research, has been with EPRI since August 1985. She was formerly a research engineer at Bat­ telle, Pacific Northwest Laboratories for two years. Between 1977 and 1983 Robin­ son worked for two companies providing reactor operation and waste transporta­ tion services at the federal government's Hanford site. She has a BS in chemical engineering from the University of Ne­ vada (Reno). • H erbert Woodson: Encouraging Ex­ ploration in Research (page 28) in­ troduces a Texas engineering professor and member of EPRI's Advisory Council who has a well-developed loyalty to R&D. Written by Ralph Whitaker, the Journal's feature editor, from an inter­ view with Woodson. • EPRI JOURNAL June 1986 3 daet iniat :50 cu enos raetni I C limate, that amalgam of nature's moods, shapes the character of places and human activity as powerfully as any other force. It influ­ ences where we live and how we like it. Climate allows corn to flourish in Iowa and causes crops to fail in Africa. It makes Siberia harsh, Tahiti gentle, and lends distinct personalities to every other place on the globe. Until recently, we took climate for granted and accepted that it would stay largely the same, at least in our life­ times and those of our children. Evi­ dence is mounting, however, that by burning fossil fuels, leveling tropical forests, and engaging in a number of other activities, humans are releasing gases to the atmosphere that could trap enough heat to raise the temperature of the earth's surface by a few degrees Celsius. No such increase has yet been detected. If and when it comes, how­ ever, even a few degrees' rise would change the climate of many regions. Some places may become sunnier, oth­ ers wetter. Today's deserts may bloom and some farming regions may wither. The timing of frosts that bound grow­ ing seasons will change, as will the fre­ quency of midsummer heat waves. Sea level could rise from several inches to a few feet, affecting coastal regions in nu­ merous ways. Some areas will change very little. Others may change a lot. The first signal of these potential changes came from measurements of the steady rise in atmospheric carbon dioxide (CO2) made by Charles Keeling of the Scripps Institution of Oceanog­ raphy. His readings from the observa­ tory at Mauna Loa, Hawaii, show that CO2 has climbed from a concentration of 315 parts per million (ppm) in 1958 to 345 ppm today. A National Academy of Sciences report released in 1983 pro­ jected that within 50 to 100 years the level of CO2 is likely to rise to 600 ppm. Because CO2 acts as a thermal blanket in the air, explains the NAS report, this doubling of gas concentration could raise average global temperature by 1. 5-4.5°C (2.7-8.l°F). A 1.5 °C warming within a century would produce the warmest climate seen on earth in 6000 years. A 4.5°C rise would place the world in a temperature regime last ex­ perienced in the Mesozoic era-the age of dinosaurs. One of the key uncertainties in pro­ jecting temperature change centers on the question of how much of the CO2 released to the atmosphere will remain there. We know from the measured in­ crease in atmospheric CO2 that 2. 5 bil­ lion tons of carbon per year, an amount equal to half the annual emissions from fossil fuel combustion, accumulates in the air. The rest, scientists believe, must be going into terrestrial plants and into the oceans. Plants absorb CO2 through photosynthesis and release it through respiration, storing some of the carbon that they metabolize in their leaves, stems, and roots. CO2 enters the ocean by dissolving out of the air into the surface waters. Some of the carbon stays dissolved as CO2 in the surface water, some is carried into the deep ocean by currents or precipitates into solid particles that fall to the bot­ tom, and some bubbles back out to the atmosphere, much like gas escaping from a bottle of club soda. Scientists do not yet agree on how the CO2 that leaves the atmosphere is apportioned among the plants and oceans, nor do they know how much more CO2 these reservoirs can absorb or how long they will hold it. Without a clear understanding of the pools and flows of carbon in the earth-water-air system, it is impossible to precisely es­ timate future atmospheric CO2 levels. Knowing how much CO2 will be in the atmosphere in the future only gets us to square one. From there we have to explore the gas's effects on temperature and ultimately on climate. Temperature and climate: A question of balance The earth's climate is driven in part by temperature, and temperature is a func- EPRI JOURNAL June 1986 5 tion of the balance between energy coming in from the sun and energy ra­ diating back to space. This balance is influenced by the atmosphere, which is transparent to visible light (between 0.4 and 0.7 µ,m) but is opaque to cer­ tain wavelengths of infrared, or heat, radiation (longer than 0.7 µ,m). Slightly more than half the light en­ tering the atmosphere is absorbed by clouds and particles or reflected back to space. The remainder is absorbed at the surface, warming the oceans and land. The warmed surfaces then reradiate this energy in the form of heat. Natu­ rally occurring water vapor and CO2 in the atmosphere absorb certain wave­ lengths of this heat and radiate some of it back to earth. This phenomenon is re­ ferred to as the greenhouse effect, be­ cause it is analogous to the trapping of heat inside a glass enclosure. The air traps enough heat to keep the earth's surface about 30°C (54°F) warmer, on average, than it would be without an atmosphere. Because naturally occurring gases in the atmosphere do not absorb all in­ frared wavelengths, some of the heat radiated by the land and sea escapes back to space. Most of this escaping heat is radiation with wavelengths be­ tween 7 and 12 µ,m, a band in which water vapor and CO2 absorb weakly. Thus the atmosphere (even with high CO2 levels) is largely transparent to in­ frared radiation in this range. This at­ mospheric window is critical to climate. Without the window, more heat would be trapped in the lower atmosphere, the earth would be warmer, and the patterns of atmospheric circulation that shape climate would be altered. Factoring in trace gases Ralph Cicerone of the National Center for Atmospheric Research (NCAR) and three colleagues recently published the results of an extensive study on the rate at which greenhouse gases other than CO2 are accumulating in the atmo­ sphere and the effects these gases 6 EPRI JOURNAL June 1986 may have on climate. The gases of con­ cern are present in minute but rising amounts, they remain in the atmo­ sphere for many years, and they are powerful absorbers of infrared radiation between 7 and 12 µ,m. The ringleaders of this group are methane, ozone, nitrous oxide, and several chlorofluorocarbons (Freons). The chlorofluorocarbons arise exclu­ sively from human activity. Although banned in the United States as aerosol propellants, these compounds are still used throughout the world as refrig­ erants and solvents and in numerous industrial processes. The sources of other greenhouse gases are not as clearly understood. Methane, for instance, has been rising at 1%/yr over the past three decades. Cicerone believes that much of the in­ crease in this gas can be traced to min­ eral exploration and mining, rapidly growing populations of cattle and ter­ mites, which excrete methane, and de­ composition of organic matter in rice paddies that for the first time are bear­ ing two and three crops annually with the expansion of modern irrigation sys­ tems. He traces the rise in nitrous oxide in part to the use of nitrogen fertilizer in agriculture. Cicerone and his coworkers calculate that the combined effect of these gases is already 60% as great as the effect from current CO2 levels and within 50 years will equal or exceed C02's influ­ ence. This additional effect could cause global temperature to rise faster than most models now predict. Not all scientists agree with the mag­ nitude of the prevailing warming pre­ dictions, however, and much of this disagreement hinges on estimates of the size and direction of various feed­ backs that may occur in response to the initial perturbation of the climate. Some responses, called negative feed­ back, will have a cooling effect that will dampen the warming trend. Posi­ tive feedback, on the other hand, will reinforce the initial warming. Understanding feedbacks One of the key feedback uncertainties concerns clouds. "If the atmosphere warms," says Richard Somerville, a theoretical meteorologist and head of the climate research group at Scripps, "more water will evaporate into the air, changing the area, height, water con­ tent, and reflectivity of clouds." Plug­ ging plausible numbers for changes in cloud reflectivity into a climate model, Somerville found that the range of pro­ jections for global warming could be either doubled or halved. He hastens to add that these results do not prove the models wrong. "Rather, they point to questions of potential importance that we have to understand better before we can rely with confidence on our model results." Another feedback uncertainty in­ volves the response of plants to rising CO2 levels. The Carbon Dioxide Re­ search Office of the U.S. Department of Energy (DOE), the government's lead agency in the study of CO2 and climate, devotes a considerable portion of its $13 million annual budget to investigating the effects of rising CO2 on vegetation. In a series of controlled studies, agri­ cultural crops grew 30-90% faster in a high-CO2 environment, pulling more carbon out of the atmosphere and stor­ ing it in plant tissues. Moreover, water­ use efficiency rose significantly in sev­ eral of the crops. Some scientists point to such findings with hope that an at­ mosphere richer in CO2 could lead to higher food yields in the future and that enhanced plant growth could slow the rise in CO2 and global temperature. Although crops may respond posi­ tively to higher CO2 when other condi­ tions are controlled, very little research has been conducted on the response of the unmanaged forests, grasslands, and tundra that cover much of the earth. Walter Oechel, director of the Sys­ tems Ecology Research Group for the California State University at San Di­ ego, conducted one of the few studies yet performed in a native ecosystem to Blocking the Atmospheric Window \ -""'· -------=::.::· "-._ ', The so-called greenhouse effect is caused by atmospheric gases that absorb certain wavelengths of heat that would otherwise radiate out to space. Carbon dioxide is not the only greenhouse gas whose concentration is rising as a consequence of human activity. Methane, nitrous oxide, ozone, Freon, and several other gases are of particular concern because they remain in the air for many years and are potent absorbers of heat within the wavelength band (7-12 µm) through which much of the earth's heat escapes the atmosphere. By blocking parts of this atmospheric window, very S(llall amounts of these gases can exert a warming effect as great as that caused by the much more abundant CO2. ------- <7 µm >12µm Atmospheric window is naturally transparent to heat with 7-12-µm wavelength. Nitrous oxide Ozone / / � Water vapor absorbs heat <7 µm. ! I •/ Heat l Light , I w,te,,apo,aod co, absorb heat >12µm. ,. I EPRI JOURNAL June 1986 7 measure long-term plant responses to elevated CO2. Oechel exposed several species of tundra plants to varying lev­ els of CO2 and temperature for three years. He found that the response in native species was less than that found for many agricultural crops. He also found significant variability in the re­ sponses of the native species examined. Some species exposed to elevated CO2 and higher temperature grew faster and stored more carbon, but other species did not. "Overall," explains Oechel, "arctic tundra exposed to higher CO2 and ambient temperatures stored more carbon than tundra under current CO2 levels, but not as much as predicted." Oechel believes that more research is needed to determine the overall response of the tundra to rising CO2. Moreover, he stresses that although much of what we learn in the arctic communities will carry over to the boreal forests and other ecosystems, additional research is necessary to understand the responses of those other communities. George Woodwell, director of the Woods Hole Research Center, agrees that we need more research in the area of plant response. "There is no ques­ tion that under experimental conditions when water and nutrients are not limit­ ing, you can show that an increase in CO2 can increase the rate of carbon fix­ ation. But there is no evidence that this is occurring in nature. In fact, the higher rates of organic matter decay that would occur in the soils of the middle and high latitudes due to cli­ mate warming, in combination with the rapid deforestation now under way in the tropics, will release large amounts of CO2 to the atmosphere. These effects will be counterbalanced to some degree by increased photosynthesis and longer growing seasons in some areas, but the net effect will be that plants will add more CO2 to the atmosphere than they absorb." The NAS report acknowledges the Carbon Cycle Is Key to Temperature Change Without man's influence, the flows of carbon (given in units of 10' t) between the air, plants, and ocean would be roughly balanced. Fossil fuel combustion adds about 5 x 10' t/yr of carbon to the atmosphere, about half of which remains in the atmosphere as rising CO2 levels. The rest is absorbed by plants and oceans, but scientists disagree on how much goes to each reservoir (stocks in units of 10' t). Some believe that plants are growing faster in the CO2-enriched atmosphere and are absorbing up to half the carbon that does not remain in the air. Others believe deforestation is outstripping any enhanced growth that might be occurring, causing plants to be a net source, rather than a sink for CO 2. Scientists must resolve this debate before they can confidently predict how CO2 levels will rise in the future. 2.5/yr stays in air 5-9/yr total 0-4/yr 8 EPRI JOURNAL June 1986 5/yr Natural balance in ocean-atmosphere exchange 2.5-6.5/yr Absorption by oceans Absorption by plants Natural balance in plant-atmosphere exchange importance of this question, explaining that if deforestation has caused terres­ trial ecosystems to be a net source of CO2 in recent years, the oceans must be taking up far more CO2 than current models suggest. If oceans are better at absorbing CO2 than scientists cur­ rently believe them to be, the study concludes, "the CO2 increase will probably occur more slowly than it otherwise would." Perhaps of greater significance than the feedback involving terrestrial plants and soils are the immense and very complex ocean-atmosphere interactions. The oceans contain 55 times as much carbon as does the atmosphere and 20 times as much as do land plants. Thus, small changes in the oceans' capacity to store carbon can dramatically alter the atmospheric concentration. In most areas, atmospheric CO2 mixes only with the top 100 m or so of sea water and is prevented from cycling downward by thermal gradients that separate the surface layer from deeper waters. However, researchers say that the only long-term way for the oceans to buffer the rise in atmospheric CO2 is to pump the carbon into the deep ocean, either as dissolved gas or as solid carbonate particles that settle in the sediments at the ocean bottom. Sci­ entists believe that CO2 is drawn into the deep ocean through currents in a few key locations like the North At­ lantic, where cold surface waters are known to sink to the bottom. Some re­ searchers speculate that a global warm­ ing could cause the oceans' currents to become more sluggish, reducing their absorption and storage of carbon and exacerbating the greenhouse effect. Potential Feedbacks to Warming Feedback related to an i n itial greenhouse warming will have important effects on long-term temperature change. Some responses (positive feedback) are expected to exacerbate the warming. Others (negative feedback) will probably have a cooling effect. The net result remains uncertain, although current estimates are for a 1.5-4.5 ° C temperature rise within a century. rming. Glacial retreat decreases reflectivity of poles. Warming: Polar thawing speeds decay and release of carbon now held i n permafrost. Cooling: Plants grow faster in high-CO2 atmosphere and absorb more carbon. Warmin9: Faster temperature rise in high latitudes reduces temperature differential between equator and poles, stalls ocean currents, and reduces pumping of CO2 to deep ocean. Warming: As oceans warm, ----�..---1 their capacity to hold 1 d issolved CO2 diminishes. Warming: Temperature rise evaporates more water, raises h umidity. A needle in the haystack Making the measurements needed to determine key parameters like up­ take, storage, and release of CO2 by oceans and plants is akin to burrowing through a haystack in search of one slender needle. Terrestrial ecosystems and oceans are huge, diverse reser- Cooling: Rising humidity leads to increased cloud cover. Warming: Deforestation releases additional CO2. Cooling: Deforestation makes tropics more reflective. EPRI JOURNAL June 1986 9 nicians are instructed to take samples voirs, and scientists are trying to mea­ only on windy days when breezes are sure minute changes in their composi­ tion. Compounding the difficulty, these blowing from the sea so as to avoid lo­ reservoirs are not at all uniform. Trees cal effects from land plants. Roughly absorb and store carbon very differently 6000 samples a year are sent to NOAA in pressurized steel containers lined on than do grasslands and tundra, and the inside with gold or other inert sub­ oceans vary widely in temperature, depth, water chemistry, and circulation. stances that will not react with the sam­ ple gas. This heterogeneity means that mea­ surements from a few locations only Tans is pursuing an ingenious new approach to resolving the question of hint at what is happening worldwide. ocean-plant uptake. He reasons that One researcher commented at a recent when oceans absorb CO2 from the air, conference that it would take 10 years they store the oxygen along with the and hundreds of people in boats gath­ carbon. When plants fix carbon, how­ ering water samples from all around ever, they return the oxygen to the at­ the world to yield even a fuzzy picture mosphere as a by-product of photo­ of oceanic absorption, storage, and re­ synthesis. Consequently, if the oceans lease of carbon. are the dominant sink for CO2, atmo­ Figuring out what is happening in plant communities is similarly challeng­ spheric oxygen levels should be falling. If plants are a major sink, however, ox­ ing. Oechel's research in the tundra is the only in situ study yet conducted on ygen levels should stay the same. Tans has spent several years developing ecosystem response to elevated CO2, equipment to measure minute changes and even this work involves controlled conditions. For the most part, however, in the oxygen content of air. He be­ researchers are trying to deduce what is lieves that his equipment, which uses an argon laser, fiber optics, and spectral happening in the plants and oceans by analysis of gas samples, will soon be measuring changes in the atmosphere ready to be put into service. more accurately. Because the atmo­ The whole point behind these ex­ sphere is better mixed and more uni­ haustive monitoring efforts is to deter­ form than the oceans and terrestrial mine the rate at which greenhouse plant communities, it is an easier gases accumulate in the atmosphere. reservoir to monitor. , This is one of the critical pieces of infor­ "If you are only interested in aver­ mation climate modelers need to study age, long-term, global trends in atmo­ how climate may change. spheric concentrations, one or two monitoring sites are sufficient," ex­ plains Pieter Tans, who directs the CO2 Modeling the climate response monitoring program for the National Climate models are nothing more than Oceanic and Atmospheric Adminis­ hundreds of linked mathematical equa­ tration (NOAA). "But if you want to tions describing how various phenom­ determine where the main sources and ena in the environment cause condi­ sinks are, you have to measure local tions to change throughout the system. Incoming sunlight, for instance, heats gradients and see how they change the earth's surface, particularly near the with the seasons. This requires many measurements, over time, from lots of equator, causing warmed air in the low sites." latitudes to rise and circulate toward NOAA operates a network of 26 re­ the poles. The earth's rotation interacts mote monitoring stations from Point with these currents to cause the trade Barrow, Alaska, to the South Pole. winds. Water evaporating off the oceans "Precision in gathering and analyzing circulates in the atmosphere and falls samples is critical," says Tans. Techto the surface as precipitation. The tilt 10 EPRI JOURNAL June 1986 of the earth's axis causes different amounts of energy to flow into the nor­ thern and southern hemispheres and leads to the annual cycles we call sea­ sons. These and related processes that determine climate are reduced to numbers and symbols in sophisticated computer programs known as general circulation models. These are the tools climatologists use to study the climate and to predict how it will change in the future. Climatologists check the accuracy of their models by running them for­ ward and backward. Years can be reconstructed in a few days of com­ puter time. If the models can recon­ struct known present or past condi­ tions, modelers feel confident in using them to project future climates. To determine if their models success­ fully recreate the past, climatologists have long been seeking reliable esti­ mates of the CO2 levels that existed in the atmosphere before the industrial revolution, when fossil fuels were first burned on a large scale. Several meth­ ods, including analysis of air trapped centuries ago in polar ice caps and carbon fixed in tree rings and slow­ growing ocean sponges, have been used to unravel this mystery. The emerging consensus is that the CO2 concentration stood at roughly 280 ppm in the eighteenth century, a level 65 ppm below today's concentration. Most of the modeling studies of COr induced climate change have been run on a few general circulation models (GCMs) in the United States. The Geo­ physical Fluid Dynamics Laboratory model at Princeton University predicts a global warming of 2°C for a doubling of CO2, although the Goddard Institute for Space Studies (GISS) and NCAR models yield about a 4°C (7.2°F) in­ crease. Climatologists are anxiously poring over temperature measurements year by year to see if a greenhouse signal is emerging from the noise of natural temperature fluctuations. Nearly all Observed and Projected Temperature The earth's average temperature rose about 0.5 ° C (0.9 ° F) over the past century, with a 0.2° C (0.36 ° F) cooling observed (black) between 1940 and 1 965. Although the temperature rise is not inconsistent with predicted effects of CO2 and other greenhouse gases, the fluctuations are still within the normal range of temperature variation. Many climatologists expect a CO2 signal to rise above the background noise in the 1990s. Projections (color) for temperature increases caused by a doubling of CO2 are 1 .5-4.5 ° C (2.7-8.1 ° F) withi n 50 to 100 years. These projections do not consider the influence of other greenhouse gases, which could speed the warming. 0.8 1 .5-4.5 ° C between 2040 and 2080 0.6 0 � Q) 0) 0.4 � :, 0.2 0 -0.2 Natural temperature variability _,, L________ __ _____ _____ _ _ _ ______� 1880 1 900 1 920 1 940 1960 1 980 2000 2020 2040 2060 EPRI JOURNAL June 1986 11 EPRI STUDIES GREENHOUSE EFFECT E PRI has sponsored a number of research efforts in the area of greenhouse gases and climate change. These projects generally focus on one of two areas: the scientific aspects of the problem and the analytic and deci­ sion-making techniques for evaluat­ ing and responding to COz-induced changes. Radian Corp. completed a study for EPRI's Coal Combustion Systems Di­ vision in 1980, assessing the scientific uncertainties of the issue, projecting CO2 emissions worldwide, and evalu­ ating the prospects for CO2 emissions control and the social and economic implications of a global warming. Not­ ing that U.S. utilities produce less than 10% of the global CO2 emissions from fossil fuel combustion, the report concludes that an international effort to reduce fossil fuel use would be re­ quired to curb CO2 buildup. Half a dozen chemical and physical processes have been proposed for cap­ turing the CO2 from fossil-fuel-fired power plant emissions. Radian con­ cluded that although some are tech­ nologically feasible, none are eco­ nomically practical. They would add several hundred dollars per kilowatt to plant capital costs, would add an undetermined amount in operating expenses, and would consume half or more of the plant's energy output. Consequently, no cost-effective means has yet been found for storing the cap­ tured carbon over long periods to keep it from returning to the atmo­ sphere. EPRI's Energy Analysis and Envi­ ronment (EAE) division has spon­ sored about half a dozen studies in the CO2 area since 1978. Rene Males, vice president and EAE division director, 12 EPRI JOURNAL June 1986 explains, "This issue is a classic envi­ ronmental risk management problem. There are large uncertainties in possi­ ble outcomes, significant environmen­ tal consequences under certain condi­ tions, and high costs for changing these conditions. The role we play is to get clearer scientific insights in or­ der to be able to more accurately eval­ uate the risks involved." Beginning in 1983, EAE sponsored a study with Scripps to resolve uncer­ tainties concerning global sources and sinks of CO2 • Although the research team refined its atmospheric and oce­ anic models in this effort, many details of the carbon cycle remain elu­ sive. Consequently, a proposal to con­ tinue this work, incorporating satellite imagery and enhanced three-dimen­ sional modeling of ocean circulation, is now under review. Another study, conducted for EPRI by the Oak Ridge National Labora­ tory, assessed how the rates of CO2 accumulation would change under dif­ ferent electricity generation scenarios and then explored the potential effects on climate, agriculture, the economy, and ecosystems. One recently completed study pro­ duced a preliminary greenhouse effect decision framework to help EPRI, util­ ities, and government research man­ agers plan research on CO2 and trace gases in a manner that focuses on the most important questions and yields the highest payoff. In another deci­ sion-related contract (cosponsored by EPRI, EPA, EEi, and NYSERDA), ICF, Inc., is exploring the potential effects of climate change on electric utilities. Using Florida Power & Light Co. as a case study, ICF is gathering informa­ tion on the utility's demand and gen- eration plans for the next 30 years. ICF will then analyze how a number of major factors may change with vari­ ations in temperature and rainfall, in­ cluding the demand patterns of major customers, the utility's operating char­ acteristics, new capacity requirements, demand management, and various fi­ nancial issues. Once the FP&L anal­ ysis is complete, the methodology will be evaluated, changed as necessary, and then applied in two additional utility case studies. The project is scheduled for completion in early 1987. EPRI's Advanced Power Systems Division has also supported a modest effort in the CO2 area. In May 1985 APS sponsored a five-day conference at Lake Arrowhead, California, at which 40 leading scientists discussed their investigations of CO2 transfer be­ tween the atmosphere, ocean, and terrestrial plants. The conference was organized by Scripps as part of a con­ tract with the APS division to improve its models of ocean-atmosphere ex­ changes of CO2 • Topics ranged from the latest techniques for estimating preindustrial CO2 levels, through eco­ system responses to elevated CO2, to new developments in oceanic and at­ mospheric modeling. As Dwain Spen­ cer, EPRI vice president and APS di­ vision director, explains, "We believe that the increases of carbon dioxide in the earth's atmosphere must be con­ sidered in planning a long-term re­ search strategy for the power indus­ try. At this time we are attempting to understand the key sources and sinks for CO2 and the important feedback mechanisms that may either amelio­ rate or exacerbate the problem in the D twenty-first century." A Global Issue 1980 figures on the distribution of CO2 emissions from fossil fuels in the United States and from around the world demonstrate the international nature of the issue. The Un ited States produces about one-quarter of global CO2 emissions, and electric utilities produce about one-quarter of U.S. emissions, or less than 8% of the global total. Trace greenhouse gases are increasing rapidly and are expected to match C02's contribution to global warming within 50 years. These trace gases are also produced throughout the world, adding to the international complexity of the greenhouse issue. Distribution of global CO2 emissions (5.1 X 10' 1/yr C) North America (27 %) Current contribution of CO2 and trace gases to expected warming Distribution of U.S. CO2 emissions (1.4 x 10' 1/yr C) Industry (29 %) Trace gases (40 %) Transportation (27 %) agree that no definite signal has yet appeared, but many expect to see it within a decade. One of the principal limitations in cli­ mate models is their ability to predict regional changes accurately-how will conditions change where people live, work, and farm? Will Iowa still be a good place to raise corn? Will the Gulf Stream stall and make Britain much colder? Will the drought-stricken areas of Africa get more rain? Some of the most sophisticated models are begin­ ning to produce regional predictions, but the jury is still out on their accu­ racy. For now, most researchers shy away from regional prognoses, but they do feel confident about their projections of global average change. So what's a few degrees? Although an average warming of a few degrees does not sound like much, it could create dramatic changes in climatic extremes. The frequency of midsummer heat waves in certain lo­ cations, for example, could rise sig­ nificantly. This could have important effects on, among other things, agri­ culture, energy consumption, and hu­ man comfort. NCAR climate specialist Steven Schneider and two colleagues found that the frequency of periods of five days or more exceeding 35°C (95°F) in the Corn Belt would grow threefold in the event of an average global warm­ ing of 1.7°C (3°F). Such conditions at critical stages in the growing season are known to harm corn and lead to re­ duced yields. A similar study con­ ducted by James Hansen, director of GISS, and his student Paul Ashcraft concluded that the number of days warmer than 32°C (90°F) and 38° C (100°F) and nights above 27°C (80°F) would rise dramatically in eight major American cities as a result of doubled CO2 • The effect on utility cooling loads could be substantial. Changes in the timing and amount of precipitation will almost certainly occur if the climate warms, affecting agricul­ ture and hydroelectric resources, among other things. Soil moisture, which is crit­ ical during planting and early growth periods, will change. Rising CO2 levels may enhance growth rates and water­ use efficiency in some crops in certain areas. Some regions will probably be­ come more productive, while other places may become less suited to agri­ culture. The North American grain belt, according to at least one climate model, will shift northward into Canada as the warming produces drier, hotter condi­ tions in the American Midwest. Of all the effects of a global warming, none has captured more attention than the prospect of rising sea levels from the melting of land-based glaciers and volume expansion of ocean water as it warms. Estimates of the rise range from several inches to a dozen or more feet by the year 2100, although the prevail­ ing weight of opinion calls for an in­ crease of about a foot and a half. The most dramatic scenario is that the West Antarctic ice sheet, which rests on land that is below sea level, could slide into the sea if the buttress of floating ice separating it from the ocean were to melt. This would raise the average sea level 15-20 ft (4.6-6.1 m). Even a 1 ft (0. 3-m) rise would have major effects on the erosion of coast­ lines, salt water intrusion into the water supply of coastal communities, flooding of marshes, and the inland extent of surges from large storms. What can be done? Many observers are beginning to dis­ cuss responses to potential climate change. Three basic views characterize the debate: we do not know enough yet about the fundamental processes to respond effectively; we should apply what we do know now to mitigate the changes; we should accept that climate change is inevitable and start immedi­ ately to adapt. Frederick Koomanoff, who directs DOE's CO2 research, holds firmly to the view that we have to conduct a lot more scientific research before we do EPRI JOURNAL June 1986 13 Measuring the Changes Through Laboratory and Field Studies The global dimensions of the greenhouse effect require a correspondingly global research effort. Air sampling at remote sites from the tropics to Alaska and Antarctica is helping to identify the sources and sinks of CO2 and other heat-absorbing gases. Field and laboratory studies on carbon uptake by plants exposed to various CO2 levels are building understanding of the factors influencing the buildup of atmospheric CO2. Sampling air for greenhouse gases Measuring plant response to CO2 14 EPRI JOURNAL June 1986 anything else. "We simply do not know enough yet to respond intelligently," he asserts. "There are vast uncertainties in the carbon cycle, in the response of vegetation, in the coupling of oceanic and atmospheric processes, and in the way these factors will interact to change our environment. There are also gaps in our data on regional and sea­ sonal climate change and in our under­ standing of the consequences of that change. We need another 10 years of concerted interdisciplinary research to resolve these uncertainties. By then we will have a clear enough understanding of the processes involved to give poli­ cymakers and the public the informa­ tion necessary to make sound deci­ sions." Many scientists believe that although our understanding is incomplete, we do know enough to mitigate the green­ house effect. "If uncertainty is a ground for no action," says Schneider, "then we would have no insurance compa­ nies and no armies." Energy policy is often identified as one area in which much could be done to reduce CO2 emissions. Greater emphasis on end­ use efficiency, certain renewable energy resources, and nuclear power are all pointed to as ways of providing energy without releasing CO2 . An examination of the sources of man-made CO2 emis­ sions, however, reveals that any effec­ tive effort to curb CO2 releases would require extensive international cooper­ ation. Moreover, the fact that trace gases are fast approaching CO2 in their green­ house potency means that they too would have to be considered in an in­ ternational effort to mitigate the warm­ ing. Controlling trace gases will not be easy. "If methane and nitrous oxide are rising as a consequence of a human ac­ tivity as essential as food production," comments Cicerone, "it's going to be very difficult to limit their increase by agreed-upon social and political changes." Dennis Tirpak directs a program look- ing at the effects of climate change for the Office of Policy Analysis in the U.S. Environmental Protection Agency (EPA). Tirpak and his colleagues main­ tain that society should begin exploring ways to adapt to the expected changes. "Ninety-nine percent of the govern­ ment's efforts in this area are directed toward research on the mechanisms that may lead to climate change," re­ ports Tirpak. "Very little emphasis is being placed on assessment and anal­ ysis of what this could mean for society and for the environment. " Tirpak cites the recommendations of an interna­ tional conference on the greenhouse ef­ fect held last October in Villach, Aus­ tria, calling for more assessment work in parallel with scientific research. "Many important economic decisions are being made today," read a state­ ment issued by the conferees, "on ma­ jor irrigation, hydropower, and other projects; on drought and agricultural land use; on structural designs and coastal engineering projects; and on en­ ergy planning-all based on assump­ tions about climate a number of de­ cades into the future. " Tirpak explains that EPA i s support­ ing some "what if" studies that make assumptions about future climate and then explore potential adaptive strate­ gies. One study now under way looks at the effects of a greenhouse warming on utilities. EPRI is a cosponsor of this research, along with EPA, Edison Elec­ tric Institute (EEI), and New York State Energy Research and Development Authority (NYSERDA). Much to learn Regardless of their views on whether we should be focusing exclusively on unraveling the scientific uncertainties or developing responses to the green­ house effect, scientists agree that we still have much to learn about the ways in which human activities are influenc­ ing climate. Keeling reflects, "The main prob­ lem with climate impact of CO2 and other gases is that we don't like to see change. We don't want to face rising sea level, for example, because it means we'll have to move a lot of peo­ ple around. The earth has seen it all be­ fore, though. The cycle was never in balance. CO2 has been far higher and far lower in the geologic past. Climate has been warmer and colder, too. We don't know how fast the changes now under way will come or how far they will take us, but we do know that the decisions we make today will have implications far into the future. The memory of the fossil fuel era will be with the earth for tens of thousands of years." • Further reading Gina Maranto. '"Are We Close to the Road's End?'" Dis­ cover, January 1986. p. 28. U.S. Department of Energy, state-of-the-art series. Decem­ ber 1 985. (Available from the National Technical Informa­ tion Service, U.S. Dept. of Commerce. Springfield, Virginia 221 6 1 .) Detecting the Climatic Effects of Increasing Carbon Diox­ ide, DOE/ER-0235. Projecting the Climatic Effects of Increasing Carbon Di­ oxide, DOE/ER-0237. Direct Effects of Increasing Carbon Dioxide in Vegeta­ tion, DOE/ER-0238. Atmospheric Carbon Dioxide and the Global Carbon Cycle, DOE/ER-0239. Glaciers, Ice Sheets, and Sea Level: Effect of a CO,­ lnduced Climate Change. Prepared by the National Re­ search Council for the U.S. Department of Energy. Sep­ tember 1985. DOE/ER/60235-1. V. Ramanathan, R. J. Cicerone, H. B. Singh, and J. T. Kiehl. '"Trace Gas Trends and Their Potential Role in C limate Change." Journal of Geophysical Research, Vol. 90, No. D3 (June 20, 1985), pp. 5547-5566. Linda 0. Mearns, Richard W. Katz, and Stephen H. Schneider. "Extreme High-Temperature Events: Changes in Their Probabilities With Changes in Mean Temperature." Journal of Climate and Applied Meteorology, Vol. 23, No. 12 (December 1984), pp. 1601-1613. Richard C . J. Somerville and Lorraine A. Remer. "Cloud Optical Thickness Feedbacks in the CO2 Problem." Journal of Geophysical Research, Vol. 89, No. D6 (October 20, 1984), pp. 9668-9672. Changing Climate: Report of the Carbon Dioxide Assess­ ment Committee. National Academy of Sciences. Wash­ ington, D.C.: National Academy Press. 1983. An Assessment of the Implications of Increased CO2 Emis­ sions and Their Control for the U.S. Electrical Utility Indus­ try. Report prepared for TPS 77-751 by Radian Corp. Jan­ uary 1980. This article was written by Michael Shepard. Technical background information was provided by Charles Hakkari­ nen and Alan Hansen, Energy Analysis and Environment Division. and Dwain Spencer, Advanced Power Systems Division. Additional support was provided by Mary Ann Allan, Seymour Alpert. Glenn Hilst, Ralph Perhac, and Thomas Wilson. EPRI JOURNAL June 1986 15 I ncreasi ng in size and efficiency at regu lar intervals, combustion tur­ bi nes have become an attractive option for uti lities seeking small increments of new capacity. A decade of work on turbi ne reliability is culmi nati ng i n advanced designs that promise higher availability in the next generation of plants. 130-150 MW �'tJ 34% .., 1990 -� _____ ,0 �# � 'i 4' \,./0 �Cf -� -�· � � Size Efficiency 10 MW 25% 1960 16 EPRI JOURNAL June 1986 10 50 MW 30% 1970 -�· 32% 1980 C ombustion turbines have been used since the late 1940s to pro­ vide electric power at times of peak demand. But because the turbines were inexpensive and because they were used only a small fraction of the time, utilities did not put as much effort into maintaining these machines as they de­ voted to their baseload plants. As a con­ sequence of minimal maintenance, de­ manding operation, and materials and design limitations, combustion turbines experienced frequent forced outages and acquired a reputation for being unre­ liable. They were also relatively ineffi­ cient, a characteristic that has improved in later models and that was of only mi­ nor concern when fuel was cheap. When oil and natural gas prices rose in the late 1970s, the operation of combus­ tion turbines became far more expen­ sive. Congress passed the Fuel Use Act of 1978, banning utilities from installing af­ ter 1992 any additional intermediate or baseload capacity fired with oil or natural gas for more than 1500 hours a year. The restrictions of this law, coupled with re­ liability concerns and high fuel prices, caused the bottom to fall out of the do­ mestic utility combustion turbine mar­ ket. Sales to American utilities plum­ meted from almost 9000 MW in 1971 to near zero between 1980 and 1982. Today, with fuel prices falling and utili­ ties anxious to minimize capital expen­ ditures and risk by adding capacity in small, modular increments with short lead times, utilities are once again pur­ chasing or considering combustion tur­ bines, both as stand-alone units and in phased combination with conventional steam turbines and coal gasifiers. If the notion of phased construction catches on, as some utilities are currently pro­ jecting, the way utilities plan and build new power plants could change dramati­ cally. A different kind of power plant Combustion turbines are versatile de­ vices. They are used by numerous chem­ ical and process industries to provide mechanical and electrical power, by the aviation industry as airplane engines, and by U.S. utilities to drive more than 50,000 MW of electric generators. Unlike coal and nuclear plants that use heat from burning or reacting fuel to gen­ erate steam that spins turbine genera­ tors, combustion turbines use air as the working fluid. Air is compressed and in­ jected at high pressure with fuel-usu­ ally natural gas or oil-into a combustion chamber, where the mixture burns and then expands through a turbine at about 1093°C (2000°F). Sometimes known as a gas turbine (referring not to the fuel but to the fact that combustion gas drives the turbine), the combustion turbine's name is misleading because the turbine is actu­ ally just one component of the power plant. Combustion turbines have grown in size and efficiency over the years. In the 1950s and 1960s manufacturers produced small turbines, of 10-25 MW, aimed prin­ cipally at the aircraft market. Typical effi­ ciencies (lowering heating value) in that period were 23-30% . Utilities bought many turbines that were essentially ret­ rofited jet engines. They were inexpen­ sive and compact, and they came on-line quickly when they were needed. Consol­ idated Edison Co. has long maintained a number of small, heavy-duty turbines on barges on the East River and fires them up to help meet New York City's demand peaks. Many other utilities put combus­ tion turbines to similar use. As utilities and other industries made more use of combustion turbines, manu­ facturers started to produce larger and more-efficient models, reaching 50 MW around 1970 and 90 MW by 1980 with efficiencies of about 32% . "Combustion turbines tend to evolve in cycles of 10 to 15 years, for both technical and market­ oriented reasons," explains Albert Dol­ bee, EPRI's senior program manager for power generation. "It takes that long to develop and test new materials and com­ ponent designs. And because utilities are not inclined to buy any equipment that hasn't been proved in extensive testing, the manufacturers are reluctant to be the first to offer new models." But changes do come, and Dolbee claims that the ma­ jor manufacturers are on the verge of of­ fering a new generation of combustion turbines for shipment after 1988, with typical sizes of 130-150 MW and efficien­ cies of 34%. To increase combustion turbine effi­ ciency, manufacturers had to make the machines run at higher temperatures be­ cause efficiency is determined in part by the temperature of the hot gas entering the turbine. Hotter operation, however, demanded more-advanced materials and cooling designs. The hot gas temperature in these machines (1260°C [2300°F] in the next-generation models) is far hotter than the turbine blades can withstand, so the metal is cooled with extracted com­ pressor air to 816-871°C (1500-1600°F). Improvements in combustion system de­ sign, better methods of cooling with air, and materials improvements in the hot section parts (combustion liner/basket, transition piece, turbine vanes and blades) have made it possible to raise operating temperatures. Utilities and other combustion turbine users have benefited extensively from the massive ($750 million in 1986) R&D effort the military devotes to improving turbine technology. Although the mili­ tary's principal concern is to improve jet engine performance, manufacturers have been able to apply many of the advances developed for military applications (such as high-temperature materials and aero­ dynamically more-efficient compressors) to commercial combustion turbines for the utility market. A flexible option One of the most flexible features of com­ bustion turbines is that they can be oper­ ated alone in what is termed an open or simple cycle, or their hot exhaust gas can be used to generate steam to drive a con­ ventional steam turbine in an approach called a combined cycle. Combined cy­ cles have a number of advantages that are making them attractive to utilities EPRI JOURNAL June 1986 17 considering capacity expansion. They are the most efficient of all ther­ mal power plant designs, converting up to 47% of the fuel energy into electricity. Combined cycles also have great fuel flexibility, as combustion turbines can operate on natural or coal-derived gas and on various grades of fuel oil. In fact, some combustion turbines can change fuels without shutting down and some can burn mixtures of fuels. Coal gasifiers can be added to a combined-cycle plant site, enabling the plant to run on syn­ thetic gas if natural gas or oil is too ex­ pensive or unavailable. Such integrated gasification-combined-cycle (IGCC) sys­ tems are receiving increasing attention in the wake of the successful operation of the 100-MW IGCC demonstration plant at Southern California Edison Co.'s Cool Water station. Because these systems can be built in stages as the need for new capacity grows, they have important financial benefits as well. Phased construction al­ lows utilities to pay for small increments of capacity as they come on-line rather than have to invest huge lump-sum amounts in baseload plants many years before they will operate. By deferring capital expenditures on new capacity until it is needed to meet incremental load growth, utilities can reduce interest payments, lower their debt burden, and protect themselves from having "impru­ dent" building costs questioned by regu­ lators. One of the greatest advantages of sys­ tems built around combustion turbines is their good environmental performance. Combustion turbines have very low pol­ lutant emissions, particularly when run on natural gas or cleaned syngas made from coal, and can operate under the strictest environmental regulations. Utility interest is growing Thanks to the flexibility and risk reduc­ tion that combustion turbines and com­ bined-cycle systems offer, many utilities are exploring these options in their gen­ eration planning. Some are interested in 1 8 EPRI JOURNAL June 1986 Combined-Cycle Power System A combustion turbine and a steam turbine can be linked in a combi ned-cycle system to increase efficiency. Compressed-air and fuel (natural gas or oil) burned in the combustion chamber produce hot gases that first spin the gas turbine in the topping cycle. The gases are then diverted to the bottoming cycle, where their heat is used to boi l water and produce steam, which turns the steam turbine l inked to a second generator. Combined cycles are the most efficient of all thermal power plants, converting up to 47% of the fuel energy into electricity. Compressor Fuel Combustion turbine Gas turbine cycle (topping) Generator ..-. '-----r Air Load Generator Steam turbine cycle (bottoming) ...____r Load Exhaust gases � 9000 8000 1 7000 1 � � 6000 1 5000 1 4000 1 Cl) � 3000 1 2000 1 1000 0 1 1970 1975 1980 1985 Combustion Turbine Shipments to Domestic Utilities In the late 1960s and early 1970s when electricity demand was growing rapidly and fuel was cheap, utilities purchased many combustion turbines, mostly to meet peak power needs. As oil and natural gas prices soared following the Arab oil embargo of 1973, American utility purchases of new combustion turbines plummeted and have stayed low ever since. Today, with oil and gas prices falling and utilities looking for ways to add capacity in small incre­ ments, many utilities are once again ordering or considering new combustion turbines. simple-cycle combustion turbines strictly for peak power needs. Texas Utilities Co. , for instance, recently received approval from its public utility commission to in­ stall 960 MW of combustion turbine ca­ pacity by 1990. In addition, the utility's current resource plan calls for an addi­ tional 500 MW of combustion turbines by 1994. According to Donald Deffebach, Texas Utilities' manager of power supply planning, "The 960 MW of combustion turbines will be used principally to meet peak power needs, but design of the sites will provide for future conversion to combined cycle. " David D 'Amico, a power engineer with Boston Edison Co. , explains that his utility is also planning to meet a growing peak demand with new combustion tur­ bines. "We have an application pending to license a site for one simple-cycle com­ bustion turbine about 85 MW in capacity. If the license is approved and the com­ pany decides to proceed with this op­ tion, we will probably install it before 1990." Other utilities, in looking ahead to their need for new baseload capacity in the 1990s, are considering phased con­ struction of combined-cycle and IGCC systems. In late 1984, 11 utilities joined EPRI in a project to analyze the merits of phased capacity expansion, starting with combustion turbines. "The eco­ nomics of phased construction are very site-specific," explains Michael Gluck­ man, who directs engineering and eco­ nomic evaluations for EPRI's Advanced Power Systems Division. "There are many ways in which to build combina­ tions of combustion turbines, steam bot­ toming cycles, and coal gasifiers, and what's good for utility A may not be use­ ful for utility B. We wanted to learn how phased construction looked under a vari­ ety of conditions. " EPRI began by preparing a data base with cost and performance figures on nine phase-in strategies. The data in­ cluded information on fuels, heat rates, availability of individual components, part load performance, equipment cost, and other factors. The information was adapted to fit with the PROMOD pro­ duction costing model that many utilities use to project costs and performance of the generating options they are consid­ ering. The participating utilities then used these data to evaluate phased construc­ tion for their systems. "Eight of them have completed this exercise," says Gluck­ man, "and each has found that some plan involving combustion turbines is the optimal choice for system expan­ sion. " Some find that combustion tur­ bines alone are what they need, others find combined-cycle systems to be their best option, and still others see IGCC as their preferred approach. Potomac Electric Power Co. (Pepco) is sufficiently convinced that phased construction meets its needs, and the utility recently launched a preliminary planning study to add two 120-MW com­ bustion turbines by 1996, followed by a 120-MW steam bottoming cycle and a coal gasifier later in the decade. "Our cal­ culations show that by building the project in phases, we can save about $120 million in present-value costs over the 30-year life of the plant, compared with building a conventional coal-fired base­ load plant of similar capital cost all at once," explains Pepco's Thomas Welle. Licensing, engineering, and environ­ mental field monitoring work on the project is already under way. Virginia Power, another utility partici­ pating in the EPRI study on phased con­ struction, also concluded that this is currently a least-cost way for it to expand its capacity. Daniel Danforth, corporate engineering adviser at Virginia Power, explains, "We expect our next new gen­ erating capacity to be a modular facility with 200 MW of combustion turbine and steam bottoming cycle . Addition of a coal gasifier, when economically justified, will provide fuel flexibility for this unit. Although decisions on generation ex­ pansion after this unit have not been made, phased, modular, combined-cycle units do figure in our planning." Florida Power & Light Co. recently de­ veloped a capacity expansion plan calling for phased construction of about 1200 MW of combined-cycle units by the year 2000. William Smith, manager of power supply planning for FP&L, states, "We want to design the systems with the op­ tion of adding gasifiers later, but for now we are just considering combined-cycle units. We expect to need our first incre­ ment of capacity, totaling about 600 MW, in 1995." Because they are small, combustion turbines have the flexibility of fitting into existing sites. This feature makes com­ bustion turbines particularly attractive to utilities that want to repower plants. Con­ sumers Power Co. recently announced a proposal to install 1160 MW of combus­ tion turbines in two phases at Unit 1 of its canceled Midland nuclear plant. Steam raised from the combustion turbine ex­ haust heat will be used to power the steam turbine already installed in the plant. The utility is now awaiting PUC approval to launch engineering studies on the project. Other utilities now facing cancelation decisions on nuclear plants will be watching the Consumers Power experience closely to see how the gas tur­ bine retrofit strategy works at the Mid­ land plant. Combustion turbine retrofits are not limited to nuclear sites. Northeast Util­ ities is considering the installation of three combustion turbines totaling about 400 MW to repower four oil-fired steam turbines at its Devon station. A coal gasification facility might be added later. It's too early to be sure, but these and other signs point to a sea change in the industry, with utilities looking very seri­ ously at combustion turbines and com­ bined-cycle systems to meet much of their demand growth for the rest of the century. But what about reliability? Despite the attractive features of low cost, fuel flexibility, modularity, and good environmental performance, there are still concerns within the utility indus- EPRI JOURNAL June 1986 19 try about the reliability of combustion turbines and combined-cycle systems. A number of factors have converged re­ cently, however, to assuage some of the utilities' doubts. The EPRl-supported Cool Water IGCC station operated at more than 50% of rated capacity in 1985. The excellent per­ formance of this first-of-a-kind plant has raised considerable interest in utility cir­ cles. Utilities are also carefully watching the performance of combined-cycle in­ dustrial cogeneration systems, which have mushroomed over the past several years. The success of these systems is undoubtedly affecting utility perceptions of combustion turbine technology. More­ over, 10 years of EPRI work in combus­ tion turbine reliability has led to greater emphasis by manufacturers on high­ reliability designs and to greater confi­ dence among utilities that with appropri­ ate maintenance, combustion turbines can be very reliable. In 1976 EPRI asked utilities for their suggestions on the kind of research the Institute should perform on combustion turbines. The utilities replied unani­ mously that their greatest concerns were in the reliability area and that was where EPRI should focus its efforts. No one had ever before made a systematic account­ ing of the causes of combustion turbine failures. Working with a group that has since become the combustion turbine task force of the Edison Electric Insti­ tute's prime mover committee, EPRI launched an effort to focus on combus­ tion turbine failure rates. This effort was inspired in part by the realization that reliability in combustion turbines has historically improved only with the discovery and correction of problems in the field-at considerable cost to utilities. EPRI's Dolbee documents this claim with a graph plotting avail­ ability against hours of operation for one model of turbine introduced in 1972. The graph shows a big dip in availability during the first five years. "What that means," he says, "is that the utilities were having to work the bugs out of 20 EPRI JOURNAL June 1986 these systems as they arose. That trans­ lates to a lot of expensive downtime and big repair costs that would have been re­ duced if the manufacturers had put more emphasis on reliability while they were designing the machines." Having borne the cost and inconvenience of compo­ nent failures and outages over the years, utilities wanted to gather the information needed to define reliability goals that manufacturers could aim for before re­ leasing new models to the market. The first round of data collection and analysis, completed in 1981, yielded a general picture of the frequency of tur­ bine outages, but it did not dig deep Inching Up the Demand Curve In an effort to reduce risk and capital expenditure in meeting uncertain future load growth, utilities today are shunning investment in large new baseload plants and are instead seeking ways to add new capacity in small increments with short lead times. Combustion turbines are well suited to this strategy, whether added alone for peaking capacity or as part of modular construction of combined-cycle or IGCC plants to meet baseload needs. Small capacity additions Large capacity additions Time The Link to Aircraft Engines Utility combustion turbines have evolved from turbines used for aircraft engines. Most of the R&D on combustion turbines is conducted by the military, which last year in the United States spent $750 million to improve these machines. Better turbine materials and designs developed by the military often find their way into commercial combustion turbines, allowing utilities to benefit directly from the Pentagon's R&D. enough to reveal the root causes of the failures. Beginning in 1982 EPRI funded development of a more detailed data base on the causes of turbine outages. Operating data from 12 utilities revealed that about 60% of combustion turbine outages are caused by failures in controls and accessories, including temperature controls and monitors, fuel supply equip­ ment, starters, and pumps for lubricating oil, water, and air. This pinpointing of problem areas en­ abled EPRI to initiate a follow-up project with General Electric Co. to examine the potential for improving controls and ac­ cessories on the next advanced model combustion turbine. Looking at forced outage data on 170 General Electric com­ bustion turbines from 1977 to 1982, the study established quantitative reliability goals for new plant designs and devel­ oped methods for analyzing the reliabil­ ity of individual components and sys­ tems. General Electric incorporated the findings of this study into its design work on new advanced models. For the first time, utility concerns about relia­ bility are being translated into equipment improvements before the machines are released. In addition to helping EPRI focus its research effort, the combustion turbine reliability data base is a tool that utili­ ties can use and benefit from directly. Charles Knauf, manager of the combus­ tion turbine division for Long Island Lighting Co., was instrumental in start­ ing the data base and has been working on it directly for the last two years as a loaned employee to EPRI. "Combustion turbines have historically been the step­ children of the utility business, " he ob­ serves, "getting just enough attention to get by." With more-systematic operation and maintenance guided by the reliabil­ ity data base, he believes utilities could obtain far more effective and reliable service from their combustion turbines. "Some maintenance procedures are per­ formed more frequently than necessary. Others aren't done often enough. Util­ ities can use the data base to develop an optimal maintenance program that will save them money in the long run. " Knauf explains that with the high operating temperatures of combustion turbines, parts can degrade and crack, sometimes breaking off small pieces of metal that can cause major damage if they pass through the turbine section of the machine. A single turbine blade can cost up to $5000; replacing the entire first row of turbine blades can cost half a mil­ lion dollars. To shut down and thor­ oughly inspect a combustion turbine, however, costs $15,000 to $25,000. The challenge then is to spend a moderate amount on maintenance so as to avoid the large expenses that arise from cata­ strophic or even moderate failures. Using industrywide figures from the data base on performance and failure rates of various turbine models and com­ ponents, utilities can draw curves plot­ ting the probability of a given part failing within a certain number of hours against the cost or number of man-hours re­ quired to repair such a failure. Focusing first on those failures that are most likely to occur and will be most costly to repair, the utility can then use the data base to determine how to avoid such failures in a cost-effective manner. Suppose the part in question has only a 20% probability of failing within 200 hours of operation but a 90% probability of failing within 400 hours-this is the kind of information the data base pro­ vides. The utility then knows it can run the equipment for 200 hours with little risk of failure but that it should inspect the part somewhere between 200 and 400 hours to avoid a far greater risk of failure. Conversely, there is little need to regu­ larly inspect a part that is statistically likely to run failure-free for 25,000 hours. This probabilistic approach to mainte­ nance planning offers no guarantees-a part can fail immediately after inspec­ tion-but it does enable utilities to oper­ ate with the odds on their side. All this work should, on paper, help to improve combustion turbine reliability. But the only way to know for sure how a piece of equipment will perform is to turn it on, wait, and watch. The final part of EPRI's combustion turbine reliability program proposes to do just that. EPRI is now seeking a host utility willing to par­ ticipate in a durability test of General Electric's new model combustion turbine. Clark Dohner, project manager for high­ reliability combustion turbines explains, "The host utility must be an early pur­ chaser of the next generation turbine. We will instrument and monitor the turbine and run it through a series of stringent duty cycles over a five-year period to test the durability and reliability of the sys­ tem and its advanced components. Our goal is to get the early failures out so they can be corrected before many units are sold to utilities." "The timing couldn't be better," says Dolbee. "The spade work utilities have sponsored in combustion turbine reli­ ability is starting to pay off with im­ proved components, designs, and O&M strategies just as the market for these ma­ chines seems to be growing again. " How closely utilities will embrace combustion turbines and combined cycles in the years ahead remains an open question. But there is little doubt that the next gen­ eration of these power plants will be more efficient and reliable than its prede­ cessors. • Further reading Planning Data Base for Gasification-Combined-Cycle Plants: Phased Capacity Additions. Final report for RP2029-1 3, prepared by Fluor Engineers, Inc., January 1 986. EPRI AP-4395. High-Reliability Combined-Cycle Program Study. Final report for RP1187-3, prepared by General Electric Co., October 1985. EPRI AP-4191. Mary Wayne. "New Capacity in Smaller Packages," EPRI Journal, Vol. 8, No. 4 (May 1983), pp. 6-13. This article was written by Michael Shepard. Technical back­ ground information was provided by Albert Dolbee. Clark Dohner, and Michael Gluckman, Advanced Power Systems Division. Additional support was provided by Charles Knauf, Ronald Wolk, and Stanley Vejtasa. EPRI JOURNAL June 1986 21 REMOTE SCANNING ,,, Of LOW-LEVEL WASTE How hot is that drum of low-level radwaste? Two new assay methods read radiation levels directly from outside the waste container and deter­ mine its content of specific radionuclides. These techniques improve accuracy and avoid the risk and high cost of extracting samples for laboratory analysis. •----------------• • 22 EPRI JOURNAL June 1986 F or nearly the last six years, utility low-level radioactive waste (LLW) management programs have been driven by regulations and laws directed to the operators of licensed commercial disposal facilities. The Low-Level Radio­ active Waste Policy Act of 1980 mandated that states assume responsibility for dis­ posal of LLW generated within their bor­ ders by January 1, 1986. As that deadline neared with only three disposal sites available, Congress gave utilities, hospi­ tals, research laboratories, and other generators of LLW an eleventh-hour re­ prieve late last year. Lawmakers extended this deadline through 1992 for states to form regional compacts and establish new disposal sites, but they also amended the LLW act to allocate remaining capacity at current sites and set progressive surcharges on waste volumes based on the degree of progress of individual states in establish­ ing new sites. While states struggle through the po­ litical and institutional difficulties of co­ operatively setting up new disposal sites, industries that produce LLW face a prob­ lem complying with separate Nuclear Regulatory Commission (NRC) rules re­ lated to the licensing of facilities for LLW disposal. These rules require utilities and other LLW producers to identify and quantify individual radionuclides prior to shipment for burial. In place since 1983, NRC's Licensing Requirements for Land Disposal of Ra­ dioactive Waste-Title 10, Section 61 of the Code of Federal Regulations (CFR)­ have posed a dilemma for many utilities operating nuclear power plants. The rules specify that wastes be classified in one of three categories based on the lim­ iting concentrations of some 16 specific (and hard to measure) radionuclides; the more hazardous categories include cer­ tain waste form and stability require­ ments. "The 10CFR61 requirements cannot be met with radiation detection equipment currently available on-site at nuclear power plants," notes Patricia Robinson, a project manager in EPRI's Nuclear Power Division. "Nuclear plants have had to sample wastes and radiochemi­ cally analyze the samples for their iso­ topic content at off-site laboratories. The cost of this approach can range from $30,000 to $200,000 a year per plant. "Not only is this practice costly, but it may result in unrepresentative samples and lead to additional personnel radia­ tion exposure, and some wastes-such as neutron-activated components-may be too radioactive to sample in a practical manner," explains Robinson. "In addi­ tion, because of sampling inaccuracies, the typical practice of estimating isotopic content from dose-rate measurements of­ ten leads to very conservative overesti­ mates of total radioactivity, which in turn lead to unnecessarily excessive disposal costs. Research results indicate that some isotopes may be overestimated by factors of 1000 to 10,000." Anticipating the need for better assay tools and techniques for LLW, EPRI is pursuing the transfer into utility use of two technologies that offer substantial improvement in accuracy and simplicity over existing sampling and radiochemi­ cal assay or calculation methods. In addi­ tion, both of these are direct-assay meth­ ods and promise to satisfy NRC criteria for keeping personnel radiation expo­ sure as low as reasonably achievable (ALARA) because they permit scanning of entire waste packages without requir­ ing workers to open containers or with­ draw samples. One, a technique for determining the content of transuranic (TRU) isotopes (those with atomic numbers greater than uranium) in a waste volume, was origi­ nally developed under a Department of Energy (DOE) contract. The second, a commercially available technology, em­ ploys collimated high-resolution spec­ troscopy to measure the gamma activity in bulk waste. Results from both types of measurement become input to micro­ computer codes that calculate individ­ ual concentrations of TRU and gamma­ emitting radionuclides. If concentrations of key isotopes are known, plant-specific scaling and dose-to-curie correlation fac­ tors can then provide reliable measure­ ment of the amounts of all isotopes in waste shipments. Both technologies have now been suc­ cessfully demonstrated in cosponsorship with Florida Power Corp. at the utility's Crystal River nuclear plant, following earlier individual testing at other plants. Used in combination, the TRU and gam­ ma-scanning techniques may revolution­ ize the technical means for compliance with federal LLW regulations and reduce sampling uncertainties and the cost of disposal in the process. "Eventually, direct assay of LLW could be commonplace at all nuclear plants," says Robinson. "Both of these technolo­ gies are acceptable to NRC as alternatives to radiochemical sampling and analysis. With the trend in federal requirements for LLW and the economic pressures tied to burial site capacity, the benefits of these technologies are clear. Our job now is to transfer the technology so that it is commercially available to the industry." Transuranic assay Various forms of LLW contain trace quan­ tities of transuranic elements that are by­ products of fission of the uranium fuel, including several isotopes of plutonium, americium, and curium. New waste clas­ sification limits require quantification of these specific radionuclides, but their de­ tection and precise measurement in a volume of waste is masked by radiation from other isotopes in the waste. Until recently, sampling and destructive radio­ chemical analysis was the only way around the problem. Under a DOE contract, Battelle, Pa­ cific Northwest Laboratories developed a technology that permits direct assay of LLW packages for TRU content by de­ tecting the neutrons emitted from the waste package. The neutrons are pro­ duced by spontaneous fission of the TRU isotopes or by the collision of alpha par­ ticles with lighter isotopes in the waste. A simple computer code converts the to- EPRI JOURNAL June 1986 23 Direct Assay for Gamma-emitting Nuclides Science Applications International Corp.'s QuantiScan system provides a d irect-assay method for determining the content in LLW of isotopes that emit gamma radiation, including those of cobalt and cesium. Tungsten shielding Lead plates Germanium-crystal detector Liquid n itrogen reservoir Lead plates The system employs a high-purity intrinsic semiconductor material-germanium-in its core that exhibits significant changes i n conductivity when exposed t o gamma rays. A tungsten collimater focuses gamma radiation entering the germanium detector, which is cooled by liquid n itrogen and otherwise shielded with tungsten and lead. Mounted on a cart and pointed at a waste drum or suspended over a liquid waste container by crane, the QuantiScan is connected to other electronic instruments and a computer that converts the gamma measurements to distributions of individual gamma-emitting isotopes. An extension to the collimater permits gamma-scanning of irradiated hardware in a spent-fuel pool. In position over liquid resin tank. 24 EPRI JOURNAL June 1986 Assay of irradiated hardware under water. Electronic instruments and computer. tal neutron count into individual isotope The underwater system, composed of 29 concentrations in units of nanocuries somewhat shorter (26-in, 66-cm) 10BF3 (billionths, 10- 9, of a curie) per gram of tubes encased in lucite moderator and waste. Radiochemical analyses of some housed in a watertight stainless steel irradiated materials scanned by the TRU box, was demonstrated at Crystal River. system indicate an accuracy within 0.1%. Six cannisters of in-core waste compo­ The prototype TRU Detection System nents with dose rates of 7000-22,000 R/h was originally built for use at the govern­ showed average TRU concentrations of ment's Hanford nuclear reservation in about 2 nCi/g-the approximate limit Washington State, but has since been of detection sensitivity for a 1000-min demonstrated at several commercial nu­ exposure. The TRU detector's proven capability clear power plants under an EPRI dem­ onstration project. The system employs a leads researchers to believe it could be cylindrical array of ninety-six 9.8-ft-long economically cost-effective if available as (3-m), 2-in-diam (5-cm) tubes filled with part of a commercial service. Given an boron trifluoride (10BF3) that surround the estimated capital cost of $300,000, few waste container. The tubes are mounted utilities could justify purchasing one. But in polypropylene that slows down, or even if a utility wanted to buy one now, moderates, neutrons from the waste. it could not. Paid for and patented by Each neutron is electronically recorded DOE, the system faces some hurdles be­ fore it sees wide use in the utility indus­ by the 10BF3 tubes. The detector can be calibrated for a va­ try. EPRI is participating in a technology riety of containers and materials, includ­ transfer effort with DOE that could soon ing solidified resin and filter media in make the system available through a nu­ concrete-lined drums. For typical mea­ clear services vendor. surement of a 55-gal drum, the container is centered inside the tube bank and Direct gamma-scanning counted for 10 min. An 8-in (20-cm) lead In addition to requirements for more­ shield around the waste container pro­ accurate assays of transuranic isotopes tects the detector from excessive gamma in LLW, NRC regulations have also in­ creased analysis and reporting require­ radiation. Demonstrations sponsored by EPRI of ments for gamma-emitting radionuclides, Battelle's TRU detector were conducted including various isotopes of cobalt and in 1984 and 1985 at Wisconsin Electric cesium. The gamma-emitters are present at Power Co.'s Point Beach nuclear plant, Philadelphia Electric Co.'s Peach Bottom fairly low concentrations and produce station, and Florida Power's Crystal low levels (from thousandths of a rad to River nuclear plant. About 40 waste con­ a few rads per hour) of radiation in typi­ tainers (with dose rates up to 2 R/h) as­ cal LLW material, but they also account sayed at the plants showed less than 10 for the much higher radiation levels in nCi/g of total TRU radionuclides-the irradiated in-core hardware, where the federal limit for the least restrictive class high neutron flux can produce gamma­ of LLW. The approximate limit of de­ dose rates of up to several tens of thou­ tection sensitivity in the demonstration sands of R/h in in-core instruments, core applications, about 1.5 nCi/g, could be barrel bolts, and control rod retainer clips. lowered to 0.5 nCi/g of total TRU by ex­ Such components are handled remotely and stored in steel cannisters underwater tending the counting time to 100 min. A similar unit has also been configured in spent-fuel pools. NRC rules require quantification of the for underwater operation to measure TRU contents associated with highly radio­ gamma-emitting isotopes in LLW, but for active components from inside a reactor some types of material, representative that are stored on-site in spent-fuel pools. sampling and radiochemical assay, again, is impractical or inaccurate because of the nonuniformity of the waste or because of high dose rates. This is true especially for irradiated hardware, where scrape sam­ pling and radiochemical analysis entail substantial uncertainties and potential personnel exposure. A device for direct assay of gamma­ emitters in waste has been commercially available since 1984 from Science Appli­ cations International Corp. (SAIC), un­ der the trade name QuantiScan. The heart of the system is a high-purity crystal ger­ manium detector cooled by liquid nitro­ gen and encased in a shielded collimator. A low-power laser assists in alignment for waste scanning. A 5-in (12. 7-cm) lead shield on the front of the unit allows it to operate in background radiation fields of several R/h. The unit is mounted on a cart and connected to a portable micro­ computer up to 200 ft (61 m) away that controls the acquisition, analysis, and re­ porting of data. The electronics package includes a high-count rate amplifier and analog-to-digital converter. SAIC has demonstrated the Quanti­ Scan system in a number of applications. The device was used in the cleanup oper­ ation at TMI-2 to help locate fuel debris in the crippled reactor. EPRI sponsored demonstrations at three operating plants in 1984 and 1985, in which SAIC assayed various waste forms, including filter car­ tridges and bulk resin in high-integrity containers, demineralizer resin transfer lines during sluicing, solidified evapora­ tor concentrates, and dry active waste in drums and boxes. In addition to these EPRI-sponsored demonstrations at Point Beach, Peach Bottom, and Crystal River reactors, the QuantiScan has been used at Virginia Power's Surry reactor and Rochester Gas and Electric Co.'s Ginna station. At Crystal River, direct samples for ra­ diochemical analysis and dose rate mea­ surements were taken for most of the wastes for comparing the results with those of the QuantiScan. Results from direct gamma assay were combined with plant-specific scaling factors to obtain EPRI JOURNAL June 1986 25 Transuranic Assay by Neutron Detection A direct assay system developed by Battelle , Pacific Northwest Laboratories determines the transuranic (TRU) isotope content of LLW by detecting neutrons emitted from the waste. The detection sy stem consists of an array of 96 tubes filled with boron trifluoride ( 10 B F3), which surround a waste container and are connected to electronic instruments that record the neutron count and translate it to TRU radionuclide distributions. Neutrons emanating from a container are slowed by moderator material before passing into the 10 BF3 tubes , where they convert molecules of boron-10 to charged ions of lithium-?. This registers as a slight electrical charge and is recorded as data for collection and analy sis. I I Radwaste drum II -� Polypropy lene moderator Stand / Casters For typical measurement of a 55-gal drum of dry LLW, the container is centered within the tube array and counted for approximately 10 min. A submersible model of the detector has also been developed for TRU assay of irradiated in-core hardware stored in spent-fuel pools. Setup of TRU detector assembly. Closing detector assembly around radwaste drum. 26 EPRI JOURNAL June 1986 Electronic instrumentation for TRU detector. was based on a very small sample and it was not known whether the transuranics on the surface of the samples were uni­ Teaming up at Crystal River form across all the hardware in the can­ Neither of the direct-assay techniques in­ nisters," explains Galen Clymer, Florida dividually is able to measure and quan­ Power's acting manager for nuclear tify all the LLW radionuclides specified waste. "But it is nonetheless clear from for analysis under current NRC rules, both results that the waste was not above but EPRI recognized that if they were the Class C limit for TRU content. Even used in combination, along with plant­ with the uncertainty, the results obtained specific scaling factors, an accurate in­ are more accurate than those obtained by ventory could be obtained for the full relying solely on computer-generated spectrum of isotopes. In late 1985 EPRI and dose-to-curie correlations." For the gamma assays, results from the and Florida Power cosponsored a project with Battelle and SAIC to demonstrate observed dose profile and QuantiScan the TRU and gamma-detectors as a team methods were in reasonable agreement at the Crystal River plant as part of a for the total curies of cobalt-60 in each broader research project to evaluate di­ cannister. When compared with conven­ rect measurement technologies for accu­ tional computer code-based calculation racy, practicality, cost, and conformance methods, the direct assay techniques with ALARA guidelines. gave results at least 50% lower. Economic benefits of the direct-assay The principal focus at Crystal River was the characterization of irradiated techniques were estimated on the basis in-core hardware that was submerged of known curie surcharges at waste dis­ in the spent-fuel pool, but other types posal sites. The conservative calculation of LLW-packages containing primary methods the utility would have been coolant system filters and resins and forced to rely on in the absence of direct compacted dry active waste-were also assay would have resulted in unneces­ assayed. About 7 ft3 (0.19 m3) of irra­ sary additional disposal costs of about diated hardware (including the canni­ $50,000. EPRI project managers are work­ ing with Florida Power and the con­ sters) were measured. The submersible model of Battelle's tractors to further analyze the potential TRU detector was used in conjunction cost savings of direct assay and to esti­ with an intrinsic germanium detector and mate generic economic benefits to the in­ dose profile measurements from a series dustry of both the TRU and gamma­ of thermoluminescent dosimeters placed scanning techniques. every 4 in (10 cm) along the length of the Although potential savings are not storage cannisters. SAIC's QuantiScan yet fully quantified, participants in the spectroscopic gamma-detector and a R0- demonstration at Crystal River were con­ 7 remote teleprobe measured the activity vinced that the increased confidence af­ from the same cannisters separately. forded by the direct assay methods Results of the TRU assay system indi­ yields obvious benefits. "Both the TRU cated transuranic concentrations in the and gamma-scanning methods allowed cannisters ranging 2-16 nCi/g-well be­ us to identify on-site what the isotopic low the Class C LLW limit of 100 nCi/g. distributions and the activity levels were, Subsequent scrape and fragment sam­ rather than relying on calculations and ples of some low-dose components were off-site analysis alone," says Clymer. "In analyzed radiochemically for compari­ typical use, they will also reduce person­ son, with results about a factor of 10 be­ nel radiation exposure. Because we've low the TRU direct assay results. proved that the lower curie measure­ "There was some difficulty comparing ments are in fact more accurate, the cost these results because the radiochemistry benefits are clear." estimated concentrations of non-gamma­ emitting isotopes as well. New tools for LLW assay Soon both the TRU and gamma-scanning techniques may be available to the utility industry for more-accurate analysis of the full spectrum of radioactive isotopes than has been possible before. The meth­ ods promise not only to revolutionize the technical means for compliance with fed­ eral LLW regulations but also to reduce personnel radiation exposure and dis­ posal costs, as well as lead to more-effec­ tive use of existing disposal site capacity because the true radionuclide concentra­ tions in waste are better known. "We expect these tools to become important options for complying with federal analysis and reporting require­ ments," notes Charles Welty, an EPRI program manager. "The more accurately utilities can determine what is in LLW, the more apparent these benefits will be." • Further reading Galen Clymer and Patricia Robinson. "10CFR61 I rradiated Component Characterization at Crystal River Unit 3." Pre­ sented at 1986 Waste Management Conference, Tucson, Ari­ zona, February 9-13, 1986. Proceedings forthcoming. Radionuclide Correlations in Low-Level Radwaste. Final re­ port for RP1557-06, prepared by lmpell Corp., June 1985. EPRI NP-4037. Taylor Moore. "The Great State of Uncertainty in Low-Level Waste Disposal," EPRI Journal, Vol. 10, No. 2 (March 1985), pp. 22-29. Environmental Radiation Doses From Difficult- to-Measure Nuclides. Final report for RP1560-03, prepared by Science Applications International Corp., January 1985. EPRI NP3840. Solid Radwaste Radionuclide Measurements . Final report for RP1568-01, prepared by NWT Corp., November 1 982. EPRI NP-2734. This article was written by Taylor Moore. Technical back­ g round information was provided by Patricia Robinson and Charles Welty, Nuclear Power Division. EPRI JOURNAL June 1986 27 Herbert Woodson PROFI LE Encouraging Exploration in Research After 32 years as an electrical engineering professor, Herbert Woodson has an understandable first loyalty to fundamental research , but committees, consulting , and advisory connections-i ncluding EPRl 's Advisory Council-keep him in touch with the practical end of the electricity busi ness. 28 EPRI JOURNAL June 1986 H erbert Woodson's long suit is elec­ trical engineering. He has three degrees from the Massachusetts Institute of Technology, teaches at the University of Texas at Austin, runs an energy research center there, and often consults with electric utilities. How does he happen to be on EPRI's Advisory Council, which specifically comprises men and women from outside the elec­ tric utility business? The answer is that he is most of all an educator-a professor and researcher since 1954, first at MIT (17 years), then at Texas (15 years), where he has also been a department chairman (10 years) and a faculty recruiter (constantly). Woodson's broad and deep experience in the academic world makes for a com­ fortable fit with his 20-plus Advisory Council colleagues, whose interests and expertise include civil rights, commerce, economics, labor, manufacturing, medi­ cine, publishing, science, and utility reg­ ulation. Together, the Council members bring a valuable range of viewpoints to bear on EPRI's management and its view of R&D purposes and priorities. Woodson has a clear personal dedica­ tion to energy R&D, and along the way he has found time to administer several grant-seeded, special-purpose R&D cen­ ters on the Texas campus. The principal one is the Center for Energy Studies, which he established a dozen years ago and has directed ever since. Born of the energy crisis, it is described as "a multi­ disciplinary research center, the central liaison for energy research, education, and public service at the University of Texas at Austin." Inquisitive on his own, as well as on behalf of his graduate students, Wood­ son has never cared whether his research was on magnetic amplifiers, supercon­ ducting windings, flywheels, pulsed­ power supplies, 765-kV transmission, or plasma torches. Thus, except for an easy affection that began when he worked briefly for American Electric Power Ser­ vice Co. 20 years ago, the utility industry is not the focus for Woodson. His mind and his imagination are on the science itself-especially on electrical phenom­ ena, and even more on the processes and apparatus that can make for better, cheaper, faster products and services. Woodson converses easily and good­ naturedly, but the twinkle in his eye is just as often a hard glint. He also talks clearly and straightforwardly, and with determination. "Let me get on a soap­ box here," he'll say, and his conviction quickly becomes evident. But he is simul­ taneously enthusiastic about and de­ tached from his interests, rewarded more by the quest than by the conquest. Woodson is and does so much because of an endlessly inquisitive and busy na­ ture. His personal energy is symbolic as well as essential to his career. He is a committed person, although he puts a different twist on the word with the observation, 'Tm overcommitted! I'm sometimes surprised that the dean doesn't change the locks on the doors while I'm gone." Woodson's bouncy, vibrant style surely had something to do with his coming to Texas in 1971. As he describes it, the style of the university mirrored his own. "The dean was new, and he wanted to move the engineering school up in the national rankings. Everybody was happy and smiling, wanting to get something done, and Texas had the resources to do it. They asked me to be chairman of the elec­ trical engineering department. The more I looked at it, the more it seemed like a golden opportunity to have a major im­ pact on a program." What does Woodson mean by impact? "Recruiting and retaining good faculty; that's the soul of an educational institu­ tion. You need to keep them profession­ ally happy, too-good equipment in the labs, good parking places! Then, of course, you want good students, and this means making sure that your oppor­ tunities are known to the kind of stu­ dents you want." Fifteen years later he is still at it. As head of a committee responsible for eight endowed chairs in electrical engineering, he continues to look for good faculty. He has a number of graduate students doing research, and he teaches at least one course per semester-"required under­ graduate classes when I can; I really en­ joy that." But he stepped down as department chairman after 10 years. 'Tm more valu­ able today as ex-chairman," he says with wry modesty, "because now I can talk to the dean about department needs in a way that would have been self-serving when I was chairman." Energy for research growth The common thread of Woodson's years at Texas is energy, and more than simply his own. The 1973 energy crisis, for ex­ ample, motivated establishment of the Center for Energy Studies. As Woodson remembers it, several Texas faculty mem­ bers had already organized a monthly discussion group on energy issues, and the university president put before them the concept of a universitywide bureau of energy. Shortly thereafter, he gave it the green light, earmarked $1 million for a four-year startup, and asked Woodson to establish and direct it. "It's gone up and down," Woodson says candidly. "We've never tried to be­ come a big contract research activity. We've worked with faculty and students and department labs; we didn't have any labs of our own until we moved into a new facility just last summer. We find someone with a good idea, and we fund him until he's to the point of getting out­ side support." As an example, Woodson cites current research that EPRI helped develop. "The work is for the chemical industry, in sep­ arations processes-you know, distilla­ tions, supercritical extractions, the use of membranes. One major performance im­ provement would be better energy effi­ ciency, and that's why we got involved. We put in the seed money and hired a couple of chemical engineers to help the faculty get things started. "Now the program is in its third year," Woodson goes on, "with 40 participants at the moment-EPRI among them-at $20,000-$50,000 each. They're all on an advisory committee, so they have first call on the results and," the professor concludes with a smile, "on the graduate students." Woodson is convinced that various in­ dustrial processes could be improved by fundamental work. He thinks especially of new electrochemical paths from iron ore to steel or from sand to semicon­ ductor-grade silicon. "It would surely be significant if we could make the silicon metal for solar photovoltaic cells less expensively." The distinction between research and development is sometimes blurred in Woodson's account of his center's work. Especially for defined end-use applica­ tions of electricity, what constitutes fun­ damental exploration? "I think there ought to be more just fiddlin' around with things," Woodson replies. "A plasma torch, for example. There ought to be op­ portunity to run stuff through it and see what happens. Try it in different ways with different chemicals and see what it can do physically and chemically." Woodson explains that a plasma torch is an electrical means of heating a gas to a much higher temperature than is possi­ ble by combustion. Chemical reactions in the hot gas, the plasma, proceed much faster. Some reactions do not even occur at lower temperatures, so there is an op­ portunity for totally new processes. EPRI JOURNAL June 1986 29 "Take plasma spray coatings, for ex­ ing & Power Co. , and Texas Utilities Co. ) ample. If you get aluminum oxide hot but he calls attention to his advisory role enough," he continues, "and you spray on nuclear power projects. "At Gulf States, for example, there are it hard enough, it will stick on a surface as a hard coating that's a good insulator, three of us, all university professors, on a resistant to corrosion and effective at nuclear safety advisory committee. Tom fairly high ambient temperatures." Pigford is chairman of nuclear engineer­ Woodson is on a soapbox now, caught ing at the University of California, and up by the possibilities for new and more­ Ned Lambremont, a radiation biologist, productive end uses of electricity. "For directs the nuclear science center at Loui­ the immediate future I have the feeling siana State. It's really a technical auditing that we will do ourselves, our economy, activity. We meet four times a year and more good by looking at the consump­ hold the engineers' feet to the fire, so to tion side of the energy equation than at speak, to make sure that the engineering the production side. We've got plenty of work has a sound technical basis." capacity. We need to use electricity in in­ Woodson describes a series of oral ex­ novative ways that will improve our aminations that have evolved. At first, they were bland recitals of work sched­ competitive position in the world." Even when he is consulting, chances ules, procedural steps, and compliance are that Woodson is teaching. In recent memoranda. Now, at the committee's in­ years he has worked with a number of sistence, they are lively exchanges on the utilities (Central Power and Light Co., physics and thermodynamics of plant Gulf States Utilities Co., Houston Light- processes. "We want to hear about the equipment they've designed and their assessment of how well it works. We give them a technical quiz on what they're do­ ing. If someone cites a commonly used factor, we ask him to derive it from first principles." The class is getting good marks, ac­ cording to Woodson. "It's come to where they relish having to make presentations to the professors. We didn't realize we were doing that much good, but man­ agement is so pleased with the height­ "We will do ourselves, our ened technical awareness that we're still economy, more good by doing it-ever since 1980." Woodson's account of similar advisory looking at the consump­ work for Houston Lighting & Power em­ tion side of the energy phasizes his affection for utilities. "When­ ever a utility asks me to do something equation than at the production side. We need I'm qualified to do, I really try to do it." And after thinking for a moment, he to use electricity in inno­ pointedly adds, "I feel that nuclear is re­ ally the finest power generation technol­ vative ways that will ogy we have right now. If we don't botch improve our competitive it up, it can be our least expensive, least position in the world. " intrusive, and safest source of baseload 30 EPRI JOURNAL June 1986 electricity; and I'd like to do what I can to make that so." The electric connections Although Woodson's professional career has unfolded in an orderly way, there has been a good bit of coincidence and serendipity involved. The important point is that he took advantage of oppor­ tunities, beginning when he graduated from high school in Lubbock, Texas, at the age of 17. That was in 1942, and the opportunity was the U.S. Navy. Woodson had only months in which to decide; at 18 he would be subject to the military draft, with no choice of service branch and few special training opportunities. He en­ listed in the navy and gambled on pass­ ing the examination for radio technician school. It was a good gamble. He passed, finished second in a class of 250, sailed a quarter-million miles on a troop trans­ port, and then decided to study electrical engineering at Texas Tech University, back in Lubbock, after the war. But a per­ ceptive navy counselor at the separation center had other ideas. Picking up where Woodson's navy travel left off, he painted a future of wide academic hori­ zons and high career goals and handed Woodson an MIT catalog. The reminiscence warms Woodson. "I was stunned, because I'd not been an ex­ cellent student. But I read the catalog, took the college board exam, applied, and was admitted. I went to Cambridge in 1947 as a 22-year old freshman on the GI Bill, and I stayed for 24 years." That outcome was another instance of serendipity. "As a doctoral student in 1956 I helped my thesis adviser compile notes for an undergraduate text in rotat­ ing machinery-my specialty-and he asked me to stay around, help him turn the notes into a book, and be coauthor. So I did. I never intended to be a faculty member, but that was the initial reason I "I helped them with the more funda- "I was impressed by how stayed on. Oh, yes, the book was pub­ mental scientific side, I think. We were thoroughly the whole all engineers, but I had an overlap into lished in 1959." Woodson recalls another person who the research community, whereas theirs advisory apparatus of was important to his increasing activity was into utility operations. I was the one committees and task in electric power. This was was Philip in the middle." forces deals with EPRl's Also in the early 1970s, Woodson Sporn, chief executive officer of Ameri­ can Electric Power Co. and a living leg­ served briefly on a committee for the program. It wasn 't just a end in the utility industry. Sporn was on White House Office of Science and Tech­ superficial approval of the MIT electrical engineering visiting nology. He thereby met the assistant di­ committee when Woodson met him and rector for energy, environment, and nat­ what the staff put up; began to learn what the power business ural resources, Richard Balzhiser, who they got down to brass had to offer. In part because Sporn was shortly to join EPRI and become its tacks. " wanted to encourage power engineering senior vice president for R&D. But before education at MIT, Woodson spent a sab­ doing so, Balzhiser organized and led a batical year, 1965-1966, as a staff elec­ 10-day trip to the USSR for Woodson and trical engineer at AEP Service Corp. , 16 others, furthering a periodic scientific Sporn's engineering subsidiary. Return­ and technical interchange. The State De­ ing to MIT, Woodson became the Philip partment officer for the group was Rob­ ert Loftness, who would later be director Sporn Professor of Energy Processing. Other professional activities in power of EPRI's Washington Office and secre­ engineering engaged Woodson in the tary of the Advisory Council. The chain of coincidence and seren­ late 1960s. Memberships on the rotating dipity in Woodson's professional career machinery and education committees of the IEEE Power Engineering Society culminated in 1984 when Loftness tele­ wasn't just a superficial approval of what (both of which continue today) are exam­ phoned Woodson and invited him to join the staff put up; they got down to brass ples that come to mind. Also, he worked the Advisory Council. The council chair­ tacks." Woodson also mentions the evident for the research division of the Edison man at the time was David White, Electric Institute, and he served on an director of MIT's Energy Laboratory, knowledge and insight of the utility reg­ American National Standards Institute Woodson's doctoral thesis adviser and ulators who hold Advisory Council mem­ bership. Like many people, Woodson Committee on generators. These set the coauthor of 28 years earlier. had a different preconception from read­ stage for his being retained in 1971 by the ing news accounts of regulatory contro­ utility industry's Electric Research Coun­ The adviser learns cil to work with its newly organized R&D Not exactly an outsider to the utility in­ versies in various states. In his own state dustry, Woodson acknowledges, "I have of Texas, for example, the utility commis­ Goals Task Force as a consultant. EPRI was built on the report of that to remind myself that the Council isn't a sion is a fairly new government body, but task force, which identified a 30-year bunch of utility people worrying about in Woodson's view it has been politicized R&D agenda for utilities, put a price on how to get a utility engineering job done. by a recent round of appointments. For Woodson, the EPRI regulator­ it, and concluded that a centralized ven­ But I've learned a good bit, especially advisers at first seemed too good to be ture was the way to go. "If you look in about what's going on at EPRI." For example, he recalls representing true. (Seven state utility regulators are the green book they put out," Woodson states, "you'll find that I was their one the Council at a meeting of the Research named to the Advisory Council by their technical consultant. I went to all their Advisory Committee (RAC), the senior professional body, the National Associ­ meetings and helped them write their technical resource drawn from EPRI's ation of Regulatory Utility Commission­ membership. "That was illuminating. I ers.) "I still suspect they aren't typical," report." Asked what he did that the task force was impressed by how thoroughly the Woodson says. "They're much more un­ members could not have done as well, whole advisory apparatus of committees derstanding of the utility industry than I Woodson searches for the right phrase. and task forces deals with the program. It thought they would be." EPRI JOURNAL June 1986 3 1 But it is on his pet subject of electricity In fact, they went on to urge that such end use that Woodson finds real oppor­ research be elevated to EPRI division tunity with the Advisory Council. Exam­ status within three years, and they fur­ ination of EPRI's end-use R&D program, ther called for this new division to com­ in fact, was the charge of a five-member mand about a third (20-40%) of EPRI's council subcommittee on which he served annual R&D budget within five years, up last year. The group conferred with from the current 8%. Further, the sub­ EPRI's Energy Management and Utiliza­ committee report endorses an improved tion Division, took on its own literature competitive stance for U.S. industry as a research and writing assignments, dis­ proper goal for EPRI "because the health cussed and edited its findings, and deliv­ of the utility industry is closely linked to ered its report last December. With some the health of the economy as a whole." Woodson sees distinct benefits from revision, it was adopted by the council and directed to Floyd Culler, EPRI's pres­ R&D that is statedly inclusive, that en­ ident, for response. compasses and integrates the interests of The report takes an aggressive view. consumers along with those of utilities. Technology, economics, and public pol­ He mounts his soapbox to make clear icy are widening the range of choice for how easily utilities can justify R&D for energy users, it points out, so electric the betterment of other industries. utilities must compete in new ways. "EPRI's work may even improve an in­ "EPRI's objectives should be shaped by dustrial process without using more elec­ a market-driven objective of increasing tricity," he begins, "but if that increases consumer choice." the level of industrial activity and makes Woodson and his colleagues affirmed us more competitive in the world, then it EPRI's recent action to consolidate its ana­ will indirectly benefit a lot of other work­ lytic and technical research in end uses. ers, too, and all the people they do busi­ ness with." The clincher comes when Woodson observes, "And everyone in­ volved in that heightened industrial ac­ tivity lives in a house that uses elec­ tricity." The tension between R and D "The way to do fundamen­ tal research is to identify some talented research­ ers in an area of interest, give them three years of funding, and tell them, 'Go do what you think best-just add to the store of information. ' 11 32 EPRI JOURNAL June 1986 Another recommendation by the Advi­ sory Council subcommittee called for de­ voting a fixed percentage of end-use re­ search funding to exploratory research­ adding to the fundamental knowledge base in relevant scientific areas. Noting the different kind of professional ex­ perience needed, the subcommittee also urged that EPRI reconsider an early plan to manage all its basic research as a sepa­ rate division. Both these points call atten­ tion to the difference, the tension, some­ times the conflict between fundamental and applied research practice. This tension draws Woodson's inter- est, and he is concerned with more than end-use R&D alone. "I think EPRI is probably best qualified to do applied work-that is, in terms of the staff it has. Look at the Cool Water project, integrat­ ing coal gasification and combined-cycle generation and demonstrating them at 100-MW scale. "But I think EPRI overplans a bit," Woodson adds. As he sees it, this is mostly a matter of a useful professional quality being carried too far, "but there's sometimes an unfortunate rigidity too. I've known some EPRI managers to have a very narrow view of what needs to be done. They've gone through their plan­ ning exercise and they seem to say, 'We've settled on this and we aren't look­ ing for anything new. If it isn't in our plan, then it's not a good idea.' Ideas have been dismissed out of hand that way," Woodson insists. The words warn of an attitude that can threaten the value and the course of ap­ plied research. But excessively detailed planning bothers Woodson mainly be­ cause it can stifle what is intended to be basic research. He elaborates this way. "If I contract with you to deliver a spec­ ified item in a specified time on a spec­ ified budget, that's not research. That's development, at best; maybe it's just production. Research means exploring the unknown. "You can sit down and think logically about where you're going and how you're going to get there. But the proba­ bility is that the real advances will be nothing you predicted." The scientific and technical literature of 25 or 50 years ago makes Woodson's point. "We're not doing today, technologically, what the experts of those days predicted. But have there been advances? You bet! Tremen­ dous advances, but not what people ex­ pected." From his long university experience Woodson insists, "The way to do funda- mental research is to identify some tal­ ented researchers in an area of interest, give them three years of funding, and tell them, 'Go do what you think best-just add to the store of information. ' "Three years later, you ask what they found out. If they learned something re­ ally significant, you fund them again. If not, you don't." Woodson relaxes a bit and admits that an orderly process is needed for guiding either basic or applied research. He men­ tions a framework-a decision matrix­ for setting priorities, making either/or decisions, and measuring results. "It's just that I think there's a tendency to overdo it," he concludes somewhat wist­ fully. Woodson also argues for a better bal­ ance between near-term applied research and exploratory, probably long-term, re­ search. "Fundamental research generally is fairly inexpensive," he points out; "so you're able to take multiple paths to be­ gin with. You take ten paths, and when one of them is fruitful, you can be happy and not worry that the others led no­ where. " But what appears to be a 90% failure rate is difficult for some scientists (and certainly for many engineers) to deal with. Woodson remembers urging his image of research management with the Electric Research Council task force 14 or 15 years ago, but without success. What stood in the way? Momentarily silent, he then admits he is picking his words carefully. "I think it was a matter of belief or, the other way around, of disbelief. A great number of the leading technical people in utilities, including CEOs, didn't easily think of R&D as a force that could be productive on a real-time basis. They weren't hostile to it; they just didn't be­ lieve in it as a way to do things more economically or efficiently during their own working years." This view of R&D was and is totally sincere, Woodson points out, and common among CEOs who come from the planning, financial, or legal side of the business today. But its paternalistic qua!ity is revealed by Woodson's figure of speech. "They fund R&D with what they think it deserves," he says. "It's like your feeling about a child's allowance: you want to give what is appropriate, but you don't want to give too much. " The adviser advises Touching again on the way power re­ search is organized and managed, Wood­ son believes that in some key areas EPRI would do well to establish and operate its own research facilities with its own re­ search staff. "I think it's difficult to main­ tain your competitive edge without having more than a monitoring role in research. I think some of the EPRI staff would be sharper if they were directly involved in it." Warming to the idea, he begins to see it in fairly focused terms, "a lab dedicated to a particular area of technology, say, turbine generators. This year the re­ search focus is bearings; next year it's shafts; and after that it could be stator insulation or stator cooling. It could move with the needs of the times." Reflecting on the number of opinions he has offered, mindful of the price tags on many of his ideas, and perhaps feel­ ing he has been shooting from the hip (a true Texan), Woodson is nonetheless a man of convictions when it comes to elec­ tric power R&D. From his soapbox he surveys a wide field of opportunity and need. "I think EPRI ought to be funded somewhere between three and six times as much as it is now. Industries that flourish put a sizable fraction of their in­ comes into research, on the order of 5% or so. And once it's up and running, re­ search is the continuous source of new "Industries that flourish put a sizable fraction of their incomes into research, On the order Of 5 °/o or so. And once it's up and running, research is the continuous source of new things that keep the industry flourishing. 11 things that keep the industry flourish­ ing." Woodson declares that R&D is a trivial part of any customer's electric bill today, "peanuts!" Even so, he lowers his sights a little as he continues. "EPRI's budget could at least be doubled or tripled, with the full consent of regulators, I believe, if the utility executives really believed in it." After more than 30 years in electrical engineering education and in close coop­ eration with the utility industry, Wood­ son sees the need in such simple terms because the subject itself is so complex. "An electric utility is technologically a very complex system," he declares, "a hell of a lot more complicated than a space shuttle. Anything that compli­ cated, in my view, is worthy of R&D at­ tention to make sure the technology is continually serviced and improved. You need to pay attention to its needs." • This article was written by Ralph Whitaker and is based on an interview with Herbert Woodson. EPRI JOURNAL June 1986 33 AT THE I NSTI TUTE Board Approves 1986 Budget Revision The num ber of EPR I member uti l ities continues to g row as revenues decrease slig htly. E PRI's Board of Directors, at its In addition, the total number of autho­ April 9 meeting in Washington, rized positions at the Institute will be re­ D.C., approved a revised 1986 duced by 5% over the next 12 months, overall budget of $311 million, which in­ primarily through attrition and retire­ cludes an R&D portion that totals $287 ments. Culler also commented that ex­ million. penditures for travel and capital equip­ Board Chairman Arthur Hauspurg re­ ment will be lowered. ported that EPRI membership continues Hauspurg called this response a "bal­ to grow and now exceeds 600 utilities. A anced and responsible management ap­ slight decline in 1986 revenues from 1985 proach to what we expect will be a short­ levels was attributed by Hauspurg to fi­ term problem." Despite the funding nancial pressures facing a few of EPRI's decrease, Culler emphasized that the In­ larger member utilities as a result of nu­ stitute's budget still provides for a vital clear plant construction programs or industrywide R&D program. "At a time weak regional economies. "Most of these when government and private funds for companies report that they expect recov­ energy R&D continue to decline," he ery in the near future," he said. said, "EPRI's work is all the more im­ Responding to the revised budget, portant for the national interest." • EPRI President Floyd Culler announced several measures to reduce adminis­ trative and support expenditures. As ap­ proved by the Board, these measures in­ clude the cancelation of executive salary The EPRI Board of Directors was ex­ increases for 1986 and some reduction panded from 15 to 24 members, and new and delay of other employee merit raises. directors were elected at the Board's New Board Members Elected 34 EPRI JOURNAL June 1 986 April meeting. New appointments to the Advisory Council were also announced at the meeting. The newly elected directors are How­ ard P. Allen, chairman and CEO, South­ ern California Edison Co.; Wilson K. Cadman, chairman and president, Kan­ sas Gas and Electric Co.; A. W. Dahlberg, president and CEO, Southern Co. Ser­ vices, Inc.; Jerry D. Geist, chairman and president, Public Service Co. of New Mexico; Girts Krumins, president, Col­ orado-Ute Electric Association, Inc.; Donald W. McCarthy, chairman and CEO, Northern States Power Co.; Nor­ man E. Nichols, assistant general man­ ager for power, Los Angeles Dept. of Water & Power; Larry W. Papasan, presi­ dent and CEO, Memphis Light, Gas & Water Division; Bernard W. Reznicek, president and CEO, Omaha Public Power District; and Stephen J. Sweeney, president and CEO, Boston Edison Co. Robert N. Cleveland, president, Buckeye Power, Inc., was reelected to the Board. Peter A. Bradford, chairman, Maine Public Utilities Commission, was ap­ pointed to the Advisory Council, and three members were reappointed: Ed­ ward F. Burke, chairman, Rhode Island Public Utilities Commission; Edward P. Larkin, commissioner, New York Public Service Commission; and Andrew Var­ ley, chairman, Iowa State Commerce Commission. • R&D Program Funding Levels Set CALENDAR For additional information o n the EPRl­ sponsored/cosponsored meetings listed below, please contact the person indicated. JUNE 24-26 Seminar: Transmission Line Design Optimization Schenectady, New York Contact: Richard Kennon (415) 855-231 1 9-11 Seminar: Plant Inspection San Antonio, Texas Contact: John Scheibe! (415) 855-2850 9-11 Workshop and Seminar: Bearing and Rotor Dynamics St. Louis, Missouri Contact: Stanley Pace (415) 855-2826 17-19 International Utility Symposium: Health Effects of Electric and Magnetic Fields Toronto, Canada Contact: Robert Patterson (415) 855-2581 JULY The following 1986 R&D funding levels 9-10 were approved for EPRI's technical di­ EPRl's Transient/Midterm Stability Computer Program visions by the Board of Directors at its Toronto, Canada April meeting: Coal Combustion Sys­ Contact: John Lamont (415) 855-2832 tems Division, $63.2 million; Nuclear Power Division, $52.1 million; Electrical 9-10 Workshop: Water Hammer in Nuclear Plants Systems Division, $36.6 million; Energy Boston, Massachusetts Management and Utilization Division, Contact: Jong Kim (415) 855-2671 $31.1 million; Energy Analysis and Envi­ 16 ronment Division, $29.2 million; and Ad­ Seminar: lnstream Flow Methodologies vanced Power Systems Division, $20.4 Seattle , Washington Contact: Edward Altouney (415) 855-2626 million. In addition, 1986 total expenditures were approved for six separately funded AUGUST programs: Steam Generator Owners Group Program II, $7.6 million; BWR 13-14 Owners Group Intergranular Stress Cor­ Software Integration for Power Systems Analysis rosion Cracking Research Program II, Wash ington, D.C. $3.2 million; Utility Acid Precipitation Contact: John Lamont (415) 855-2832 Study Program, $0.8 million; Hydrogen Control Program, $0.6 million; Nuclear 25-27 Nuclear Plant Life Extension Studies Fuel Industry Research Program, $0.5 Alexandria, V i rginia million; and Seismicity Program, $0.5 Contact: Melvin Lapides (415) 855-2063 million. 27-28 Total authorized funding was in­ Software Integration creased for five other programs to the fol­ for Power Systems Analysis lowing approximate levels: PWR Steam Seattle, Washington Contact: John Lamont (415) 855-2832 Generator Reliability, $30 million; Robot Applications for Nuclear Power Plants, $8.5 million; Amorphous Metal Power SEPTEMBER Transformer, $7.4 million; Turbine Blade Life Improvement, $6.2 million; and Ag­ 3-5 Workshop: Technology Transfer ing of Extruded Dielectric Power Cables, Fort Worth, Texas $6.2 million. • Contact: Conway Chan (415) 855-2099 22-25 Seminar: Partial-Discharge Testing and Radio-Frequency Monitoring of Generator Insulation Toronto, Canada Contact: James Edmonds (415) 855-2291 23-26 Workshop: Gas Turbine Procurement and Repowering Pittsburgh, Pennsylvania Contact: Henry Schreiber (415) 855-2505 24-25 Industrial Applications of Adjustable-Speed Drives Washington, D.C. Contact: Marek Samotyj (415) 855-2980 OCTOBER 7-9 1986 Fuel Oil Utilization Workshop Philadelphia, Pennsylvania Contact: William Rovesti (415) 855-25 1 9 14 Seminar: Coal Transportation Costing and Modeling San Diego, California Contact: Edward Altouney (415) 855-2626 14-16 Seminar: Solid-Waste Environmental Studies Technology Transfer Milwaukee, Wisconsin Contact: lshwar Murarka (415) 855-2150 15-16 6th Annual EPRI Contractors' Conference on Coal Gasification Palo Alto, California Contact: Neville Holt (415) 855-2503 EPRI JOURNAL June 1986 35 TEC H N O LOGY TRANSFER N EWS Getting People I nvolved: PSl's Key to Tech Transfer A ccording to David Odor, a research coordinator for Public Service Co. of Indiana (PSI), "EPRI is doing a super job, but technology transfer is a two-way street, and the utility should take the initiative." PSI has taken this philoso­ phy to heart and has developed an inte­ grated technology transfer program that works with only minimal maintenance. The successful program turns on two key points: a bottom-up approach that stresses PSI staff involvement and an easily accessible data base used to track EPRI work in PSI target areas. The bottom-up approach, a corner­ stone of the effort, was built into the pro­ gram from the very beginning. PSI's Re­ search Department conducted a survey and staff interviews to establish the ini­ tial level of technology transfer within the utility. Armed with that information, PSI identified EPRI research projects that matched the utility's application prior­ ities. Information on each of these proj­ ects was condensed and made available on a utility data base accessible through­ out the company. EPRI Journal R&D sta­ tus reports on PSI priority projects have also been condensed to one or two pages and matched with staff interest areas. With this system the utility aims to identify PSI-EPRI priority projects at the earliest point, to track their progress, and then to apply the research results. PSI 36 EPRI JOURNAL June 1986 staff members can routinely access the system to obtain relevant EPRI informa­ tion from the data base as it is needed. The information stored in the data base has also been an essential element of the four benefit assessment studies con­ ducted by PSI since 1983. PSI combines all its available resources to effectively promote technology trans­ fer and to involve people within the util­ ity. Members of the PSI-EPRI Advisory Committee serve as industry advisers on EPRI committees, and within PSI they provide guidance on utility research objectives and disseminate information on EPRI's structure, funding, programs, and personnel. In addition, PSI staff members help identify the EPRI re­ search projects that will match corpo­ rate priorities. Company needs can quickly and easily be matched with staff resources by using information from the Technical Interest Profile (TIP) system, the PSI distribution list for the condensed R&D status re­ ports, and the PSI library distribution list for final reports on PSI-EPRI priority projects. Stored in the PSI computer sys­ tem, the TIP records are used to distrib­ ute EPRI seminar announcements in ad­ dition to the 1500 EPRI report summaries that are distributed monthly to over 100 PSI employees. Care is also taken to track reassigned employees within the utility and to ensure that all data bases contain the correct information. The effectiveness of a utility's tech­ nology transfer depends on a good, well­ executed plan rather than on enormous resources. It takes only three profes­ sional staff members and one clerk in the Research Department to develop and implement this technology transfer pro­ gram, even though PSI, with 4000 em­ ployees, is the largest investor-owned utility in the state. Janie Farrington, the research clerk, is a key ingredient to the success of the PSI program. According to her, "Technology transfer is more than data and reports-the exciting, and pos­ sibly critical, part is interacting with peo­ ple and responding to issues that impact company needs." Because of automation she now spends only 25% of her time keeping the program running smoothly. One result of this program is the PSI staff's familiarity with EPRI and its re­ search products. Farrington describes the EPRI resources and the availability of information in presentations made to PSI personnel in power plants and field offices. In addition, EPRI videotapes are presented at widely advertised luncheon programs, and copies are distributed throughout the company. As the PSI staff is more aware of EPRI's value as a resource, it increasingly turns to EPRI with technical problems. In 1984, for example, station personnel noted that the Gibson No. 5 chimney liner ap­ peared to have shifted. Ronald Richard, the chemical superintendent at Gibson station, contacted Dorothy Stewart, a project manager in the EPRI Desulfuri­ zation Processes Program, and one week later EPRI and its contractors were on site to evaluate the situation and to help es­ tablish operating guidelines until the problem could be solved. Asked what suggestions he would make to other utilities considering a similar technology transfer program, Thomas McCafferty, PSI's research man­ ager, emphasizes that they should "pick a staff with a broad understanding of the utility's internal organization. Then give each member the freedom to go from department to department and the charge to get people involved. " After all, he reasons, "if the company were going to start work on a power plant project costing millions of dollars, it would make a long-term assignment of a top­ notch team to see that the job was done right. Technology transfer deserves no less." • Indicator Available for Stability Limits P ower plant control room operators can now avoid a unit trip by using a new electronic instrument that provides a clear indication of the approach to the generator's stability limit. This new de­ vice, the power angle instrument for large synchronous machines, has been produced for commercial use by Eumac, Inc. The instrument operates by taking the armature voltage zero crossing as a reference point and relating an electrical signal derived from the rotor position to that reference point. Using a magnetic pickup and a toothed wheel on the shaft, the device provides digital and analog readouts of the voltage proportional to the angular difference between signals. EPRI licensee: Eumac, Inc., 221 W. Sur­ rey Avenue, Phoenix, Arizona 85029. • EPRI Contact: Dharmendra Sharma (415) 855-2302 Guidelines for Controlling Corrosion-Causing Products A ustenitic stainless steels or nickel­ base steels used in nuclear power plant components are subject to stress corrosion cracking or intergranular at­ tack when placed in contact with materi­ als or chemicals containing certain con­ taminants. Steels can be inadvertently damaged by expendable materials or chemicals, such as lubricants, solvents, tapes, and nondestructive evaluation substances. Although no harm will result if concentrations are kept at safe levels, there have until recently been no stan­ dard industrywide practices. After reviewing the wide variety of current practices, EPRI saw the need to create a recommended program for the industry, incorporating the most effec­ tive industry practices. Guidelines for Con­ trol of Expendable Products (EPRI NP-4449) is now available to help utilities develop such a program to identify and control the contacts between contaminants and the vulnerable steels in nuclear power plants. Written in a clear, easy-to-read style, the recommended program identifies the contaminants, specifies their maximum safe concentrations, and describes test methods for determining contaminant levels in each batch or lot of a product. In addition, the program provides a model for a procurement specification, a justifi- cation of acceptable contaminant levels, and a sample product list. Utilities can use this document to evaluate their cur­ rent practices and develop a more com­ prehensive program for expendable product control. • EPRI Contact: John Carey (415) 855-2105 Manual for Rating Cable Terminations U ntil recently, emergency ratings of cable terminations in high-voltage underground transmission systems were often inaccurate because they were based on limited testing supported by experi­ ence. A new manual, Emergency Ratings of Cable Terminations (EPRI EL-4293), now provides two computerized methods to calculate these ratings-a network pro­ cedure and a matrix procedure-as well as emergency rating tables. Besides help­ ing transmission system designers and operators provide maximum safe oper­ ation during emergencies, these meth­ ods can be used to eliminate overdesign and unnecessary conservatism in plan­ ning and operating utility systems. Quick and easy to use, the network procedure produces acceptable ratings with a minimum of time and effort. The matrix procedure requires more com­ puter time and is more complicated to use, but it produces more accurate rat­ ings. When time is truly of the essence, the manual's rating tables can provide utility personnel with an emergency rat­ ing immediately. Applicable to all widely used cable sizes and voltages under typi­ cal operating conditions and termination environments, the tables provide a quick estimate, saving time and eliminating the need to have an experienced computer programmer available. Although the ta­ bles do not include ratings for forced­ cooled pipe-type cable systems, these can be calculated by using the computer­ ized methods described in the manual. • EPRI Contact: John Shimshock (412) 722-5781 EPRI JOURNAL June 1986 37 R&D Status Report ADVANCED POWER SYSTEMS DIVISION Dwain Spencer, Vice President LINE-CONNECTED PHOTOVOLTAIC SYSTEMS Photovoltaic (PV) power plants, which convert sunlight to electricity, may offer utilities a vari­ ety of advantages as potential bulk power sources: they emit no pollutants, they are mod­ ular, they can be constructed quickly, and they are highly reliable. Although PV costs are still too high tor bulk power applications, they have dropped dramatically in the past decade and will probably continue to decline. Over the past five years a number of test installations have been set up in the United States, and state-of­ the-art megawatt-scale installations have gone from initial design to line-connected operation in less than a year. It has been the objective of an EPR/ project to monitor and analyze the operation of representative line-connected PV systems and to track the performance of indi­ vidual modules produced by different manu­ facturing techniques. Since the project began, 13 line-connected systems, ranging in size from 18 kW to 1000 kW, have been monitored, along wfth more than 20 modules from a vari­ ety of manufacturers. Systems and modules The PV power plants examined in this project (RP1607) were assembled from one of two major types of modules: concentrator modules or flat-plate modules. Concentrator modules must track the sun in two axes, whereas flat­ plate modules can operate in either a fixed, single-axis tracking mode or in a two-axis tracking mode. The flat-plate modules tested are further characterized by the type of material used in the cells that constitute the modules, either single-crystal or polycrystalline silicon. Single­ crystal silicon is used in most concentrator modules, although data are also reported for a Varian Associates concentrator module in which single-crystal gallium arsenide cells are used. Table 1 shows the design features of the line-connected systems included in the study. Modules are composed of cells, each pro- 38 EPRI JOURNAL June 1986 ducing about half a volt, that are connected in series-parallel combinations. Three of the plants-the ones at Hesperia, California; Lov­ ington, New Mexico; and Beverly, Massachu­ setts-are composed of separate but identi­ cal segments. The sunlight that powers PV modules is measured in watts per square meter. Two com­ monly measured components of incoming so­ lar radiation (insolation) are the direct-normal, which comes directly from the sun in parallel rays, and the global, which includes the direct­ normal plus the diffuse, or indirect, insolation that is refracted and reflected from clouds, dust particles, water vapor, and the ground. Because d irect-normal insolation arrives i n parallel rays, i t can b e focused and concen­ trated; diffuse insolation arrives from all direc­ tions and hence cannot be concentrated. On a clear day the total global insolation re- ceived by a surface perpendicular to the sun's rays typically is about 1000 W/rn2 ; the direct­ normal component of that total (i.e., the com­ ponent useful in concentrator modules) is about 850 W/m2 . These levels of insolation are used as the standard condit ions against which PV modules are rated. The systems examined can be distin­ guished by the sunniness of the location (be­ cause the power output depends on insola­ tion) and by the operating characteristics of the equipment. The available global insolation varies substantially from one location to an­ other and with the tracking mode employed­ for example, from 3.64 kWh/m2 average per day at Georgetown, D.C. (a fixed system), to 9.31 kWh/m2 per day at Hesperia (a two-axis tracking system). Two of the PV sites monitored by EPRI are particularly sunny: Phoenix and Hesperia. These sites are ideal for either type Table 1 LINE-CONNECTED PV PLANTS MONITORED IN 1 984 Site Beverly, Massachusetts 1 Dallas, Texas2 El Paso, Texas1 Georgetown, o.c.s Hesperia, California3 Lovington, New Mexico1 Oklahoma City, Oklahomas Phoenix, Arizona2 1 Sing!e-crystal silicon. 2Concentrator. 3Polycrystalllne silicon. 4Based on calculated ratings. Year On-Line Rating (kW) Manufacturer Calculated Module Efficiency (%) Tracking Capacity Factor (%)4 1 981 1 982 1 981 1 984 1 982 1 00 27 18 300 1000 94.9 1 8.7 1 4.3 231 .0 729.0 5.3 8.3 5.3 6.3 6.8 Fixed Two-axis Fixed Fixed Two-axis 5.7 1 9.8 24.9 1 2.0 32.9 1 981 100 88.6 5.2 Fixed 20.6 1 982 1 982 1 35 225 78.4 1 51 .0 5.0 6.7 Fixed Two-axis 1 6.1 9.8 of module; flat-plate modules are installed at Hesperia and concentrators at Phoenix. Table 1 also shows key performance results from the line-connected sites; the results con­ firm that efficiency is generally higher for single-crystal modules, for newer sites {which benefit from improvements in cell manufac­ turing techniques and design), and for con­ centrator sites (AP-2544, -3244, -3792, -4466). Moreover, researchers find that historical test data provide a solid basis for determining how these plants respond to insolation, ambi­ ent temperature, and wind speed. Wind speed affects module cooling, and PV conversion efficiency is a function of cell temperature, be­ ing greater at lower temperatures. Because PV systems are modular, the per­ formance of large, megawatt-scale systems can be predicted by using data obtained from smaller systems and modules. Regression equations based on past performance predict output levels with a high degree of confidence. Evaluation of the regression equations for each site at standard conditions yields the calculated ratings {which are all lower than the manufacturers' ratings) shown in Table 1. Photovoltaic modules produce de, and all but one of the large systems monitored in this project use inverters to convert the de to ac; only in the 1 8-kW system operated by the El Paso Electric Co. is de used d irectly for charg­ ing batteries. In a cooperative program with EPRI, Pacific Gas and Electric Co. (PG&E) monitored 26 PV modules over a three-year period. The results confirm those reported in the analysis of line­ connected systems. Module output levels again failed to match manufacturers' ratings in most cases, but statistical d ata can be used in regression equations to predict with confi­ dence what the output levels will be as a func­ tion of insolation and temperature. The results from PG&E's fixed flat-plate test program indicate efficiencies that are higher than those shown for the line-connected sys­ tems because some of PG&E's modules were constructed by using more-advanced manu­ facturing techniques and because the line­ connected systems incur some losses in inter­ connections and inverters, as well as from mismatches in currrent-voltage curves. Differences in tracking mode also affect the amount of insolation converted to de. PG&E found that in San Ramon, California, two-axis tracking modules receive about 34% more in­ solation than do fixed flat-plate modules and therefore produce about 34% more energy output. For utilities with peak loads in midafter­ noon, the more uniform output pattern of the two-axis tracking throughout the day suggests a greater capacity credit than can be given for Figure 1 Dallas (color) and Lovington-1 (black) show relatively high availabilities throughout the year. However, capacity factors in Dallas peak during the summer when days are longer, but not in Lovington. In Lovington the skies are clearer during the winter (less haze and fewer clouds than in summer). Efficiency and reliability 100 � g_, g parabolic trough concentrators are u nsatisfac­ tory for PV applications, largely because of reflectivity degradation and nonuniformities in concentrated solar flux. However, both the line-focus and point-focus Fresnel modules show promise for large-scale utility applica­ tions, and EPRI monitored one of each. 80 "iii 60 l co .£ co 1984 fixed flat-plate modules. PG&E checked the ef­ fect of seasonally adjusting the fixed flat-plate modules but could achieve only a 3% gain in energy output. There is substantial variation in insolation with the seasons, as shown by the monthly ca­ pacity factor data from Dallas and Lovington-1 in Figure 1. The days are longer in June and July, and as would be expected, the insolation (and thus the capacity factor) is g reatest dur­ ing these months in Dallas. But at Lovington, the insolation peak occurs not in midsummer, when skies are hazy and cloudy, but in the early part of the year, when skies are more of­ ten clear. Because lifetimes of 20 to 30 years are antic­ ipated for PV plants, a key issue in the fea­ sibility of using PV as a bulk-power source is whether performance will degrade over time. PG&E found 2�7% degradation in four early modules over a three-year period. No degra­ dation was observed for most of the modules. The performance of one of the line­ connected systems did deteriorate over a period of several years because of the in­ verter's failure to follow the modules' peak power point, which varies with insolation and module temperature. Concentrator modules have been based on three designs: line-focus parabolic troughs, line-focus Fresnel lens modules, and point­ focus Fresnel lens modules. Although they ap­ pear to work well for collecting thermal energy, Concentrator systems are potentially more cost-effective than crystalline silicon flat plates, not only because they require less of the high­ cost crystalline silicon but also because the cells can be more efficient at high concen­ trations. This efficiency gain is partially offset by a decline in efficiency caused by the higher cell operating temperatures; as a result, cool­ ing is an important aspect of concentrator design . The line-focus system at Dallas extracts heat from the cells by pumping coolant past the cells; the heat is then used in a nearby hotel complex. Most other concentrator systems, in­ cluding the one operated by Arizona Public Service Co. at Phoenix, use passive cooling, with air-cooled heat sinks mounted behind the cells. The concentrator systems showed substan­ tially higher efficiencies (calculated for stan­ dard conditions) than the flat-plate systems, ranging from 8.8% at Phoenix to 1 6.0% for a 1000 x gallium arsenide test module built by Varian and evaluated by PG&E. The high reliability projected for PV equip­ ment offers a major potential advantage in util­ ity applications. PG&E noted that PV modules were more reliable than the equipment used to monitor them. Even in the larger line-conected systems there are few moving parts (none in fixed flat-plate systems), and with experience and improved designs, the actual perfor­ mance has shown steady improvement. Some problems do persist. For example, throughout 1984 and early 1 985 the Phoenix site experienced a variety of component prob­ lems, including an inverter failure, a blocking­ diode failure and ground fault that caused a fire, moisture inside the modules, and data acquisition system outages. Availability, defined here as the ratio of sys­ tem operational hours to the number of hours of sufficient insolation to operate the system, is a measure of equipment reliability. Figure 1 shows that system availability was always above 80% at the Dallas and Lovington-1 sites. This suggests that future PV equipment can be expected to be quite reliable, so a plant's output will depend almost exclusively on the insolation at the site and on the efficiency with which sunlight can be converted into electricity. EPRI JOURNAL June 1986 39 ADV/lJ\ICED SYSTEMS DIVISIOI\I Because of the inherently reliable design of line-connected PV systems, their operating and maintenance costs are expected to be low. Preliminary results from plants that have passed the period of early failures and engi­ neering changes confirm this expectation. Estimates of these costs fo r 1985 range from about 0.3¢/kWh for Lovington to 0.85¢/kWh for Hesperia to as much as 2.8¢/kWh at Dallas, where the circulating coolant system has required much attention. Project Manager: John Schaefer USE OF LIGNITE IN T EXACO IGCC POWER PLANTS The Texaco coal gasification process uses a controllable, high-pressure coal-water slurry transport system to feed coal to the gasifier. Because of the high equilibrium moisture con­ tent of lignite (compared with that of bitumi­ nous coal), the total amount of water in a slurry feed with lignite is undesirably high. The in­ creased water content results in higher oxygen consumption, with an attendant increase in in­ tegrated gasification-combined-cycle (IGCC) power plant costs and a reduction in net power output. The major objective of the research de­ scribed in this report is to identify and evaluate alternative methods for feeding lignite to the Texaco gasifier that would eliminate the dis­ advantage of carrying surplus water into the gasifier. The Texaco coal gasification process has been demonstrated at commercial scale with bituminous coals at the Cool Water IGCC facil­ ity near Barstow, California, and at Tennessee Eastman Co. 's coal-to-chemicals facility in Kingsport, Tennessee. The Texaco process uses a controllable, high-pressure water slurry transport system to feed coal to the gasifier. The process is at a disadvantage, therefo re, when gasifying a lignite coal because of the lignite's high-equilibrium moisture compared with that of bituminous coal and the corre­ spondingly high total water content of the slurry feed. The higher water content of the feed results in higher oxygen consumption, with an attendant increase in IGCC plant cost and a decrease in net power output. EPRI funded a project with Energy Con­ versions Systems, Inc. (ECS), to address the problems associated with using Texas lignite in Texaco IGCC power plants (RP2221-1). The major objectives of the research were to iden­ tify and evaluate alternative methods of feed­ ing lignite to the Texaco gasifier that would eliminate the disadvantages of carrying sur­ plus water into the gasifier and to determine the relative cost of producing electricity in 40 EPRI JOURNAL June 1986 Table 2 STUDY CASES AND RESULTS Case Coal Illinois No. 6 2 Lignite 3 Lignite 4 Lignite 5 Lignite 6 Lignite 7 Lignite Oxygen/ Coal Carbon* 0.47 Plant Cost ($/kW)t 1 41 0 Electricity Cost (mills/kWh)t 43.9 0.43 1 430 33.8 Conventionally dried lignite is fed to the gasifier in a dense-phase pneumatic flow with recycle CO2 carrier at a concentration of 96 wt% wet solids. 0.42 1 440 33.8 Conventionally dried lignite is slurried in liquid CO2 at a concentration of 88 wt% wet solids; slurry is heated to vaporize the CO2 before the mixture enters the gasifier. 0.42 1 500 35.4 As-mined lignite is slurried in water at 50 wt% dry solids; slurry water is vaporized; about 50% of steam is removed (skimmed) in a cyclone; lignite is fed to the gasifier with steam at a concentration of 66.7 wt% dry solids. 0.42 1 460 34.6 Lignite (dried with hot water under pres­ sure) is fed to the gasifier in a water slurry at a concentration of 60 wt% dry solids. 0.52 1 630 38.3 As-mined lignite is fed to the gasifier in a water slurry at a concentration of 50 wt% dry solids. 0 59 2305 42.0 Gasifier Feed System As-mined coal is fed to the gasifier in a water slurry at a concentration of 66.5 wt% dry solids. Conventionally dried lignite is fed to the gasifier in a dense-phase pneumatic flow with recycle syngas carrier at a concentration of 98 wt% wet solids. 'Oxygen (100%)/coal carbon (molar ratio). !Total capital requirement (e.g., contingency, working capital, AFDC) in mid-1984 dollars. 'Thirty-year levelized mid-1984 constant dollars, investor-owned utility. Illinois No. 6 and lignite coal cost $2.25 and $1.15 per million Btu, respectively. /GCC power plants that use Texas lignite and bituminous (Illinois No. 6) coal. ECS researchers designed and developed capital and operating costs for seven Texaco IGCC power plants (Table 2). As can be seen, the base case reference plant (case 1) was fueled with Illinois No. 6 coal and used a con­ ventional water slurry for transporting the coal to the gasifier. The remaining six plants were fueled with lignite coal and used three different systems for transporting coal to the gasifier-a pumpable water slurry, a pumpable liquid car­ bon dioxide (CO2) slurry, and a d ry dense­ phase transport with either gaseous CO2 or re­ cycle syngas. Researchers used two approaches to re­ duce the water content for those plants em­ ploying a lignite-water slurry feed. In one ap­ proach (case 5) they heated the slurry to vaporize the water and concentrated the coal solids in a cyclone separator before feeding to the gasifier (a skimming operation). In the sec­ ond approach (case 6) they reduced the equi­ librium moisture content of the lignite by drying with hot water under pressure. A plant with a conventional lignite-water slurry feed (case 7) was included to complete the study. The slurry water requirement was com­ pletely eliminated in cases 2, 3, and 4 by using either dense-phase pneumatic transport or liquid CO2 slurry transport for feeding the lig­ nite fuel. Dense-phase transport by syngas carrier (case 2) requires recycling a small quantity of carrier product gas back to the gasifier with the feed coal, which results in a slight increase in gasifier oxygen consumption and gas plant mass throughput. For dense-phase transport with gaseous CO2 (case 3), the required CO2 carrier gas must be recovered from the cleaned product gas in a separate acid gas removal unit (Amine Guard process). For slurry transport by liquid CO2 (case 4), the required CO2 carrier fluid must also be recovered from the product gas in a separate acid gas removal unit, and an­ other separate system is required for liquefy­ ing the recovered CO2 gas prior to slurrying with coal. The design bases for the seven IGCC plants evaluated in this study were primarily selected to maintain continuity with previous EPRI stud­ ies. All the plant designs and costs for the seven study cases were factored from a prior Texaco IGCC plant design (EXTC-MOD2) that used an Illinois No. 6 coal feed (AP-3129). EXTC-MOD2, in turn, was based on the origi­ nal design and cost estimates for I GCC plant EXTC-79 (AP-1624). Early in the ECS study, trade-off studies showed that significant improvements could be made to Texaco IGCC plants by the judi­ cious selection of commercially available processes for acid gas removal and tail gas treatment. Specifically, the trade-off studies indicated that Union Carbide Corp.'s H-S and Sulften processes were superior to the Selexol (acid gas removal) and Beavon-Stretford (tail­ gas treatment) processes used in EXTC­ MOD2. Following the trade-off studies, the design for base case plant EXTC-MOD2 was modi­ fied to incorporate the H-S and Sul/ten pro­ cesses. The revised plant (case 1 in this study) is considered the reference base case plant design. The designs for the six evaluated lig­ nite IGCC plants (cases 2-7) also incorporate the H-S and Sulften processes. The Chicago area was the site for the Illinois No. 6 coal plant design (case 1) and the Dallas area was the site for the Texas lignite plant designs (cases 2-7). The Illinois No. 6 coal was assumed to be delivered to the plant site in a washed and sized condition, and the lig­ nite coal, in an as-mined state. On an as­ received basis, the mo·1 sture content of the Illinois No. 6 coal was 12 wt%, the sulfur con­ tent was 3.8 wt%, and the higher heating value was 11 ,241 Btu/lb; for the lignite coal, the mois­ ture content, sulfur content, and higher heating value were 34.9 wt%, 0.89 wt%, and 6826 Btu/lb, respectively. The lignite in cases 2, 3, and 4 was dried from 34.9 to 12.0 wt% moisture in conventional rotary drum dryers prior to grinding and then pulverized in unheated ring roller mills. The heat required for drying was supplied by sen­ sible heat available in the heat recovery steam generation (HRSG) stack gases. The lignite in case 6 was dried from 34.9 to 12.0 wt% moisture at 1200 psi (8 MPa) and 626°F (330°C) with hot water obtained from the HRSG economizer. The design of the lignite drying equipment is based on results of an EPRI project with the Energy Research Center, University of North Dakota (AP-4262). Data from the research show that drying high­ moisture-bearing, low-rank coals with hot water or steam under pressure can effectively reduce the equilibrium moisture content of the coal by as much as two-thirds. In addition, rhe­ ological studies indicate that slurries of water and lignite dried under pressure can attain a dry solid concentration of 60 wt% (the value used as the design basis for case 6). As-mined lignite-water slurries are limited to about 50 wt% dry coal content (the value used as the design basis for case 7). The design of the coal-liquid CO2 slurry equipment for case 4 was based on data from an EPRI project with Arthur D. Little, Inc. (RP2469-1). Recent data from AOL's pilot plant pipe loop viscometer indicate the maximum at­ tainable wet solids content of coal in a pump­ able CO2 slurry to be 88 wt%. The power block design for all seven evalu­ ated cases is based on currently available combustion turbines having a firing tempera­ ture of 2000°F (1 100°C) at the combustor exit. The steam cycle is a 1450 psig (10 MPa) sys­ tem (turbine throttle) with 900°F (480°C) super­ heat and a single 900°F reheat. A single steam turbine generator system is employed. Effi­ ciencies of the individual back-pressure and condensing ends of the steam turbine are the same as those used in the EXTC-79 study (AP1624). To facilitate the factoring of the gas turbine design from IGCC plant EXTC-MOD2 and to achieve comparable costs, the quantity of total heat content in the fuel gas to the gas turbine combustor was selected as the common de­ nominator for all seven cases. Each design case, therefore, had a total clean fuel gas heat rate of 7380 million Btu/h (lower heating value), and the coal feed rates were selected to match that production. In all plant designs for this study, the only net plant products are electricity and elemental sulfur. Total gaseous emissions of NOx and sul­ fur compounds are designed to be within the limits set by the revised EPA standards (June 11, 1979), which apply to a new coal-fired boiler plant. On the basis of these regulations, 90% of the sulfur contained in the feed coal is recovered as elemental sulfur. Stack gas NOx is controlled by humidifying the fuel gas to a steam concentration of 40 wt% prior to com­ bustion in the gas turbine. Table 1 gives selected results from the eval­ uation of the seven Texaco IGCC power plants. The major conclusions from the study are as follows. o The electricity costs for lignite cases 2-6 are substantially lower than those for bituminous coal, case 1 . Therefore, it may be stated that with properly designed coal feed systems, Texaco IGCC power plants using lignite have the potential to generate power at costs that are lower than those using bituminous coal. o As can be seen from cases 2-7, the cost of electricity for a given coal is proportional to the oxygen consumption (i.e., the higher the 02 consumption, the higher the electricity cost). Also, lignite cases 5-7 demonstrate that oxy­ gen consumption and the cost of electricity increase with an increasing water content of the gasifier feed. Case 7, a conventional water slurry feed, is clearly not cost-competitive with any of the other lignite study cases. o Dense-phase pneumatic feeding of lignite coal with either recycle syngas (case 2) or gaseous CO2 (case 3) is the lowest-cost op­ tion for generating electricity. CO2 slurry feed (case 4) is also attractive, comparing favorably within the accuracy of these cost estimates with a skimmed-water slurry feed (case 5). Al­ though the liquid CO2 and skimmed-water slurry feed cases have predicted costs that are slightly higher than the dense-phase transport cases, the slurry systems have an advantage in that they provide conditions that are amena­ ble to flow control of the feed coal to the gasifier. It should be noted that the lower cost of elec­ tricity for lignite cases 2-6, compared with that for bituminous coal (case 1), reflects the lower cost assumed for the lignite coal-$1.15 per million Btu for lignite versus $2.25 per million Btu for I llinois No. 6. In the instance where lignite costs are comparable to bituminous coal costs (e. g . , where lignite mining and transportation costs are higher than those as­ sumed in this study), the bituminous coal plant (case 1) would give the lowest cost of elec­ tricity for all the cases studied. It should be further noted that the proposed schemes for reducing or eliminating the water content of the slurry feed for lignite cases 2-6 need further development. For example, re­ searchers have to demonstrate that dry dense­ phase feed systems (the feed systems with the lowest electricity cost) provide the flow control required for feeding coal to an entrained gasifier. RP2469-1 yielded some rheological data on low-rank coal-liquid CO2 slurries; and limited data are available to support the water skimming concept and the hot water drying of coal at pressure (AP-4262) . All these pro­ posed schemes, however, have to be tested on a larger scale in a Texaco gasifier with lig­ nite feed. Project Manager: Michael Epstein EPRI JOURNAL June 1986 41 R&D Status Report COAL COMBUSTION SYSTEMS DIVISION Kurt Yeager, Vice President FOSSIL FUEL PLANT LIFE EXTENSION Fossil fuel generating units have traditionally been built with an assumed life of 25-30 years. The expectation was that these units would be replaced with new units to meet load growth requirements and, through advances in tech­ nology, to produce power at lower cost. This expectation has not been realized because of low load growth, high interest rates, escalating construction costs, and uncertain regulation. As a result, the average age of existing gen­ erating capacity is increasing, and utilities are interested in keeping units in service for 40-60 years or longer. This report describes EPRl's current and planned research to develop ge­ neric guidelines to help utilities devise strat­ egies for keeping units in service as long as economically practical. As units age, critical components may de­ grade through such mechanisms as erosion, corrosion, creep, fatigue, and creep-fatigue in­ teraction. The condition of these critical com­ ponents and the time remaining befo re re­ placement or major repair is required are major considerations in the cost of keeping older units in service. In addition, equipment degradation in older units may lead to in­ creased maintenance requirements in order to attain desired levels of availability and per­ formance. Determining unit cost is further complicated if operating strategies are imple­ mented that sacrifice unit perfo rmance to ex­ tend the life of critical components. The objective of a life extension program is to identify and implement a strategy fo r keep­ ing a generating unit in service as long as is economically practical. The development of such a strategy requires the cooperation of corporate planners, financial and system plan­ ners, and plant operations and engineering personnel. The implementation of life exten­ sion affects most aspects of subsequent plant operation through a phased program of in­ spections, capital expenditures, and oper- 42 EPRI JOURNAL June 1986 ating and maintenance (O&M) practices. In its effort to develop generic guidelines for fo ssil fuel plant life extension, EPRI has spon­ sored six projects to explore and document activities that are part of the life extension process. o RP2596-2 and -3 supported the develop­ ment of reference guidelines for life extension at Boston Edison Co. 's Mystic Unit 6 and Ni­ agara Mohawk Power Corp.'s Huntley Unit 67. o RP2596-4 and -5 supported Consumers Power Co. in developing a set of generic guidelines for analyzing plant life extension from the standpoints of corporate planning, financial planning, environmental planning, and risk analysis. o RP2596-6 supported an engineering evalu­ ation of the boiler and turbine at Pacific Gas and Electric Co. 's Pittsburg Unit 5. o RP2596-1 supported the integration of the findings of these EPRI projects and other in­ dustry programs around the world into a set of generic guidelines that a utility can use to de­ velop its own unique life extension strategy. EPRl's generic guidelines describe the life extension process in terms of three steps­ corporate planning, life extension planning, and life extension plant operation. These are not discrete, sequential steps but are related in time and by information requirements. Basi­ cally the same steps are necessary whether a utility decides on an incremental (phased) ap­ proach or a front-end refurbishment approach, which entails an initial extended outage. To integrate the steps, the guidelines in­ clude a life extension program management system. This system, which features an inte­ grated life extension decision (ILED) model, will support the data gathering, organization, and analysis required for a plant- or system­ wide program. It facilitates the scheduling of component inspections and any necessary re­ pairs or modifications to minimize outage and maintenance costs and maximize system and unit generation. Corporate planning At the corporate planning level, the life exten­ sion option must compete for funds against a variety of other alternatives-among them, demand-side planning, new generation tech­ nologies, power purchases, and new con­ ventional central station plants. From a system planning perspective, the value of the life­ extended plant to the utility system not only must exceed the costs of life extension but also must compare favorably with the other generation alternatives. The financial effects of life extension projects must meet the utility's financial goals, and provisions must be made for complying with current and expected envi­ ronmental regulations. The selection of plant life extension goals includes defining the life extension period, the required plant availability and efficiency, the power output, the energy costs, and the ex­ pected total costs of the tasks necessary to extend plant life. Historical O&M data for com­ ponents from the candidate plants will be aug­ mented with generic component data from the industry to provide a basis for initial estimates of the plants' remaining useful operating life and the projected costs of life extension. These costs are then used in a more de­ tailed system planning study to verify the rank­ ing of the units in terms of their suitability for life extension. Finally, these rankings, and the as­ sociated costs and benefits of potential life extension projects, are used by corporate planners in comparisons with other major in­ vestment alternatives. The preliminary findings of the Consumers Power study (RP2596-5) suggest that a company's historical O&M philosophies and practices will to some degree determine the approach it should follow in evaluating life ex­ tension. Initial projections also indicate that with the exception of major environmental con- trol projects to comply with changing regu­ lations, the incremental costs associated with extended-life operation at Consumers Power are relatively small. Because the costs are not large, the ranking of units is determined almost entirely by the benefits calculated in the sys­ tem planning analysis. and the value of the information to be obtained. The process a utility uses to select com­ ponents can itself become expensive because of the number of components involved. To fa­ cilitate this process, the generic guidelines present simple plant models to serve as a con­ sistent evaluation mechanism fo r all plants. Life extension planning Life extension plant operation Once a unit has been selected fo r life exten­ sion, a program plan must be developed for the scope and schedule of life extension activ­ ities. The objective of the plan is to minimize the cost of the life extension effort and to max­ imize plant ava·1 1ability and efficiency for a se­ quence of repairs and inspections within a given cost The most important operating decision is to select which components in a unit are to be managed through the life extension program. Components fall ·1 nto two categories on the ba­ sis of failure mechanisms and their conse­ quences. One category consists of compo­ nents critical to continued plant operation; these should be considered at this stage for all units. The other category consists of compo­ nents selected in reaction to current or antici­ pated problems; these will be unit-specific. Components in the first (critical) category are as follows. Once a utility has completed the planning activities and has selected the components to be evaluated, detailed scheduling is neces­ sary so that the critical component inspec­ tions and modifications can be performed within the framework of normal utility operation and maintenance. Methods for assessing the remaining useful life of plant components are a critical part of EPRl's effort to formulate life extension strate­ gies for fossil fuel power plants. They are nec­ essary for determining not only remaining life but also appropriate operation and inspection intervals. The generic guidelines delineate life assessment methods and, through the pro­ gram management system (specifically the ILED model), provide a tool for controlling the scheduling and costs of residual life inspec­ tion and analysis. EPRI continues to support work to develop and improve residual life as­ sessment techniques. The components cur­ rently being addressed are boiler pressure parts and tubing, steam pipes, and steam tur­ bine rotors. The research on boiler pressure parts has made significant progress in relating creep damage to the extent of creep cavitation. I n combination with replication techniques, this information provides a nondestructive, quanti­ tative life assessment tool (RP2253-1). In other efforts, currently used accelerated rupture tests for assessing remaining life have been evaluated and refined; miniature specimens have been developed that can yield the same information now provided by large specimens (RP2253-1 ); and a fracture-mechanics-based methodology for estimating creep crack growth has been established (RP2253-7). It is expected that the integration qf all these efforts will represent a major advancement in life as­ sessment (RP2253-1 0) . EPRI i s also investigating acoustic emission monitoring of cracking (RP734-6), using re­ tired headers and field testing. Preliminary results from the monitoring of a cracked header at Wisconsin Public Service's Pulliam station indicate that ligament cracking may be detectable. In the area of boiler tubing, the measure­ ment of steam-side oxide scale thickness to predict remaining life is being evaluated (RP2253-5); researchers are also comparing o Steam generator: drums (steam and lower), superheater headers, reheater headers, wa­ terwall headers, downcomers, main steam piping, hot reheat piping , economizer inlet headers o Turbine: rotor, shell, steam chest, casing o Generator: rotor, stator windings, insulation, transfo rmer o Balance of plant: intake and discharge structures, structural steel, stack liners, station main transformers For both component categories, critical and reactive, an initial inspect'1 on plan is developed on the basis of the failure mechanisms and the design and operating history of the compo­ nents. The consequences of failure are identi­ fied and quantified in dollars. Then, after each O&M cycle, the d ata collected are analyzed and used to revise the plan. I nformation on plant condition and the need for repair or mod­ ification of the components managed through the life extension program will be provided to corporate planners for use in their annual re­ view of resource allocation decisions. It is important that the scope of the initial inspections for both component categories be selected so as to complement current informa­ tion. An initial phased inspection plan should be developed on the basis of inspection costs this technique with a variety of methods based on wall thickness measurements, microstruc­ tural information, and destructive tests to de­ termine its relative merits (RP2253-10). In an­ other project a code called PODIS (prediction of damage in service) has been developed for assessing the condition of dissimilar-metal welds on the basis of nondestructive analysis (EPRI CS-4252) . Much other work i s being sponsored to im­ prove the nondestructive evaluation of boiler tubing. Ultrasonic testing (UT) guidelines for assessing boiler tube damage have been de­ veloped under RP1865-5. Techniques for as­ sessing wall thickness, cracking, pitting, and hydrogen damage are covered, as well as sur­ face preparation, personnel qualifications, available equipment, and costs. A computer code for analyzing UT data and determining optimal inspection intervals has also been developed. An alternative to UT boiler tube thickness measurement-the electromag­ netic acoustic transducer-is being investi­ gated in RP1865-3. This technique is generally limited to wall thicknesses greater than 0.1 in (2.54 mm). EPRI is also evaluating the use of acoustic leak detectors for monitoring boiler and feedwater heater tube damage; data from the first U.S. utility installations are being col­ lected and the benefits assessed (RP1863-2). With respect to steam pipes, preliminary guidelines for currently available inspection and life assessment technology have been prepared (RP2596-7). The capabilities and limitations of ultrasonic and radiographic tech­ niques for detecting cracks in steam pipes are being investigated by using pipe samples with a variety of known embedded flaws. Also, a computer code that will predict remaining life under creep crack growth conditions has been developed for longitudinally welded pipes (RP2596-7). In addition, new inspection and monitoring techniques that can potentially reduce in­ spection costs and improve coverage are be­ ing pursued (RP1893-4), including such auto­ mated inspection systems as internal pipe crawling devices, UT flaw-sizing techniques, radiography, and acoustic emissions for mon­ itoring crack growth and leak formation. These advanced techniques will be incorporated into the guidelines fo r the inspection, monitoring, and evaluation of the entire steam line system. In research on steam turbine rotors, several nondestructive techniques fo r assessing the toughness degradation of in-service rotors, such as Auger analysis, chemical etching, and the use of single Charpy specimens, have been developed (RP559, RP2257-1). Methods for estimating toughness on the basis of chem­ ical composition are being investigated in EPRI JOURNAL June 1986 43 COMBUSTIOl\i DIVISION RP2481-2. A previously developed computer code caller SAFER (stress and fracture evalu­ ation of rotors) is being updated in RP2481-3; the improved version will be available in the next 18 months. A materials data handbook on rotors is also in preparation (RP2481-4). The early detection of problems can often minimize or eliminate forced outages and can facilitate the scheduling of spare parts and manpower for planned outages. Diagnostic monitoring offers the ability to predict and detect equipment failures in their incipient stages and thereby to avoid costly damage or even catastrophic failure. EPRI is sponsoring the development and demonstration of several diagnostic monitoring systems. Of particular interest are the current demonstrations of a boiler stress analyzer at Consolidated Edison Co. of New York (RP1893-1) and continuous vibration signature analysis systems at Phila­ delphia Electric Co. and United Illuminating Co. (RP1 864-1 , -2). A boiler stress analyzer tracks the long-term creep-fatigue damage in headers, drums, and steam lines. Both turbine and boiler stress analyzers can extend the life of heavy-walled components by helping oper­ ators minimize thermal stresses during fre­ quent startups and load swings. A computer­ based continuous vibration monitoring system can curtail major failures in the turbine gen­ erator, feedwater pumps, and fans by de­ tecting such damaging conditions as steam whirl, oil whip, shaft cracking, misalignments, blade rubbing, and bearing deterioration. Program management system The objective of all the plant evaluations is to collect data to support decision making on life extension alternatives. In order for the data to be useful, they must be analyzed to determine how the condition of the equipment affects plant performance and the equipment's re­ maining useful life. This analysis must provide estimates of life and cycle life expended and of remaining useful life; indicate the operating parameters that can be controlled or limited to extend remaining useful life; make recommen­ dations regarding replacement under current, projected, and/or modified O&M practices; and indicate the degree of uncertainty inherent in the analysis results. Data from the O&M and inspection cycles and from accelerated testing will be analyzed by using the program management system's !LED model to determine the minimum-cost approach. The life extension plan and the maintenance plan will then be revised to effect the minimum-cost life extension program to meet the goals set for the specific plant. The !LED data on the costs and performance of the life extension program can be fed back to car- 44 EPRI JOURNAL June 1986 porate planners to be used in evaluating con­ tinued funding of the program. When an analysis of the results of various inspections indicates that major expenditures are required, the life extension program man­ agement system helps utility planners identify and analyze their potential options. The !LED model includes five submodels whose outputs are the decision measures used by manage­ ment in allocating resources to any life exten­ sion program: power (kW); availability (h/yr); energy costs (mills/kWh); total costs ($/yr); and schedules (h/yr). This much-needed life extension tool enables a utility to evaluate any number of units in its system over a period of time. In summary, the EPRI generic life extension guidelines developed under RP2596 provide the basis on which a utility can develop its own life extension strategy. Plant life extension starts with the current state of the plant as de­ termined from plant design and O&M data and from generic industry information. These data are used in corporate planning as the initial data set from which to estimate the plant's ex­ pected remaining useful operating life. If, after evaluating future energy-producing options, corporate planners select the life extension option, they prepare a plan that will meet the goals set. The operations group then develops the initial life extension plan, defining the nec­ essary tasks. The life extension management system, featuring the !LED model, supports the implementation of the plan. Each subse­ quent plant O&M cycle produces nel/'4 data, which are analyzed with the model, and the plan is revised in terms of cost and schedule for work required to keep the plant at the de­ sired performance levels. Although the life extension tools have been developed for aging fossil fuel plants, utilities can and should use them on newer units. Like­ wise, the residual life assessment techniques developed under RP2253 and RP2481 should be used on equipment as early in its operating life as possible. EPRI plans to work closely in the future with a number of utilities to demonstrate these methods and operating philosophies-espe­ cially the program management system. Project Manager: R. B. Dooley ESP INTERMITTENT ENERGIZATION There are over 1000 electrostatic precipitators (ESPs) in service at utility plants for fly ash emission control. The power consumption of the transformer-rectifier (TR) sets of these pre­ cipitators is affected by both ESP size and application. For high-efficiency precipitators, EPRI data indicate that average ESP power consumption costs 0.238 mill/kWh. For a 500MW unit, the power consumed by the ESP's TR sets is equivalent to about $675,000 a year on the basis of a 65% capacity factor. A recently developed method for controlling ESP power supplies-intermittent energization (IE)-has the potential to reduce the power consump­ tion of most precipitators and, under some cir­ cumstances, to improve performance as well. Many currently installed ESP controllers can be reset for IE at no cost and with no disruption of service. For controllers not now designed for IE, new solid-state circuits are available that can enable reliable, inexpensive controls to be built in for as little as $2000-$5000 per TR set. Conventional ESP controls normally operate by applying unfiltered full-wave or half-wave rectified de voltage to the discharge elec­ trodes in the precipitator. When IE controls are used, the voltage to the precipitator is turned off during preselected half- cycles or whole cycles of operation. All voltage-current control occurs on the primary side of the TR. Figure 1 a shows, in simplified form, the ac voltage applied to the primary side of the TR, the de voltage applied to the discharge elec­ trodes, and the current flow with normal de en­ ergization. Figures 1 b and 1c show voltage and current for two common IE schemes. I n Figure 1 b the primary voltage i s turned o n for one half-cycle and off for two half-cycles; in Figure 1 c it is turned on for two half-cycles and off for four. The de voltage on the discharge electrodes does not decay to zero in either IE scheme, even when the primary voltage is turned completely off, because the precip­ itator acts as a capacitor during the very short off-cycle time periods. Such IE schemes result in the consumption of much less current and therefore less power. The pattern in Figure 1 b i s said t o have a n IE ratio of 1.3; that i n Fig­ ure 1c, 2.6. These are the most commonly used IE ratios. Other ratios can also be used effec­ tively under certain site-specific conditions. To determine the effects of IE control on both ESP power consumption and perfo rmance for a wide range of operating conditions, a com­ prehensive pilot-scale test program was initi­ ated under RP1835. The work was conducted by Southern Research Institute at the Ara­ pahoe Test Facility in Denver, Colorado, using a pilot ESP that can treat flue gas at flow rates of 3500-5000 actual ft3/rnin (1 .65-2.36 m3/s). Controls capable of I E operation were installed on this ESP, and tests were conducted from July through December 1 985. The pilot ESP was equipped with an S03 conditioning sys­ tem so that the resistivity of the fly ash, which is normally high, could be varied from 1012 il-cm to approximately 2 x 1010 il-cm, the optimal COAL COMBUSTION Figure 1 The use of IE controls can reduce ESP power consumption, as illustrated by voltage and current waveforms for conventional full-wave rectified de energization (a) and two IE schemes (b, c). a. Normal full-wave rectified de energization �1� 11• �1� 11@ Time Time gl� g l___ Time b. IE with voltage on for one half-cycle, off for two half-cycles Time Time gl� g l___ Time c. IE with voltage on for two half-cycles, off for four half-cycles Jlil Time 2.5 2.0 0.5 Low-resistivity ash ( - 1 09 0-cm) High-resistivity ash ( - 1 0 12 0-cm): severe back corona o'"::--�-:c"::�--=-��..,.....,�----,-c,,-�� 1 .0 0.8 0.6 0.4 0.2 Fractional Power Consumption 0 Figure 2 Pilot-scale tests were conducted to deter­ mine the effect of IE on ESP outlet gas opacity and power consumption for fly ashes of different resis­ tivities. Fractional opacity is the ratio of IE opacity to opacity with conventional energization; fractional power consumption, the ratio of IE power consump­ tion to conventional power consumption. ESP voltage (de) g� g Time value for ESP operation. In a series of parametric tests, the opacity of the outlet flue gas was measured to determine ESP performance trends over a wide range of conditions. In a smaller number of compre­ hensive tests, total mass and particle-size­ dependent collection efficiencies were deter­ mined to thoroughly characterize performance under selected conditions. Figure 2 shows typ­ ical results from testing IE patterns of the types illustrated in Figure 1. The tests demonstrated that the effect of IE depends on the resistivity of the ash being col­ lected in the precipitator. When fly ash resis­ tivity was low, the outlet gas opacity increased slightly with any decrease in power consump­ tion. For moderate-resistivity ash, however, power consumption could be reduced signifi­ cantly with no increase in opacity. The data obtained when the precipitator was operating in severe back corona, a result of the collec­ tion of high-resistivity fly ash, are more im­ pressive. Under these conditions, opacity de­ creased and collection efficiency increased at the same time that power consumption was reduced. DIVISION R&O ST/\TUS It should be noted that IE effectiveness is more easily quantified if the ability of IE to re­ duce ESP power consumption and improve performance is related to electrical condi­ tions in the precipitator rather than to the ash resistivity that influences these conditions. Electrical conditions-defined by the voltage­ current relationship in an ESP-can be mea­ sured more precisely than ash resistivity. Mea­ suring the voltage-current relationship can help determine where in the range between spark limit (optimal, nondegraded conditions) and back corona (degraded conditions, as in­ dicated by a voltage-current plot with a nega­ tive slope) the ESP is operating. The IE tests are the first ESP tests conducted at the Arapahoe Test Facility in which a de­ crease in power consumption was accom­ panied by an improvement in performance. Flue gas conditioning and pulse energization have been used to improve precipitator per­ formance, but each process has usually re­ sulted in increased power consumption. The trends in opacity were confirmed by measurements of collection efficiency. These data show that when the precipitator was col­ lecting fly ash with either low or moderate re­ sistivity and was operating at spark limit (i.e., with little or no back corona), power con­ sumption could be reduced by almost 50% with little or no loss in performance. During the tests when the ESP was operating in back co­ rona (i.e., under degraded conditions) be­ cause of high-resistivity ash, emissions were reduced by approximately 40% at the same time that power consumption was reduced by 50%. Two factors appear to play an important role in explaining why IE can reduce ESP power consumption without significantly increasing emissions. First, only a relatively small fraction of the current in a precipitator actually charges the fly ash particles. Thus, it is at least the­ oretically possible to reduce the current with­ out sign·1ficantly reducing the charge on the fly ash particles. Second, the average voltage (and the current in the electric field) in a pre­ cipitator does not decrease dramatically when selected pulses to the primary side of the TR set are blocked. This is explained by the fact that the precipitator acts as a capacitor during the off cycles. Future IE investigations will include full-scale tests to verify the results obtained at Arapahoe and to develop full-scale application and oper­ ating guidelines. Also, IE technology may be combined with other ESP technologies that are being investigated by EPRl-for example, wide plate spacing and flue gas condi­ tioning. Project Managers: Walter Piu//e and L. F. Rettenmaier EPRI JOURNAL June 1986 45 R&D Status Report ELECTRICAL SYSTEMS DIVISION John J. Dougherty, Vice President TRANSMISSION SUBSTATIONS HVDC modulation control The unique ability to control the flow of power in HVDC systems has been exploited since the earliest days of HVDC, particularly for improv­ ing the dynamic performance of ac systems. The reactive power consumption of the de converters, however, especially those with weak ac systems, has limited the controlling actions of HVDC systems. The advent of a modern control design technique in tandem with the high-speed computation capabilities of microprocessors has significantly increased the performance capabilities of HVDC modu­ lation systems. The objective of this project is to increase dynamic performance of ac/dc systems by modulating the HVDC system's active and re­ active power flows (RP1426-4). This permits higher ac line loadings for a given reliability index. Previous projects resulted in a central modulation controller, which modulates de power and voltage in response to rectifier bus frequency to increase the steady-state stability of the ac system. This controller was imple­ mented in contemporary digital hardware (In­ tel 8086 base), and its performance was dem­ onstrated on a General Electric Co. simulator. However, the central controller requires the communication of the de voltage modulation signal from the rectifier to the inverter terminal and thereby may affect the reliability of the modulation system. The current project has ef­ fectively eliminated the communication re­ quirement by developing a decentralized con­ trol system (Figure 1 ) . Also, the small-signal modulation system was extended to include large-signal modulation to improve the tran­ sient stability of generators near the converter terminals. The control techniques of the previous projects were developed and demonstrated on small-scale synthetic ac systems with real­ istic performance limitations. Future effort is planned to demonstrate the control techniques 46 EPRI JOURNAL June 1986 on an actual large-scale power system (Mid­ Continent Area Power Pool). The current proj­ ect was extended to demonstrate the feasibil­ ity of deriving a modulation controller design model from a large-scale stability studies data base. Test results of this model will be verified by comparison with field measurements. EPRI then plans to design, build, and field-test a commercial-grade digital modulation control system for the Square Butte de system. Proj­ ect Manager: Selwyn Wright Pyrolysis and combustion of PCBs PCB involvement in utility fires has become a prime consideration in seeking to reduce po­ tential public liability. This project was initiated in response to the twin problems of contam­ inated mineral oil and high-level (up to 5%) contamination of retrofill fluids during part of the retrofill period. New York State Health Dept is examining pyrolysis and combustion products of PCBs at 100% concentration and as 5000, 500, and 50 ppm contaminants in mineral oil, silicone, and tetrachloroethylene (RP2028-4). We speak of pyrolysis here as partial oxidation at elevated temperature in an oxygen-deficient atmo­ sphere. Combustion is defined as partial or complete oxidation in the presence of a flame with adequate oxygen supply. Tetrachloro­ ethylene, although it is normally considered a replacement fluid, may (under some circum­ stances) be found in ce rtain retrofill processes. Results to date for pyrolysis and combustion of Aroclor 1254 in all three test fluids show roughly linear conversion yields of PCBs to polychlorinated dibenzofurans (PCDFs) as the PCB concentration in the feed mixture is de­ creased. Maximum yields occur around 550°C in all cases. Below this temperature, conver­ sion to PCDF is slow, while at higher tempera­ tures, PCDF is destroyed faster than it forms. Modulation t controller +--- frei;i:�i� Modulation � rectifier frequency _____., controller * C o� ""§ � o O � � D o �o Ac system HVDC line � Rectifier * Inverter - Ac system Ac line Figure 1 A decentralized control system for modulating the power and voltage of HVDC line terminals elim­ inates the need for a communication link between the inverter and rectifier of an HVDC line connecting two ac systems. This improves the overall reliability of the interconnection, while taking advantage of the in­ creased performance capabilities of modern modulation systems (high-speed computation permits rapid re­ sponse to power swings). Work on an in vitro bioassay to integrate the PCDF-like physiologic activity of all the partial oxidation products formed has run into some difficulty because of interference with the test by the bulk dielectric fluids present. Steps are being taken to mitigate this problem. The results of our laboratory trials are found to be significantly different from those of other workers in the field. This fact brings with it a word of caution in applying any of these results to real-world PCB fires (except for use as a guiding principle) because each real fire, un­ der completely random conditions, is different from all others. Even within a given fire, there will be an infinite range of competing reac­ tions. 11 can be assumed that only a small por­ tion of the combustion process has optimal conditions and that the combustion process varies with time, leading to the partial destruc­ tion of the various combustion by-products. Thus, all the research being done must be considered as setting a boundary or worst­ case condition that gives a general direction to the investigation of the real world. Another contamination material to be exam­ ined in this project will be trichlorobenzene/ tetrachlorobenzene, sometimes used as a dil­ uent for PCB. The pyrolysis and combustion products of pentachlorophenol, used in large quantities as a wood preservative, will also be examined. Project Manager: Gilbert Addis UNDERGROUND TRANSMISSION Computerized data base on dielectric materials As described in the October 1984 issue of the Journal, the intent of this project (RP7897-5) is to provide a central source of reliable data on the technical, commercial, and environmental properties of all dielectric materials. The rea­ son for establishing such a data base is that the yearly worldwide addition of some 3000 pertinent documents to the existing body of over 60,000 documents does not permit an in­ dividual involved in research, development, and manufacture of dielectrics and electrical equipment to adequately stay abreast of new developments. The data base covers several hundred properties and information items in the following categories. o Physical properties o Electrical properties and phenomena o Thermal properties and phenomena o Chemical properties and phenomena o Optical and thermoradiative properties o Mechanical properties o Flammability properties and information o Health hazards o Processability o Availability and cost o Usefulness and applications o Material characteristics o Parameters and other information One particular strength of this data base is that in addition to containing the raw numerical data as published, it includes an analysis of the data; as raw data may vary in quality, re­ liable preferred values are also available. At the user's option, access to the data base will range from written or telephoned requests to direct and interactive access from the user's computer terminal by telephone lines. Users will be able to extract data in whatever format they desire, tabular or graphic, and in what­ ever units they prefer. Further, the system will support on-line statistical analysis and mathe­ matical manipulation of the data. The data base will be exceedingly simple to use. No knowledge of computer operations is required of the user i n the interactive mode because the user is guided by a simple menu system. The initial phase of this project was the development of a pilot data base limited to dielectric fluids as a proof of concept. This phase has been successfully completed, and the data base on d ielectric fluids is now up to date. Analysis and synthesis of these data have commenced. Currently in progress is the loading of data on gases and selected solids. Commercial operation of the d ata base is ex­ pected to commence early in 1 987. This work is being carried out by the CIN­ DAS group (Center for Information and Numer­ ical Data Analysis and Synthesis) at Purdue University under the direction of C. Y. Ho. Project Manager: Felipe G. Garcia DISTRIBUTION Distribution communication, automation, and load control In the September 1982 issue of the EPR/ Journal we described a unique pilot AM broad­ cast two-way communication concept that was demonstrated in Los Angeles, California (RP1535-1). The success achieved in this southern California pilot encouraged the EPRI staff and the Distribution Advisory Task Force to proceed with a 1000-point (500 points out­ bound, 500 points return) large-scale test (RP1535-3). Philadelphia Electric Co. (Peco) was selected as the host utility because of its proximity to the heavily populated area around New York City and the inherent high volume of radio transmission activity. To use an old cliche, "if it works here, it should work any­ where." The unique communication concept em­ ployed involves synchronizing a network of re­ mote receivers to the highly reliable and accu­ rate radio carrier frequency of an existing AM broadcast (in this case station WCAU). In some applications, the return link transmitters are involved as well. The forward (outbound) link digital signals are superimposed on the broadcast signal in a noninterfering manner and are communicated at 16 bits per second to the remotely located transmitters/receivers (transceivers). Because high-power AM sta­ tion ranges exceed 100 mi (161 km), one broadcast station can cover a large portion of most utility service areas. The return link con­ sists of a VHF transmitter module (of the re­ mote transceiver) at each customer location that is synchronized with a central radio re­ ceiver. A central receiver can usually receive the VHF signals from transceivers within a 15-20-mi (24-32-km) radius. The receiver can provide several narrow-band subchannels, which, in turn, improve the signal-to-noise ratio and channel utilization of the return link by a large factor. Thus data can be returned at a rate exceeding 1 000 bits per second. Since 1982 the contractor, McGraw-Edison Co. , has been occupied with the design, test, modification, manufacture, and shipment of the system hardware. Concurrently, system al­ gorithms have been developed and computer software has been coded into operating com­ puter programs. The system computer, a VAX 1 1-730, is operational, and the installed trans­ ceivers and central receivers are being exer­ cised by the system computer to generate per­ formance data. Peco tests each transceiver as it is received and befo re it is installed in the field. One hun­ dred and fifty transceivers will be installed at customer meters and another 350 in distri­ bution substations. The transceivers are dis­ persed over a large area of the Peco system. Four central receivers are needed to effec­ tively and reliably handle the VHF return-link­ transmitted data from the appropriate trans­ ceivers. The four central receivers are installed and operating. Transceiver installation started in mid November and should be completed in January of 1986. Ten distribution automation control units have been added to the project. These DACUs will demonstrate the capability of the system to acquire analog systems data (such as current, voltage, status) convert them to digital data, and then transmit the data to the central computer. EPRI JOURNAL June 1986 47 ELECTRICAL SYSTEMS DIVISION R&D STATUS REPORT The system test will continue through June 1986. The data will then be analyzed and system performance evaluated. This EPRl­ sponsored test and demonstration of the con­ cept at Peco should confirm the ability of this technology to make more-efficient use of the frequency spectrum available for transmission of data. The possibility of using this technology to transmit data from low-cost, return-link transmitters used in distribution automation or load management functions should also be re­ vealed. Project Manager: William Shula Figure 3 The 38-kV, 1 200-A three-phase CLP that will be tested on a utility distribution system to confirm its reliability under actual service conditions. A 38-kV-class current-limiting protector A current-limiting protector (CLP) provides high continuous current-carrying capability under normal operating conditions, given ultrahigh-speed current limitation under fault conditions. The time required for a CLP to limit the fault current after sensing it is about 400 µ,s. The current is then reduced to zero within the first quarter cycle for a symmetrical fault and within the first half cycle for an asym­ metrical fault. Figure 2 shows the current wave allowed by a CLP for a 40-kA fault current, compared with that allowed during a normal three-cycle b reaker operation. Completed research established the princi- r n pie of operation and resulted in the successful field trials of a 15-kV-class device for use on distribution systems (RP1 1 42). The results of this work are reported in two EPRI final reports (EL-1 250, December 1 979 and EL-2724, No­ vember 1982). This research resulted in the test installation of numerous three-phase CLP systems at voltages ranging from 4 . 1 6 to 23 kV, which included 8 utility applications and 12 in­ dustrial plant applications to date. Following the development of the 15-kV CLP, research continued to extend the rating of the device to 38 kV. This work is cosponsored by the Empire State Electric Energy Research Corp. and conducted by Phoenix Electric Corp. The present development project (RP11 42-3) has established the following tar­ get ratings. o Rated voltage: 38 kV (rms recovery voltage across a single device) o Rated continuous current: up to 1 .2 kA rms o Ultrahigh-speed bus sectionalizing in distri­ bution substations o Prospective fault current: up to 40 kA rms symmetrical o Transformer protection to limit fault currents that can damage internal windings o Fault-sensing level: 3-15 kA instantaneous o A cost-effective solution for distribution sys­ tems where short-circuit currents exceed the interrupting ratings of protective equ ipment o Environment: indoor/outdoor V Figure 2 Short-circuit current of 57 kA peak through a three-cycle current breaker (black) and a similar fault current reduced by a current-limiting protector (color) to 21 kA. The CLP also reduces the fault energy let through from 80 to 1 .9 MJ. 48 EPRI JOURNAL June 1986 operation if the predicted fault current level is within the ratings of series-connected circuit breakers or reclosers, which will operate in­ stead of the CLP. In December 1985 the 38-kV-class design successfully passed both preprototype and prototype high-power short-circuit current test­ ing. Preparations for final design performance tests are now under way. Figure 3 is a perspec­ tive drawing of the final design. On completion of the development program, extensive field test and demonstration will be conducted on a 34-kV distribution system at a large northeastern utility. This demonstration will include staged fault testing and long-term operational testing to confirm the reliability of the CLP system under actual field conditions. CLP application satisfies several significant needs and provides the following benefits. The triggering device used to date operates the CLP when fault current exceeds the set instantaneous current triggering level (3-15 kA). Under development in the current phase is a fault-sensing system that distinguishes be­ tween symmetrical and asymmetrical currents and predicts the ultimate rms symmetrical value that the fault current would have if it were not limited by the CLP. This sensing capability will allow the CLP control logic to prevent CLP o Significant reduction in operating costs through the elimination of losses of current­ limiting reactors that are paralleled with the CLP o A replacement for fuses susceptible to fa­ tigue or incorrect operation caused by re­ peated overcurrents close to their protective level Project Manager: Joseph Porter SYSTEMS DIVISION R&D STATUS REPORT OVERHEAD TRANSMISSION Reliability-based design of transmission line structures A practical nondestructive evaluation (NOE) procedure has been developed that provides an accurate prediction of the statistical strength distribution of a group of wood poles in a transmission line (RP1352-2). This NOE method is valuable in supporting the develop­ ment of reliability-based design procedures; however, this first-generation NOE method cannot predict individual pole strength. Early project results indicated that a pole's visual appearance and even the age of the wood are not necessarily strength indicators, but that advanced sonic NOE techniques could be a very reliable means of predicting the strength of individual wood poles (Figure 4). The goal of RP1352-4 is to develop a sec­ ond-generation NOE method into a reliable and practical tool to determine the strength of individual new and in-service wood poles. The principal ongoing efforts for this project are (1) development of advanced NOE techniques to determine strength and stiffness characteris­ tics of individual wood poles, and (2) testing of wood poles removed from in-service lines to assess the rate of strength deterioration and long-term performance and to obtain calibra­ tion data required by advanced NOE methods. A second-generation NOE method under development is based on the sonic waveform spectral analysis method. This method in­ volves assessment of the sonic wave charac­ teristics as the wave travels through the wood pole. This far more powerful NOE method has the potential to accurately grade new wood poles according to their strengths and to as- sess in situ, in-service pole strengths. On the basis of evaluations to date, the second­ generation NOE method can predict the strength of an individual pole with far greater accuracy than is currently possible with visual inspection and decay detection methods. For both new and in-service poles, the typical de­ viation for the relationship between measured pole strength and predicted pole strength by NOE measurements is 600 psi (4 MPa). If a utility is to use the second-generation NOE method to determine the strength of indi­ vidual in-service poles, the NOE method must be calibrated for the specific geographic area and for pole species by a limited number of full-scale destructive pole tests (Figure 5). The need fo r additional in-service wood pole data for the second-generation NOE method and the need of utilities fo r in-service data on wood pole strength and rate-of-degradation assessments can both be satisfied with a com­ bined test effort that provides all the data needed by both parties at a cost saving to each. Ongoing EPRI testing with Engineering Data Management, Inc. (EDM), and utilities, which collects this information on in-service poles from various geographic regions of the United States, has been under way since early 1985 and will continue until mid 1 986. Through EDM, utilities can arrange for in­ service poles to be tested to destruction to provide the needed data. This effort can be an integral part of a wood pole management pro­ gram. As part of the basic pole test program, additional measurements and NOE parame­ ters are recorded for evaluation as part of the development of the second-generation NOE method. The total cost of testing to the utility is $250 per pole, plus the cost of shipping the Figure 4 Full-scale testing of a new wood pole determines its mechanical properties. Figure 5 The data necessary for determining in­ service pole strength is recorded by a second­ generation NDE method. poles to Fort Collins, Colorado. Utility funding covers the direct cost of testing the wood pole to failure and reduction of data. EPRl's funding of approximately $450 per pole and the addi­ tional support by EDM covers indirect test costs, acquisition of additional NOE data, and further development of the sonic waveform NOE method. More than 150 in-service poles were tested as part of this program in 1985. Approximately 300 additional poles are promised by utilities throughout the United States for 1986. In addi­ tion, several utilities have already begun ex­ tensive field programs that use the second­ generation NOE method to determine the individual strength of in-service wood poles. Their participation in the in-service wood pole test program provided the necessary cali­ bration data for the NOE method. The practical application of this break­ through in technology is now available for field use by interested utilities. Existing laboratory NOE equipment and off-line evaluation of the NOE data are being used to provide individual wood pole strength values. EPRI expects to have sufficient new and in­ service wood pole correlation data available so that development of commercial second­ generation NOE hardware-a soundwave an­ alyzing black box-can begin in 1986. This hardware will provide instantaneous, direct readout of individual pole strength and elimi­ nate the need for the present off-line data re­ duction to determine pole strength. Project Manager: Paul Lyons EPRI JOURNAL June 1 986 49 R&D Status Report ENERGY ANALYSIS AND ENVIRONMENT DIVISION Rene Males, Vice President ADVANCED PLANNING RESEARCH The external business environment of utilities has become less predictable and more com­ petitive. Further, traditional capacity options are not necessarily as desirable or as feasible as in the past, and utilities are broadening their strategies to include options with which they have less experience. In light of these changes, the tasks of decision making, policy formu­ lation, and strategic planning have become less routine and more complex for utilities. New analytic tools and quantitative information are desirable to support these tasks. By help­ ing define the implications of decisions, they can serve to confirm intuition or else change the "mental map" of the decision maker. The communication of useful information to sup­ port decisions helps bridge the gap between the decision maker and the analyst. The ad­ vanced planning research summarized in this report, one focus of the strategic planning sub­ program of EPRl's Energy Resources Pro­ gram, is aimed at bridging this gap. Utility modelers and analysts have expressed disappointment at the apparent lack of impact that good, carefully written analysis has on ma­ jor decisions. Surprisingly, the same disap­ pointment has been voiced by top utility offi­ cials. This phenomenon has also been noted in many other industries and policy-making pro­ cesses. In the professional literature, it is often referred to as a crisis in management science or operations research. Apparently, a good quantitative analysis of a problem as perceived by the analyst is not enough; factors perceived as important by the client (the decision maker) must also be ad­ dressed, and the results must be communi­ cated-a two-way process not completely un­ der the analyst's control. To be effective, then, the analyst not only must conduct relevant analysis, developing new quantitative tools as necessary, but also must devote a large amount of time to communication. The strategic planning subprogram devel- 50 EPRI JOURNAL June 1986 ops tools and information to support strategic problem solving in utilities: it analyzes the ef­ fect of nontraditional technology options, de­ velops new tools for risk analysis, and identi­ fies and studies risk management strategies. The usefulness of the resulting research prod­ ucts depends on how well they fit with utility planning and decision processes and on how well the information they produce is commu­ nicated-that is, on their ability to bridge the gap between the analyst and the decision maker. Thus the subprogram's research on ad­ vanced planning methods is an important component of the effort to support utility strate­ gic problem solving. Described below are sev­ eral projects that seek to develop quantitative and nonquantitative approaches to the prob­ lem of bridging the gap. This research, mo­ tivated by a utility-articulated need, will also provide insights to EPRI to help it place its re­ search products into practical use. Utility Modeling Forum Under RP1303 the Utility Modeling Forum (UMF) has applied existing utility computer models to a range of current issues with the objectives of stimulating the transfer of mod­ eling capability across the industry, promoting effective interaction between model develop­ ers and users, and identifying research needs. In the course of this work, the notion contin­ ually surfaced that how models are integrated into utility decision and planning processes is as important as the technical characteristics of the models themselves. In a 1981 UMF report comparing utility cor­ porate models (EPRI EA-2065), the way pro­ cess issues affect the usefulness of models was evaluated by examining the planning ap­ proach of the user organization, model credi­ bility, model design and development, and model applications. In particular, the UMF working group noted the importance of as­ sumptions testing, communication skills, and broader direct user involvement with models. The UMF catalyzed the development of many of the research projects discussed be­ low, and its key findings have been confirmed time and again by in-depth analyses of the role of models in the decision-making process. Planning methods used outside the utility industry The objectives of this research are to identify planning methods in other industries that have possible value for the utility industry, to apply such methods to specific utility issues, and to develop a rigorous understanding of the infor­ mation required to support complex, important utility decisions (RP1634). To date, a catalog of planning tools (EA-3793) and two reports on utility case study applications involving mar­ keting planning (EA-3736 and EA-3794) have been published. By 1984 utility analysts and managers had given EPRI much evidence suggesting that quantitative studies typically had little effect on decision making. To see if the analysts' as­ sessment was shared by the users of the anal­ yses, nine utility chief executives were sur­ veyed as part of RP1634. The survey results (which were reported in Energy Analysis and Environment Division technical newsletters in February and June 1985) confirmed the ana­ lysts' observations. In general, the executives stressed that most important decisions were driven by qualitative factors and that judgment and intuition, rather than technical analyses, played the dominant role in the decision pro­ cess. They were unanimous in suggesting that analysts needed to improve their commu­ nication skills. It is clear from the survey results that if exec­ utive decision processes are to be supported, the nature of these processes must be under­ stood. The early work in RP1634 helped define the current research activities, which include the development of a descriptive model of ma­ jor decision-making processes in utilities and the preparation of a concept paper supporting the need for a new decision-making paradigm in the utility industry. In another important ef- fort, researchers are defining a new category of techniques to support ill-structured, unique decisions-para-analytic (or nonquantitative) techniques. These will augment quantitative tools for problem solving and planning but will focus more on improving the decision process in terms of group decision making, innovative­ ness, and communications. These techniques will be tested this summer for a number of rep­ resentative and diverse utility problems. A guidebook on their use is also being devel­ oped. This research area continues to offer com­ plex challenges, but it has the potential in the long run to create a new type of professional discipline for supporting decisions, one that combines an understanding of technical anal­ ysis with an understanding of the processes involved in quality decision making. EPRI is receptive to any additional participation in this project by its utility members. Simplified models for planning support The original focus of this effort was to develop a comprehensive approach for integrating fuel and investment decisions in utilities (RP2372). Four major challenges to integrated planning and decision making were identified: uncer­ tainty from many sources, functional integra­ tion across many different departments, multi­ ple criteria for choosing a course of action, and the dynamic of the decision environment. The project team concluded that current planning tools were, by and large, inadequate for effec­ tively meeting all these challenges. It cited the obstacles to communicating complex informa­ tion within organizations as an especially big problem. Several ways to surmount the above difficul­ ties were recommended, including the use of simplified but representative models of the de­ cision problem called response models. There are two kinds of response model. One is a unique model created by an analyst or an ex­ ecutive to describe the decision system. The other is derived by using various kinds of re­ gression equations to mimic the relationship of inputs and outputs of a large, complex model. Designed fo r particular issues, response mod­ els are accurate for the intended purpose but may not have many other uses. The objective of the current research is to develop a guide­ book that describes the development and ap­ plication of response models for problem solv­ ing and planning and that provides examples of their use in real utility situations. The project's first report (EA-4166) presents a general discussion of integrated fuel and in­ vestment planning. Significantly, the project team concluded that the ultimate challenge is to successfully apply existing analytic tools to decision making within current organizational structures. A discussion of planning concerns cited by individual utility planners is included in the report. Planning for nontraditional options As part of a project on demand-side manage­ ment (DSM) funded by the Demand and Con­ servation Program (formerly in the Energy Analysis and Environment Division and now in the Energy Management and Utilization Division), the strategic planning subprogram managed a study of the way DSM program objectives are developed in a utility organiza­ tion and the role that analysis and quantitative techniques play in setting those objectives (RP2548). The study's final report (EA-4220) empha­ sizes the importance of organizational issues in the initial planning of DSM programs. Many of the utility personnel interviewed cited cul­ tural, institutional, and process problems, rather than analytic problems, as the primary barriers to implementing DSM programs. As the report notes, "Many of the issues faced by demand-side planners are soft and subjective and do not lend themselves to traditional utility analytical planning techniques." The report also includes a case study of the DSM objec­ tive-setting process at Carolina Power & Light Co. Extending power plant life appears to be an­ other attractive nontraditional option for utilities in the present business environment, and a study similar to the DSM study is under way for plant life extension programs (RP2074). The project team will document how utilities cur­ rently link corporate objectives and life exten­ sion strategies and will develop options for im­ proving this process. About 10 utilities are participating in the study. Interviews being conducted with their personnel focus on four areas: the business environment, application of quantitative tools, organizational structure, and process issues. A final report is expected this summer. Other advanced planning methods Utilities have used scenario analysis for some time to assess the effects of alternative futures on the outcomes of current decisions. The most obvious example is the high-medium-low set of load forecasts typically produced by many utilities. However, there has been little systematic scenario planning focused on the testing of assumptions and the selection of a population of scenarios that explore, in a self­ consistent way, the potential unfolding of events. An early effort to compare scenario analysis with probabilistic approaches was conducted by the UMF. A scoping study is now under way to explore how the practice and methods of scenario planning might be applied to utility planning (RP2379-12). The study's tasks in­ clude a literature review, a workshop to bring senior utility planners and experts from other industries into a creative discussion about scenario-planning applications, and a repo rt based on the workshop findings. Another advanced analytic approach under study is the multiple perspectives approach. Its developer, Harold Linstone, recognized that many important real-world business and policy decisions deal with situations that are ill struc­ tured and systems that are sociotechnical. The purpose of the multiple perspectives approach is to bridge the gap between analysis and the real world of ill-structured systems. Three types of perspectives, all dealing with the "how" of seeing a problem rather than the "what," are used: technical/analytic, organizational/soci­ etal, and personal/individual. Each perspec­ tive gives important insights not obtainable with the others. The motivation for the multiple perspectives approach is similar to EPRl's motivation in de­ veloping a research area on advanced deci­ sion-making, policy-making, and planning ap­ proaches, and a small scoping study on the potential application of multiple perspectives in the utility industry has been sponsored (RP2379-13). The study's final report (available early this summer) will describe the multiple perspectives concept, present a number of applications in typical utility decision-making situations, and suggest how the concept can be more generally applied in the utility industry. An agenda for further research will also be de­ veloped, if appropriate. Another project is focusing on the devel­ opment of an integrated technical approach to utility planning (RP2807). Called CHOICE, this approach consists of three steps: developing risk profiles of technology options, analyzing the effect of these options on a utility's corpo­ rate objectives (taking into account the uncer­ tain business environment), and developing risk management approaches for combining options into risk-resilient strategies. To be useful in practice, the CHOICE ap­ proach must consider the process issues as­ sociated with integrated planning as well as the technical issues. Thus EPRI expects that the findings of the planning-process research projects described above will play a major role in supporting the CHOICE project and ensur­ ing that the finished product will be useful to individual utilities. In summary, several projects in the strategic planning subprogram address aspects of the process of decision making and planning, and a significant number of research results will be reported in 1986. EPRI believes that this re­ search will provide critical suppo rt to the utility EPRI JOURNAL June 1986 51 ANALYSIS AND Ef\lVI RONMENT DIVISION R&D STATUS REPORT industry at a time when its decision situations involve an increasing number of uncertainties and are very ill structured. In addition, the re­ search findings will influence the type of quan­ titative tools developed by EPRI and by man­ agement science and operations researchers in general. Acting Program Manager: Domi­ nic Geraghty SOLID-WASTE ENVIRONMENTAL STUDIES A major concern associated with the land dis­ posal of utility solid wastes is the release and migration of solutes to groundwater. Protecting groundwater from contamination is a principal objective of regulations being developed un­ der the Resource Conservation and Recovery Act, EPA's Groundwater Protection Strategy, the Safe Drinking Water Act, and various state and local rules. Requirements that affect waste disposal are of concern to the electric utility industry because it generates over 80 million tons of solid wastes annually. To decide when and to what extent control technologies should be applied, reliable predictions of the mobili­ zation and environmental fate of leachates are needed. Actions based on a less than ade­ quate understanding could result in controls that are either more or less stringent than re­ quired for safe disposal; in either case costs could increase for the utility industry. In re­ sponse to this need, EPRI began research to develop methods for determining how the dis­ posal of solid residues from fossil fuel com­ bustion influences groundwater quality, and in 1983 this work was consolidated into the solid­ waste environmental studies (SWES) project (RP2485). The goal of SWES is to develop and validate methods (including new data) for predicting the release, transformation, transport, and ulti­ mate fate of chemicals from utility solid wastes. Planned activities under SWES include evalu­ ating existing predictive models, assembling interim models, and conducting experiments to quantify waste leaching behavior, physical transport, and chemical attenuation. In addi­ tion, a special study is quantifying the un­ certainty associated with field measurement methods and groundwater sampling tech­ niques. Over the longer term, researchers will develop improved geohydrochemical models and validate them with field data collected under the SWES project. Leaching chemistry studies To be able to predict waste solute concen­ trations in groundwater, we mustfirst be able to predict the release (concentrations and du­ ration) of solutes from a waste disposal site. Yet few data are available for making such pre- 52 EPRI JOURNAL June 1986 dictions. The SWES project seeks to develop a mechanistic understanding of the processes that govern waste dissolution. Toward that goal A. D. Little, Inc., conducted a limited number of pilot studies to test the feasibility of using slurry batch experiments and elevated temperatures to quantify long-term leaching (RP2485-4) . Three different waste types-coal ash, oil ash, and flue gas desulfurization (FGD) sludge­ were leached, with and without leaching solution renewal, at either 20° or 90°C. The results (reported in EPRI EA-4215) indicate that simple washout, equilibrium sorption reac­ tions, equilibrium dissolution-precipitation re­ actions, and rate-limited dissolution processes are i nvolved in d etermining which chemicals are released. In June 1985 Battelle, Pacific Northwest Laboratories i nitiated further studies on waste leaching chemistry (RP2485-8). One objective was to develop data on the reproducibility of EPA's extraction procedure (EP) and toxicity characteristics leaching procedure (TCLP). In round-robin testing completed last February, three laboratories evaluated the procedures on three fly ashes, two bottom ashes, and two FGD sludges. This testing has found that compared with the EP method, the new TCLP method generally yields higher concentrations in the extracts for the inorganic elements ana­ lyzed. Approximately 55% of these differences fall in the ± 25% range; about 19% of them are 100% or larger in magnitude. The TCLP method is easier to implement and more pre­ cise than the EP method. A report on this study is scheduled for publication this summer. Other work is under way to develop equi­ librium and kinetic data for use i n predicting waste leaching rates and waste interstitial solution concentrations for fly ash, bottom ash, scrubber sludges, oil ash, and fluidized-bed combustion wastes. These studies will use approximately 100 waste samples from some 40 power plants. The planned laboratory- and field-scale experiments will be conducted through 1991, and reports will be published as various parts of the research are completed. Subsurface transport studies Once waste solutes enter the subsurface environment, they are transported by advec­ tive and dispersive processes. Advection i n groundwater systems is relatively well under­ stood; dispersion, which can significantly af­ fect solute concentrations, is not. To quantify dispersion in the subsurface environment, re­ searchers need a better fundamental under­ standing of its causes. Most groundwater solute transport models represent dispersion mathematically by using dispersivity coefficients, which are assumed to be constant and independent of the perme- able media. Recent work indicates, however, that dispersion in groundwater most likely re­ sults from variations in the hydraulic conduc­ tivity of the permeable media and the inter­ mingling of water flow paths. Dispersion appears to increase with distance of travel from the solute source, but this apparent in­ crease and its relationship to the properties of the permeable media have not yet been quantified. EPRI EA-41 90, which is a review of subsurface solute transport processes, pre­ sents details. Under SWES two fundamental studies are focusing on dispersive processes. In the Mac­ rodispersion Experiment (MADE), researchers from the Tennessee Valley Authority and the Massachusetts Institute of Technology are studying dispersion in the saturated ground­ water zone (RP2485-5) . By performing tracer experiments in a well-characterized aquifer, the researchers hope to relate dispersion to measurable physical properties of the subsur­ face environment so that proper predictive for­ mulations can be developed. EA-4082 de­ scribes the experimental design. A heterogeneous aquifer was selected for MADE in order to represent typical utility site conditions. The field site is located at Co­ lumbus Air Force Base in Mississippi in an area with regulated access. During 1 985 soil coring, geophysical measurement, and pumping tests were performed to quantify the aquifer's char­ acteristics, including the spatial distribution of hydraulic conductivity. Also, to aid in experi­ mental design, simulations of the planned tracer injection were made with existing math­ ematical models. Seven tracers are to be in­ jected in the aquifer, and monitoring of the plume's movement will continue for two to three years. A special gas chromatography­ mass spectrometry probe system, developed in part under the SWES project, will be used for the direct determination of plume tracer con­ centrations in the field. Dispersion in the unsaturated zone is being investigated by researchers from the Univer­ sity of California at Riverside in a study co­ funded by Southern California Edison Co. (RP2485-6). Two sites in southern California, Etiwanda and Moreno, are being used. An analysis of data from previous tracer experi­ ments (a steady-state bromide application at Etiwanda and chloride applications at both sites) has indicated that with downward seep­ age, dispersion increases with distance over at least the upper 3 m of the soil. A pilot tracer experiment is now being set up at the Moreno site, and full-scale experiments at both sites are to begin late this year. The depth to the saturated zone, about 100 m, is representative of dry conditions in the West, where unsatu­ rated flow takes on greater importance. ENERGY ANALYSIS AND Ei\JVIROI\IMENT D IVISIOI\I R&D STATUS REPORT Chemical attenuation studies Besides the physical processes discussed above, chemical interactions with the g round­ water and the permeable media can cause large changes (typically decreases) in solute concentrations. Some solutes, like aqueous lead, may undergo concentration decreases of several orders of magnitude when passing through less than 1 m of soil; others, like chlo­ ride, may undergo no measurable concen­ tration decrease over distances of hundreds of meters. Most solutes experience soil-chemical interactions between these two extremes. Which solutes will be removed from solu­ tion? To what extent will removal occur and under what conditions? How can it be reliably predicted? These questions are being ad­ dressed by Battelle, Pacific Northwest Labora­ tories i n the SWES chemical attenuation stud­ ies (RP2485-3). The SWES data collected thus far indicate that chemical attenuation is due mostly to three processes: precipitation, co­ precipitation, and adsorption. These pro­ cesses have been shown to be highly de­ pendent on the chemical characteristics of the groundwater and the soil materials (EA-3356). Adsorption is important tor solutes that do not readily precipitate and also, to a lesser ex­ tent, fo r some solutes that do precipitate. It is typically represented by a distribution coeffi­ cient (Kd ), which is the ratio of the concen­ tration of the adsorbed chemical to the con­ centration of the chemical i n solution; higher Kd values indicate g reater chemical attenuation. The adsorption of solutes is strongly de­ pendent on such factors as the solution's pH and ionic composition. SWES data on the ad­ sorption of the leachate species chromate on a common soil constituent (Fe203 · H 20), for ex­ ample, show that Kd increases as pH de­ creases (Table 1)- The data also show that some of the common macroions compete with chromate for adsorption sites, thereby reduc­ ing chromate adsorption. These results indi­ cate that Kd is not a constant and that a more sophisticated approach is necessary to prop­ erly model the chemical attenuation of solutes. SWES research has also shown that chro­ mate concentrations decrease because of the conversion of chromium(VI) to chromium(III) by very small amounts of reduced iron present in hematite and biotite, common soil minerals. Chromium(III) in the subsurface environment was found to be controlled by precipitation­ dissolution reactions. Chromium hydroxide and chromium iron hydroxide were identified as the two solid compounds important i n keep­ ing chromium(III) concentrations at levels well below the drinking water standards in the pH range of 5 to 10. A complete thermodynamic and kinetic data base has been developed for chromium (EA- Table 1 CHROMATE ADSORPTION ON AMORPHOUS IRON OXIDE Distribution Coefficient (Kd ) Solution pH Simple Solution* Complex Solutiont 5.65 478,600 1 8,600 6.94 1 1 2,200 5,370 7.80 6,600 130 '0.1 molar NaNO, solution. tsolution with Na, N03, Ca, 804, and CO2 ions. 4544). Several soils from utility sites have been used in the experiments to ensure that the re­ sults have practical application. Similar data bases are being developed for arsenic, boron, cadmium, selenium, vanadium, and zinc. Sev­ eral reports will be published during the course of this research, which is scheduled for completion by the end of 1990. Field sampling methods Model calibration, validation, and assessment at specific waste disposal sites all require sampling and analysis of the subsurface envi­ ronment. Sampling is generally more costly for groundwater than for surface waters because of the cost of installing wells. The analysis of groundwater solutions is also more difficult, in part because of higher total dissolved solids concentrations. Further, groundwaters are of­ ten supersaturated or undersaturated with re­ spect to atmospheric gases. It has long been suspected that the sampling methods used may significantly alter what is measured. To date, little effort has been devoted to quan­ tifying the uncertainties introduced by sam­ pling methods. The first phase of the SWES research in this area (RP2485-7) focused on quantifying the errors in groundwater quality measurements caused by sampling devices (EA-4118). The researchers found that some commonly used sampling devices and procedures introduce enough oxygen to convert dissolved iron(II) to iron(III), which is relatively insoluble and rap­ idly precipitates from solution. Solutes, partic­ ularly metals that are regulated, either copre­ cipitate with the i ron(III) or adsorb onto the freshly formed amorphous iron oxide precip­ itates; thus the measured solution concen­ trations of these chemicals are reduced. Various sampling and measurement tech­ niques are being field-tested in a 30- by 30-m coal ash test cell at the Montour power plant of Pennsylvania Power & Light Co., which is cofunding the research. Preliminary results indicate that both tensiometers and neutron probes provide reliable measurements of the infiltration of moisture in an ash landfill. A re­ port on this work will be published later this year. Interim model development The models being improved under SWES will be used to predict the groundwater concen­ trations of solutes leached from utility solid wastes . SWES researchers have reviewed and tested about 100 existing groundwater mod­ els; the results are presented in EA-3417. Ef­ forts are now focused on the development of an interim model to simulate both the hydro­ logic and geochemical behavior of utility waste leach ates. Called FASTCHEM, this interim model is be­ ing developed by Battelle, Pacific Northwest Laboratories and is scheduled for comple­ tion in late 1987 (RP2485-2). FASTCHEM will simulate the flow of water and the transport of solutes through saturated and unsatu­ rated subsurface environments. The hydro­ logic component of the model uses a two­ dimensional finite-element solution technique. Groundwater flow will be routed along stream tubes. Chemical reactions will be simulated by using a geochemical subroutine. The coupling of the hydrologic and geochemical routines represents a major technical effort. The stream tubes will be divided into series of discrete cells for modeling the geochemical reac­ tions; an advective transport step will follow. Dispersive transport will be simulated by using a Markov model. The probability distribu­ tion function can be selected to accommo­ date either Fickian or non-Fickian dispersion representations. In March 1986 EPRI released an IBM PC code called MYGRT for use i n analyzing solute migration in g roundwater (EA-4543). Screen­ ing studies with MYGRT can determine if more­ detailed analysis is necessary. Developed by Tetra Tech, Inc. (RP2485-1), MYGRT contains both a one-dimensional code and a two­ dimensional code. A prerelease version was tested by several utilities, and recommended improvements have been incorporated into the final version. In conclusion, the solid-waste research of EPRl's Environmental Physics and Chemistry Program has produced results that are already being put to use by the utility industry and by researchers and regulators. A total of 29 re­ ports h ave been published to date. Two tech­ nology transfer seminars have been held, and a third i s scheduled for October 1 986. Project Managers: lshwar P. Murarka and Dave A. McIntosh EPRI JOURNAL June 1986 53 R&D Status Report NUCLEAR POWER DIVISION John J. Taylor, Vice President RELIABILITY-CENTERED MAINTENANCE AT NUCLEAR PLANTS The nuclear power industry is continuously striving to improve plant reliability and avail­ ability without compromising stringent safety standards. Reliability is manifested by the ex­ tent to which operations are continuous and trouble-free; availability is evidenced by a low level of unscheduled downtime. Improvements in design, in operating procedures, in techni­ cal specifications for operation, and in mainte­ nance are some of the ways in which reliability and availability can be positively affected. Re­ cent attention has focused on improving strat­ egies for preventive maintenance (PM), which is maintenance performed at regularly sched­ uled intervals to prevent failures, to detect in­ cipient failures, or to check for hidden failures in off-line systems. A PM approach used successfully by the air­ lines and the military is reliability-centered maintenance (RCM). Simply stated, RCM is a systematic consideration of system functions and the ways in which functions can fail, plus a priority-based consideration of safety and eco­ nomics that identifies applicable and effective PM tasks. A 1984 EPRI study identified RCM as one of seven candidate areas for technology transfer from commercial aviation to the nuclear power industry (NP-3364). Using RCM, the airlines had no increases in unit maintenance costs over a 1 6-year period despite increases in the size and complexity of the aircraft they oper­ ated. And during the same period, the airline safety record steadily improved. EPRl's study concluded that the airlines' control of mainte­ nance costs and their improving safety record were due in large measure to the systematic identification of applicable (i.e., it works) and effective (i.e., it's worth the cost) PM tasks that result from the RCM process. EPRI and its utility advisory boards recom­ mended that the usefulness of RCM be evalu­ ated in trial applications at selected nuclear 54 EPRI JOURNAL June 1986 Information collection (step 1 ) is the gather­ ing of system design and operation documen­ tation, existing PM procedures and practices, corrective maintenance records for the sys­ tem, and information related to failure experi­ ence with the system and with other, similar systems. The latter information is of particular value in highlighting areas where new or im­ proved PM tasks can have significant effect. Identification and partitioning (step 2) re­ quires that system boundaries be identified and that inputs and outputs across the bound­ aries be defined. Each system is then sepa­ rated into functional subsystems, which en­ sures that appropriate functions and related functional failures are clearly distinguished as inputs to succeeding steps. The requirements analysis (step 3) is the heart of the RCM process. In the analysis, functionally significant items are identified, a functional failure analysis is performed for each item, and dominantfailure modes associ­ ated with these failures are established in a failu re modes and effects analysis. Dominant failure modes are those that occur frequently power plants. Florida Power & Light Co. (FPL) and Duke Power Co. were the hosts for two pilot applications. The contractors were Los Al­ amos Technical Associates, American Man­ agement Systems, and Saratoga Engineering Consultants. The ACM technique The RCM process comprises three major ac­ tivities. o Identifying functions and functional failures o Establishing the importance of functional failures and failure modes (by using a decision logic tree) o Defining PM tasks, but with the specific ob­ jective of following the established priorities (selecting only those PM tasks that are both applicable and effective) The RCM approach (Table 1) is imple­ mented in five basic steps: (1) information collection, (2) identification and partition, (3) requirements analysis, (4) PM task selection, and (5) packaging. Table 1 RCM IMPLEMENTATION STEPS Information Collection Identify RCM analysis team Identify data sources Collect/compile data 2 Identification and Partition Describe systems Identify constituent system elements Define zonal locations (if necessary) Define system boundaries and interfaces 3 Requirements Analysis 4 5 PM Task Selection Packaging Identify functionally significant items Apply RCM decision logic Compare with existing tasks Define functions Identify potential PM tasks Detail task instructions Define functional failures Identify dominant failure modes or functional failures Select applicable and effective tasks Revise task schedules Establish task intervals Install revised plan Establish causeeffect data Identify design modifications Audit Pursue redesign options or are so injurious or costly that they must not be allowed to occur at all, at least insofar as that is possible. I n addition, the dominant fail­ ure modes must be identified at a level at which a preventive task can be perfo rmed (an emphasis that is basically different from that in the design process). An RCM decision logic tree is used for each dominant failure mode in order to select the PM tasks (step 4). By using the decision logic, each dominant failure mode is addressed by an appropriate PM task, by a design change, or by judging the failure mode to be accept­ able. The process specifies (1) if the PM task should be done at a fixed interval, for example, time or cycles (time directed); (2) if it should be done at some predefined level of equipment performance or condition (condition directed); or (3) if a surveillance or failure-finding task is needed. In step 5, the PM tasks are packaged; that is, they are defined in detail sufficient for their implementation into actual maintenance pro­ cedures. RCM applications The first application of RCM was in the com­ ponent cooling water systems (CCWSs) of FPL's Turkey Point units 3 and 4 at Florida City, Florida (NP-4271 ) . The CCWS removes heat from plant components during normal oper­ ation, as well as after potential accident­ initiating events. CCWS was selected fo r the pilot study for two reasons: PM tasks that already existed involved several CCWS main­ tenance areas-electrical, mechanical, and instrumentation and controls. Further, the cor­ rective maintenance (CM) load on the systems had been high. These high levels of PM and CM indicated the possibility of major benefits as a result of an RCM effort. All five steps of the RCM procedure were completed for the CCWSs at the Turkey Point stations. The pilot study identified 180 func­ tional failures from 66 separate functions per­ formed by the system; this, in turn, generated about 350 dominant failure modes, which were reduced (on the basis of no significant safety or productivity effect) to 50 fo r analysis by means of the decision logic tree. The study recommended 24 tasks that differed from the existing ones. Of the tasks recommended, 6 were time directed (TD), 7 were condition di­ rected (CD), 4 involved failure finding (FF) and 7 suggested design changes. A project team then drafted changes to the Turkey Point plant procedures for the TD and CD preventive maintenance tasks that were identified. This package of suggested PM changes will be a part of the evaluation of the plant PM program now under way at Turkey Point. The application of RCM to the main feed­ water system (MFWS) at Duke Power's William B. McGuire station, north of Charlotte, North Carolina, was begun near the end of the Turkey Point pilot study. The MFWS at McGuire had recently been modified in response to oper­ ational problems and corrective maintenance early in the plant's life. The RCM pilot study team investigated the system as it was before those modifications were made. The results of this RCM application (although not final at the time of writing) identify 22 possible CD tasks and 28 FF tasks. In addition, the study team recommended possible plant changes in two areas, one of which had been implemented earlier independently of the RCM study. Pre­ liminary comparisons of the recommended RCM tasks and the current PM tasks, as iden­ tified by interviews of plant personnel, indicate that many of the RCM tasks are not now being addressed. Of the RCM-identified tasks that were already covered by current PM, the RCM team recommended that 25% be changed from TD to the more effective CD tasks. Thus, it appears that the RCM process not only identified the major elements of the cur­ rent PM program at the McGuire station but that it also developed desirable candidate CD tasks as replacements for existing TD tasks. The RCM process also verified the need for action in areas in which the plant is known to have been modified. In summary, the RCM process provided means of achieving the same PM program in a potentially more cost-effective manner, and it also suggested two design changes and addi­ tional FF tasks that concern the prevention of operational surprises. These trial applications demonstrate that RCM can be a valuable part of an overall plan to develop or evaluate a PM program at a nuclear power plant. Because RCM investigates dominant failure modes, it can help eliminate ineffective tasks, often changing costly time-directed tasks to less fre­ quent condition-directed ones and identify­ ing missing tasks that can prevent significant failures. With the completion of these pilot studies, the way is clear for further applications of RCM. The FPL pilot application is reported in NP-4271 ; the Duke Power application report is in preparation. These EPRI reports document in detail the analyses, including system de­ scriptions, analysis worksheets, and specific recommendations. The reports make possible a thorough evaluation of RCM by prospective utility users. In addition, the reports point out practical considerations and limitations of RCM, including a way to select systems for which RCM is most likely to be cost-beneficial and the importance of involving plant person- nel in the RCM analysis process. Manager: John Gaertner Project CHEMICAL CLEANING UPDAT E EPRI and the Steam Generator Owners Group (SGOG) have been working to resolve steam generator problems since 1977. In a recent major accomplishment, Northeast Utilities used a slightly modified version of an SGOG­ developed chemical process to clean the sludge pile regions of the steam generators at Unit 2 of the Millstone Nuclear Power Station. The cleaning removed harmful corrosion prod­ ucts that accumulate in steam generators dur­ ing normal operation. The Millstone cleaning by Northeast Utilities was the successful cli­ max of a seven-year, multimillion-do/lar SGOG­ sponsored laboratory and mockup testing project. As a result of that work, utilities now have an additional tool with which to combat steam generator degradation. The impurities and corrosion products that ac­ cumulate in steam generators contribute to corrosion-induced damage on the secondary side of pressurized water reactor (PWR) gen­ erators. Corrosion products form in the steam, feed, and condensate system, and then they are transported by the feedwater into the steam generator. There they are deposited on generator tubes and support structures. Sev­ eral hundred pounds of this sludge can be deposited each year. The sludge in mixed­ alloy systems consists principally of magnetite (Fe304) , copper, and copper oxides (CuO and Cu20). Chemically active impurities from such sources as condenser cooling water leakage concentrate in the sludge and corrode steam generator tubes and support structures. The result can be stress corrosion cracking, inter­ granular attack, tube deformation (denting), wastage, and tube pitting. Removal of the sludge can significantly improve steam gener­ ator reliability. Water chemistry improvements at many util­ ities have reduced the amount of sludge enter­ ing steam generators, but previously depos­ ited sludge, coupled with the smaller amounts still entering the generators, remains a serious problem at many plants. Water-lancing to re­ move the sludge can be effective in removing the softer deposits, but much of the harder sludge is unaffected by lancing. Chemical cleaning can, however, complement and in­ crease the effectiveness of sludge removal. To deal with this problem, SGOG developed a two-step chemical cleaning process that uses the chelating agent ethylenediamine­ tetraacetic acid (EDTA). Table 2 lists the con­ stituents of the generic sludge-removal sol- EPRI JOURNAL June 1986 55 NUCLEAR POWER D IVISION R&D STATUS REPORT Table 2 SGOG GENERIC SLUDGE REMOVAL SOLVENTS Iron Oxide Solvent 200°F Copper Oxide Solvent 100° F 1 0% EDTA* 5% EDTA 1 % hydrazine 2-3% hydrogen peroxide o.5% cc1-so1t Ammonium hydroxide to pH 7.0 Ammonium hydroxide to pH 7.0 Ethylenediamine to pH 10 *Ethylenediaminetetraacetic acid. tNonproprletary corrosion inhibitor, vents for iron and copper. Before applying these solvents, the utilities generally perform site-specific qualification testing. Cleaning preparation Pitting of steam generator tubes at Millstone Unit 2 was first detected in 1981 at the end of the fourth fuel cycle. The pitting was limited to the region below the first tube support within the sludge pile area. Water-lancing was only partially successful in removing the accumu­ lated sludge, and plans were made to use a chemical cleaning process. The SGOG ge­ neric solvent was selected as the basis for i qual fication, followed by significant utility­ sponsored, site-specific qualification. Ulti­ mately, the SGOG iron solvent and a slightly modified version of the SGOG copper solvent was used at Millstone. Northeast Utilities was the overall program manager and sponsor for the Millstone clean­ ing. Supporting contractors were Combustion Engineering, Inc., for solvent qualification, on­ site corrosion monitoring, laboratory analyses, and cleaning-process consultation; Pacific Nu­ clear Services for hardware procurement and installation assistance, solvent mixing, solvent handling and recirculation, and the process in­ strumentation and control system; and London Nuclear Associates for waste disposal. The chemical cleaning system used at Mill­ stone was designed to provide fo r (1) primary­ side recirculation for temperature control, (2) chemical mixing and storage, (3) secondary­ side recirculation, (4) waste holdup and de­ mineralized water storage, (5) vent collection/ scrubber, and (6) rinse waste cleanup by demineralization. 56 EPRI JOURNAL June 1986 All system materials were qualified in the materials testing programs. Separate re­ circulation systems were used for the primary and secondary sides of the steam generator, thus allowing other outage activities, such as refueling, to proceed during the chemical cleaning. Temporary dams were used in the hot and cold leg nozzles of the steam gener­ ator channel heads. Steam generator heat-up and cool-down times were also minimized by using the primary-side recirculation system. A chemical mixing system was provided fo r preparing and establishing the temperature of the iron and copper solvents, as well as the demineralized water rinse and passivation solutions prior to their introduction into the steam generators. Measured volumes of solu­ tion were prepared in the mix tanks as a backup to the steam generator level indication to control the volumes injected into the recir­ culation system. In this way, the solvent levels in the steam generator were at all times con­ trolled and maintained below the first tube sup­ port. The cleaning solvents, rinse, and pas­ sivation solutions were recirculated through the steam generator, using the secondary-side recirculation system. An air-operated pump was used to drain residual solvent from the steam generators and to provide for solvent removal should an electrical failure occur. Pro­ cess control was provided by an automatic data acquisition system that monitored and recorded critical process variables. Automatic circuits shut down the recirculation pumps on high and low pressure, as well as on high steam generator level. Around-the-clock efforts were begun sev­ eral months befo re the actual cleaning. Dur­ ing the early stages of equipment installation and testing, several hundred workers were re­ quired; during the actual cleaning operation, however, the manpower needs were signifi­ cantly reduced. Cleaning The chemical process, consisting of recirculat­ ing the iron and copper solvent described pre­ viously, was applied to the two steam gener­ ators in succession. Two iron-solvent stages were applied to both generators. Six copper­ solvent stages were applied to the first gener­ ator and four to the second. Samples of the solvents were analyzed for copper and iron (as appropriate), in addition to several other pa­ rameters, to determine when each step had progressed to completion; that is, when no fur­ ther significant increase in iron or copper in the solvent was observed. About 209 hours were Figure 1 Steam generator tubes after chemical cleaning. The cleaning removed over 550 lb (250 kg) of corrosion products from two steam gener­ ators. required to clean the first generator, whereas the second required only about 103 hours from the initial rinse to the prepassivation rinse. The cleaning of the second generator was accom­ plished in less time than the first because fewer copper-solvent stages were used (four instead of six) and because experience and confidence in the process and equipment had been gained in cleaning the first unit. The corrosion products removed by the sol­ vents included iron, copper, zinc, and nickel. About 310 lb (141 kg) of these substances were removed from the first generator and about 257 lb (117 kg) from the second (as deter­ mined by analyses of the solvents). Several cleaned steam generator tubes are shown in Figure 1. Approximately 50% of the corrosion prod­ ucts removed by the chemical cleaning was copper metal. The cleaning process genera­ ted about 1200 gal (4.5 m3) of waste for each solvent-and-rinse stage. The total volume of solvent liquid wastes that had to be processed was 24,000 gal (90 m3). Rinse wastes were reused after cleanup by ion-exchange. The chemical cleaning of the two steam gen­ erators of Millstone Unit 2 was successful in removing over 550 lb (250 kg) of sludge and demonstrated the effectiveness of a large­ scale chemical cleaning effort based on the SGOG-developed solvent process. As a re­ sult, several utilities are planning similar steam generator cleaning projects. Project Man­ ager: C. Lamar Williams New Contracts Funding and Duration Contractor/ EPRI Project Manager $84,900 8 months Quantum Consulting, Inc./ A. Faruqui Effects of Decontamination on BWR Fuel (RP2296-12) $607, 1 00 23 months Babcock & Wilcox Co./ J. Santucci Effects of Decontamination on BWR Fuel (RP2460-1 ) $200,000 9 months Babcock & Wilcox Co./ J. Santucci Funding and Duration Contractor/ EPRI Project Manager Project Quant'lfication and Verification of Enhancements to the PROMOD Combined-Cycle Module (RP2699-7) $46,000 12 months Energy Management Associates, Inc./A. Lewis Commercial Demand-Side Management Program Impact Analysis (RP21 52-4) Underground Coal-Gasification Test (RP2735-01 ) $1 50,000 34 months Gas Research Institute/ N. Hertz Nuclear Power Depolymerization of Coal (RP8003-3) $50,000 10 months Lawrence Berkeley Laboratory/L. Atherton Project Advanced Power Systems Energy Management and Utilization Coal Combustion Systems Boiler Tube Failure Manual (RP1 890-7) $220,300 22 months General Physics Corp./ B. Dooley PWR Pilot Plant Life Extension Program (RP2643-1 ) $255,300 2 4 months Virginia Power/M. Lapides In-Place Inspection of Turbine Blades (RP1 957-6) $1 26,000 13 months Reinhart & Associates/ J. Scheibe/ BWR Pilot Plant Life Extension (RP2643-2) $1 50,000 24 months Nortt\ern States Power Co./M. Lapides Technical and Economic Guidelines for Evaluating Power Plant Water Management Options (RP2 1 1 4-5) $237,000 14 months Sargent & Lundy Engineers/W. Micheletti LMFBR Technical Integration Studies (RP2658-6) $520,200 1 3 months Westinghouse Electric Corp./D. Gibbs Field Testing of Behavioral Barriers for Cooling-Water Intake Systems (RP221 4-6) $474,400 14 months Lawler, Malusky and Skelly Engineers/ W. Micheletti Preliminary Conceptual Design Study for a Small LWR (RP2660-6) $299,000 9 months Babcock & Wilcox Co./ B. Sugnet Evaluation of Utility Experience With Probabilistic Safety Assessment (RP2682-2) $84,700 1 1 months Delian Corp./J. Gaertner Demonstration of Heat Stress Management and Development of a Personal Monitor for the Nuclear Power Industry (RP2705-5) $247,600 1 5 months Westinghouse Electric Corp.jJ. O'Brien Reevaluation of Nuclear Plant Seismic Margin (RP2722-1 ) $651 ,600 29 months NTS Engineering/ R. Kassawara Steam Generator Sludge Landing: Evaluation of Current Practices (RP2755-9) $48,500 6 months Combustion Engineering, lnc./C. Williams Steam Condensation Effects on Reactor Containment Building Aerosol Behavior (RP2802-1) $90,900 36 months Electrical Systems Evaluation of Active Filter Concepts for HVDC Converters (RP21 1 5-15) $34,200 1 2 months S. Wright Development of Advanced Float-Zone Silicon Material (RP2737-4) $613,000 25 months Westinghouse Electric Corp./H. Mehta Power Circuit Breaker Diagnostics (RP2747-1 ) $1 78,200 24 months University of Buffalo Foundation, lnc./S. Wright Feasibility Study for MOS-Controlled Thyristor (RP8001-2) $250,000 12 months General Electric Co./ H. Mehta un·,versity of Minnesota/ Stanford UniversityI F. Rahn R&D Staff Energy Analysis and Environment Market Potential and Commercialization Strategy for Lead-Acid Battery Systems for Load Management Applications (RP1084-21 ) $57,600 1 1 months Battelle, Columbus Laboratories/D. Rigney Small Cogeneration Technology Directory and Project Support (RP1 276-27) $1 1 5,300 12 months Science Applications International Corp./S. Hu Development of a Framework to Construetively Involve Stakeholders in a Comprehensive Evaluation of Alternatives Involving Complex Technological Issues (RP1 433-2) $301 ,200 21 months University of Southern California/T. Wilson Potential Impacts of Climate Change on Electric Utilities (RP2141 -1 1 ) $50,000 6 months ICF Incorporated/ C. Hakkarinen Advanced Filters for Pressurized Fluidized$6,024,600 27 months Bed Combustion Pilot Plant: High-Pressure High-Temperature Demonstration (RP1336-8) Central Electricity Generating Board (England)/ J. Stringer Materials Damage From Acidic Deposition: Phase 1 , Planning Study (RP2071-4) $106,500 14 months Martin Marietta Environmental Systems/B. Syrett, A. Silvers Steam Turbine Rotor Life Assessment and Extension: Material Properties Handbook (RP2481-4) $79,200 10 months The Metal Properties Council/A. Viswanathan Low- Pressure Model Rotor With Relaxed Specification (RP2741-3) $71,600 12 months Vereinigte Edelstahlwerke Ag/R. Jaffee Development of Clean Steels (RP2741-4) $283,000 23 months Bethlehem Steel Corp./ R. Jaffee EPRI JOURNAL June 1986 57 New Technical Reports Requests for copies of reports should be directed to Research Reports Center, P.O. Box 50490, Palo Alto, California 94303; (415) 965-4081. There is no charge for reports requested by EPRI member utilities, U.S. universities, or government agencies. Others in the United States, Mexico, and Canada pay the listed price. Overseas price is double the listed price. Re­ search Reports Center will send a catalog of EPRI reports on request For information on how to order one-page summaries of reports, contact the EPRI Technical Information Division, P.O. Box 10412, Palo Alto, California 94303; (415) 855-241 1. ADVANCED POWER SYSTEMS Experimental Testing of a Catalytically Treated Coal in a Moving-Bed Gasifier AP-4506 Final Report (RP2656-2); $25.00 Contractor: Battelle, Columbus Division EPRI Project Manager: R. Frischmuth Study to Determine the Cost-Effectiveness of Using Steam-Injected Gas Turbines With Coal Gasification Equipment AP-4507 Final Report (RP2029-6); $25.00 Contractor: The Ralph M. Parsons Co. EPRI Project Manager: B. Louks Thermal Reactivity Studies of Coal-Derived Residuum AP-451 3 Final Report (RP1604-4); $32.50 Contractor: University of Wyoming EPRI Project Manager: N. Stewart COAL COMBUSTION SYSTEMS Methodology for Predicting the Corrosion of Underground Residential Distribution Cables Using Modeling Techniques EL-4448 Final Report (RP2200); $32.50 Contractor: Harco Corp. EPRI Project Manager: H. Ng Speech Recognition and Synthesis for Electric Utility Dispatch Control Centers EL-4481 Final Report (RP2473-1); $25.00 Contractor: Honeywell Technology Strategy Center EPRI Project Manager: C. Frank Modeling of External Power System Networks for On-Line Security Analysis EL-4496 Final Report (RP1999-4); $25.00 Contractor: Arizona State University EPRI Project Manager: C. Frank Arc Products of Transformer Insulating Systems Containing Tetrachloroethylene EL-4497 Final Report (RP1499-4, -5); $32.50 Contractors: Springborn Laboratories, Inc.; Westinghouse Research and Development Center EPRI Project Manager: G. Addis Thermal Upgrading of Low-Rank Coal : Process-Screening Study AP-4435 Final Report (RP2221-11); $25.00 Contractor: Bechtel Group, Inc. EPRI Project Manager: N. Hertz Life Assessment and Improvement of Turbogenerator Rotors for Fossil Plants CS-4160 Proceedings (RP2481-1); ordering information available from Research Reports Center EPRI Project Manager: R. Viswanathan PG&E Photovoltaic Module Performance Assessment AP-4464 Final Report (RP1607-2); $32.50 Contractor: Pacific Gas and Electric Co. EPRI Project Managers: J. Schaefer, R. Taylor Modeling Sulfur Capture by Recycle Fines in Fluidized-Bed Combustion CS-4442 Final Report (RP1179-12); $25.00 Contractor: GA Technologies, Inc. EPRI Project Manager: C. Derbidge Photovoltaic Field Test Performance Assessment: Technology Status Report AP-4466 Final Report (RP1607-1); $32.50 Contractor: Boeing Computer Services Co. EPRI Project Manager: J, Schaefer Study of Utility Boilers for a Coal-Water-Slurry Demonstration Test CS-4473 Final Report (RP1895-8); $40.00 Contractor: Burns and Roe, Inc. EPRI Project Manager: R. Manfred Geohydrochemical Models for Solute Migration : Evaluation of Selected Computer Codes EA-3417 Final Report (RP2485-2); Vol. 3, $47.50 Contractor: Battelle, Pacific Northwest Laboratories EPRI Project Manager: I. Murarka Processing of Coal With Microorganisms AP-4472 Final Report (RP2606-1 ) ; $25.00 Contractor: National Bureau of Standards EPRI Project Manager: L. Atherton Proceedings: 1985 EPRI PCB Seminar CS/EA/EL-4480 Proceed ings (RP2028); $55.00 EPRI Project Managers: R. Kamai, V Niemeyer, G. Addis Proceedings: New Directions on the Extrapola­ tion of Health Risks From Animals to Man EA-4447 Proceedings (RP2378-7); Vol. 1, $32.50; Vol. 2, $32.50 Contractor: Meeting Planning Associates EPRI Project Managers: A Silvers, G. Newell Combustion Turbine Materials Problems AP-4475 Final Report (RP2382-2); $32.50 Contractor: Battelle, Columbus Laboratories EPRI Project Manager: R. Viswanathan Materials for Large Land-Based Gas Turbines AP-4476 Final Report (RP2382-1); $32.50 Contractor: National Research Council EPRI Project Manager: W Bakker Proceedings: Conference on Life Prediction for High-Temperature Gas Turbine Materials AP-4477 Proceedings (RP2382-3) ; $55.00 Contractor: Syracuse University EPRI Project Manager: W. Bakker Prediction of Emissions and Performance of Coal Liquids in Tangential- and Wall-Fired Boilers AP-4491 Final Report (RP1412-9); $47.50 Contractor: Combustion Engineering, Inc. EPRI Project Manager: W Rovesti 58 EPRI JOURNAL June 1986 ELECTRICAL SYSTEMS Design and Economic Evaluation of Higher-Strength and Corrosion-Resistant Generator Retaining Rings: Metallurgical Studies and Ring Fabrication EL-3169 Final Report (RP1876-1); Vol. 2, $32.50 Contractor: General Electric Co. EPRI Project Managers: R. Viswanathan, D. Sharma Long-Life Cable Development: Cable Materials Survey EL-4398 Final Report (RP2439-1 ); $32.50 Contractor: University of Connecticut EPRI Project Manager: B. Bernstein Transients on Transmission Systems: Data Presentation and Analysis EL-4403 Interim Report (RP751-1); $40.00 Contractor: Westinghouse Electric Corp. EPRI Project Manager: S. Nilsson State-of-the-Art Review: Pyrolysis and Combustion of PCB Substitutes EL-4503 Final Report (RP2028-12); $25.00 Contractor: SCS Engineers, Inc. EPRI Project Manager: G. Addis ENERGY ANALYSIS AND ENVIRONMENT Techniques and Models to Estimate Health Benefits of Controlling Toxic Substances Emitted from Coal-Fired Power Plants EA-4490 Final Report (RP1826-1 ); $32.50 Contractor: SRI International EPRI Project Managers: P. Ricci, R. Wyzga ENERGY MANAGEMENT AND UTILIZATION Demand-Side Management for Rural Electric Systems EM-4385 Final Report (RP2381-4); $32.50 Contractor: Battelle, Columbus Division EPRI Project Managers: A Faruqui, C. Gellings, 0. Zimmerman Commercial End-Use Metering Workshop EM-4393 Proceedings (RP1216-10); $32.50 Contractor: Synergic Resources Corp. EPRI Project Manager: A Faruqui Lighting Handbook for Utilities EM-4423 Final Report (RP2285-6); $40.00 Contractor: Enviro-Management & Research, Inc. EPRI Project Managers: S. Pertusiello, G. Purcell Zinc Bromide Battery Development EM-4425 Final Report (RP635-3); $32.50 Contractor: Energy Research Corp. EPRI Project Manager: W Spindler Proceedings: 1985 EPRI Cogeneration Symposium EM-4432 Proceedings (RP1276-25); $32.50 Contractor: Synergi c Resources Corp. EPRI Project Manager: D. Hu Assessment of Restaurant Heat Recovery and Load Leveling EM-4461 Final Report (RP1087-3); Vol. 1, $25.00 Contractor: Applied Energy Systems, Inc. EPRI Project Manager: L, Harry Evaluation of a Novel Dual-Fuel Heat Pump EM-4479 Final Report (RP2033-19); $25,00 Contractor: ETL Testing Laboratories, Inc. EPRI Project Manager: P. Fairchild COMMEND Planning System: National and Regional Data Analysis EM-4486 Final Report (RP1216-6); $32.50 Contractor: Georgia Institute of Technology EPRI Project Managers: A Faruqui, J, Wharton Induction Melting of Metals: State-of-the-Art Assessment EM-4508 Final Report (RP2613-3); $25.00 Contractor: Battelle, Columbus Laboratories EPRI Project Manager: L. Harry Performance of Electronic Ballasts and Other New Lighting Equipment EM-4510 Final Report (RP2285-4); $25.00 Contractor: Lawrence Berkeley Laboratory EPRI Project Manager: G. Purcell Proceedings: Food Service Equipment Research Workshop EM-4517 Proceedings (RP2034-17); $25.00 Contractors: W L Whiddon & Associates, Inc,; Hart, McMurphy & Parks, Inc. EPRI Project Manager: G, Purcell Advanced Commercial Survey Methods (COMSURV): Demonstration of Tailored Versus General Questionnaires EM-4519 Final Report (RP1216-9); Vol. 1 , $32,50 Contractor: Applied Management Sciences, Inc. EPRJ Project Managers: A Faruqui, J. Wharton Electrotechnology Reference Guide EM-4527 Final Report (RP2613-5); $32,50 Contractor: Resource Dynamics Corp. EPRI Project Manager: L. Harry Trends in the Energy Efficiency of Residential Electric Appliances EM-4539 Final Report (RP2034-9); $25.00 Contractor: Science Applications International Corp. EPRI Project Managers: S, Braithwait, 0. Zimmerman NUCLEAR POWER Diagnostic Training for Nuclear Plant Personnel: Implementation and Evaluation NP-3829 Interim Report (RP2294-1); Vol. 2, $32.50 Contractor: Search Technology, Inc. EPRI Project Manager: J. O'Brien BWR Full Integral Simulation Test (FIST) Program: TRAC-BWR Model Development NP-3987 Final Report (RP495-1); Vols, 1-3, $85,00 for three-volume set Contractor: General Electric Co. EPRI Project Manager: S, Kalra BWR Full Integral Simulation Test: Phase 2, Test Results and TRAC-BWR Model Qualification NP-3988 Final Report (RP495-1); $40.00 Contractor: General Electric Co. EPRI Project Manager: S . Kalra Long-Term Inspection Requirements for Nuclear Power Plants NP-4242 Final Report (RP2057-1); $32.50 Contractor: Nutech Engineers EPRI Project Managers: M. Lapides, T Marston, M. Behravesh Fuel Performance Evaluation in 16 x 1 6 Assemblies a t Arkansas Nuclear One, Unit 2 NP-4250M Final Report (RP586-1); $25,00 Contractor: Combustion Engineering, Inc, EPRI Project Manager: J. Santucci Power Plant Alarm Systems: Survey and Recommended Approach for Evaluating Improvements NP-4361 Interim Report (RP2011); $32.50 Contractor: MPR Associates, Inc. EPRJ Project Manager: J. O'Brien Two-Phase Jet Modeling and Data Comparison NP-4362 Final Report (RP1733-4); $25,00 Contractor: S. Levy, Inc, EPRI Project Manager: A Singh Effects of Impurity Segregation, Alloy Composition, and Microstructure on the Stress Corrosion Cracking and Temper Embrittlement of Rotor Steels NP-4440M Final Report (RP1929-7); $25.00 Contractor: General Electric Co, EPRI Project Managers: J, Gilman, A Giannuzzi Guidelines for Control of Expendable Products NP-4449 Final Report (RP2410-3); $25.00 Contractor: Stone & Webster Engineering Corp. EPRI Project Managers: J, Matte, J, Carey Proceedings: Sixth Information Exchange Meeting on Debris Coolability NP-4455 Proceedings (RP1931-1); $55.00 Contractor: University of California at Los Angeles EPRI Project Manager: M. Merila Steam Generator Tube Denting: Field Study NP-4456 Topical Report Contractor: S, Levy, Inc. EPRJ Project Manager: T Oldberg lntergranular Corrosion of Steam Generator Tubes: Field Study NP-4457 Topical Report (RPS306-1); $40,00 Contractor: S. Levy, Inc, EPRI Project Manager: T Oldberg Proceedings: 1983 Workshop on Secondary-Side Stress Corrosion Cracking and lntergranular Corrosion of PWR Steam Generator Tubing NP-4458 Proceedings (RPS302-22) ; $70.00 Contractor: Dominion Engineering , Inc. EPRI Project Manager: P. Paine Microstructure, Microchemistry, and Microdeformation of Alloy 600 Tubing NP-4465 Final Report (RPS303-9); $25.00 Contractor: Battelle, Pacific Northwest Laboratories EPRI Project Manager: A Mcllree Operation of Hydrogen Water Chemistry for 18-Month Cycle at Dresden-2 NP-4470M Interim Report (RP1930-7); $25,00 Contractor: General Electric Co, EPRI Project Managers: M. Naughton, C. Wood BWR Radiation-Field Control Using Zinc Injection Passivation NP-4474 Final Report (RP819-2); $25.00 Contractor: General Electric Co, EPRI Project Manager: C. Wood Proceedings: 1 984 Workshop on Secondary-Side Stress Corrosion Cracking and lntergranular Corrosion of PWR Steam Generator Tubing NP-4478 Proceedings (RPS302-22); $70.00 Contractor: Dominion Engineering , Inc, EPRI Project Manager: P, Paine Cobalt Deposition in Oxide Films on Reactor Pipework NP-4499 Final Report (RP1445-4); $25.00 Contractor: UK Atomic Energy Authority EPRI Project Manager: C . Wood Manual of Recent Techniques for LWR Radiation-Field Control NP-4505-SR Special Report; $25.00 EPRI Project Manager: C, Wood Lifetime of PWR Silver-Indium-Cadmium Control Rods NP-4512 Final Report (RP1628-4); $40.00 Contractor: Westinghouse Electric Corp, EPRI Project Manager: J . Santucci Two-Phase Flow Regimes and Carry-Over in a Large-Diameter Model of a PWR Hot Leg NP-4530 Final Report (RP2393-2); $32.50 Contractor: Science Applications International Corp. EPRI Project Managers: J, Kim, M, Divakaruni PLANNING AND EVALUATION Prospects for U.S. Basic Industries, 1986-2000: Implications for Electricity Demand P/EM-4502-SR Special Report; $40.00 EPRI Project Manager: 0, Yu EPRI JOURNAL June 1986 59 New Computer Software The Electric Power Software Center (EPSC) provides a single distribution center for computer programs developed by EPRI. The programs are distributed under license to users. No royalties are charged to nonutility public service organizations in the United States, including government agencies, u niversities, and other tax-exempt organizations. Industrial orga­ nizations, including nonmember electric utilities, are required to pay royalties. EPRI member utilities, in paying their membership fees, prepay all royalties. Basic support in installing the codes is available at no charge from EPSC; however, a consulting fee may be charged for extensive support. For more information about EPSC and licensing arrangements, EPRI member utilities, government agencies, universities, and other tax-exempt organi­ zations should contact the Electric Power Software Center, UCCEL Corp., 1930 Hi Line Drive, Dallas, Texas 75207; (214) 655-8883. Industrial organiza­ tions, including nonmember utilities, should contact EPRl's Manager of Licensing, P.O. Box 10412, Palo Alto, California 94303; (415) 855-2866. EPAM: Sampling Design for Aquatic Ecological Monitoring Version 1 .0 (IBM PC); EA-4302-CCMP Contractor: University of Washington EPRI Project Manager: J. Mattice ESPRE: EPRI Simplified Program for Residential Energy Version 1 .2 (IBM PC) Contractor: Arthur D. Little, Inc. EPRI Project Manager: S. Braithwait EXPOCALC: Exposure Assessment Tool for Transmission Line Electric Fields Version 1 .1 (IBM PC) Contractor: Enertech Consultants, Inc. EPRI Project Manager: R. Patterson HARMFLO: Harmonic Power Flow Program Version 3 . 1 (IBM); EL-4366 Contractor: Purdue University EPRI Project Manager: J. Mitsche HFRAME: Analysis or Design of Transversely Loaded Planar Wood Transmission Structures Version 84 (CDC, IBM); EL-4097 Contractor: Research Institute of Colorado EPRI Project Manager: P. Lyons ADEPT: Acid Deposition Tree HPOF: Transmission System Economic Evaluation Program Version 2.2 (IBM, IBM PC); EA-2540 Contractor: Decision Focus, Inc. EPRI Project Managers: T. Wilson, R. Richels Version 1 . 0 (IBM); EL-2833 Contractor: Systems Control, Inc. EPRI Project Manager: R. Samm ASK: PCB Economic Risk Management Model Version 2.3 (IBM PC) Contractor: Decision Focus, Inc. EPRI Project Manager: V. Niemeyer LOADSIM: Load Shape Simulation Version 2.0 (IBM, IBM PC); EM-3287-CCM Contractor: Energy and Control Consultants, Inc. EPRI Project Manager: T. Yau MMS-ASCL: Modular Modeling System COGEN3: Design, Costing, and Economic Optimization of Cogeneration Projects Version 1 .2 (IBM CMS, IBM MVS) Contractor: Mathtech, Inc. EPRI Project Manager: D. Hu Version 1 .QR1 (IBM, CDC); CS/NP-301 6, CS/NP-2086, CS/NP-2945, CS/NP-2988, CS/NP-2989 Contractor: S. Levy, Inc. EPRI Project Manager: M. Divakaruni COIL: Contaminated-Oil Economic Risk Management Model NPHASE: Fault Protection for High-Phase-Order Transmission Lines Version 2.3 (IBM PC) Contractor: Decision Focus, Inc. EPRI Project Manager: V. Niemeyer Version 1 .0 (IBM); EL-3316 Contractor: Auburn University EPRI Project Manager: J. Lamont CONTRACTMIX: Fuel Contract Mix Decision Analysis Methodology TOPP: Transmission Outage Performance Prediction Macroprogram Version 1.0 (IBM PC) Contractor: Decision Focus, Inc. EPRI Project Manager: H. Mueller Version 1 . 0 (IBM); EL-3880 Contractor: Commonwealth Research Corp. EPRI Project Manager: N. Balu DTAC: Data Transfer and Conversion UPM: Utility Planning Model Version 1 .0 (IBM); EL-4292 Contractor: Boeing Computer Services EPRI Project Manager: J. Lamont Version 2.0 (IBM CMS, IBM MVS); EA-3384 Contractor: Arthur Anderson & Co. EPRI Project Manager: V. Niemeyer 60 EPRI JOURNAL June 1986 E L E CT R I C POW E R R ES E A R C H I N STI T U T E Post Office B o x 1 0 41 2 , Pa l o A l t o , C a l i f o r n i a 9 4 3 0 3 NONPROFIT ORGANIZATION U.S. POSTAGE PAID PERMIT NU MBER 60 SUN NYVALE, CALIFORNIA EPRIJOURNAL ADDRESS COR RECTION R EQUESTED