ML20086T839
| ML20086T839 | |
| Person / Time | |
|---|---|
| Site: | Harris  |
| Issue date: | 02/27/1984 |
| From: | Koope R CAROLINA POWER & LIGHT CO. |
| To: | |
| Shared Package | |
| ML20086T831 | List: |
| References | |
| OL, NUDOCS 8403070059 | |
| Download: ML20086T839 (24) | |
Text
- - _ -
O JEG;-...d.U G...;;cz UNITED STATES OF AMERICA 00tvETED NUCLEAR REGULATORY COMMISSION E' '* 2
' _j p'm ' --'
BEFORE THE ATOMIC SAFETY AND LICENSING BOARb' In the Matter of CAROLINA POWER & LIGHT COMPANY )
and NORTH CAROLINA EASTERN ) Docket Nos. 50-400 OL MUNICIPAL POWER AGENCY ) 50-401 OL
)
(Shearon Harris Nuclear Power )
Plant, Units 1 and 2) )
AFFIDAVIT OF ROBERT H. KOPPE
~
4 County of Bopider )
) ss.
State of Colorado )
Robert H. Koppe, being duly sworn according to law, deposes and says as follows:
Q. Please state your name, title, and address.
A. My name is Robert H. Koppe. I am Manager of Reliability and Safety Projects for the S. M. Stoller Corporation, 191914th Street, Suite 500, Boulder, Colorado 80302.
Q. What is your educational background?
A. I received a BS from the State University of New York at '.,yracuse in 1965. I received an MS in Nuclear Engineering from Ohio State University in 1966 and completed all course work toward a PhD in Nuclear Engineering at the Massachusettes Institute of Technology.
l Q. What is your occupational experience? -
A. From 1968 to 1974, I was employed by the Consolidated Edison Company of New York, in the Nuclear Engineering Division. From 1970 to 1974, I was manager of that Division which comprised about 15 engineers and was responsible for safety 1 .
I
- S403070059 840228 PDR ADOCK 05000400 9 pg l
cnalysis, lictnsing cnd op: rations suppcrt for nuclear systems at all Con Ed
~
l 1
nuclear plants. Operations support included review of operating procedures and
)
operational events, assistance in solving operating problems, and design for l several major plant modifications. Much of this work was related to Indian l
Point 2 and 3, two large (873MW and 965MW) plants with Westinghouse Nuclear Steam Supply Systems and Westinghouse turbine generators.
I joined the S. M. Stoller Corporation (SMSC) in 1974. At SMSC I have performed and/ar directed a large number of projects related to nuclear plant safety and the availability of nuclear and fossil power plants. Many of these projects have been related to the collection and analysis of power plant operating data.
Q. What is the S. M. Stoller Corporation?
A. SMSC is an engineering and technical consulting firm, which specializes in the electric power industry. Examples of the types of projects undertaken by Stoller include: engineering analysis and consulting in the technology and economics of the nuclear fuel cycle, including waste disposal; capacity factors of nuclear and fossil plants; decommissioning of nuclear plants; economic analysis and avall-ability improvement programs for nuclear and fossil plants; and computer technolcgy applied to security and hea'th physics control in nuclear plants.
Q. What has SMSC done in the area of power plant performance?
Under my direction, SMSC maintains a data base called OPEC (Operating Plant Evaluation Code) covering the operating history, including the cause of every outage and der ting, of all nuclear power plants in the United States over 400 MW, from their start of commercial operation through the present. Maintenance of OPEC is funded by the Electric Power Research Institute (EPRI), the Institute
( for Nuclear Power Operations (INPO) and SMSC EPRI is funded by most of the b
utilities in the United States to perform a wide variety of research and development related to the generation, transmission and distribution of electric energy, including many projects designed to improve the availability and reliability of nuclear and fossil-fired power plants. EPRI makes extensive use of OPEC to focus R&D funds in areas which are likely to provide the largest improvements in plant availability. INPO was formed by US Nuclear Utilities 1
1 I 2 l
tit:r the tecid;nt et Thrce Mila Island, as a focus fcr industry-wida efforts to improve the quality of plant operations and maintenance, as they affect both safety and plant reliability. INPO uses OPEC to track overall industry experience and to help focus their audits of individual plants by identifying areas where performance is above or below average. Under my direction, SMSC has provided extensive assistance to EPRI and INPO in analyzing and interpreting operating data for both nuclear and fossil plants.
Q. Has SMSC published any reports on power plant operating experience?
A. Much of SMSC's work is contained in reports to individual utilities and in informal reports to EPRI and INPO. Published reports include:
o EPRI-NP1191 Nuclear and Large Fossil Unit Operating Experience, September,1979.
o EPRI-NP2092 Nuclear Unit Op-rating Experience - 1978 and 1979 Update, October,1951.
o A series of four reports, EPRI-NSAC 9, 35, 37, and 49, entitled Screening and Evaluation of (1979/First Half 1980/Second Half 1980/First Half 1981) Licensee Event Reports.
Two additional reports are now in preparation and will be published by EPRI in early 1984. One covers nuclear plant operating experience for 1980-1982 while the other covers the performance of all fossil plants 300MW and larger, through 1981. I participated in the preparation of all these reports and directed the work on all except the four related to Licensee Event reports. In addition I have presented two technical papers on power plant availability:
o "A Power Plant Availability Improvement Methodology Based on the New NERC Generating Availability Data System (GADS)"
Presented at the annual meeting of the Institute of Electrical and Electronics Engineers in St. Louis in October,1981.
o " Developing an Availability Improvement Program" Presented at the annual meeting of the American Nuclear Society in Washington, D.C. in November,1982.
3
- ~
_ I was clso chairman of a session on p:wcr plint availability et the winter r
, meeting of the American Nuclear Society in San Francisco in November,1977.
Q. What other types of work has Stoller performed relative to plant availability?
A. Some of the work which SMSC has carried out under my direction ir. recent years includes:
o Establishing performance goals for fossil and nuclear plants, o assisting utilities with their plant availability improvement programs, o conducting workshops on availability improvement, o evaluating equipment suppliers in terms of component reliability, and o evaluating the reliability of nuclear safety systems.
Q. What is the purpose of this affidavit?
A. The NRC Staff's Final Enivronmental Statement assumes a capacity factor of 55% to calculate the net electric generation to be produced by the Shearon Harris plant. The purpose of this affidavit is to present my evaluation of the 4
reasonableness of that assumption.
Q. What is capacity factor?
- A. Capacity factor is the percentage of the theoretically maximum possible output of a plant which is actually produced in a given time period. It is calculated as:
(Plant Output)
(Plant Rating)(Time)
Confusion sometimes arises because a number of different plant ratings may be used in calculating capacity factors. For example one may use design ratings or maximum dependable capacity and either of these may be expressed in terms of gross output or net output.
Q. What ratings do you believe are correct?
a
, A. There is no such thing as on2 absolutely correct rating to us2 in calculating capacity factors. What is important is that, whatever choice is made, it is used consistently. In.this proceding, we are looking at historical capacity factors for ,
operating nuclear plants and using these to calculate expected output from Shearon Harris. If we were to calculate historical capacity factors using gross '
design ratings for operating units, we would multiply these by the gross design rating of Shearon Harris to obtain its expected gross output. On the other hand, we could calculate historical capacity factors based on net design ratings and multiply these by the net design rating of Shearon Harris.
Q. What ratings have you chosen to use in your evaluation?
A. I have chosen to use net design ratings. The choice of net rating is dictated by the fact that the benefit of any plant is its net output rather than its gross output.' Traditionally, plant design ratings are based on a set of ideal conditions '
and most plants actually operate at (average) outputs which are somewhat below their design ratings. For this reason the so called Maximum Dependable Cr- :.y (MDC) for many plants is less than the design rating. For my evaluation I have chosen to use design rating. All the historical capacity factors
! quote in this affadavit are based on plant design ratings and the resulting capacity factors should be multiplied by the Shearon Harris design rating (900 mw net) to obtain expected net-electric output. There are two reasons for my choice of design rating. First, we know the design rating for Harris, but its MDC will not be known until af ter it operates. Second, there are inconsistencies from utility to utility in determining MDCs. Capacity factors based on MDC are useful because they measure the performance of a plant against the rating which l the utility expected that it could actually attain. However, the inconsistencies
! In determining MDCs reduce the usefulness of capacity factors based on MDC for comparing units or projecting the performance of one unit based on the experience of others.
Q. Since the MDC for Harris is likely to be lower than its design rating, would you not get a lower expected output from Harris if you used MDC7 A. No. The average capacity factor, based on net design rating, fc all modern
. Westinghouse Plants has been 61.8%. If we assume that Shearon Harris will have 5
- w - ,.v., ,_ , .,-,,v, ..,-.y- 3.r , - . ,.,,.m,-- ,,,.=--r v v v = crw --w--tw'r-+-'e-teve-w---
, e e-v t- - ----* re w+-*e-- -vYT-"e- '
.- - -_ - . - .- . =. . .. - . . . _ . --
a the same CF, it will produce a net electric output of - -
6 (8766X.618X900) = 4.88 x 10 MW hours per year. (Note, that 8766 is the average number of hours in a year). For operating Westinghouse plants, MDC is, on the average,98% of design rating. If we assume the same will be true for Shearon Harris, it would have an MDC of (.98X900) = 882MW. However, the capacity factor for all Westinghouse plants, based on MDC is (h) = 63.06% Multiplying through yields (8766X.6306X882) = 4.88 x 10' MW hours per year which is, of course, the same result.
i Q. What is the basis for the 61.8% capacity factor you just quoter'?
A. Unless otherwise stated, all capacity factors which ! quote in this affidavit were calculated using the following conventions:
- o. Data for each unit begins with the first full calendar month after the unit was declared to be in commercial operation, and continues through Sept. 30, 1983, the last date for which data was in the OPEC system at the time this affidavit was prepared.
o When capacity factors for different units are averaged, each is weighted by the time period represented. For example, if one unit ran two years at 60% and another ran one year at 70%, the average would ber
((2X60) + (IX70))/3 = 63.3%
o The design ratings of a a few units have changed with time. For example, the Ginna plant was initially designed to operate at 410 MW (net). After two years of operation, Ginna was allowed to increase its reactor output which increased its net electric output to 490MW.
Where such changes have taken place, capacity factors for each time period were calculated using the rating in effect at that time.
Q. How e,d you evaluate the reasonableness of the 55% capacity factor for Shearon
' Harris?
A. - First I looked at the historical performance of plants which are generally similar
- to Shearon Harris and then I looked at the major differences between Shearon Harris and its predecessors.
6 c
,w-r, , -~ = ,-, wry , y.,., ,.,,,,,,->,-----w, ,,-.-,,-,-,--e,- ,-.,-.,,,--,,,,,y,..,4,.--,, .r.,-,,-Tw mm- , , , , - -
Q. Hrw did you d7ttrmina which plants tre most similar to Hrrris?
A. There are no two nuclear plants which are truly identical. Therefore, it is necessary to choose plants which are generally similar in those design features which have the largest effect on capacity factor. Since approximately three-fourths of all plant outages are related to the Nuclear Steam Supply System (NSSS), it is clear that comparison should focus on NSSS design.
Q. What operating plants are there with NSSSs most similar to Shearon Harris?
A. Shearon Harris has an NSSS supplied by Westinghouse. As of September,1983, there were 31 units in commercial operation in the U.S. which have Westinghouse NSSSs. One of these, Yankee Rowe, started up in 1961, is very small, and has a design, very different from subsequent Westinghouse units. Two others, Connecticut Yankee and San Onofre 1 started up in 1967 and differ from the plants which followed, in a number of significant ways. The remaining 28 units started up over the time period 1970 through 1982 and are all generally similar to each other and to Shearon Harris in their basic designs. Therefore, I will focus the remainder of my discussion on those 28 units, which are listed in Exhibit RHK-1. There are still many differences in design detail among these units and between them and Shearon Harris, and some of these have had a signif; cant impact on capacity factors. However, these 28 are the currently operating units which are most similar to Shearon Harris. I should point out that the capacity factors of the three units I am omitting have been equal to or greater than the average of the 28 units.
Q. What has been the experience with these 28 Westinghouse units?
A. As of September 30,1983, the 28 units had accumulated a total of 210.7 unit j years of operation. The average capacity factor over all those unit-years was l
61.8 %
Q. Is it not true that the 28 units cover a broad range of sizes?
A. The design net electric ratings of the 28 units range from 490 MW to 1180 MW, with an average (both the mean and the median) of about 360 MW. The design net electric rating of Shearon Harris is 900 MW.
7 l
Q. Hrs there been any difference in captcity factor between smaller rnd inrger Westinghouse plants?
In the Westinghouse design, plant size is increased by increasing the number of primary coolant loops. As shown in Exhibit RHK-1, the six smallest plants have two primary coolant loops and all have Design Ratings around 500MW. There are 10 plants with three loops, ranging in size from about 700 to about 900 MW, and 12 plants with four loops, ranging in size from about 900 to about 1200 MW. ,
Average capacity factors for these units have been:
Number Of Capacity Loops Factor (%)
2 73.3 3 58.0 4 55.3 All 61.8 Q. Would it not be,more appropriate to compare Harris to the three-loop plants rather than ali plants?
A. I do not believe so. I have examined the performance of each nuclear plant component in detail and I am convinced that size, per se, has a relatively small effect on capacity factor. One indication of this is the fact that three and four-loop Westinghouse plants cover the size range from 700 to 1200 MW and show I little variation in capacity factor as a function of size. Another indication is the experience of Boiling Water Reactors which show only small variations in capacity factor over the entire size range from 500 to 1100 MW. If size had an Important effect on capacity factors, there would be a steady decrease in PWR capacity factors with increasing size, and-there is not. Also there would be a decrease in BWR capacity factors with increasing size, and there is not. It appears to be just a coincidence that the six smallest Westinghouse units happen t'o all have above average capacity factors. Also, I should point out that even if l
size had a substantial effect on capacity factor, the average capacity factor for all units would still be appropriate to Shearon Harris since the size of Shearon l
! Harris (900 MW) is very close to the average size of all operating units (860 MW).
- Q. What do you conclude from the preceding?
l l^
8 f
L
[
A. I conclude that historical capacity factors for plants most similar to Shenron Harris have been between 58.0% and 61.8%, depending on whether one looks at all units or foc'uses only on three-loop units, and that, on balance, 61.8% is more appropriate.
Q. How have capacity factors varied with unit age?
i A. Exhibit RHK-2 shows average capacity factors as a function of unit age, far all
. units (two, three, and four-loop) and for three-loop units alone.
, -Q. What has been the cause of the decrease in capacity factors in years 9-14?
A. This decrease has been entirely due to problems with corrosion of steam generator tubes and tube supports. Exhibit RHK-3 shows the capacity factors which would have been achieved in the absence of those problems. The numbers in RHK-3 were calculated by taking actual capacity factors and adding 75% of the CF loss attributed to steam generator corrosion. (In the past, Westinghouse units have operated at about 75% capacity factor during those times they were not shutdown for plant modifications or steam generator repairs and I have assumed that they would have continued to do that if they had not had steam generator corrosion problems.)
Q. Why have the three-loop plants been more affected by steam generator corrosion than other Westinghouse units?
A. While many factors affect steam generator performance, the two factors which
! have had the biggest negative impact have been use of phosphates for feedwater treatment and use of saltwater for condenser cooling. There are only five units which are located on saltwater and used phosphates. Four of these five units happen to have three-loops. These four three-loop plants (Surry 1 and 2 and
- Turkey Point 3 and 4) are also the only units to have replaced steam generators within the time period covered by the data. While other units have had steam gGerator problems, the seawater units have had the worst problems and the comcidence that most seawater units have three-loops is the reason for the lower than average capacity factors of three-loop plants in years 9-14.
l
[ 9
Q. Whtt does the experience to date indicato about the performance of these units
- in their 15th through 30th years?
A. Exhibit RHK-3 shows that for the units now operating, performance of equip-ment other than steam generators has steadily improved over the first 14 years of operation. I expect performance of this equipment in years 15-30 to be comparable to that in years 9-14 and this would result in average capacity factors in years 1 through 30 being somewhat higher than the average for years 1 through 14. Exhibits RHK-2 and RHK-3 show that steam generator performance ,
was much worse in years 9-14 than in years 1-8. This is largely due ' the outages for steam generator replacement at Surry and Turkey Point. Since + se units now have new steam generators, it is reasonable to expect that, at the worst they' will now repeat. their earlier performance, resulting in capacity factors over a thirty-year life which are essentially the same as what they have experienced to date. Actually I believe that improvements in steam generator design and water chemistry will result in plants such as Surry and Turkey Point (and Shearon Harris) doing better in the future. I will elaborate on this later in this affidavit. However, it is important to understand that, even if we assume no improvements, it is reasonable to expect that capacity factors over a thirty-year life will be at least equal to those experienced to date.
- Q. Is it reasonable to project the performance of a plant such as Shearon Harris l based only on the experience of currently operating units?
A. No it is not. Even if there were an infinite amount of data from operating plants so the statistical confidence in the data were perfect (and it is actually far from perfect), the data could only be directly applied to Shearon Harris if the design
- of Shearon Harris, and the _ conditions under which it would operate, were identical to those of the operating plants. In fact the Shearon Harris design differs in many details from the designs of its predecessors, and some technical judgment is required. (Even if one simply applies historical capacity factors to
~
l Shearon Harris, one is making a technical judgment. Specifically one is
. assuming that even though Shearon Harris differs significantly from its prede-cessors, its performance will be the same. This may or may not be true, but it is certainly not obvious.)
10
Q. What era some of the design differsnces bntween Shearor. Harris and its predecessors?
A. There are a number of changes which have been incorporated in the Shearon Harris design to reduce or eliminate problems which have affected operating plants. Some of these are:
o Several plants have had problems with cracking of thermal sleeves in the reactor coolant system and have removed the sleeves to elimi-nate the problem. Thermal sleeves have already been removed from Shearon Harris.
o A number of plants experienced cracking of feedwater inlet piping near the steam generators due to thermal cycling. Shearon Harris has a modified design to correct this problem.
o Many units experienced long outages, or longer.than normal refuel-ings, during 1979-1982 to inspect and modify seismic pipe supports, as required by NRC. The Harris design will meet all of these require-ments.
o A number of plants have had to replace control rod guide tube support pins, which have been cracking, with improved pins. The improved pins have already been installed in Shearon Harris.
o Many plants have had problems with cracked or failed low-pressure turbine blades. The Shearon Harris turbine has much more rugged low-pressure blades to reduce levels of vibrations and therefore reduce cracking.
o Many plants have had problems with cracking of low-pressure turbine discs at highly stressed keyways in the area where moisture first condenses in the turbine. The Shearon Harris turbine rotor design eliminates the keyways in this area.
o Shearon Harris has a spare set of low-pressure turbine rotors while most plants do not. This will significantly reduce the length of any 11
cut ges which may result from problems with low-pressure blades or discs.
o There are a number of modifications which have been incorporated in Shearon Harris to reduce the likelihood of steam generator corrosion.
These will be discussed in more detail on a following page.
o There are a number of other changes which have been incorporated in
- Shearon Harris which should reduce outage time. These include
- improvements to the areas of fuel densification, reactor coolant pump flow splitters, feedwater spargers, "TMI modifications", and fire protection. In addition, plants with ice condenser containments b
have had more outage time caused by problems with containments than other plants and this has reduced average capacity factors.
Shearon Harris does not have an ice condenser. Similarly, plants on seawater have had more outage time caused by problems with containment cooling and service water than other plants, and Shearon
, ' Harris is not on seawater.
Q. How do you know that all these design improvements will actually eliminate the problems they are intended to solve?
A. All of these design improvements have been extensively analyzed and/or tested and most of them have been incorporated in some operating plants and have worked well. Of course, this does not prove that every one will work perfectly
. but there is a high probability that each one will work and a virtual certainty that most of them will work.
l . Q. What are the steam generatcr improvements which have been incorporated in
'Shearon Harris?
A. Earlier, I said that the worst steam generator corrosion problems have occurred at plants which are on seawater and/or used phosphate chemistry. Since Shearon Harris is not on seawater and will not use phosphates, its steam generator performance would be expected to be better than the historical average for all Westinghouse plants, even if nothing else changed. To date the three-loop plants 12
have lost arp average of 8.0% of capacity factor due to steam generator corrosion. At t.he same time, those Westinghouse units which never used phosphates and are not located on seawater (there are 12 such units; one two-loop unit, five three-loop units and six four-loop units), have lost an average of only 0.8%. Several of these 12 units have now operated for seven to nine years and they have not shown evidence of the major problems which affected other Westinghouse units at similar ages. There are several other improvements which have been made in the design of Shearon Harris which increase the likelihood that it can operate with capacity factor losses more like 0.8% than 8.0%. One important change is the elimination of the crevices between the tubes and the tube sheet at Harris. It is the presence of crevices (along with the use of phosphates) which is responsible for most steam generator problems at existing nonseawater plants, including Point Beach I and CP&L's H. B. Robinson 2, both of which*are now undergoing steam generator replacement.
Q. ~ What would the performance of operating Westinghouse plants have been had they originally started up with all the improvements you have described?
A.. Exhibit RHK-3 shows that elimination of steam generator corrosion problems alone would have improved the capacity factors of three-loop plants from 58.0%
to about 64.0%. Elimination of pipe-support problems would have increased this to about 67% and elimination of all the problems I described earlier would have increased it to somewhere in the range of 75 to 80%.
Q. Can we reasonably expect Shearon Harris to have a capacity factor of 75% or better?
(. . A. While this is possible, it does not appear likely. There is a significant probabliity
!- that Shearon Harris will experience some problems which have not been 1
j experienced to date. As I noted earlier, there are a number of new design l , features in Harris, and the-e is some probability that any of these may result in new problems. I believe that new problems will affect Harris at a rate which is lower than has been experienced in the past.
l
.Q. Why do you believe that Harris will have fewer new problems than its prede-l
( cessors?
13
~ . . .- . ____ _ __ _ _ , _ _ _
A. It is important to remember that essentially every design feature in the plants which started in the 1970's was new. Few of these design features had received extensive analysis or testing and, in retrospect, it is not surprising that there were problems. About the only component which received thorough analysis and testing in the 1960's was the nuclear fuel, and its performance through the 1970's has been incredibly good. The designs of steam generators, and most other components, were simply scaled up from very small prototype or military !
reactors with minimal analysis and testing. It is only in the last ten years that >
there has been thorough realistic testing of these designs, both in the laboratory and in operating plants.
Many of the design features in Shearon Harris are identical to ones which have worked well in operating plants. Tne simple fact that Shearon Harris has fewer new' features than earlier plants means that the probability of new problems is reduced even in the absence of extensive testing. In fact, the new design features in Shearon Harris have generally been based on more analysis and testing than was the case 10 or 15 years ago and have often been tested as a result of hm.nc been incorporated as backfits in existing plants. This further reduces the probability of new problems.
- . Q. Are there other reasons why Shearon Harris might not reach or exceed 75%
capacity factor?
A. - A number of the problems which I mentioned earlier (pipe supports, TMI modifications, and fire prcitection) resulted from changing NRC requirements.
. New NRC requirements have been imposed continually over the last fifteen or :
I more years and it appears inevitable that they will continue to be imposed at some level in the future. However, the fact that so many changes have already been required (and incorporated in Shearon Harris) means that there are fewer possible changes remaining. This is especially true since -NRC requirements
, usually go far beyond the specific occurrences which triggered them. (For L example, in recent years NRC has required provisions for dealing with a wide range of possible accidents, not just accidents similar to the one at TMI.) It is
.also worth noting that NRC has recently shown more flexibility in allowing utilities to schedule plant modifications and this could result in higher plant
( capacity factors even if the rate of occurrence of new requirements does'not
( diminish.
I 14
_ . _ , _ _ _ ~ , ~ . . - . . . . _ . - _ . _ . . . _ _ _ ._._._.. - - _
. ~ --. _ _ _. ._ -- - - = - - .. _- .. . - __ -
Q. Are there any other reasons to expect that capacity factors for Shearon Harris might be higher than historical averages?
A. The power output of most plants is limited by tae license limit on core thermal power. .That is to say, if the license limit on these plants were increased, they would be capable of producing additional output up to some level, which is usually determined by the capability of the turbine. The license limit on most plants is' determined by the rules contained in Appendix K to 10CFR100. Since ,
there is now a large amount of data indicating that these rules are extremely conservative, it is reasonable to anticipate that they will someday be relaxed. If i
-this were to happen, Shearon Harris would be able to produce at least 4% more output at full power than will be possible at the present license limit. All other things being equal, this would increase the capacity factor (based on the 900 MW I desfgn outing) by two to three percent.
4 Q. What do you conclude based on the preceding?
A. 'I conclude that there are good reasons to expect that Shea:on Harris will not l experience many of the problems which have affected its predecessors and that ;
the rates at which new problems occur and new NRC requirements are imposed will be lower in the future than they have been in the past. ,
Q. Your testimony is based on the performance of all Westinghouse units or all 1 three-loop Westinghouse units. Mr. Eddleman has said that estimates of Shearon Harris capacity factors should be based on the performance of selected units such Beaver Valley 1 or North Anna 1 or McGuire 1. Would this be logical?
A. No, it would not. Most of the problems which affect nuclear (or fossil) power plants occur randomly. In any short time period (say 3 to 5 years) a plant may have a capacity factor which is 10 or 20% above or below average. As plants ,
accumulate 10 or 20 or 30 years of operation, the random events average out and the capacity factors of all plants get closer and closer to the average. (If all events were random, the standard deviation of the distribution would decrease as the square root of the number of years each plant had operated). Since many of
' the plants now operating have operated only a few years, it is not surprising that some have capacity factors which are well below average and some which are [
15
+ , _ , , - , ~ .,_.,_m, m,., .- . . , , . , _ , _ , , _ , , . , , . , .,.,m,,.%._,myym- . , ,, - , .. ,,,,
...,.,._.r..,.. ....._-m, _ _ ,.. . _4_...
well above average. As these plants all accumulate 30 years of operation, their capacity factors will all draw closer to the average.
Even if we did not expect the distribution of capacity factors to narrow with time, it would not be logical to focus on a few units. Cost-benefit evaluations are supposed to be based on reasonable estimates of the costs and the benefits.
Both logic and statistical theory indicate that the most reasonable estimate of the value of a parameter is not the highest or the lowest value it could have, but the average value it would have if it were measured many times. It is this
, average, or "best estimate" value of capacity factors which I have addressed.
Q.' Are Beaver Valley I or North Anna 1 or McGuire 1 more similar to Shearon Harris than other units?
A. No, they are not. Beaver Valley I and North Anna 1 are three-loop plants like Shearon Harris, but they are no more like Shearon Harris than other three-loop plants. Both Beaver Valley 1 and North Anna 1 have Model 51 steam generators
' while Shearon Harris has Model D-4 steam generators. McGuire 1 is a four-loop plant, rather than three-loop, and has Model D-3 steam generators, rather than Model D-4.
Q. How have these units performed?
A. Because I am dealing here with just a few units, I examined their performance through 12/31/33. Beaver Valley I had a capacity factor of 37.7%. Much of the outage time at Beaver Valley was due to pipe supports and a problem with safety pumps. The pipe support problems were industry wide but had a particularly large effect on Beaver Valley. Shearon Harris will meet current NRC require-ments and will not have similar problems. The safety pump design problem was ,
unique to Beaver Valley. North Anna 1 had a capacity factor of 58.0%, in spite of long outages due to thermal sleeves, flow splitters and control rod guide tube pins. During its brief history to date, McGuire I had a capacity factor of 41.6%.
This plant ran very well but was limited to 50% power due to the Model D-3 vibration problem and then (in early 1983) had an outage for steam generator A
modifications. In the absence of the steam generator vibration problem M 3 McGuire l's capacity factor would have been about 70%, and during the last half f
. 16
of 1983 following correction of the stsam gen::rator problem it actually did run at about 70%. While the performance of Beaver Valley and McGuire was below average, there were, of course, other units whose capacity factors were above average. Two of these happen to be the most recent three-loop plants; North Anna 2 with a capacity factor of 65.0% and Farley 2 with a capacity factor of 80.7%. One could certainly argue that since Farley 2 and North Anna 2 are the most recent three-loop plants, they should be more similar to Shearon Harris than Beaver Valley 1 or North Anna 1 or McGuire 1.
With regard to the steam generator problems at McGuire, I would note chat the Model D-4 steam generators at Shearon Harris are different from the Model 0-3 steam generators at McGuire. The only operating unit with the Model D-4 is Krsko in Yugoslavia. Modifications, similar to those planned for Shearon Harris, have. been incorporated in the Krsko steam generators and have reduced vibration levels to within acceptable limits. According to " Nucleonics Week,"
Krsko had a capacity factor of 67% in 1983.
Q. Mr. Eddleman also has contended that increased involvement by CP&L in the Shearon Harris design and construction will lead to lower capacity factors. Does this make sense?
A. No, it does not. Eighty to ninty percent of all nuclear plant outage time involves the NSSS and the turbine generator. Since the designs of the NSSS and the turbine generator are outside the control of the architect engineer or the utility, it follows that the utility or A/E influence on design can have only a small impact on capacity factor. In any case, it is my experience that utility involvement generally results in increased design margin. This is particularly l
evident in the provision of space to make it easier to maintain equipment. Of all the nuclear plants I have been in, the ones which were the most crowded were the turnkey units, where there was no utility involvement. Finally, I might point out that if one believes that the A/E portion of the job has an impact on capacity l
l factor, one should take account of the fact that the four operating nuclear units l
designed by Ebasco have lifeline capacity facto s which are 7.2% higher than the average for all other units.
Q. Do you believe that we should expect higher capacity factors from Shearon Harris simply because it was designed by Ebasco?
l 17 l
b
\
A. No, I do not. However, this does illustrate that number games can be used to give unrealistically high capacity factors, as well as unrealistically low ones.
Q. What about the effect of construction on Capacity Factors? !
A. I have to way of knowing what effect CP&L's involvement has had on the construction of Shcaron Harris. I do know that most nuclear plant outages are the result of equipment failures, such as steam generator tube corrosion or turbine blade cracks, and that these result from some combination of design, operation, and maintenance. Since only a very small amount of lost capacity factor results from construction deficiencies it follows that the way in which a plant is constructed has very little effect on capacity factor, one way or the cther.
Q. Can plant management have an effect on capacity factor?
A. Yes, it can and I am convinced that it does. I am also convinced that much of the' variability which is observed in capacity factors is due to chance, rather than management. I do not believe that it is possible to reliably quantify the effect of management on capacity factor.
Q. Have you considered CP&L's management experience in ooerating its Brunswick and Robinson nuclear plants in your evaluation of Shearon Harris?
A. I am, aware of the historical capacity factors at Brunswick, which have been below average, and of the historical capacity factors at Robinson (which is a three-loop Westinghouse unit and generally similar to Shearon Harris), which have been above average in spite of steam generator problems. Specifically, Robinson 2 had a cumulative capacity factor of 62.4% since entering commercial operation in March,1971. I am also aware that CPal believes the company is
, making significant management improvements at Brunswick and that this experience will benefit Shearon Harris. I have not however given any penalities or credits based on management, and my evaluation effectively assumes that, in the long term, the management of Shearon Harris will be average. It is relevant to note that the capacity factors of all plants get closer to average as time goes on and that I expect the average capacity factors for plants such as Shearon 18
Harris to be considerably above 55%. It would therefore be quite reasonable to assume that Shearon Harris will achieve a 55% capacity factor even if its management were below average.
Q. The NRC staff has applied a 55% capacity factor to a net electric rating which is lower than the Shearon Harris Design Net Rating of 900 MW. Does this appear r casonable?
A. The staff used a capacity factor of 55% in conjunction with a net electric rating of 868 MW. This is equivalent to using a capacity factor of about 53%, based on the design net rating of 900 MW (since .55 times 868 equals .53 times 900.) I have shown that average capacity factors for plants most similar to Shearon Harris have been in the range of 58.0% to 61.8%, also based on design rating.
Even if'orie expects that Shearon Harris will do no better than its predecessors, the staff's estimate is five to nine percent low. Given the reasons for anticipating better performance from Shearon Harris, the staff's estimate clearly must be regarded as not only reasonable but conservative.
- fA Robert H. Koppe Sworn to and subscribed before me this J27Mday of February,1984.
4 State of Colorado )
- ss.
County of Boulder )
QWbkt W Q. ' MQ Notary Public My Commission expires on 9-//-?7 p.
19
EXHIBIT RHK-1
' OPERATING PLANTS WITH MODERN WESTINGHOUSE NSSSs Plant First Full Month Number of Design Net Design Rating of Commercial Operation Loops Electric Rating Ginna- 4/70 2 490 Point Beach 1 1/71 2 497 Robinson 2 4/71 3 700 Surry 1 1/73 3 788 Turkey Point 3 1/73 3 693 Point Beach 2 5/73 2 497 Surry 2 ' 5/73 3 728 Turkey Point 4 10/73 3 693 Prairie Island 1 1/74 2 530 Zion 1 1/74 4 1040 Indian Point 2 7/74 4 873 Kewaunee 1 7/74 2 535 Zion 2 10/74 4 1040 Pralrie Island 2 1/75 2 530 Cook 1 9/75 4 1054 Trojan 6/76 4 1130 Indian Point 3 9/76 4 965 Beaver Valley 1 3/77 3 852 Salem 1 7/77 4 1090 Farley 1 12/77 3 829 Cook 2 7/78 4 1100 North Anna 1 7/78 3 907 North Anna 2 1/81 3 907 Sequoyah1 7/81 4 1148 Farley 2 8/81 3 829 Salem 2 11/81 4 1115 McGuire 1 12/81 4 1180 Sequoyah 2 6/82 4 1148
- 20
EXHIBIT RHK-2 CAPACITY FACTORS FOR WESTINGHOUSE UNITS, AS A FUNCTION OF AGE Age (Years) 1-4 5-8 9-14 Overall All Units 59.8 63.9 63.0 61.8 Three-Loop Units 58.2 59.5 34.8 58.0 e
a 4
21
O EXHIBIT RHK-3 CAPACITY FACTORS FOR WESTINGHOUSE UNITS IN THE ABSENCE OF STEAM GENERATOR CORROSION Age (Years 1-4 5-8 9-14 Overall All Units 62.0 67.0 72.2 65.9 Three-Loop Units 61.0 65.5 69.2 64.0 i
- O 22
,r. U, t n . , . , :. ~ ' ' ~
, UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION BEFORE THE ATOMIC SAFETY AND LICENSING BO RM D,,9 -1 pg ,.00 In the Matter of ) II'h:,[;,. g.- , ,-
) Erl Ah'CU" '
CAROLINA POWER & LIGHT COMPANY )
AND NORTH CAROLINA EASTERN MUNICIPAL )
POWER AGENCY )
) Docket Nos. 50-400 OL (Shearon Harris Nuclear Power Plant, ) 50-401 OL Units 1 & 2) )
CERTIFICATE OF SERVICE I hereb'y certify that copies of " Applicants' Motion for Summary Disposition of Eddleman Contention 15AA,"" Applicants' Statement of Material Facts as to Which There Is No Genuine Issue to be Heard on Eddleman Contention 15AA," and " Affidavit of Robert H. Koppe," were served this 28th day of February,1984 by deposit in the United States mail, first class, postage prepaid, to the other parties on the attached Service
, List. '
Dale E. Hollar Associate General Counsel Carolina Power & Light Company Post Office Box 1551 Raleigh, North Carolina 27602 (919) 836-8161 Dated: February 28,1984
P SERVICE LIST James L. Kelley, Esquire John D. Runkle, Esquire Atomic Safety and Licensing Board Conservation Council of North Carolina U. S. Nuclear Regulatory Commission 307 Granville Road
-Washington, D. C. 20555 Chapel Hill, North Carolina 27514 Mr. Glenn O. Bright M. Travis Payne, Esquire Atomic Safety and Licensing Board Edelstein and Payne U. S. Nuclear Regulatory Coramission Post Office Box 12643 i Washington, D. C. 20555 Raleigh, North Carolina 27605 Dr. James H. Carpenter Dr. Richard D. Wilson Atomic Safety and Licensing Board 729 Hunter Street U. S. Nuclear Regulatory Commission Apex, North Carolina 27502 Washington, D. C. 20555
, Mr. Wells Eddleman Charles A. Barth, Esquire 718-A Iredell Street Mymn Karman, Esquire Durham, North Carolina 27705 Office of Executive Legal Director U. S. Nuclear Regulatory Commission Thomas A. Baxter, Esquire Wcshington, D. C. 20555 John H. O'Neill, Jr., Esquire Shaw, Pittman, Potts & Trowbridge Docketing and Service Section 1800 M Street, N.W.
Office of the Secretary Washington, D. C. 20036 U. S. Nuclear Regulatory Ccmmission Washington, D. C. 20555 Dr. Phyllis Lotchin 108 Bridle Run Mr. Daniel F. Read, President Chapel Hill, North Carolina 27514 Chapel Hill Anti-Nuclear Group Effort Bradley W. Jones, Esquire 5707 Waycross Street U. S. Nuclear Regulatory Commission RIleigh, North Carolina 27606 Region II 101 Marietta Street Dr. Linda Little Atlanta, Georgia 30303 G vernor's Waste Management Board - -
513 Albemarle Building Robert P. Gruber 325 Salisbury Street Executive Director Rcleigh, North Carolina 27611 Public Staff North Carolina Utilities Commission ,
Ruthanne G. Miller, Esquire Post Office Box 991 Atomic Safety and Licensing Raleigh, North Carolina 27602 Board Panel U. S. Nuclear Regulatory Commission
. Wcshington, D.- C. 20555 I
~ww a v ,,----nes - ,w e r m-- a v v ,nen