ML20003A160
| ML20003A160 | |
| Person / Time | |
|---|---|
| Site: | Palisades  |
| Issue date: | 01/31/1981 |
| From: | Robert Murray, Nelson T, Wesley D LAWRENCE LIVERMORE NATIONAL LABORATORY |
| To: | |
| References | |
| CON-FIN-A-0233, CON-FIN-A-0415, CON-FIN-A-233 NUREG-CR-1833, UCRL-53015, NUDOCS 8101290785 | |
| Download: ML20003A160 (187) | |
Text
{{#Wiki_filter:. NUREG/CR-1833 UCRL-53015 Seismic Review of the Palisades Nuclear Power Plant Unit 1 as Part of the Systematic Evaluation Program l Manuscript Completed: December 1980 Date Published: January 1981 Prepared by T. A. Nelson, R. C. Murray, D. A. Wesley,' J. D. Stevenson' Lawrence Livermore Laboratory 7000 East Avenue Livermore, CA 94550 i Prepared for i Office of Nuclear Resetor Regulation l U.S. Nuclear Regulatory Commission Washington, D.C. 20555 NRC FINS A0233, A0415
' Structural Mechanics Associates BLolk M8E 9
l I I FOREWORD The U.S. Nuclear Regulatory Commission (NRC) is concucting the Systematic Evaluation Program (SEP), which consists of a plant-by-plant limited reassessment of the safety of 11 operating nuclear reactors that received construction permits between 1956 and 1967. Because many safety criteria have changed since these plants were initially licensed, the purpose of the SEP is to develop a current documented basis for the safety of these older f acilities. The 11 SEP plants were categorized into two groups based upon the extent to which seismic design was originally considered and the quantity of available seismic design documentation. Unit 1 of the Pa!isades Nuclear Power Plant, the subject of this report, was categorized under drpup 1. A detailed evaluation of plant structures and the hundreds of individual components within each Group 1 plant has not been performed. Rather, the evaluations rely upon limited analysis of selected structures and sampling of representative components from generic groups of equipment. The component sample was augmented by walk-through inspections of the facilities to select additional components based upon their potential seismic fragility. This limited assessment of the Palisaces facility relied in large part upon the guidance, procedures, and recommendations of recognized seismic design experts. Accordingly, a Senior Seismic Review Team (SSRT) under the direction of N. M. Newmark was established. Members of the SSRT and their affiliations are
-I Nathan M. Newmark, Chairman j Nathan M. Newmark Consulting Engineering Services Urbana, Ill.
William J. Hall Nathan M. Newmark Consulting Engineering Services Urbana, Ill. Robert P. Kennedy Structural Mechanics Associates, Inc. Newport Beach, Calif. iii
John D. Stevenson
~ Structural Mechanics Associates, Inc.
Cleveland,; Ohio Frank J. Tokarz (member until September 30, 1980) Lawrence Livermore National Laboratory Livermore, Calif. The SSRT was charged with the following responsibilities: e To develop the' general philosophy of review, setting forth seismic design criteria and evaluation concepts applicable to the review'of older nuclear plants, and to oevelop an efficient, yet comprehensive, review process for NRC staff use-in subsequent evaluations, e To assess the safety of selected older nuclear power plants relative to those designed under current standards, criteria, and procedures, and to recommend generally the nature and extent of retrofitting to bring . these plants to acceptable levels of capability if they are not already at such levels. The SSRT developed its general philosophy and presented it in the first SEP report, which reviews Unit 2 of the Dresden Nuclear Power Station. The assessment of Palisades reported here is the third in the series of SEP seismic reviews of Group 1 plants. This report' provides partial input into the SEP seismic evaluation of Unit 1 of the Palisades Nuclear Power Plant. The results of the seismic evaluation will be documented in a Safety Assessment Report prepared by the NRC staff that will address the capability of the Palisades systems to respond to seismic events or to mitigate the consequences of such events. A limited peer review of this report was concucted by the SSRT to ensure consistency with the review philosophy established during the SSRT's review of Drescen Unit 2 and to review the results of the limited reanalyses of plant 'tructures and the component sample. Safety for seicnic excitation implies that certain elements and components of an entire system must continue t'o function under normal operating and test loads. The SSRT did not review all aspects of the plant's operation and the safety margins available to ensure that those elements and components needed for seismic safety would not be impaired beyond the point iv
for which they can be counted on for seismic resistance because unusual of operating conditions, sabotage, operator error, or other causes . These aspects will have been studied by others. However, where unacceptable risks of essential elements not being able to function properly to resist mc seis i events were noted or inferred, greater margins of safety or provision for redundancy in the design of these elements are considered by the necessary. e SSRT to b The authors wish to acknowledge M. Nitzel of EGr.G, Idaho,Falls piping analysis. for his The authors also thank T. M. Cheng, technical monitor of this work at the NRC, for his continuing support . v
ABSTRACT A limited seismic reassessment of Unit 1 of the Palisades Nuclear Power Plant was performed by the Lawrence Livermore National Laboratory for the U.S. Nuclear Regulatory Commission as part of the Systematic Evaluation Program. The reassessment focused generally on the reactor coolant pressure boundary and on those systems and components necessary to shut down the reactor safely and to maintain it in a safe shutdown condition following a postulated earthquake characterized by a peak horizontal ground acceleration of 0.2 g. Unlike a comprehensive oesign analysis, the reassessment was limited to structures and components deemed representative of generic classes. Conclusions and reconsendations about the ability of selected strts t>ttes and equipment to withstand the postulated earthquake are presented. 1 a i i r vii
. ~ _ . -
S CONTENTS iii Foreword . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . vii Abstract . . . . . . . . . . .
. . . . . . . xiii List of Illustrations . . . . . . . . . .
. . . . . . . . . . xvi List of Tables . . . . . . . . .
i 1 Chapter 1: Introduction . . . . . . . . . . . . . . . . 1 1.1 Scope and Depth of Review . . . . . . . . . . . . 3 1.2 Plant Description . . . . . . . . . . . . . . . Seismic Categorization . . . . . . . 4 1.2.1 . . . . . 5 1.2.2 Equipment and Structures . . . . . . . . . . . 5 1.3 Organization of Report . . . . . . . . . . . . . 8 Chapte: 2: Summary and Conclusions . . . . . . . . . . . .
. . . . . . . . . 8 z.1 S tr uctures . . . . . . . .
9 2.1.1 Containment Building . . . . . . . . . . . . 2.1.2 Auxiliary Building . . . . . . . . . . . . . 9 9 2.1.3 Field-Erected Tanks . . . . . . . . . . . . . 2.1.4 Underground Piping . . . . . . . . . . . . . 10 2.2 Mechanical and Electrical Equipment, and Fluid and Electrical Distribution Systems . . . . . . . . . 10 11 2.3 Piping . . . . . . . . . . . . . . . . . . . , 2.4 Concluding Remarks . . . . . . . . . . . . . . . 12 i Chapter 3: Basis of Reevaluation of Structures and Equipment . . . . . . . . . . . . . . . . 14 3.1 General Approach to Reevaluation . . . . . . .. . . . 14 3.2 Geology, Seismicity, and Site Conditions . . . . . . . 15 15 3.3 Structures . . . . . . . . . . . . . . . . . 3.3.1 Response Spectra, Damping, and Inelastic Behavior . . . . . . . . . . . . . 16 3.3.2 Analysis Models and Procedures . . . . . . . . . 17 ix
4 3.3.3 Normal, Seismic, and Accident Loadings . . . . . . 17 3.3.4 Forces, Stresser, and Deformations . . . . . . . . 18 3.3.5 Relative Motions . . . . . . . . . . . . . . 18 3.4 Equipment and Distribution Systems . . . . . . . . . 18 3.4.1 Seismic Qualification Procedures- . . . . . . . . 19 3.4.2 Seismic Criteria . . . . . . . . . . . . . . 19 3.4.3 Forces, Stresses, and Deformations . . . . . . . . 19 3.5 Miscellaneous Items . . . . . . . . . . . . . . 19 3.6 Evaluation of Adequacy . . . . . . . . . . . . . 20 Chapter 4: Previous Seismic Analyses . . . . . . . . . . . 21 4.1 Introduction . . . . . . . . . . . . . . . . . 21 4.2 Design Earthquake Motion . . . . . . . . . . . . . 21 4.3 Seismic Analysis . . . . . . . . . . . . . . . 24 4.3.1 Methods of Analysis . . . . . . . . . . . . . 25 4.3.2 Damping . . . . . . . . . . . . . . . . . 27 4.4 Stress Criteria . . . . . . . . . . . . . . . . 27 4.5 Seismic Analysis of Structures . . . . . . . . . . . 33-4.5.1 Containment Building and Internal Structures . . . . 33 4.5.2 Auxiliary Building . . . . . . . . . . . . . 40 4.5.3 Turbine Building . . . . . . . . . . . . . . '43 4.6 Seismic Analysis of Piping . . . . . .- . . . . . . 45 4.7 Seismic Analysis of Equipment . . . . . . . . . . . 47 ! 4.7.1 Control Rod Drive Mechanism
. . . . . . . . . . 47 i
4.7.2 Other Reactor Internals . . . . . . . . . . . 47 4.7.3 Other Major Class 1 System Equipment . . . . . . . 47 4.7.4 Other Class 1 Equipment . . . . . . . . . . . 48 l l Chapter 5: Reassessment of Selected Structures . . . . . . . . 50 5.1 Introduction . . . . . . . . . . . .'. . . . . 50 5.2 Design Earthquake Motion . . . . . . . . . . . . . 50 5.2.1 Peak Ground Acceleration . . . . . . . . . . . 51 5.2.2 Ground Motion Characteristics . . . . . . . . . 51 5.3 Seismic Design Methods . . . . . . . . . . . . . 55 5.3.1 Soil-Structure Interaction . . . . . . . . . . 55 x
5.3.2 Combination of Earthqr.ake Directional Components . . . . . . . . . . . . . . . . 60 5.3.3 Combination of Earthquake and Other Loads . . . . . 62 5.4 Confirmatory Analysis of Structures . . . . . . . .. . 63 5.4.1 Containment Building . . . . .. . . . . . . . 6:. 5.4.2- Auxiliary Building . . . . . . . . . . . . . 71 5.4.3 Field-Erected Tanks . . . . . . . . . . . . . 79 5.4.4 Underground Piping . . . . . . . . . . . . . 83 5.5 Evaluation of Critical Structures . . . . . . . . . . 84 5.5.1 Containment Building . . . . . . . . . . . . 85 5.5.2 Auxiliary Building . . . . . . . . . . . . . 86 5.5.3- Building-to-Building Interaction . . . . . . . . 87 5.5.4 Field-Erected Tanks . . . . . . . . . . . . . 90 5.5.5 Underground Piping . . . . . . . . . . . . . 93 5.6 Seismic Input Motion for Equipment and Piping . . . . . . 94 i Chapter 6: Seismic Evaluation of Mechanical and Electrical Equipment and Fluid and Electrical Distribution Systems . . . . . . . . . . 96 6.1 Introduction . . . . . . . . . . . . . . . . . 96 6.
1.1 Purpose and Scope
. . . . . . . . . . . . . 96 6.1.2 Description of Selected Components . . . . . . . . 98 6.2 Seismic Input and Analytical Procedures . . . . . . . . 101 6.2.1 Criginal Seismic Input and Behavior Criteria . . . . 101 6.2.2 tieismic Input for SEP Reassessment . . . . . . . . 102 6.2.3 SEP Acceptance Criteria . . . . . . . . . . . 103 6.3 Evaldation of Selected Components . . . . . . . . . . 104 6.3.1 Mechanical Equipment . . . . . . . . . . . . 104 6.3.2 Electrical Equipment . . . . . . . . . . 119
( . . i 6.4 Piping . . . . . . . . . . . . . . . . . . . 124 6.5 Summary and Conclusions . . . . . . . . . . . . . 126 l References . . . . . . . . . . . . . . . . . . . . . 129 xi
Appendix As Evaluation of the Effects of Assuming 204 Maximum Modal Damping . . . . . . . . 135 A.1 Introduction . . . . . . . . . . . . . . . . . 135 i A.2 Method of Analysis . . . . . . . . . . . . . . . 136 A.3 Building Response . . . . . . . . . . . . . . . 138
- A.4 In-Structure Response Spectra . . . . . . . . . . . 141 References for Appendix A . . .- . . . . . . . . . 147 1
Appendix B: In-Structure Response Spectra . . . . . . . . . . 149 e l t l xii
4 LIST OF ILLUSTRATIONS 6 1-1 Plan view of the Palisades plant . . . . . . . . . . . 1-2 Plan view of containment building, auxiliary building, and turbine building, showing the arrangement of major . equipment . . 7 4-1 Seismic response spectra used for the OBE analyses of Palisades, for various levels of damping . . . . . . . . . 22 4-2 Seismic response spectra used for the SSE analyses of Palisades, for various levels of damping .- . . . . . . . . 23 4-3 Ground response spectrum from the Taf t 1952 earthquake, compared to the SSE response spectrum for 4% damping . . . . . 24 4-4 Floor response spectrum (OBE, 0.5% damping) for the ' 646-f t level of the containment building . . . . . . . . . 26 4-5 Transverse section of the containment and turbine buildinga . . 34 4-6 (a) Preliminary single-stick model of the containment i shell, with node weights; (b) inertial forces, shear envelope, and moment envelope from an OBE~ analysis of the
-preliminary model . . . . . . . . . . . . . . . . 35 4-7 Mathematical model used in the second analysis of the-containment building . . . . . . . . . . . . . 36 4-8 Response envelopes from a response-spectrum analysis j using the model of 4-7 and an OBE of 0.1 g . . . . . . . . 36 4-9 Two-stick, lumped-mass model of the containment building and internal structure . . . . . . . . . . . . 37 4-10 Principal mode shapes for the two-stick containment building model shown in Fig. 4-9 . . . . . . . . . . . 38 4-11 Response envelopes from an OBE response--spectrum
- analysis using the model in Fig. 4-9 . . . . . . . . . . 39 ' 4-12 Mathematical model used in the analysis of the nor'i-south seismic response of the auxiliary building . . . . 41 4-13 Mathematical model used in the analysis of the
~
east-west seismic respcase of the auxiliary building . . . . . -42 4-14 Maximuru displacement, moment, and shear envelopes for the auxiliary building . . . . . . . . . . . . . 43 i l { 5-1 Comparison of original 7.5% Housner design spectrum i to R.G. 1.60 spectra for 10% and 20% of critical damping . . . 54 i ! xiii (
5-2 Two-stick, lur. ped-mass model of the containment building and internal structure . . . . . . . . . . . . 56 5-3 Containment building model used in_the SEP reevaluation . . . . 64 5-4 Mode shapes for the first four modes of the containment building for the median soil case .. . . . . . . 66 5-5 Shear distribution in the e u.tainment building for the lower, median, and upper sail cases, compared to the
- distribution calculated in the original analysis . . . . . . 68 5-6_ Moment distribution in the containment building for the lower, median, and upper soil cases, compared to the distribution calculated in the original analysis . . . . . . 69 5-7 Three-dimensional model used in the reanalysis
! of the auxiliary building . . . . . .- . . . . . . . . 72
- 5-8 Mode shapes for the four most important modes of the l
auxiliary building for the median soil case . . . . . . . . 74 5-9 Distribution of E-W shear in the auxiliary building for the lower, median, and upper soil cases, compared to the results of the original analysis . . . . . . . . . 75 5-10 Distribution of E-W moment in the auxiliary building for the lower, median, and upper soil cases, compared l to the results of one original analysis . . . . . . . . . 76 5-11 Distribution of N-S shear in the auxiliary building for the lower, median, and upper soil cases, compared to the results of the original analysis . . . . . . . . . 77 5-12 Distribution of N-S moment in the auxiliary building for the lower, median, and upper soil cases, compared to the results of the original analysis . . . . . . . . . 78 5-13 Response to overturning moment in Palisades storage tanks . . . 82 5-14 Schematic plan of auxiliary feedwater line . . . . . . . . 84 5-15 Shear wall plan for the auxiliary building at El 590 ft . . . 88 5-16 Relative displacements of the containment and auxiliary buildings, assuming sof t-soil conditions . . . . . 92 5-17 Linear and nonlinear tank wall force distributions resulting from seismic overturning moment M . . . . . . . . 92 A-1 Response spectrum (2% damping) used for the time-history ! analysis of the containment building, superposed on the 1 corresponding smoothed spectrum from R.G.1.60 . . . . . . . 137 A-2 Shear distribution in the containment building as determined from the response-spectrum analysis of Chapter 5 and two time-history analyses . . . . . . . . . 139 A-3 Moment distribution in the containment building as determined from the response-spectrum analysis c_ Chapter 5 and two time-history analys'es . . . . . . . . . 140 xiv
J 7
.-A-4 Comparison of base slab. response spectra .. . . . . . . . . 142 A-5 Comparison of response spectra for the. concrete
! internals (El 649 ft) - . . . . . . . . . . . . .. . . 143 A-6 Comparison of. response spectra for the containment Vessel- . . . 144 1
-A-7 Comparison of' response spectra for the ; concrete internals
-(El 649 f t) , for upper-bound and median soil conditions ~ . . .- . 145_
! A-8 Comparison of response spectra for the concrete
. internals (El 616 f t) . - . . .~ . . . . . . . . .- . . . 146
. B-1 Comparison of . horizontal in-structure - response
_ spectra for the containment building base slab . - . . .. . . . 150 B-2 Horizontal in-structure response spectra for the i
! containment building base slab, based on R.G.1.60 spectra . . . . 151 4
B-3 Vertical in-structure response spectra for the containment ' building base slab, based or. R.G.1.60 spectra . . . 152 E-4 Vertical 'in-structure response spectra for the containment building base slab, based on R.G. 1.60 spectra . . . 153 B-5 Comparison of horizontal in-structure response spectra for the internal structures (El 616 f t) . . . . .. . 154 e EB-6 ' Horizontal in-structure response spectra for the internal structures - (El 616 f t) , based on R.G. 1.60 spectra . . . . . 115
. . B-7 Herizontal in-structure response spectra for the internal a
structures (El 649 f t), based on R.G. 1.60 spectra . . . . . 156 i . 4 B-8 . Horizontal in-structure response epectra for the internal , s structures (El 649 f t), based on R.G.1.50 spectra . .. . . .. 157 i B-9 Horizontal in-structure response spectra for the I containment shell (El 730 f t), based on R.G. 1.60 spectra . . . 158 l 4 B-10 Horizontal in-structure response spectra for- the [ containment shell (El 730 f t), based on R.G. 1.60 spectra . . . 159 B-ll Comparison of E-W horizontal in-structure response spectra for tia auxiliary building base slab . . . . . . . . . . 160 . B-12 Comparison of N-S horizontal in-structure response spectra , for .the auxiliary building base slab . .- . . . . . . . . 161 i B-13 East-west horizontal in-structure- response spectrum for the
. auxiliary building ?(El 610 f t),. based on R.G.1.60 spectra . . . 162 B-l4 North-south horizontal in-structure response spectrum for the auxiliary building .(El 610 f t), based on' R.G.1.60 spectra . . . . 163-B-15 ' Comparison of E-W horizontal in-structure response spectra for 'the auxiliary building (El 640 f t) . . . . . . . . . 164 B-16 East-west horizontal in-structure response spectr s for the auxiliary building L(El-640 f t), based on R.G. 1.60 spectra . ._ 165:
B-17 _ Comparison of N-S horizontal .in-structure . response spectra for the auxiliary building -(El 640 f t) . . . . . '. . . .. 166 xV I _ .. . . . - ..m.- - . . . . ., .. , . . - . _ - . . , __.. - ,. _ _
l r BrI8 North-south' horizontal in-structure response spectra for the auxiliary building (El 640 f t), based on R.G. 1.60 spectra . . . 167 B-19 Vertical in-structure response spectra for the auxiliary building (El 640 f t), - based on R.G.1.60 spectra . . . 160 i b-lu . Vertical in-structure response spectra for the auxiliary building (El 640 f t), based on R.G.1.60 spectra . . .- 169 I l LIST OF TABLES 3-1 Damping values from R.G. 1.61 compared to those j recommended for the SEP evaluation . . .. . . . . . . . . 16 4-1 Seismic analysis methods used to evaluate Class 1 systems . . . 25 l l 4-2 Damping values used in the design of Palisades . . . . . . . 28 4-3 Summary of original load combinations and allowable stresses 29 l . . 4-4 Member properties of the two stick containment building model shown 1.n Fig. 4-9 . . . . . . . . . . . . . . 39 l 4-5' Maximum permissible spans for rigid piping systems . . . . . 46 l 5-1 ^ Original and currently recommended damping ratios, expressed as percent of critical damping . . . . . . . . . 53 l 5-2 Comparison of median-soil spring rate constants and geometric. damping values for Palisades containment building, calculated in different ways . . . . . . . . . 58 5-3 Comparison .of median-soil spring rate constants and I geometric damping values for Palisades auxiliary building . . . 61 5-4 Containment building response frequencies, in Hz . . . . . .. 65 l 5-5 Modal damping ratios for the containment buildina ! (median 0o11 case) .. . . . . . . . . . . . . . . . 67 i ! 5-6 Ratios ef the loads calculated for the containment building l and the concrete int ernals in the original analysis to those calculated in the present reanalysis . . . . . . . . 70 5-7 Auxiliary building response f requencies, in Hz .- . . . . . . 73 5-8 Modal damping ratios for the auxiliary building (median soil) . . 74 j l 5 Ratios of the loads calculated for the auxiliary building in the original analysis to those calculated in the present reanalysis . . . . . . . . . . . . . . . . 79 5-10 Descriptions of field-erected tanks . . . . . . . . . . 81 xvi
82 5-11 Field-erected tank static design accelerations . . . . . . . 83 5-12 Response characteristics of field-erected tanks . 5-13 Typical member loads and capacities at El 590 ft in the 89 auxiliary building for the postulated 0.2-g SSE . . . . . . 6-1 Mechanical and electrical components selected for . . 99 seismic evaluation and the basis for selection . . 6-2 SEP acceptance criteria for determining seismic design adequacy of passive mechanical and electrical . . 105 equipment and distribution systems . . . . . . . . . 6-3 Stress levels induced in supporting pipes by . . 113 motor-operated valves . . . . . . . . . . . 6-4 Forces and moments in the steam generator supports caused 116 by loads due to the SSE and a main coolant pipe rupture . . . . 6-5 Forces and moments in the primary coolant inlet and outlet 117 nozzle due to seismic and pipe rupture loads . . . . . . . 6-6 Original electrical and instrumentation seismic 120 design qualifications . . . . . . . . . . . . . . .
. 127 6-7 Summary of conclusions . . . . . . . . . . . . . .
xvii
CHAPTER 1: INTRODUCTION This report describes work at the Lawrence Livermore National Laboratory (LLNL) to reassess the seismic design of Unit 1 of the Palisades Nuclear Power Plant. -This limited reassessment includes a review of the original seismic design of selected structures, equipment, and components, and seismic analyses of ' selected items, tsing current modeling and analysis methods. The LLNL work is being performed for th U.S. Nuclear Regulatory Comraission (NRC) as part of the Systematic Evaluation Program (SEP) . The purpose of the SEP is to develop a cu-rent documented basis for the safety of ! 11 older operating nuclear reactors, including Palisades. The primary objective of the SEP seismic review program is to make an overall seismic n safety assessment of the plants and, where necessary, to recommend backfitting in accordance with the Code of Federal Regulations (10 CFR 50.109) . The important SEP review concept is to determine whether or not.a given plant meets the " intent" of current licensing criteria as defined by the Standard Review Plan --not to the letter, but, rather, to the general level of safety that these criteria cictate. Additional background ,information about the SEP-can be found in Refs. 4 and 5. 1.1 SCOPE AND DEPTH OF REVIEW This review of Palisades is considerably different in scope and depth f rom current reviews for construction permits and operating licenses. Its focus is limiteo to toentifying safety issues and to providing an integrated, balanced ) approach to backfit considerations in accoroance with 10 CFR 50.109, which f specifies that backfitting will be required only if substanfial ad< tional protection can be oemonstrated for ~the public health and safety. Such a finding requires an assessment of broad safety issues by considering the interactions of various systems in the context of overall plant safety. Because individual criteria do not generally control broad safety issues, this review is not based on demonstrating compliance with specific criteria in the Standard Review Plan or Regulatory Guides. However, current licensing
4 c.iteria do establish baselines against which to measure relative safety f a: tors to support the broad integrated assessment. Therefore, we compare the seismic resistance of the Palisades f acility n, a qualitative f ashion to that dictated by the intent of today's licensing criteria in order to determine whether Palisaces meets acceptable levels of safety and reliability. References in this report to load ratios and safety f actors do not refer in an absolute sense to acceptable minimums, but to cesign-based levels thought to be realistic in light of current knowledge. In general, original levels do not represent maximum levels because such unclaimed factors as low stress and a structure's ability to respond inelastically contribute to seismic resistance. In particular, resistance to seismic motions does not mean the complete absence of permanent deformation. Structures and equipment may ceform into the inelastic range, and some elements and components may even be permitted to suffer damage, provided that the entire system can continue to function and to maintain a safe shutoown conoition. The Palisades assessment focuses on the integrity of the reactor coolant pressure bounuary--that is, components that contain coolant for the core, and piping or any component not isolable (usually by a double valve) from the core--and the capability of essential systems and components required to shut cown the reactor safely and to maintain it in a safe shutdown condition during and after a postulated seismic disturbance. The assessment of this subgroup of equipment can be used to infer the capability of other safety-related systems, such as the emergency core cooling system (ECCS) . To review the selected systems, an evaluation was made of the reactor contair; ment building (together with its internal structures) and the auxiliary building to demonstrate structural adequacy and to obtain seismic input to equipment. Fielo-erected tanks and a typical buried pipe were also analyzed. A zero-period peak horizontal ground acceleration of 0.2 g was employed along with R.G. 1.60 response spectra for t'io structural evaluations. Mechanical and electrical equipment representative of items installed in the reactor coolant system and safe shutacwn systems at the Palisades facility were examined for structural integrity and for electrical and mechanical functional operability. In order to develop a basis for evaluating the estimated lower-bound seismic capacity of mechanical and electrical components and aistribution systems, we visited the site and identified components for review that potentially have a high degree of seismic fragility. The methods 2
of selection of the representative equipment for. this limited assessment are
' described in detail in Chapter 6.
The safe shutdown earthquake (SSE) is the only earthquake level considered because it represents tne limiting seismic loading to which the plant must respond safely. Present licensing criteria sometimes result in the operating basis earthquake (OBE) , which is usually one-half the SSE, controlling the design of structures, systems, and components for which operation, not safety, is at issue. Because a plant designed to shut down safely following an SSE will be safe for a lesser earthquake, investigation of the effects of the OBE was deemed unnecessary. Safety for seismic excit; tion implies that certain elements and components of an entire system must continue to function under normal operating and test loads. The seismic review team did not review all aspects of the plant's operation and the safety f actors available to ensure that vital elements and components would withstand unusual operating conditions, sabotage, operator error, or other nonseismic events. The report addresses structures, systems, and components in the as-built condition and considers those modifications since the issuance of the operating license that have been made to the selected Class 1 components. Information about structures, systems, and components was primarily obtained from the Palisades docket (Docket 50255) maintained by the NRC in Bethesda, Maryland. Additional information was supplied by the utility and the architect-engineer either through correspondence or during site visits. 1.2 PLANT DESCRIPTION 1 l The Palisades plant is part of the Michigan power pool, providing energy j to ooth the Consumers Power Company (CPCo), which owns and operates the plant, l and the Detroit Edison Company. It is located on the east shore of Lake Michigan, about 5 mi south of South Haven, Michigan, and about 16 mi north of Benton Harbor, Michigan. The plant's Unit 1 is a pressurized light-water moderated and cooled nuclear reactor, commonly designated as a PWR. T' e plant was designed to produce 2650 MW of heat and 845 MW of net electrical power. Combustion Engineering, Inc., designed and supplied the nuclear fuel system and the nuclear steam supply system (NSSS), which includes the reactor vessel, steam generators, pressurizer, and pumps, plus auxiliarv system 3 i
i o components, instrumentation, and the reactor protective system. This NSSS was the first supplied by Combustion Engineering. Bechtel Corporation and its affiliate, Bechtel Power Company, designed and supplied the remaining plant structures, systems, and equipment. Bechtel Corporation actually constructed the entire plant, including the NSSS, for which Combustion Engineering gave technical advice. Westinghouse Electric Corporation supplied the turbogenerator. The ' Atomic Energy Commission issued Construction Permit No. CPPR-25 to CPCo on March 13, 1967. Provisional Operating License No. DPR-20 was issued on March 24, 1971. CPCo filed for a full-term operating permit on January 22, 1974. 1.2.1 Seismic Categorization According to Appendix A of the Final Safety Analysis Report (FSAR) , the plant equipment and structures were categorized in one of three seismic classes: j e Class 1: those structures, systems, and equipment whose f ailure could cause uncontrolled release of radioactivity, or those essential for immediate and long-term operation following a loss-of-coolant accident (LOCA), e Class 2: those structures, systems, and equipment that can sustain limited damage without endangering safe shutdown of the NSSS following a reactor trip .c normal shutdown. The f ailure of Class 2 items could not result in the uncontrolled release of radioactivity, o Class 3: those structures and components whose failure would not , result in the release of radioactivity and would not prevent reactor i shutdown, but may interrupt power generation. ~ l Note that these classifications differ f rom those in Regulatory Guide 1.29,7 which was issued af ter the cesign of Palisades. 4
1.2.2 Equipment and Structures Inherent to the design of a PWR, a closed-cycle reactor, are four barriers that prevent fission products from reaching the environment: e Fuel matrix. e Fuel cladding. e Reactor vessel and coolant loops. e Reactor containment building. The reactor core comprises uranium dioxide pellets enclosed in Zircaloy tubes
-with welded end plugs. Two closed reactor coolant loops, connected in parallel to the reactor vessel, constitute the reactor coolant system (RCS) .
Each loop has two reactor coolant pumps and a steam generator. The reactor containme t building is a post-tensioned, prestressed concrete cylinder and come structure. The walls are prestressed vertically and circumferentially, and the dome is prestressed with three groups of tendons oriented at 120 to each other. The containment building is described in more detail in Sec. 4.5.1. The structurally independent auxiliary and turbine buildings are located to the north and west of the containment building (Fig. 1-1). They are described briefly in Secs. 4.5.2 and 4.5.3, respectively. A schematic plan view of these three major structures is shown in Fig.1-2. 1.3 ORGANIZATION OF REPORT The report has six chapters. Chapter 2 is a summary of the overall assessment of the ability of Palisades b. resist the stipulated SSE event. Included is an evaluation of the sigt icance of any identified deficiencies or areas that may require further study. Chapter 3 contains a description of the general basis for reevaluation of structures and equipment. Chapter 4 includes a presentation of the original f acility seismic design methods, models, anc criteria for structures, equipment, and piping; it also sunmarizes the available original calculated seismic responses. Chapter 5 contains a comparison of the seismic loadings and responses for which the f acility structures were originally designed with corresponding seismic loadings and 5
L h
- g -
U _ Turbine Auxiliary - l building , building -
,,, Access road Discharge h I f g
[ h L_ c:::::::t inlet I Intake " # structure J'l /
/
Containrnent building FIG. 1-1. Plan view of the Palisades plant. i responses derived using techniques thought more realistic in light of current knowledge. Chapter 6 uses the in-structure response spectra generated in chapter 5, as well as other available information, to evaluate the capability of mechanical and electrical equipment and of fluid and electrical cit,trioution systems to resist seismic loads and to perform their necessary safety functions. 6
S Spent fuel poe' Railroad track 7 Corn, dor J
\ !\ -
Warehouse Control J (below) room
'\ 0.
H. P. turbine s " -t' i [ :! *: " Radwaste Operating floor - s trea m nt s . b ' / ! \ h l l 4 :: N
~
Water treatment building \ y[ 'f=2 I
/
P O Primary I coolant j
" I (below) p 3 - pump Moisture ,
{_ _'{ m
\(N f
/ Containment separator h * \
_ _ -/, / - \ vessel i Water intake 3 - ? y % ,, structure C \ Main steam (below) _N ,,
% L. P. turbine Aux. bay roof 3, -
,, ' Generator
- ' Exciter FIG. 1-2. Plan view of containment building, auxiliary building, and turbine building, six) wing the arrangement of major equipnent.
CHAPTER 2:
SUMMARY
AND CONCLUSIONS within the limited scope of this reevaluation, we examined typical structures, equipment, components, and systems individually, to e Assess the adequacy of the existing plant to function properly during and following an SSE.
- e Qualitatively judge the overall factor of safety with regard to seismic resistance.
e Make specific recommendations on upgrading or retrofitting, as appropriate.
, 2.1 STRUCTURES We evaluated the containment building (and its internal structures) and the auxiliary building to demonstrate structural adequacy and to obtain seismic input te equipment. We also reexamined the structural adequacy of
, fielo-erected tanks and a typical buried pipe. For the SSE structural evaluation, a peak horizontal ground acceleration of 0.2 g was used along with R.G. 1.60 response spectra. New analytical models were developed for the containment and auxiliary i buildings that accounted for the soil-structure interaction effects of the layered site. A fairly broad range of soil properties was used to account for the uncertainties in the soil characteristics. For each structural analysis, seismic response loads were ;alculated and were compared to the seismic loads used in the design of the structure. Where ~ the loads based on the reevaluation guidelines were less than those used for design, the structure was juaged to be adequate without additional evaluation. Where the recalculated loads significantly exceeded the design loads or where the original design loads were not available, stress analyses of the controlling structural elements were conducted. Where the seismic stresses were found to be low compared to yield, the structure was again judged to be adequate. For structures with higher stresses, the effects of structure auctility were included as requitec. 8
1 I
-l l
l 2.1.1 Containment Building The. loads developed in the containment _ building for the stiffer soil conditions exceed those used in cesig1; however, we found the resulting stresses to be well below yield for both the concrete internal structures and the containment vessel. Based on relatively low structural damping ratios consistent with the low seismic stresses, minimum factors of safety of 1.5 for the internals and 2.1 for the containment vessel were computed. We conclude that the containment bui] ding is capable of withstanding the 0.2-g SSE. 2.1.2 Auxiliary Building Using a new thtee-dimensional analytical model, we found that the loads
,ceveloped in the auxiliary building exceed the original seicmic design loads.
Based on the minimum concrete design strength but neglecting the $ factor for workmanship,.our analysis showed that all eleme,ts of the auxiliary building structure remain well below yield for the SSE. If the $ factor of 0.85 is included, one shear wall at the northeast corner of the building at El 590 f t is expected to experience light cracking. The auxiliary building is considered capable of withstanding the 0.2-g SSE with no loss of function. We also considered the consequences of possible interactions between the containment, auxiliary, and turbine buildings. No structural damage sufficient to cause any loss of function is predicted. 2.1.3 Field-Erected Tanks Three field-erected tanks were evaluated for the SSE in terms of SEP guidelines. These were the T-2 concensate storage tank (CST) , the T-58 safety injection and refueling water tank (SIRWT), and the T-81 primary water supply make-up storage ' tank (PWSMST). The T-2 CST and T-58 SIRWT wete both founc to be adequate to withstand the 0.2-g SSE with no stresses above yield. However, the SSE will produce loads in the T-81 PWSMST that will stress the anchor
. bolts well above yield. Failure of these bolts is expected to result in .taak -
wall buckling and pcssible base plate f ailure, leading to a loss of water. If this tank is confirmed to be essential for safety, modifications should be implemented to increase the anchor capacity. t l 9
.2.1.4- Underground Piping I We evaluated the auxiliary feedwater line to assess the adequacy of buried pipes at Palisaces. Conservative assumptions were made concerning the soil strains ez?ected during the SSE. Stresses were computed at discontinuities and at penetrations (where relative motion between structure and soil could occur), as well as in the straight-run sections. No stresses above ASME Code allowables were calculated in the auxiliary feedwater line.
4 Assuming it to be a typical buried pipe, we conclude that critical buried I pipelines will not fail as a result of the 0.2-g SSE. 2.2 MECHAMCAL AND ELECTRICAL EQUIPMENT, AND FLUID AND' ELECTRICAL DISTRIBUTION SYSTEMS As discussed in Chapter 6, typical mechanical and electrical equipment components were selected for review in large part on the basis of the judgment and experience of the SEP seismic review team comprising the authors and certain SSRT and NRC staff members. The vocumentation that exists regarding the original specifications applicable to procurement of equipment, as well as documentation concerning qualification of the equipment, varies greatly. In some cases the qualification for an item of equipment is quite specific,
- whereas in other cases the qualification pertains only to a class of equipment.
1 Because we lacked essential seismic design and qualification data, 'our review of tne seismic design adequacy of mechanical and electrical equipment is incomplete. Additional data in the form of analysis or test results must be developed before cefinite conclusions can be drawn. Therefore, based upon the design review and independent calculations made for this reassessment, we were unable to confirm the capability of the following mechanical and electrical components to withstand the 0.2-g SSE without loss of structural integrity and required safety function e Essential service water pumps oesign details unavailable. e Auxiliary feedwater pumps: design details unavailable. e Diesel generator. oil storage tanks: no evaluation performed because of lack of information.
- e. Safety injection tanks additional analysis of support structure required.
10
e Motor-operated valves: further analysis needed, o Control rod urive mecnanism: further analysis required to ensure active function. e Steam generators: design details unavailable. e Reactor coolant pumps: design cetails unavailable. o Reactor vessel supports and internals: design details unavailable. e Battery racks: lateral bracing should be replaced or strengthened. e Motor control centers design details unavailable. e Switctigear: confirmation of anchorage design details necessary; other design details unavailable. e Control room electrical panels: licensee to verify seismic design adequacy. e Transf ormers: end units should be securely anchored; other cesign aetails unavailable, e Electrical cable raceways: analysis of support systems needed. 2.3 PIPING Pending completion of the final piping analysis report, only preliminary results are available. Portions of four piping systems were analyzec. Throughout, it was assumed that suitable stress analyses of the supports and suostructure were performed for the original loads. In addition to the conclusions below, concern was expressed about the adequacy of pipe wall thicknesses tor certain pipe sizes. e Residual heat removal system. Seismic stresses in the piping were found to be well within allowable limits; however, support and anchor loads based on the SEP acceptance criteria are generally higher than those aetermined in the original analyses. Further analysis is warranted. Further consideration should be given to the nozzles, since the anchor moment loading is significantly greater than the cesign loading. e Component cooling system. Piping stresses are well within allowable limits. Support loads were found to be generally higher than svailabic design loaos; however, no failure is anticipated during the postulated SSE. Anchor loads cetermined in the reanalysis were also 11
l generally higher than original loaos. Further consideration should be given to the nozzles in cases where the anchor loads have increased significantly. i e Auxiliary feedwater system (three models, including the steam line to l the P-8B turbine). Reanalysis results for two portions of the system l show that ASME Code stress limits will be exceeded during an SSE. l Additional lateral and vertical dynamic supports are needed in I sections of the steam line to pump P-8B; further analysis is necessary
- for overstressed support Rz27C near pump P-8A. Insufficient data were available for anchor loao comparisons. Analysis of the third auxiliary feedwater model showea no excessive stresses; however, adequate data on original support and anchor loadings were not available for comparison.
! e Regenerative heat exchanger letdown. Code allowable stresses will be exceeded in the piping, primarily oue to a deficiency in axial and i lateral restraint in a single vertical leg. Insufficient data were available for comparisons of support and anchor loads, i I l 2.4 CX)NCLUDING REMARKS l Based on the combined experience and judgment of the authors and the l SSRT, the reviews of the original design analyses, and comparisons with similar items of equipment and components in other more recently designed reactors, we conclude that e Structures and structural elements of the Palisades f acility are l adequate to resist an earthquake with a peak horizontal ground acceleration of 0.2 g, provided that the anchor capacity of tne T-81 PWSMST is increased (Sec. 5.5.4) . l e In view of the limited amount of both analysis and test oocumentation, l no definitive statement can be made about the overall seismic design adequacy of mechanical and electrical equipment or of the piping systems. More data must be developed before equipment seismic design I adequacy can be determined in accordance with evaluation criteria in this report. l 12
We therefore recommend that e Modifications be made as necessary to the mechanical and electrical equipment items listed in Sec. 2.2, to the auxiliary feedwater piping, and to the regenerative heat exchanger letdown piping; and where definitive conclusions could not be drawn, that additional analysis be per formed. e All safety-related electrical equipment in the plant be checked for
- adequate engineered anchorage; that is, the anchorage should be found to be adequate on the basis of analysis or tests employing design procecures (load, stress and deformation limits, materials, tabrication procedures, and quality acceptance) in accordance with a recognized structural cesign code.
e A general reconnaissance of the plant be made to identify items that are (1) . overhead or suspenced, (2) on rollers, or (3) capable of sliding or overturning. All such items, whether permanently installed or not, that could dislouge, f all, or displace curing an earthquake and impair the capability of the plant to shut down safely simuld be upgraced so that they no longer jeoparoize the plant. 1 13
CHAPTER 3: BASIS OF REEVALUATION OF STRUCTURES AND EQUIPMENT 3.1 GENERAL APPROACH TO REEVALUATION The seismic reevaluation part of this study centers on e Assessment of the general integrity of the reactor coolant pressure boundary. e Evaluation of the capability of essential structures, systems, and components required to shut down the reactor safely and to maintain it in a safe shutdown condition (including the capability for removal of residual heat) during and after a postulated seismic disturbance, which in this case is the SSE. To accomplish this level of reevaluation, it is necessary to assess the i f actors of safety of essential structures, components, and systems of the older plant, relative to those designed under current standards, criteria, and i
- procedures. Such evaluation should help to define the nature and extent of
! any retrofitting required or possible to make these plants acceptable, if they are not already at acceptable levels. f As used in the previous paragraph, the term " relative" is not to be construed as implying an evaluation based on the norm of current criteria, l standards, and procedures, but instead, an assessment made in the light of knowledge that led to such a level of design. It would be irrational to assume that an older plant would consist of structures, equipment, components, and systems that would meet current criteria in every instance; even so, those items that do not meet current criteria may be entirely adequate in the sense of meeting acceptable safety and reliability criteria. Within the smpe of the investigation, it was impossible to reexamine every item in detail. On the other hand, by examining structures, equipment, components, and systems individually, it was felt that it would be possible to assess their idequacy and general f actors of safety f.. meeting the selected ; i SSE hazard. Thereaf ter, on the basis of an evaluation of the structures, l? l
items of equipment,.or systems, as appropriate, it should be possible to provide e Judgmental assessment of the adequacy of the existing plant to function properly during and following the SSE hazard, including judgmental. assessment of the overall f actor of safety with regard to seismic resistance. e Specific comments pertaining to upgrading or retrofitting as may be i appropriate. The detailed basis of the reevaluation approach to be followed generally is presented in Refs. 4 and 5. The specific bases of reevaluation are described next.
- 3. 2 - GEOIDGY, SEISMICITY, AND SITE CONDITIONS The seismicity information forms the basis for arriving at the effective peak transient ground motions (acceleration, velocity, and displacement) for use in deriving response spectra, time histories, etc., in the reevaluation.
Thus, one important initial basis of reevaluation is a comparison of the original seismic design criteria with those selected for reevaluation. Another important basis for reevaluation is the treatment of soil-structure interaction (SSI) . More accurate methods for computing SSI are available today than were in use when Palisades was designed. This is especially true of layered sites such as Palisades. Existing soils information was thus used in the reevaluation. 3.3 STRUCTURES The first task in examining structures is to stannarize the nature and makeup of the structures, based on knowledge about original design criteria and information on the as-constructed plant. Also required are summaries of the cesign analysis approaches employed, including loading combinations, stress and deformation criteria, and controlling response calculations. Chapter 4 provides these summaries. In evaluating the seismic design criteria, ic is generally necessary to have information concerning the 15
response spectra used originally, the applicable levels of damping, and the modeling approach used in the analyses. Also needed are details of input and methods of analysis used in designing mechanical equipment, piping, and electrical system supports. Thereaf ter, with the seismic criteria applicable to the reevaluation known, and with knowledge of other normal loading criteria deemed necessary, it is possible to estimate the response to the seismic excitation. In some cases, it may be necessary to carry out new seismic analyses with the original model or new models, as deemed appropriate. The final bases for evaluation will involve consideration of many factors, including the following items. 3.3.1 Response Spectra, Damping, and Inelnstic Behavior 4 One basis for evaluation will be comparison of the original response spectra with the response spectra applicable to the reevaluation, taking account of appropriate damping values. The damping values specified in R.G. ! 1.61 and those recommended in NUREG/CR-0098 for reevaluation purposes are summarized in Table 3-1. The reason for permitting higher damping values TABLE 3-1. Damping values from R.G.1.61 compared to those recommended for the SEP evaluation. 2 Damping (4 of critical damping) j R.G. 1. 61 (SSE) NUREG/CR-0098a l Reinforced concrete 7 7 to 10 Prestressed concrete 5 5 to 7 7 to 10 c Welded assemblies 4 5 to 7 Bolted and riveted assemblies 7 10 to 15 Piping 2 or 3 2 to 3
" Recommended for yield level. No prestress left.
Without complete loss of prestress. l l 16 l 1
is discussed in Ref. 4. Although there are limited data on which to base damping values, it is known that the R.G.1.61 values are conservative to ensure that adequate dynamic response values are obtained for cesign purposes. The lower values recommended in J4UREG/CR-0098 are, in most cases, close to the R.G. 1.61 values. The higher values in the NUREG/CR-0098 column are best-estimate values believed to be average or slightly above average values for typical structures. It is recommended that these higher values be a used in design or evaluation for stresses at or near yield, and when moderately conservative esthmates are made of the other parameters entering into the design or evalcation. A second basis for evaluation is the level of inelastic response exhibited by the structures, as measured by ductility factors. It is recommended in Ref. 4 that low ductility f actors (1.3 to 2) be used for conservatism and to help ensure that no gross deformation occurs in any critical safety elements. This, in turn, ensures that system ductility is maintained at a low value. Local inelastic behavior, which arises from deformation of a number of interconnected elements, may result in larger local cuctility factors than those predicted at the system level. Ar. assessment of the local element deformation and its role in system performance needs careful evaluation and is largely judgmental. Local element ductility should be permitted in equipment only if it can be clearly demonstrated that functional-ability is not impaired and that a significant margin of strength still remains. 3.3.2 Analysis Models and Procedures The reevaluation also considers the adequacy of the originni analysis models, and assesses the possible effects of SSI, overturning, and torsion. Analysie procedures used in the reevaluation should be in keeping with the state of the art. In general, response-spectrum or time-hidtory analyses are used unless other reasons dictate other approaches more or less sophisticated. f 1 3.3.3 Normal, Seismic, and Accident Loadirj{s 1 The loading combinations of particular mportance in the reevaluation process incorporate normal loadings (dead lo#ad, live load, pressure, I l 17 l l
temperature, etc., cs appropriate) with seismic loadings. Design basis accident load eff ects were not considered; however, one criterion examined was that the reactor coolant pressure boundary be maintained to preclude an earthquake-initiated LOCA. 3.3.4 Forces, Stresses, and Deformations A significant aspect of the reevaluation involves assessment of the reasonableness of the forces (axial an' shear forces, and moments) , together with associated stresses and deformations, used in the original design and their adequacy in the light of the seismic criteria applicable to the Jeevaluation. This assessment considers effects arising f rta horizontal and vertical excitation and takes into account the proportion of. total effects attributed to seismic factors. Also, the amount of limited inelastic behavior that is to be accommodated is evaluated as may be appropriate. 3.3.5 Relat.ive Motions The reevaluation takes account of the eff ects of any gross relative motions that might influence piping entering buildings or spanning spaces between buildings, the eff ects of tilt, and other interaction effects. 3.4 EQUIPMENT AND DISTRIBUTION SYSTEMS Of particular importance in the reevaluation process is the assessment of the adequacy of critical mechanical and electrical equipment, and of fluid-and electrical-distribution systems. The reevaluation centers on those items or systems essential to meeting the general criteria described earlier. A major task of the reevaluation process is to identify the critical saf ety-related systems and the criteria originally used for procurement and seismic qualification of equipment. For the systems selected, represectative items or systems were identified on the basis of e Physical inspection of the f acility (where specific items were identified as possibly having nearly lower-bound seismic resistance) . e Representative sampling. I 18
Af ter' identifying appropriate systems or items, and af ter ascertaining the nature of the seismic criteria used during procurement or qualification, ' the reevaluation effort turns to a detailed assessment of the original design in the light of current kr.,owledge about equipment vulnerability to seismic excitation. Specifically, the evaluation involves consideration of the following items. 3.4.1 Seismic Qualification Procedures The initial reevaluation assessment is concerned with the original seismic qualification of the equipment item or system, in terms of the seismic test 'perfonsance (level and extent of testing), or analyses that may have been made, or both. 3.4.2 Seismic Criteria s The second major aspect of reassessment involves comparison of the original seismic design criteria with those currently applicable. Consideration is given to such items as the in-structure response spectra, dynamic coupling, and dcuping. I 3.4.3 Forces, Stresses, and Deformations For.those items of equipment for which loads, stresses, or deformations may be a major f actor in design and performance, the reevaluation involves e Examination of the original loading combinations and analyses. e Calculation or estimation of the situation that exists under the reevaluation criteria. Particular attention is directed to the effect l of any increase in the seismic component of load, stress, or deformation. l 3.5 MISCELLANEOUS ITEMS In a subsequent step of the reevaluation, it may be appropriate to
~
evaluate such items ar. sources of water for emergency core cooling and to 19
l l assess whether or nc~ ~,y potential problems could occur with regard to dams, i intake structures, cooli- water piping, etc. I 3.6 EVALUf h OF ADEQUACY On the basis of the reevaluation assessments made as a part of the foregoing studies, an overall evaluation of the adequacy of the critical j structures and representative equipment items and systems is made. Such an evaluation takes into account judgmental or f actual assessment of the factor of safety, as the case may be, and consideration of the adequacy of ind.',vidual items in a system in terms of overall system performance. l i i l l l l l l 20
CHAPTE' 4: PREVIOUS SEISMIC ANALYSES
4.1 INTRODUCTION
This c.boter presents the original seismic design criteria for
! Palisades. The seismic loadings and allowable stress criteria for Class 1 structures, equigenent, and piping are defined, and the calculated seismic I
responses of critical structures are described. The data presented in this chapter are used to define the design basis and to form
- he basis for comparison with SEP acceptance criteria in Chapters 5 and 6. Most of the j information has been drawn f rom the FSAR; detailed references are given later, in- the sections describing the individual analyses.
4.2 DESIGN EARTHQUAKE MOTION Palisades was designed for an operating basis earthquake (OBE) with a peak ground acceleration (A, )-of 0.10 g and for a safe shutdown earthquake (SSE) with an A,,g of 0.20 g. (The OBE values are called " design earthquake loads" in the FSAR, and SSE values are called "maxistan credible earthquake" values.) Tim .=rtical conwnent of acceleration, when considered in the dynamic analyses, was asstuneo to be two-thirds of the horizontal component. Response spectra for structural design were developed f rom spectra in Ref. 9. Several acceleration .sectra--including Taf t 1952, Olympia 1949, El Centro 1934, and El Centro 1940-were normalized to A, = 0.33 g, combined, and smoothed, then plotted on tripartite graph paper. The resulting response spectrtsa was multiplied by appropriate scaling f actors, corresponding to the OBE and SSE accelerations. The results were the design spectra shown
'in Figs. 4-1 and 4-2. In-structure spectra developed from the Taf t record alone were also used in several equipment analyses. Figure 4-3 compares the ground response spectrtsa from the Taf t earthquake to the smoothed design spectrum for 44 damping.
21
100 v fx i v i i xf i fg i v i v i fx i v,
,sE 50 - t 0 -
' + fo
\ 0.0 %
g f 20 - 0.5 % -
% 8
/ \
4 ' 10 g ~k, -
/ s / / / -
a \ O
;5 - 0-% d I3 /
0 2% s
\
o.,[% s s. 5% /
>2 -
gg 10 % -
.s'
/
1 Te D
/
0.5 $ SJ O O s, s-0.2
@D -
/ \
0.1 \ l \/ l\ /l\ /I \/ l\ /l\ /l \/ I / 0.01 0,02 0.05 0.1 0.2 0.5 2 1 5 10 Period (s) FIG. 4-1. Seismic response spectra used for the OBE analyses of Palisades, for various levels of damping (f rom FSAR, Appendix A) . 22
100 i f3 ;y ;y ; 4 ; v g v g fg ; y
#y %
50 E #O - 00%
/ so p, y s 0.5 %
' 0 1%
20 -
% '\ -$
/
10 h 4 # - 2%
/ \ /
5% ' b ' _ N y5 - O o$ jo y - C
/ O P y a s .
s Q o/- O-
>2 -
gt 0$ lO O' T b 02 W O - og
% / %
e 0.5 -
<g, n O
/ O
\ 0l p )
O@' 0.2
/07 g-4 O-
\
0.1 I \/ I\ /l\ /l \/ I\ /l\ /I \/ I 0.01 0.02 0.05 0.1 0.2 0.5 1 2 5 10 Period (s) FIG. 4-2. Seismic response spectra used for the SSE analyses of Palisades, for various levels of damping (from FSAR, Appeadix A). 23
100 pl /NI XIAl/\l AlXl /- IX 50 h O ( T 7 aft record ,O ) 20 - g
?
d b Housner O s h Z
= 5 (4% damping) o N
2 (- s 09 Og o Y h
-f O
1 ds O. oqs 0 # D 0.5 0 oo% O
% 8s9 -
0.2 %,'i o 9.% c 0.1 \/AAAV #^VA 0.1 0.2 0.5 1 2 5 10 20 50 100 Frequency (Hz) FIG. 4-3 Ground response spectrum from the Taf t 1952 earthquake, compared to ! the SSE response spectrum for 4% damping (f rom Ref.10) ., 1 4.3 SEISMIC ANALYSIS Table 4-1 is a list of the Class 1 systems at Palisades, showing the seismic analysis method by which each was evaluated. In tne brief descriptions of these analysis methods that follow, the emphasis is on the general characteristics of each method. Later, in the more detailed l descriptions of the individual analyses, note will be taken of applicable details.
-??
TABLE 4-1. Seismic analysis methods esed to evaluace Class 1 systems. Item Type of analysis Model Structures: Containment building and Response spectrum Lumped mass and beam internal structure Auxiliary building Response spectrum Lumped mass and frame Turbine bililding Response spectrum Lumped mass and beam Electrical penetration Response spectrum Lumped mass and beam enclos ure Intake structure and auxiliary Equivalent static Unknown feed pumps enclosure Piping and equipment: Flexible piping Response spectrum Lumped mass and beam Rigid piping Equivalent static Lumped mass and beam Reactor internals Unknown Unknown Control rod drive mechanism Unknown Unknown Spent-fuel storage racks Response spectrum Lumped mass and grid Other major equipnent Equivalent static Unknown Other equipnent Several (see Sec. Unknown 4.7.5) 4.3.1 Methods of Analysis 4.3.1.1 Dynamic Methods 1 l Dynamic response-spectrum analyses, using the smoothed response spectra l shown in Figs. 4-1 and 4-2, were performed on the containment building (with its internal structure), the auxiliary building, the turbine building, and the electrical penetration enclosure in the containment. Response-spectrum analyses based on floor spectra derived from the 1952 Taft earthquake record were performed. on all flexible piping systems and several pieces of Class 1 equipnent. The dynamic structural analyses and the analysis of the spent-fuel storage racks considered only horizontal motion. (For the structures, 25
horizontal responses were then' ceabined with responses to static vertical loads in calculating the stresses. In the analysis of the spent-fuel storage racks, the vertical and ' horizontal stresses were combined by tne square-root-of-the-sum-of-the-squares [SRSS] method.) Each piping system was analyzed twice, once for each principal direction of horizontal excitation; in each analysis, vertical excitation was applied sir.ultaneously. The stress at each point was taken as the maximum obtained f ran the two analyses. Information is unavailable for the dynamic analyses of Class 1 appendages. Proprietary Bechtel computer programs were used for most of the dynamic analyses. The spent-fuel storage . racks were analyzed using the STARDYNE code. Floor response spectra were generated from time-history analyses.. The input was the record of the 1952 Taf t earthquake. The floor spei,,tra s.tre , smoothed using straight-line segments, so that the peaks were broadene i at .the natural frequencies of the corresponding structures. A typical floor spectrum 7 is shown in Fig. 4-4. Such floor spectra were used either as the basis for . 1.26 . ,,
,,,,,g , ,
,,,,,q , , , , , , , ,
1.11 - - 0.96 - - 3
; 0.81 - -
.9 3 0.66 - -
8 0.51 - - , m _y 0.36 - - 2 o + f 0.21 - - I
. 0.06 - -
-0.09 - -
-0.24 -
, ,, ,,,,,i , , ,,,,,,, , , , , , , ,7 0.1 1 10 100 Frequency (H7) l t
FIG. 4-4. Floor r'esponse spectrum - (OBE, 0.5% damping) for the 646-f t level of the a>ntainment. building . '(f rom Ref. 11) . j 26 i
dynamic analyses or as' the source for accelerations to be osed in static analyses. For piping or' equipment Detween two or more floor levels, the floor spectrum for the level immediately above the center of mass was used. 4.3.1.2 Equivalent-Static Methom The' equivalent-static method depends on seismic coefficients (in g's) to obtain static lateral forces for structural oesign. The forces are simply the products of the seismic coefficients and structural weights. In the analysis of Palisaces, the coefficients used for rigid piping and for eq- paent were at least equal to the peak floor accelerations. In lieu of a dynamic analysis, some Class 1 piping was designed for a horizontal static load equivalent to i the peak of the horizontal spectrum, combined with two-thirds of this peak value applied vertically. I 4.3.2 Damping Damping values specified for the design of Palisades are given in f Table 4-2, along with a:Sping values for various Class 1 items. ' The
! values acecant for both structural camping and, where applicable, soil damping.
J 4.4 STRESS CRITERIA Stresses resulting from the 0.10-g OBE Excitation, in combination with 2 stresses imposed by nonseismic loaos, were held to coce-allowable levels. In addition, it was required that yield stresses not be exceeded during a 0.20-g SSE. The load combinations used in the design, compiled from the FSAR and Ref. 12, are listed in Table 4-3.~ (Revised stress criteria recently submitted I by the licensee are currently being reviewed by the NRC.) The loading combinations designated as " design loads" for the containment
, building were the basis of a " working stress" oesign. Stress criteria were generally those of ACI Code 318-63,13 with the exceptions outlined in a Appendix B.1 of the FSAR. Other loading combinations for the containment building were-designated 4 'vield loads." These combinations include factored loacs, which retle + evaluations as to the loads most , - ,n r n ~ ~ - - - + . . .
TABLE 4-2. Damping values used in the design of Palisades. l t of critical damping For OBE For SSE ltructural types: Welded steel-plate assekbiles 1.0 1.0 Welded steel-frame structures 2.0 2.0 Bolted steel-frame structures 2.0 2.0 Concrete equipment supports on 2.0 2.0 another structure Reinforced concrete structures 5.0 7.5 on soil Prestressed concrete structure -- -- on soil Steel piping 0.5 0.5. Class 1 items: Containment building and --* -- internal structure b b Auxiliary building -- -- l Turbine building Unk- Unk j Electrical penetration enclos'me Unk Unk Flexible piping 0.5 0.5 Spent-fuel storage racks 4.0 7.5 Class 1 appendages Unk Unk l ar he final analysis of the containment building and internal
. structure used the following values for damping: structural j
damping--24 (OBE) and 5% (SSE); soil damping--54 (OBE) and 104 (SSE); and composite modal damping--54 (OBE, modes 1 and 2) , 24 (OBE, modes 3 and 4), and 7.5% (SSE, all modes). b The OBE analysis of the auxiliary building used St damping for modes associated with reinforccd concrete structures and 0.5% damping for modes associated with steel structures. SSE responses were inferred from the OBE results. 28
TABLE 4-3. Summary of original load combinations and allowable stresses. Load combinations" Design criterion Containment building
- a. D + F + 1.15P Design loads (see text) . Thermal loads are due to the i - temperature gradient through the wall and to expansion
- b. D+F+P+T A of the liner.
- c. 1.05D + F + 1.5P + Tg
- d. 1.05D + F + 1.25P + 1.25E + Tg ,, Yield loads (see text) . Thermal loads as above.
- e. D + F + P + E' +T A rntainment internal structure y a. D+L+E Allowable stresses specified in ACI Building Code, ACI
- 318-63 (Ref.13), or AISC Manual of Stee3 Construction, 6th ed.
- b. D+L+T A Stresses to be less than 133% of (a) above.
- c. D + P + R + E' Local yielding allowed, but not to interfere with safe shutdown.
continued aAbbreviations are explained at the end of the table. 4 i
'AABLE 4-3 continued.
Load or.ubinations Design criterion Other Class I structures
- a. 1.25D + I; + 1.25E Stresses limited to yield strengths of the effective load-carrying structural materials, reduced by an appropriate yield capacity reduction factor. Yield strength for steel was taken from ASTM specifications.
Concrete structures were designed for ductile behavior whenever possible.
- b. 1."25D + 1.25H + 1.25E Same as above, except if the dead load decreases the total stress, 0.9D was used in place of 1.25D.
- c. 1.25D + 1.25H + 1.25W Same as (b) above, w d. D + R + E' Same as (a) above.
o
- e. D + H + E' Same as (a) above.
Reactor internals
- a. D+E Stress criteria of Sec. III, ASME Boiler and Pressure Vessel Code, Article 4.
- b. D+E' Small amount of yielding permitted.
- c. D + R + E' Permanent deformation is permitted.
- continued
l TABLE 4-3' continued. Load combinations Design criterion Spent-fuel racks
- a. D+L Allowable atress,
- b. D+E Allowable stress.
- c. D+E+T o 150% of allowable stress.
- d. D + E' +To 160% of allowable stress.
- e. D+E+T 3 160% of allowable stress.
- f. D + E' + Tg 170% of allowable stress,
- g. D+Tg + stuck fuel
- 160% of allowable stress,
- h. D+Tg + fuel assy drop 160% of allowable stress.
Other Class I systems and equipment
- a. MOL + PTT + E APPl icable code-allowable stress.
- b. MOL + MTT + E Minimum yield stress at appropriate temperature.
- c. MOL + MIT + E' 110% of minintaa yield stress at appropriate temperature.
continued
' TABLE 4-3 continuec.
Abbreviations: D = Dead loads of structures and equipment, plus any other permanent loading that contributes to stress, including hydrostatic or soil loads. In addition, a portion of the live load was added when it included items such as piping, cables, and trays suspended from floors. An allowance was made for future additions to the permanent load. E = Design earthquake load (equivalent to OBE) . E' = Maximum earthquake load (equivalent to SSE). F. '= Effective prestress loads. H = Force on the structure due to the thermal expansion of pipes under operating conditions. L = Live loads. MOL = Maximum normal operating load, including design pressure, design temperature, and piping and support reactions. MTr = Maximum thermal transients during emergencies such as full-power reactor trip, turbine generator trip, loss of auxiliary power, and the design accident. U P = Design accident pressure loads. PTT = Normal thermal transients such as those associated with start-up, shutdown, and load swings. R = Force or pressure on the structure due to the rupture of any one pipe. To = Operating thermal loads. TA = Thermal loads due to the design accident. W = Wind load.
L subj ect to variation and most critical to safety. The stress criteria for the yield loads.were based on the yield and ultimate stress values of ACI Code - 318-63 and the appropriate ASTM specifications. These allowable stress limits were reduced by appropriate yield capacity reduction f actors ($ factors) to account for "small adverse variations in material strengths, workmanship,
. dimensions, control, ' and degree of supervision."13 4.5 -SEISMIC ANALYSIS OF STRUCTURES This section presents the results of analyses used in the original design and- is the' basis of comparisons in Chapter 5 with SEP acceptance criteria.
The original design analyses were not verified as part of this program. 4.5.1 Containment Building and Internal Structures (FSAR, Secs. 5.1.2, } 5.1.3, and 5.1.9; FSAR, Amendment 14, answer to question 5.10; FSAR, Amendment 15, answers to questions 5.8, 5.20, and 5.21; Ref. 12, answer to question 2. A) The reactor containment building, which houses the NSSS, is a vertical, cylindrical, reinforced concrete structure (Fig. 4-5) . Its inside diametet is 116 ft, and its inside height is 189 f t. Containment walls are 3.5 f t thick, the come is 3 f t thick, and the base slab thickness varies between 8 and 13 ft. The containment building was the first in the United States to be post-tensioned with fully prestressed walls and dome. Each of the 845 tendons, stressed to about 800,000 lb, comprises ninety 1/4-in.-diameter, high-tensile steel wires. The building was designed to withstand the internal' pressure (55 psig at 283 F) that would r9sult if the largest primary pipe ruptures. A 1/4-in. carbon-steel liner plate on the inside surface of the containment concrete ensures leak tightness. The primary structures inside the containment building ate the supports for the reactor vessel and steam generators, and the walls and slabs surrounding the steam generators and primary loop piping. l .Three separate seismic analyses were perfcxmed on the containment building and internal structure, each analysis reflecting the stste of the art and the design requirements at that stage in the engineering process. In early 1967, during the preliminary design stage, the containment shell 33
- a-,- c. -, - - - - . . . . - - , . , ,.
l l Cont:inment building El 782' 0"
-~
.P.
Turbine building El 683' 9" Steam generator t < J O Moisture % MJ I I ( f
~
El649'0" separator h Primary- , I w ]/ pumps \f Turbine C / A \ t uctu \ FW eacto t ) heater - vessel _ O Condenser __ V . O __ _ eis30 0-1 I 1 " CircuTation L El 570' 0"
]A "
water L - ~ Engineered pump safeguard access pump rooms tunnel FIG. 4-5. Transverse section of the containment and turbine buildings.
.- . . - . - . =
was modeled separately from the internal structure as a single-stick, lumped-mass , fixed-base model (Fig. 4-6) . The response-spectrum technique was used to generate modal responses that were added absolutely to get the structural responses. The forces so generated were used in the structural design of Palisades. Figure 4-6 shows the response envelopes for this OBE analysis, which used 2% structural damping. Later, soil-structure interaction (SSI) was included in a model of the containment (Fig. 4-7). This modified model incorporated translational soil springs and offset vertical springs to account for rocking stiffness. The mass of the internal ' structure was lumped at the base of the containment. This model was analyzed by the response-spectrian technique, .and the modal responses were combined by the SRSS method. Damping was 44 for the OBE analysis and 7.5% for the SSE analysis for all modes. The results shown in Fig. 4-8 were reported in Sec. 5.1.3 of the FSAR. 16.73 5.20 44 f t o 4o .27
-- 0 7.39 37 ft
-h0 7.39 m -
7.2o 4.s4 38 f t
-LO 7 39
" - 8 o8 7.09 37 ft j 407.39 8.82 95 19 ft s t ~ \10.99 A w,777 77n?77 Weights in 103 kips inertial force (103 kips) boear (103 kips) Moment (105 kip-ft)
FIG. 4-0. (a) Preliminary single-stick model of the containment shell W h noce weights; (b) inertial forces,. shear envelope, and moment envelope from an OBE analysis of the preliminary model (from Ref. 12) 35
l ? 16.73 44 f t t
--()7.'39 37 ft E = 5.50 X 103 kips /in.2
' G = 2.35 X 103 kips /in.2 i
" -() 7.39 l = 4.84 X 1010 in.4 38 4 A = 1.89 X 105 in.2 l
()7.39 37ft l 5
--l(T)7.33 kH = 1.54 X 10 kips /in.
- l19 f t; 4.25 ft 40.0 Ky= 0.57 X 105kipsiin.
l 120 f t---l i FIG. 4-7. Mathematical model used in the second analysis of the containment l building (f rom FSAR, Sec. 5.1.3). The mass of the internal 3 structures is l lumped at the base of the model. Weights are given in 10 kips. ! 4.83 l 4.82 1.21 2.13 l , 6.05 1.21 4.44 6.59 1.31 6.94 6.46 l l 1.33 7.26 9.46
-8.90 ---
16.10 10.33 Inertial. force (103 kips) Shear (103 kips) Moment (105 kip-ft) FIG. 4-8. Response envelopes f rom a response-spectrtsa analysis using the model of Fig. 4-7 and an OBE-of 0.1 g (from FSAR Sec. 5.1.3). 36
~. . . - - - - -- -. ._-
J In June 1969, the third 'and final seismic analysis was completed. Both SSI and coupling effects between the containment shell and the internal structure were included in a two-stick model (Fig. 4-9 and Table 4-4), which J was analyzed by the response-spectrum technique. Responses of the first four modes (illustrated in Fig. 4-10) to OBE excitation were combined by the SRSS method to produce the envelopes shown in Fig. 4-11. Other details of the analysis are presented in Amenchment 15 to the FSAR and in the response to question 2.A in Ref. 12. Amendment 15 states: This combined system produced lower forces on the containment building than were produced by the single-branch model of the containment used for oesign purposes. The design seismic forces on the containment building , are therefore conservative. 4 This third model was also used to generate the in-structure response spectra for equipment qualification, as discussed in Sec. 4.3.1.1. o [ Elevation, f t 765 - 0 Containment
' vessel l
721.25 - o 2 u 683.75 - o 3
- o -
o - 649
': 646.25 ~ " Concrete 6
4 internal structure ] _ J 608.75 - o "
\
\ M-
/
I 590 s a Kg = 1.54 X 108 lb/in.
,my-{
-- ,-- -- - g-lJ Lf Ky j12 53j Ky = 0.57 X 108 lb/in.
m 120 ft - FIG. 4 . Two-stick, lumped-mass model of the containment building and internal structure (f rom FSAR, Amenchnent 15) . 5 1- 37
,~ - .- - _. - - .
Containment
- O O
\
O O Internals O O C O O I a I Mode 1 Mode 2 1.940 Hz 5.521 Hz
- O O O O 0 0 0 l
0 0
- l -
Mode 3 Mode 4 12.958 Hz 20.284 Hz FIG. 4-10. Principal mooe shapes for the two-stick containment building model shown in Fig. 4-9 (f rom FSAR, Amendment 15) . 38
S TABLE 4-4. Member properties of the two-stick. containment building model shown in Fig. 4-9 (from FSAR, Amendment 15) .a E, G, Member A,ft2 x Ay ,ft2 I,ft z 4 10 3 kip /ft 2 103 kip /ft 2 1-5 1,310 874 2,340,000 792 338 6 1,429 814. 833,513 576 246 7A 1,451.2 865 465,209 576 246 7B 1,188.7 811 384,312 576 246 8 1,376 865 336,503 576 246 9 - 11 278,000 185,000 9,999,999 792 338 12, 13 10.0 6.67 100.0 683 -- 14 10.0 6.67 100.0 1,840 --
" Abbreviations: A, shear area; I *, moment of inertia; E, modulus of elasticity; G, shear modulus.
0.205 - 3433 Containment
/0.148 -
4506 150,200
- 0.112 -
5232 319,060 Internals 0.126 - 809 i 0.099 .- 5667 514,540 . l 0.120 - 1700 21,854 0.109 - 5890 -724,720 0.119 - 2810 48,960 0.119 - 0.119 -
, 0 93,620 Shear in base slab 9428 Accleration (g) Shear (kips) Moment (kip ft)
FIG. 4-11. Response envelopes from an OBE response-spectrum analysis using the model in Fig. 4-9 (f rom Ref. 12) . 39
m The containment was analyzed for axisymmetric and nonaxisymmetric loads. The axisymmetric loads--dead loads, pressure and thermal loads associated with an accident, and the prestress loads--were calculated using the FINEL code and an axisymmetric, finite element model of the cc atainment that neglected buttresses, penetrations, brackets, and anchors. Loads arising from these nonaxisymmetric features, seismic and wind loads',' and concentrated loads were all considered in the nonaxisymmetric analysis described in Sec. 5.1.3.2 of the FSAR. Seismic loads were larger than wind loads in all cases. The resultant stresses are documented in Table 5-1 of the FSAR. 4.5.2 Auxiliary Building (FSAR, Sec. 5.2; FSAR, Amendment 15, answers to questions 5.8 and 5.16; Ref.11, answer to question A.1; Ref.12, answer to question 2.C) The auxiliary building comprises a reinforced concrete system of interconnecting floors and wclls to an elevation of about 649 f t, above which rise structural steel framing and attached nonstructural walls. The building is constructed on a mat foundation. Two two-dimensional models of lumped masses and beam elements were developed for the auxiliary building--one each for the north-south and east-west directions (Figs. 4-12 and 4-13) . At the base of the structure, as for the containment building, offset vertical springs a.M translational springs simulate SSI. Detailed model properties are tabulated in Ref.11. Stiff ness coefficients at the mass locations, which were determined from the model properties, were input to a Bechtel code that calculated system f requencies and mode shapes. The mode shapes and materials were evaluated to determine OBE modal damping values--0,5% for modes associated with steel portions of the structure and about 5% for those associated with reinforced concrete portions. The latter value accounts for the effect of soil damping. An OBE analysis of both models, using the response-spectrum technique, resulted in the maximum structural responses shown in Fig. 4-14. Medal responses were combined by the SRSS method. SSE responses were calculated by doubling the OBE values--a simple method that produced conservative values because of the lower OBE damping. ] Torsional effects, which arise from the asymmetry of the building, were considered in the design by distributing design horizontal loadings (obtained 40
Elevations: 695'8"
,, @[ N Mass Weight pt. (kips) 1 7,430 670' \ / 2 3,260 3 '10,788 4 13,639 674* N
***' 6" T/
5 6,677
, . N/ *
E 659'
'^. N/
649' / 640' =@ m
'624' =
=@
610'- :
-601' s 589' -
b " i 1 l 579' ' I I
-50' - 1 1.5' O 11.5' 50' GO' FIG. 4-12. Mathematical model used in the analysis of the north-south seismic response of the auxiliary building (f rom FSAR, Amendment 15) .
4 from the decoupled seismic system analyses) in accordance with the actual shear-wall rigidity calculation. No dynamic coupling was considered between the auxiliary and turbine buildings because the connections between them at the mezzanine and operating l floors were designed with slotted bolt holes to provide a 3-in. gap. 41 e
e Elevations: 695'8" , 663'6" : : Steel framing i @ 649' ll ,, ,'l l Reinforced concrete 640' - 624' : -- : 610' : :- 601' : :- O 589'h Mathematical model used in the analysis of the east-west seismic l FIG. 4-13. response of the auxiliary building (from FSAR, Amendment 15) .
)
l 42
4 0.092 - 0.440 0.090 - 4.1 1.617 O.0L9 29.8 3.734 0.085 - 82.0 5.220 0.084 129.0 5.634 0.083 I 196.5 Shear in base 6.555 Maximum displacement (in.) Shear (103 kips) Mo' ment (103 kip-f t) FIG. 4-14. Maximum displacement, moment, and shear envelopes for the auxiliary building (from FSAR, Amendmct 15). Each is a composite of the north-south and east-west seismic analyses, showing the maximum value calculated at nodes 1 through 6. 4.5.3 Turbine Building (FSAR, Sec. 5.3.1; FSAR, Amendment 15, answers to questions 5.8, 5.16, and 5.21; FSAR, Amendment 17, answer to question 8.0; Ref. 11, answer to question A.1) The turbine building (Fig. 4-5) is a steel-frame building with insulated siding. Within the building, reinforced ccacrete enclosures house CL.ss 1 equipnent. The building was analyzed for SSE excitation (A, 0.20 g) using the response-spectrtzn technique. The building was modeled only for the less-rigid east-west direction; the model was a system of lumped masses and stiffness coefficients. Three cases were analyzed: 43
e Case 1. The turbine building frame was considered to be restrained solely by its tie to the auxiliary building--by encasement of the secondary columns of the frame in the auxiliary building wall. It was concluded that this tie will not cause f ailure of the auxiliary building wall or the roof over the turbine building auxiliary bay. e Case 2. The turbine building was considered to be a rigid frame, supported at the ground-floor level and unrestrained at the operating-floor level. Maximum frame deflection at the 625-f t elevation was calculated to be 3.4 in., enough to close the 0.75-in. gap between the turbine generator _ pedestal and the operating floor and to cause the pedestal to act as a restraint. This closure necessitated analysis of the third case. e Case 3. The turbine generator pedestal was treated as a restraint to the building frame at the operating-floor level. It was concluded that the resulting lateral force would not affect the pedestal. l In all three cases, the crane was assumed to be unloaded and located at any bay of the turbine building. Mode shapes from the three analyses indicate that the crane support columns move in the same direction and have a maximum deflection of 2.75 in, at the roof. Based on the uniform movement of the columns at the crane rail and the f act that column stresses are below those allowed, it was concluded that the building will not collapse and that the crane bridge and trolley will remain in place. In addition, steps were taken to ensure
- hat the 2.75-in. deflection would not close the gap between the turbine and auxiliary buildings, inducing additional forces that would have to be borne by the concrete shear walls of the auxiliary building. Section 5.3.1 of the FSAR states:
The resul,ts of the turbine building dynamic analysis for the 0.2-g earthquake showed that the auxiliary building floor slab would be overstressed due to the direct connection of the turbine building girders to the auxiliary building wall columns and the large openings in the auxiliary building floor slab. The overstress condition has been eliminated by cutting the associated turbine building girders, providir; vertical' supports with sliding surf aces at the girder cut points, and providing a 3" gap between Ethe girders and the auxiliary building wall columns. The 3" gap is greater than the seismic displacement of the turbine building at the 625-ft elevation. i 44 1
The' earthquake-induced deflection would also close the gap between the turbine building and the intake structure on the west side of the turbine building. However, at the level of the intake structure roof, the predicted deflection is less than 2.75 in. The intake structure walls can absorb the induced stresses without exceeding 85% of the yield stress. 4.6 SEISMIC ANALYSIS OF PIPING (FSAR, Secs. 4.3.6, 6.1.2, 9.1 through 9.4, 9.7, 'and 10.1; FSAR, Amendment 15, answers to questions 5.4 and 5.8; Ref.12, answers to questions 2.F and 2.G) The following piping systems were considered in the original seismic analysis: o Primary coolant piping. e Saf ety injection system piping, e Steam and power conversion system piping. Only part of this system was designed to Class 1 requirements. e Service water system piping. Only part is Class 1. e Reactor primary shield cooling system piping. e Component cooling system piping. The portion of the system outside containment is Class 1. o Fuel pool cooling system piping, o Auxiliary feedwater system piping. Only part is Class 1. Large piping, having an inside diameter greater than 3 in., was idealized as a series of lumped masses separated by elastic members. The masses were located to represent the dynamic and elastic properties of the system. For example, all valves, supports, elbows, tees, and other connections were represented by lumped masses. Seismic responses were calculated using the response-spectrum method and 0.5% damping. The floor response spectra were developed as described in Sec. 4.3.1.1. Typically, all modes with frequencies less than 20 Hz (up to a maximum of 10 modes) were considered. Modal ! responses were combined using the SRSS technique. Smaller Class 1 piping systems were analyzed as rigid systems (defined as having fundamental frequencies greater than 20 Hz) . The rigidity of these systems was ensured by permitting no unsupported insulated-pipe spe,ns greater
~
45
i than .those indicated in Table 4-5. Restraints were also placed as near as possible to bends and concentrated loads auch as valves. Stresses on rigid systems were then calculated on the basis of static loads corresponding to the peak acceleration of the appropriate floor. All static and dynamic piping analyses were carried out twice, once for each of two orthogonal horizontal excitations. A simultaneous vertical excitatioa equal to two-thirds of the horizontal was applied in each case. l Piping between two structures was also designed to withstand stresses imposed l by the maximum differential movement between horizontal restraints. These l stresses were combined with other seismic stresses by the SRSS method. Detailed stress results were reported for only two piping systems: the f auxiliary feedwater system and the main steam line (steam and power conversion l system). The maximum OBE and SSE stresses on each were between 714 and 99% of allowable values. Stresses on rigid piping were reported to be "relatively l 10w." l TABIE 4-5. Maximum permissible spans for rigid piping systems (from Ref.12, answer to question 2.F) . l l Piping i.d., in. Span 0.5 4'3" 0.75 5' 1 5'10" 1.5 7'2" 2 8'2" 2.5 9' 3 9'10" 4 11'3" 46
4.7- SEISMIC ANALYSIS OF EQUIPMENT 4.7.1 Control Rod Drive Mechanism (FSAR, Sec. 3.3.4 and Amendment 15, answer to question 5.23) I The control. rod drive mechanisms are mounted on flanged nozzles atop the reac.or vessel closure heao. Horizontal interconnections provide lateral stability and restrict the deformation of the control rod shrouds. It.was determined by experiment that a shroud deformation greater than 0.76 in, would make it impossible to insert the corresponding control rod into the core. A conservative allowable deformation limit of 0.51 in. was adopted.
- Under extreme loading,.the. shrouds of the first row of control rods were
! found'to deform more than 0.51 in. Safe snutdown would not be jeopardized, however, since all other shroud deformations were within the allowable limit. Other Reactor Internals (FSAR, Sec. 3.3.4 and Amendment 15, answers'to
~
1 4.7.2 questions 5.8 and 5.12) t i The reactor internals comprise the core support barrel, the upper guide structure, and the flow skirt. Among their functions is the transmission of control.roa dynamic loads and other loads to.the the reactor vessel flange. 7-The method used for analyzing the internals was not clearly specified, but- the-i input acceleration was given as the acceleration of the concrete mass at the ! reactor support. Vertical and h'orizontal accelerations were applied simultaneously. 4 In all cases, the stresses due to pipe rupture were much larger than those due to seismic loLos. U.. der extreme loading (D + R + E'; see
. Table 4-3), yielding was fo'ad to occur in the first row of control rod shrouds. Other: stres:::_. were within allowable limits.
4.7.3 Other Major Class 1 System Equipment (FSAR, Secs. 4, 6.1 through 6.3, , 9.1 through 9.4, 9.7, and 10; FSAR, Amenchnent 15, answers to questions
- 5. 8, 5.34, and 7. 7)
The following Class 1 systems were analyzed using an equivalent-static +
. methods i
! 47
-. _ . -_ _ - . _ . ~ - . - _ _ _
e e e Primary coolant system. 1 e Safety injection system. I e Containment spray system. e Service water system. e Reactor primary shield cooling system. e Component cooling system. e Auxiliary feedwater system. Components were modeled as single-degree-of-freedom masses, and natural frequencies were estimated from the deflections produced by known loads. - Based on these natural frequencies, seismic loads were then taken from the appropriate floor response spectra. In some cases, loads greater than those indicated by the floor responses were used in the analysis. Vertical and
, horizontal seismic accelerations were applied simultaneously. (The cerivation of the floor response spectra is discussed in Sec. 4.3.1.1.)
No detailed analysis results are available; however, all components were reported to have been conservatively designed to withstand seismic forces , greater than those specified. In several cases, seismic loads were insignificant compared to loads imposed by other postulated accident conditions. 4.7.4 Other class 1 Equipmenc Other Class 1 equipment incluces: 4 e Electrical cable conduits and trays (FSAR, Amendment 15, answer to question 5.6; Ref.12, answer to question 2.K) . e Appendages to Class 1 systems, piping, and equipment (FSAR, Amendment 15, answer to question 5.17) . i e Battery racks (Ref.12, answer to question 2.K) . { Detailed analyses of the cable conduits and trays are not available; 4 however, it is expected that seismic activity would only crack the concrete encasements of underground wires. The flexibility of the conduits and wires, both above and below ground, ensures that electrical continuity would not be disturbed. (Since 1971, cable tray supports have been explicitly designed to 48
resist horizontal seismic ground accelerations of 0.05 g.) ; Appendages were analyzed independently of the Class 1 components to which they are attached; however, their masses were accounted for in the analyses of the major components. Appendages to Class 1 piping and equipnent were analyzed statically; appendages to major Class 1 systems were analyzed by the response-spectrum method. Battery racks were analyzed by the equivalent-static method. For the OBE analysis, horizontal and vertical seismic acceler stions were 0.15 and 0.067 g, respectively; for the SSE analysis, twice these values were used. Additional information on the original seismic design qualification of electrical equipment is stamarized in Chapter 6 (Table 6-6) . l I. l 49
l t i I CHAPTER 5: REASSESSMENT OF SELECTED STRUCTURES l s.1 INA'RODUCTION In this chapter, the seismic loacs and responses on which the Palisades structural designs were cased (see Chapter 4) are comparea to corresponding aeismic loads and responses derived using SEP seismic evaluation methods. This comparison is made to icentify those structures that essentially meet SEP seismic criteria and those that need.to be investigated further. Seismic-loadings and responses are examined for the containment and auxiliary ouildings, field-erected tanks, and a typical buried pipe. In addition, seismic uesign loadings are compared to seismic input motion based on current oesign practice for locations throughout the containment and auxiliary builoings where equipment and piping are supporteo. Since the completion of the Palisades plant, a number of changes in seismic ucaign methous and qualitication criteria for structures and equipment l have occurred. These changes do not necessarily imply that old seismic qualification criteria were inadequate, merely that the criteria are now better defined ano require less interpretation by the designer . The general i j trend has oeen to increase i e Allowable stresses for the specified seismic loading function. i e Allowable damping. l e Number of loading conditions to be considered simultaneously. ! e Degree of cophistication to be used in the analyses. e Quality assurance requirements. 5.2 DESIGN EARTHQUAKE MOTION 1 In describing the design earthquake motion for a given site, several items of information are required: o Peak ground acceleration, together with either design ground response i spectra or a uesign time history. ; i l SC
e How and where in the structure the oesign inputs are specified (such as at the base slab, in the free field, etc.) . e Simultaneous directional components. e Duration or. number of strong motion cycles. Tnis section compares the ground motion parameters specified for Palisaces with the SEP acceptance criteria. 5.2.1 Peak Ground Acceleration The regulation currently go*ierning scismic design of commercial nuclear power plants is 10 CFR 100, Appendix A. It sets forth the principal seismic at.d geological consideraticns to be used in determining such design bases as requirements for the OBE and SSE, for peak acceleration levels, and i for design response spectra. As discussed in C.. apter 4, Palisaces structures were designeo for an OBE and an SSE with peak ground accelerations of 0.1 and 0.2 g, respectively. A simultancci.s vertical component of earthquake motion equal to two-thirds of the horizontal component was considered in the plant cesign. For this reevaluation, an SSE characterized by a 0.2-g peak horizontal ground acceleration also was employed. Although a probabilistic evaluation of the seismicity of the Palisades site may justify a lower value, it was consicered unlikely that a level higher than 0 2 g would be required. 5.2.2 Ground Motion Characteristics In addition to the peak ground acceleration, either a design time history (or histories) or ground response spectra are necced to uefine a cesign earthquake. Typical current practice is to specify either site-dependent spectra or, more of ten, ground response spectra like these in R.G.1.60. 6 Tnese latter spectra are cased on the mean plus one stancard ceviation of spectra generated from a series of strong-motion earthquake records that include horizontal and vertical components for both rock and soil sites. I currently, time-history analyses are based mostly on artificial earthquakes wnose _ response spectra envelop the smocthed R.G.1.60 cesign spectra. Rather than directly compare response spectra for equal damping ratios, 51
. .- - _ . ._m ~ . . . . . ..
4
'it is more informative to'consicer also the ' damping esed in the design of a
Palisades. Table 5-1 lists the damping ratios used for Palisades, together
.with those from R.G. 1.61 for the SSE and those recommended in
- j. -NUREG/CR-0098 for structures at or below the yield point and at
.approximately one-half the yield poir' In. general, the damping ratios used in the design of Palisades are lower
- than those now in use for SSE cesign analyses. The amount of soil damping allowed by the NRC to be used in conjunction with R.G.1.61 structural or component damping values has been the subject of extensive discussion in the Past. However, including any reasonable amount of soil damping results in
]. .nigher damping tor all systems than was used in cesign. l In the original analysis, 7.5% of critical damping was used for the SSE for all moces for both reinforced and prestressed concrete structures. In the reanalysis, different values of damping were used for different modes, because more sophisticated methods for computing geomecric (radiation) damping in the soil for layered sites'were available. Composite modal damping was used to account for the variations between soil and structural uamping. Tae actual geometric and composite modal damping values used are discussed in Secs. 5.3 . and 5.4.
- In general, the composite modal damping values used in the reanalysis range from 10% to 20% of critical. To obtain a rough estimate of the expected
] variation in response, the original Housner 7.54-damped ground response spectrum used for cesign is shown in Fig. 5-1, t.ogetner with the 10% and 204
- (interpolated) spectra from' R.G.1.60. For the median soil case, the l fundamental frequency of the containment ouilding is approximately 2 Hz. As is apparent from Fig. 5-1, the-response resulting from use of the original l 7.St-camped spectrum is. approximately the same as would be expected if .the 20%
I R.G.1.60 spectrum were used for all modes. However, for modes with closer to 10% damping, an increase in response of as much as 50% can occur in the frequency range of interest. Apparently, no analysis of buried pipe was conducted during the original-1 cesign. Also, the incomplete soils cata available do not include any determination of soil strains expected during the SSE. For the SEP reevaluation of the buried pipe, the soil strain (c ) was conservatively
-4 * ""*
determined to be approximately 2 x 10 in./in., as obtained from the relationship 4 52
-- - . - _ - - ~__
l l TABIE 5-1. . Original and currently recommended damping ratios, expressed as percent of critical damping. R.G. 1.61 NUREG/CR-0098' NUREG/CR-0098 Original DBE Original SSE (SSE) (yield levels) (working. stress). Welded steel-plate 1 1 4 5 to 7 2 to 3 assemblies Wel6ed steel-frame 2 2 4 5 to 7 2 to 3 structures Bolted steel-frame 2 2 7 10 to 15 5 to 7 structures toncrete equipment 2 2 7 7 to 10 3 to 5 supports on another structure
- c. Reinforced concrete 5 7.5 7 7 to 10 2 to 3
" structures on s il (+ soil) (+ soil) (slight cracking)
(+ soil) Prestressed concrete 2 to 5 7.5 5 7 to 10 2 to 3 containment structure (+ soil) (w/ loss of (w/o loss of on soil prestress) prestress) (+ soil) (+ soil) Steel piping 0.5 0.5 2 to 3 2 to 3 1 to 2
Frequency (Hz) 100 50 20 10 5 2 1 0.5 0.2
/i\ IX 19 /IN K XJ /i\ A XI /
200 ( 100 50 2 y R.G.1.6010% damping 3 gO R.G.1~60 20% damping / 20 g Housner (7.5% damping) g A
\ / O 10 Q % $
5 5 y 4 # - D N / A 09 93 s2 4 o o,
^
1 y f % 0.5 y g o## O-a os
)
O/ 9 O je ) 0.2 ( 9 Cb
@ Tc gol D
0.1
' # % fE 0.05 A A A\ ^ ^^A A\ /
0.01 0.02 0.05 0.1 0.2 0.5 1 2 5 10 Period (s) FIG. 3-1. Comparison of origitaal 7.5% Housner design spectrum to R.G.1.60 spectra for 10%_and 20% of critical damping.
" max I"a max
- C g where Vmax is the maximum ground velocity and C is the apparent longitudinal hor'izontal propagation speed of the seismic waves with respect to the 6 t.r ucture. No geotechnic investigation was conioucted to cetermine the )
apparent longitudinal horizontal-wave propagation speed. It was 54
l l i conservatively estimated at approximately 4000 f t/s, which accounts for the
.p roximity to bedrock.
5.3 SEISMIC DESIGN METHODS As previously discussed, seismic analysis methods have changed since the oesign of.Palisaces. Palisades was cesigned for both OBE and SSE conditions;
~ for the concrete structures founded on soil, increased damping at the SSE response levels was considered. On the other hand,.the conservative OBE damping values were used for both the OBE and SSE analyses of all other structures and components.
The analysis of the Palisades structures was conducted using two-cimensional mocels simulating the N-S and E-W responses. No vertical dynanic analysis was conducted. Planar 2-D models are adequate to compute the response of symmetric structures such as the containment ouilding; however, such planar models cannot be used to calculate out-of-plane or torsional responses, which may be significant for such nonsymmetric structures as the auxiliary building. Current analytical techniques and compute:: models are considerably more sophisticated, ano current licensing requirr.ments would typically require additional load combinations resulting free other transients. Because these combinations are unlikely, however, our reevaluation has concentrated on the original design combinations, with primary attention devoted to the seismic margins. Several current assumptions and criteria are discussed below and compared with those used in the original oesign and analysis of Palisaces. 5.3.1 Soil-Structure Interaction The response of the Pal'isades structures was originally calculated on the basis of lumped-mass, two-dimensional models of the containment and auxiliary nuildings. The structural models were supported on f requency-independent springs representing the soil flexibilities. These soil flexibilities, in turn, were calculated using elastic half-space theory, assuming a soil shear 6 , mooulus G of 6.4 x'10 lb/ft and a Poisson's ratio of 0.25. How these values were obtained is unclear. The original design models were developea on the assumption of a rigid 35 I
Dese slab. A horizontal spring simulating the horizontal soil: stiffness was attached to the base slab, and the rocking stiffness was simulated by two
. equivalent vertical springs attached at opposite ends of the slab (Fig. 5-2) .
The vertical springs were used only to simulate the rocking stiffness; no vertical analysis was conducted. Also, no torsional response was computed for either the containment building or auxiliary building. Much more sophisticated methods of analysis currently exist for the calculation of SSI effects, particularly for layered sites such as Palisades. However, it was decioed that the soils information currently available for Palisades does not warrant a finite element or frequency-dependent compliance funct. ion analysis as part of the SEP. Rather, a frequency-incepencent analysis that included the effects of the layered site was selected as being consist-ent with the level or soils information available, as well as the sophistication of the structural models. Uncertainties were treated by consioering a fairly broad range of soil properties. Elevation, f t 0 765 - o j Containment vessel
- 721.25 - o l 683.75 - o 3
I - 646.25 - "
~
Concrete 6 4 internal structure f o - 622 3 608.75 - 5 o[-606 8 Kg = 1.54 X 108 lb/in. I j D LI ] 10 9 11 "[4" I K y g12 13q]Ky = 0.57 X 108 lb/in.
+ - 120 f t . . FIG. 5-2. Two-stick, lumped-mass model of the containment building and internal structure (f rom FSAR, Amendment 15) .
56
Among the most useful of the current SSI analysis methods for -layered sites are those of Kausel ' and Luco. ' Both have advantages and uisauvantages when applied to the Palisaues site. The original study by Kausel et al. does not trett vertical response, while Luco's metrod does not directly' include the effects of embedment and leads to a freqacncy-uependent solution for fairly wide parameter ranges. However, by utilizing
.octh methods and ootaining a check of the parameters that can be computed inaepenoently by the two methods, a solution can be obtained that reflects the la}ered-site characteristics had is consistent with the information available for the soil.
In calculating the soil spring rates for the site, Poisson's ratio was chosen as 0.46, in accoruance with recommencations in Refs. 20 and 21 for 6 saturated sands. A meaian soil shear modulus of 4.38 x 10 lb/ft was calculated using the approach outlinea by Seea and Idriss to account for the recuction in stiffness as a function of increased soil shear otrain tar the 0.2-g SSE. The radius of the containment building base slab is 60 f t, and toe septh to ucdrcck is approximately 150 f t. An emoedment cepth of 17 ft was useo in the analysis, which corresponds to the depth f rom grade to the bottom of the base slao, out neglects the sloping backtill on cne side of the str uc ture. In using Kausel's method, only half the embedment effects were used in computing the horizuntal and rocking spring rates. .This accounts for the fact that.a cohesionless soil is unable to develop any significant tensile capacity. Cracks woulo, therefore, be expected to form between the soil and the vertical structure surf aces on the tensile side during an earthquake in tue range of 0.2 g. Only the effects on the compression sice would be felt. Similarly, no embedment eff ects were incluced for the vertical spring rate. Atter one or more strung lateral response cycles, cracks between the soil and the vertical surf aces could be expected over much of the total area for much of a vertical response cycle. Also, since the enbedment is relatively small, the coupling that accounts for the interaction between the horizontal and rocking moces was neglectec. Table 5-2 presents a comparison of the frequency-indepenaent containaant oailding spring retes computed by the various techniques. Also shown for comparison are the horizontal ana rocking springs used in the original analysis. The norizontal spring rate calculated using Kausel's method without
.embeument is virtually identical with that computed using Luco's 37
TABLE 5-2. Comparison of median-soil spring rate constants and geometric damping values for. Palisades containment building, calculated in different ways.a The subscripts h, 4, and v represent horizontal, rocking, and vertical modes, respectively. Values marked by asterisks were used in the reanalysis. Original design Elastic half space Kausel Kausel- Luco (static) (no embedment) (no embedment) (no embedment) (w/ embedment) (no embedment) Spring constants: kh, lb/f t 1. 85 x 109 1.3 x.109 1.63 x 109 *1. 92 x 109 1.63 x 109 kg, lb-f t/ rad 4. 92 x 1012 4.58 '< 1012 4.89 x 1012 *6.59 x 1012 4.77 x 1012 ky , lb/f t -- 1.97 x .109 2.89 x 109 3.25 x 109 *2.87 x 109 Damping constants:b Dh, t -- 34 34 34 -- w
- D, t --
2.5 4 1 1 0 D,y % - 60 - 6 3 aValues in the four rightmost columns are based, from lef t to right, on Refs. 21; 17; 17 and 18; and 19. b 5% soil material damping used for all modes.
approacia. The -rccking springs aitfer by less tnan 34. These spring rates are
-higher than the values computed using elastic half-space theory. Including the embedment effects and a higher Poisson's ratio further stiffens the system. On the other hand, the lower shear modulus tends to sof ten the system compared to the original cesign values. The results for the median soil case
.are that. the horizontal spring rate is approximately 44 higher than that originally usea, while the rocking spring is about 344 higher. Most of the increase is directly attributable to including the embedment effects. The vertical spring rate used in the SEP reanalysis, calculated using Luco's' methoo (assuming no frequency dependence) as for the horizontal and rocking springs, is almosti identical to that determined using Kausel's method. To further account for uncertainties resulting from the lack of soils information and the analytical approximations described above, the = structural evaluation of Palisaces was conoucted assuming a +50% variation in the soil shear modulus. This results in a range of soil modulus values from 30% to 90% of the low-strain value. The extreme soil values and the best estimate are referred to as the upper- and lower-bound soil cases and the median soil case.
Geomectic (or radiation) damping considerations are also very important i in tne solution of SSI problems. However, such considerations are complicated by tue layereu-site characteristics at Palicaces: Energy tends to be
- reflected from the soil-bedrock interface. The result is significantly lower damping for some moces than in an equivalent elastic half space. This is particularly true for the vertical response. In addition, the damping values i
tend to be more sensitive to small frequency variations than equivalent half-space values. ~ Depending on the f requency of the system and the stiffness of the beorock, the damping may be limited to only the internal (material) camping of the soil, essentially no geometric damping may exist. The geometric damping l values calculated for the Palisades containment building are shown in Table 5-2. Horizontal and rocking damping ratios of 344 and 14, respectively, were calculated in accorcance with Ref. 17. The horizontal camping f corresponds to the half-space value, whereas the 14 rocking damping is sumewhat below the 2.5% for an equivalent elastic half space. Luco's method indicates virtually no radiation damping for the rocking mode. For the vertical response, the variation in the geometric damping is much more pronounced for the layered site, since the underlying bedrock tends to 59
# -- .< .. = ,- ~
reflect most of the elastic wave energy back into the structure. TLu vertical camping for the Palisades site, as calculated from Kausel's recommended empirical relation, is 6% of critical. Although none of the cases analyzed by Kausel corresponds exactly with the Palisades site parameters, the vertical damping obtained for the case closest to Palisades indicates that the 6% value is conservative. Using Luco's method ' and a rigid bedrock interface, no geometric damping is indicated for the vertical response. However, when using the actual bedrock properties and an equivalent-static approach consistent with that used to compute the vertical spring rate, we obtaineo a vertical geometric damping of approximately 3%. (These values may be compared to the 60% vertical camping for an elastic half space.) Small changes in f requency result in very large changes in the calculated vertical damping for the layerea-site values. Unfortunately, the limited soils information available does not permit an exact calculation of the system trequencies. Also, the results of Luco do not include a case for Poisson's ratio of 0.45. In view of these uncertainties, a vertical geometric damping of 5% of critical was used in the reanalysis. A similar procedure was followed for the auxiliary building, with an equivalent rectangular base slab in place of the circular base slab used for the containment building. Table 5-3 shows the soil spring rates and geometric damping values calculated for the median soil case. No embedment exists for the auxiliary building. In addition to the geometric damping values discussed above, 5% soil material damping was used for all moues for both structures, in j accordance with recommendations in Ref. 21. l 1 5.3.2 Combination of Earthquake Directional Components
)
The design of Palisades was based on the absolute addition of one horizontal and one vertical load component. Current recommenced practice is to combine the responses for the three principal simultaneous earthquake directions by the SRSS method, as described in R.G.1.92. Alternatively, it is recommendeo by Newmark and Hall that directional effects be combined by taking 100% of the ef fects due to motion in one direction and *0% of trae
- effects from the two remaining principal directions of motion,
- j. the result of an SRSS combination of three components, compared to an absolute addition of two, depends on the relative magnitudes of the responses 60
i TABLE 5-3. Comparison of median-soil spring rate constants and geometric damping values.for Palisades auxiliary building. The subscripts h, $, v, and 0 represent horizontal, rocking, vertical, _ and torsional modes, respectively. Elastic Elastic Original design Original design half space half space Kausela Kausela (E-W) (N-S) (E-W) (N-S) (E-W) (N-S) Spring constants: Kh, lb/ft 1. 91 x 109 2.01 x 109 1.42 x 109 1.49 x 109 1.90 x 109 1.90 x 109 i K ,. lb-f t/ rad 1.19 x 1013 6.35 x 10 12 9.84 x 1(,12 5.06 x 10 12 1.07 x 1013 6.49 x 1012 4 Ky, lb/ft -- -- 1.99 x 109 1.99 x 109 3.42 x'109 3.42 x 109 Kg, lb-f t/ rad -- -- 8.82 x 1012 8.82 x 1012 8.82 x 1012 8.82 x 1012
, Damping constants:
s D, 67 67 67 67 h % D, % -- -- 27 27 1.4 1.7 4 D,y % -- -- 1.'. 6 116 5 5 Dg, % 26 26 26 26 ELuco spring used for vertical cases. h
and the geometry of the structure. For instance, if the two horizontal load components are approximately equal and the vertical component is negligible, the SRSS method results in an increase in stress of approximately 404 for a square plant structure and On for a circular one. Combining the effects oy the 1004/404/406 method produces a 40% increase in stress (compared to direct aduition) for a square building at.d an increase of about 84 for a circular structure, such as the Palisades containment building. If the two horizontal loao components are approximately equal and result in stresses about equal to ' that f rca the vertical component, both SRSS and 1004/404/40% stress combinations are less than the aosolute sum of one horizontal component und the vertical component. Conditions typically f all somewhere between the two cases discussed aouve. For various structures, with eitner rectangular or circular plans, the stresses calculated using current methods of load combinations can vary from about 404 less to *0% greater tnan the stresses on which the original uesign was based. 5.3.3 Combinations of Earthquake and Other Loads The cesign and analycis of Palisades used the load combinations for Class 1 structures and equipment shown in Table 4-3. Load combinations are now specitied in applicable design coues and standards, such as ASME Sec. III, Div. 2, and ACI-349. These codes, which describe the load combination procecures and cases to be consiocreo, tend to be system cepenaent. In general, the trend has been to allow increased stresses for the specified seismic loading, out also to simultaneously increase tne number of load conditions to be considereo. The NRC has endorsed these load combinations, with the exceptions noted in Sec. 3.8 of the Standard Review Plan. Because stresses resulting trom 1 cad cases ano combinations of loads from these more recent criteria are not available, this reevaluation has considered only the etiects of changes in seismic -loading prouuced by applying more' recent methods. The allowable stresses used in the design of Class 1 structures are discussed in Chapter 4. These allowable stresses were used as the basis for seismic margin comparisons. l 62 l
5.4 CONFIRMATORY ANALYSIS OF STRUCTURES
. I
.the original mathematical models of the containment and auxiliary !
. I buildings were two-dimensional planar representations. Such models are ;
normally adequate for symmetric structures, but are not appropriate for structures or equipment where significant asymmetry exists. Because of the high degree of symmetry, a two-dimensional model was considered adequate for the Palisaaes containment building. Therefore, a new two-dimensional model of this structure was developed, incorporating the SSI effects discussed in the previous section. To evaluate the asymmetric effects of the auxiliary building, however, a new three-dimensional model was developed. These models were used to develop new structural loads from a response-spectrum analysis using R.G.1.60 spectra, as well as new in-structure response spectra consistent with the SEP guidelines (see Ref. 4) .
' For the most part, the new models.were similar to the original models, with the exception of the treatment of the SSI effects for the layered site, Mass and stif fness properties were computed and compared with the original values to the extent possible. While the mass of the structure could be computed from availsble drawings, neither the equipnent weights nur the original design calculations were available. The masses used in the original calculations were, therefore, compared with the tributary masses of the structure, combined with a representative value simulating the weight of the equipment. In all cases, the original values agreed closely with the estimated masses; therefore, the original mass values were used in the new models. A similar approach was used for checking member stif f nesses, except that new torsional stiffness values were calculated for the 3-D auxiliary building model. In view of the close agreement between the dynamic properties ! criginally used and those calculated in this reevaluation, the uncertainties in the structural models are expected to be overshadowed by the uncertainty in the soil properties. This greater uncertainty is accounted for by using a fairly wide range of soil values.
5.4.1 Containment Building The model developed for the confirmatory analysis of the containmant building is shown in Fig. 5-3. This model is essentially the same structural 63
Y o Elevation, f( 765 - O Containment vessel 730 O e 721.25 - O 683.75 - O t> 649 Concrete internal structure O 622 ~ 0 616 608.75 O O 606 590- (H -x fM k and k, h 581.25
; d v
i FIG. 5-3. Containment building model used in the SEP reevaluation. The open circles are massless nodes. The horizontal spring symbol depicts both a linear hori; ontal spring and a rocking spring (k4)--see Table 5-2. model that was used for cesign except for nea soil springs thu account for tht- la';ered-site characteristics. Several new massless nodes were also addei at locations where in-structure response spectra were desired. The first several "ibration frequencies of the containment building - 4 64
, -,,v
~ - . ~. .- . -- . - . ~ . . - .
_ . founded on the layered soil site are shown in Table 5-4 for the upper, median, I and lower soil cases. Virtually all of the' lateral-' response occurs as a
~ result of the four modes shown. The vertical response is essentially a single-
. frequency. response with very little amplification'through the structure._ Also shown for comparison are the frequencies of the first four response modes used in-the original. design analysis. The mode shapes for the median soil case are shown'in Fig.'S-4.
. In the original design, a rodal damping ratio of 7.5% of critical was i used for the SSE for all moc.es. In the SEP reevaluation, the.large p aif terences in the. levels 'of geometric ano structural damping for the various modes associated with SSI required the calculation of ccumposite damping ratios. The method selected was originally developed by Roesset, Whitman, and Dobry. In this approach, the composite-damping ratios comprise two terms l based'on energy. proportioning. One term accounts for energy assumed to
' dissipate in viscous form, and the other term is based on the energy assumed i
{ to dissipate in.a hysterctic form. In the SEP reevaluation, the viscous portion is assmed to consist of the horizontal and vertical components of the soil camping. -The remaining portion of the soil and structural damping is proporcioned on the asstumption that the energy is dissipated hysteretically. In computing the composite moaal damping ratios, the structural damping values recommended in NUREG/CR-0098,4 ' and shown above in Table 5-1, were i used. Composite damping ratios were originally calculated on the assumption 4 s ~ TABIE 5-4. Containment building response frequencies, in Hz. 4 Original Upper Median Lower Mode design soil case soil case soil case
-1 1.94 2.42 2.06 1.52 2 (vert) ~-- 5.78 4.79 3.44 5.52 5.78 5.78 4.28 3
4 12.96 13.24 12.35 12.46 5 (vert) -- 17.93 17.74 17.56
'6 20.28 20.23 20.02 19.81 1
- 65
, y-- y,---.- <wv w- - = - - w- - -*- e 'e-----eg-
- y y-- .-, rv.
i 1 o f I I I ( o o , ,,, i l e o o 4 4HP l I I o o O OO oo O 8
, o 4 l , l l 1 o
- 88 8
> , o $ o o o oo Mode 1 Mode 2 Mode 3 Mode 4 2.06 Hz 4.79 Hz 5.78 Hz 12.85 Hz (vertical)
FIG. 5-4. Mode shapes for the first four modes of the containment building for the mecian soil case. The unperturbed model is shown with dashed lines, that the cynamic response in the containment building would produce stresses at or near yield in the concrete. Based on the loads developed using this assumption, however, the stresses proved to be substantially less than the yield levels. The final analysis to determine the loads for the containment building was based on 3% of critical, in accordance with recommencations in Ref. 4 for prestressed concrete at about one-half the yield point. In view of the large amount of geometric damping predicted for the horizontal moces, a maximum composite modal damping ratio of 20% of critical was assumed in the analysis. (Appendix A discusses the consequences of this assumption.) Table 5-5 shows the composite modal damping ratios used in the analysis of the containment building for the median soil case. The dynamic resportse of the containment building, conputed using the methods and criteria discussed in the previous sections, is presented in tne 66
1ABLE 5-5. Modal damping ratios for the containnent building (median soil case) . Damping, Mode Frequency. % of critical 1 2.06 8 2f 4.79 10 3 " L;e 5.78 20 4 12.85 10 5 17.74 4 6 20.02 6 form of shear and bending moment diagrams for the containment vessel and concrete internal structure. Results for the median soil case, as well as for the upper and lower-bound soil conditions, are shown in Figs. 5-5 and 5-6. Also shown in these same figures are the corresponding values used in design. 4 As is evident from these figures, the dynamic loads resulting in the structure from application of the SEP acceptance criteria are, for the lower-bound soil case, not significantly different from the original design loads. The recalculated shears are less than the original loads throughout the containment vessel, and the recalculated bending moments are very slightly less than the original-values for elevations below 600 f t. For the icwer-cound soil condition, the new dynamic loads throughout the internal J structur'e are essentially the same as the design values.
~
For both the median and upper-bouna soil cases, the dynamic loads calculated in the SEP reanalysis exceed the original values throughout both the containment ve,ssel and the concrete internal structure. Seismic loads resulting from vertical response were apparently not calculated in the original aesign analysis. Based on the present reanalysis, the vertical response throughout the containment building is 0.34 g for the lower-bound soil caso and 0.27 g for both'the median and upper-bound soil conditions. Table 5-6 is a stanmary of the load ratios for the containment vessel and concrete internals. (The load ratio is defined as the ratio of the load calculated in the original analysis to that derived in the SEP reanalysis.) 67
l~ l 765 i - i l i 1 i e i
/e Original - l Lower soil l l
r Median soil l
.740 -
i Upper soil ~ i i
!! i
_L__7._. t, 715 - l l . l
~'
I l Containment 1 l
, i
; .l' vessel l -
i l I
= 690 -
l -
; l i t__L
$* \
.__._ .) -
l l
.!! i l
- i I
665 - g ! I - I l l
! I I i F 640 - L _ ~1 _. i__.1 1
. i Internal -
i 8 l structure I I ! l ! I i . i l ! l IH I} 615 - l i
, l g, t
l
-3 I l i I
l 11 590 ' ' ' ' ' l 0 10 15 20 0 5 l i. Shear (103 kip) l FIG. 5-5. Shear distribution in the containment building for the lower, median, and upper soil cases, compared to the distribution calculated in the original analysis. I i er' 68 [ l ! . ._m - . _ , - _ _.__ _ _ - _ . . -
65 , , , , , I I k l - 740
\. \ l
\\ Containn.et i \\ vessel l
l 715 -
\\.
's g i -
i \\
\ \\ l
\ \\
\ g\ Lower soil -
\ l 690 - t
\. Or. .igina, g j l
Median soil 7 s . I .9 s I lii \ g [ Upper soil
! 665 -
\s \s g\
I I
\ \.
\ \ \ j g \.
\ \.
'- \ \ ; -
640 -
\
'\
\ \ l
\ \. l
\
g \. Internal
\ \ structure 615 - \ .\ l
\
\ \. l
\ \. I
\
\
\ k g \ ! Yg I ~
I
" I I N 590 3 0 0.08 0.16 0.24 0 1 2 Moment (106 kip ft)
FIG. 5-t,. Moment distribution in the contairment building for the lower, meolan, and upper soil cases, compared to the distribution calculated in the original analysit.. 69
TABLE b-6. Ratios of the loads calculated for the containment building and the concrete internals in the original ana)ysis to those calculated in the present reanalysis. Soil condition 4 Lower bound Median Upper bound Containment vessel (El 590'): Shear 1.07 0.78 0.69 Moment 0.96 0.73 0.05 Containment vessel (El 610'): Shear 1.01 0.76 0.68 M(.anent 1.0 0.82 0.66 Containment vessel (El 645'): Shear 1.01 0.78 0.68 .i Moment 1.03 0.78 0.68 Containment vessel (El 685'): Shear 1.05 0.81 0.71 Mament 1.07 0.81 0.69 Containment vessel (El 720'): Stear 1.14 0.89 0.78 Moment 1.23 0.96 0.83 Concrete internals (El 590'): Shear 1.02 0.84 0.83 I Munent 1.0 0.82 0.81 Concrete internals (El 605'): Shear 1.03 0.83 0.78 Moment 1.0 0.81 0.70 l Concrete internals (El 625'): Shear 0.82 0.64 0.73 Moment 1.0 0.83 0.69 I l i 70
- , - -I
The original loads were based on the Housner response spectra with 7.5% modal damping; the recalculated loads are based on the R.G.1.60 spectra with composite damping computed as previously discussed. The upper-bound soil case gives the highest response for the containment building. Load ratios of approximately 0.7 are typical for this condition. The effects that loads of this level have on the capacity of the structure are discussed in Sec. 5.5.1. 5.4.2 Auxiliary Building The model developed for the reevaluation of the auxiliary building is a three-dimensional, lumped-mass model, as shown in Fig. 5-7. The locations c' ' he masses in the model account for major discrete masses such as the refueling pool, but other equipment weights were taken to be uniformally distributed throughout the floor slabs. Both translational mass and rotational moments of inertia were included. Stiffnees and center-of-rigidity calculations were based on the concrete shear walls and steel framing of the upper elevations, but any masonry-block wall stiffness values were neglected. All floor diaphragas were assumed rigid. New soil springs were based on the layered-site characteristics and on a rectangular base slab with the same area as the slightly irregular base slab of the actual structure. The natural frequencies below 33 Hz for the auxiliary building are shown in Table 5-7 for the three soil cases considered. Mode shapes for the more important response modes are shown in Fig. 5-8 for the median soil case. No auxiliary building frequencies or mode shapes were available for the original cesign analysis. Composite modal damping ratios based on the same energy-proportioning method used for the containment building were calculated for the auxiliary building. The response was originally calculated using structural damping values for steel and reinforced concrete on the assumption that the response would be close to yield. However, the stresses developed throughout the structure using these values were substantially less than yield. Consequently, the final loads developed for this reevaluation were based on 3% of critical damping ter reinforceo concrete and 74 for bolted steel, which are consistent with Ref. 4. As in the case of the containment building, a maximum moual damping ratio of 20% of critical was used in the auxiliary building analysis. (See Appendix A for discussion.) The composite modal damping ratios 71
4 4 Elevation, ft U 695.7 J l O 668.5 l eN 649.0 640.0 VK e 624.0 W I 610.0 2
' > 601.0
=
Y 590.0 X khand k 4 , k3and k, M l k, and k0 anrn l FIG. 5-7. Three-dimensional model used in the reanalysis of the auxiliary building. All nonvertical elements above 590 f t lie in the x-y plane. Each spring symbol depicts both a linear spring and either a rocking spring (kg) c'. , torsional spring (kg)--see Table 5-3. 72
TABLE 5-7. Auxiliary building response frequencies, in Hz. No frequencies were available for the original analysis. Primary response Upper Median Lower direction Mode soil case soil case soil case (median soil) 1 2.24 2.24 2.23 N-S 2 2.28 2 28 2 27 E-W 3 3.76 3.76 2.60 Coupled 4 5.39 5.06 3.69 N-S 5 6.09 5.12 3.78 N-S 6 6.20 5.30 3.83 E-W 7 6.35 5.41 5.40 N-S 8 7.34 7.29 5.54 E-W 9 9.38 7.74 7.27 Vertical 10 .10.4 10.4 10 4 Coupled 11 22.9 19.6 14.5 N-S 12 25.1 21.0 16.9 E-W . 13 31.1 30.8 30.5 E-W for the auxiliary building for the median soil case are shown in Table 5-8. The shear and bending moment distributions throughout tSe auxiliary building are shown in Figs. 5-9 through 5-12 for botn the N-S and E-W responses and for the three soil conditions. Also shown are the original design values, which were taken as the maxima of the N-S and E-W responses at each location in the structure. The lateral loads in the auxiliary building increase as the assumed soil modulus increases. In general, however, the response of the auxiliary building is considerably less aff ected by soil stiffness than is the ! containment building. The ratios of the auxiliary building design loads to the loads calculated in the present reanalysis for the three soil cases are shown in Table 5-9. The minimum load ratios are at El 640 f t for bending moment; however, this is not a critical location as far as stress in the structure is concert.ed. No cesign loads were available for the steel framing 73
l l T i T f I I I I I I t> t> <> t> I I I O > > 0 Il q ,d O <> <> 0 > 0 dp0 f O <l> q) O t p 6 00 4 O O () O t HD thp t > 0 t> 0 ( f> hl IE h O h (E Mode 4 Mode 5 Mode 6 Mode 7 5.06 Hz 5.12 Hz 5.30 Hz 5.41 Hz N-S direction N-S direction E W direction N S direction FIG. 5-8. Moce shapes for the four most important moces of the auxiliary building for the median soil case. The unperturbed model is shown with dashed lines. TABII 5-8. Modal damping ratios for the auxiliary building (median soil) . Damping, Damping, Mode Frequency % of critical Mode Frequency % of critical 1 2.24 7 8 7.29 9 2 2.28 7 9 7.74 10 l 3 3.76 7 10 10.4 7 4 5.06 20 11 19.6 20 5 5.12 20 12 22.0 20 6 5.30 20 13 30.8 6 7 5.41 9 l l 74
i~ 6% [ , , i 668 - E 649 q - I e 9 I l
) 640 , -
u I I r Lower soil 624 - Median soil T _ f Upper soil N Original li 11 610 - q - 601 3 i I 590 I b I O 10 20 E-W shear 3 (10 kip) FIG. 5-9. Distribution of E-W shear in the auxiliary building for the lower, median, and upper soil cases, cor. pared to the results of the original analysis. l I 75
.c 000 l l l
668 h I i hI. 649 k E ii e N - 3 a 640 ' b w
\ Original Upper soil _
624 m.K
% Median soit
- k. g, Lower so.li V
610 - T N
%'(N -
601
%.%'*% %.b.
I I %b .. 2 4 0 E W moment (106 in.-kip) l l FIG. 5-10. Distribution of E-W moment in the auxiliary building for the l l lower, median, and upper soil cases, compared to the results of the original analysis. 76
696 g l l I
.I I
I I I 668 'k - i _ 649 - - E 8_
% 640 - . -
t g I' Lower soil Median soil 624 - I ~
.f! Original l
. Ij Upper soil I 610 - -I
;g j -
ili. !' l 601 - i{ y-] I - il l i i1 - i I 590 I^i ' I I O 10 20 N S shear (103 kip) FIG. 5-11. Distribution of N-S shear in the auxiliary builoing for the lower, median, and upper soil cases, compared to the results of the original analysis. 77
696 i i l
- Il l l l
' l 668 Y - tn l . 649 - i =
= '$
C r O l j 640 - B ID Ooginal l 624 - \)L Lower soil Median soil ! Upper soil
.,y s.
610 - '>- - w.g 601 - -
,R w %..
...y'..
l l 0 2 4 N S moment (106 in. kip) FIG. 5-12. Distribution of N-S moment in the auxiliary building .for tne lower, median, and upper soil cases, compared to the results of th original snalysis. 78
( TABLE 5-9. Ratios'of the loads calculated for the auxiliary building in the original analysis to those calculated in the
~
present reanalysis. Soil condition-Lower bound ' Median . Upper bound (N-S) (E-W) (N-S) (E-W) (N-S) (E-W)
- El 590's f ~ Shear 1.09 1.04 1.03 1.05 0.96 1.03 I
Moment .0.92 0.86 0.87 '0.87 0.85 -0.85 J El 601's j Shear 1.10 1.05' 1.04 1.06 0.98 1.04 Moment 0.87 0.79 0.83 0.82 0.82 0.79 El 610's Shear 1.10 1.07 1.04 1.07 0.97 1.04 Moment 0.98 0.87 0.89 0.89- 0.87 0.85
, lEl 624's
- Shear 1.10 1.10 . l. 0 1.10 0.99 1.0 Moment 0.93 0.97 0.99 0.99 0.88 0.92 El 640's Shear 1.0 1.0 1.0 1.0 1.0 1.0
,' Moment 0.8 0.8 0.8 0.8 0.7 0.7 l 1 . i above El 049 f t. Also, no vertical response w;s calculated in the original oesign analysis of the auxiliary building. .'.ssed on the SEP reanalysis, the vertical response throughout the structure is 0.23 g for the upper soil case, 0.24 g for the median case, and 0.28 g for the lower-bound condition. 2
-Vi rtually no maplification occurs within the structure for vertical excitation.
- 5.4.3 Field-Erected Tanks i
.i
, The capability.of three field-erected tanks to withstand seismically -
1 induced stresses was-evaluated with regard to current methods and guidelines as set 'forth in Ref. 4. The tanks wete the T-2 condensate storage tank (CST), i t . 79
- . __ ~ _u a . _ _ _ _ _ _ . . .. _ . . . _ . _ _ _ _. ..
the T-81 primary water supply make-up storage tank (PWSMST), and the T-58 safety injection and refueling water tank (SIRWT) . Descriptions of these tanks are given in Table 5-10. The T-2 CST and T-81 PWSMST are located at ground level near the northwest corner of the turbine building. Part of the T-2 CST is located above the cover slab for the common valve pit for these two tanks. The T-58 SIRWT is located on one of the auxiliary building roof slabs at El 640 ft. Several potential f ailure modes were investigated in the course of the SEP reevaluation of the three tanks. Among these were buckling of the tank sidewalls due to the seismic overturning moment; yielding, fracture, or pullout of the anchor bolts; collapse of the tank roof; sliding of the tank at the base, with subsequent rupture of connections; and f ailure due to high tensile stresses in the hoop direction that might result from hydrodynamic pressures occurring simultaneously with the hydrostatic pressure. Because of the geometry of the Palicaces tanks, actual overturning is not a potential problem. However,- if the anchor bolts fail, the overturning moment must be resisted by the tank walls and the weight of the internal fluid. For the
- fluid to be eff ective, a portion of the tank must separate from the foundation i
as shown in Fig. 5-13. This portion of the tank is highly stressed and is a potential source of f ailure in the presence of nonductile behavior (which might be caused, for example, by poor velds) . Calculation of the overturning moments and shears requires consideration of two important effects. The first is the bupulsi"e force due to the tank shell and part of the fluid moving together. The second arises from a convective mofe that occurs when part of the water near the free surface sloshes back and forth inside the tank. Table 5-11 shows the original design acceleration levels, which were the basis of an equivalent-static design. The SEP reevaluation of the field-erected tanks was conducted using methods outlined by Veletsos to calculate the impulsive mode frequencies. This approach is based on Rayleigh's method and includes both the shear and flexural deformations of the tanks. The convective (sloshing) f requencies were calculated follcwing recommenaations in Ref. 29. Table 5-12 lists the impulsive and convective frequencies for the three tanks. Spectral accelerations are also shown, together with the assumed modal damping ratios used in the analysis. Because it is unlikely that the maximum modal responses will occur sbaultaneously, the SRSS method was used to combine the impulsive
' and convective forces for all field-erected tanks.
80
TABIE 5-10. Descriptions of field-erected tanks. Wall thickness Base plate Number, diameter, and Tank Diameter Height and material thickness and material material of anchor bolts T-2 28 ft 28 ft 3/16 in., 1/4 in., 14, 2 in., A307 CST A36 steel A36 steel (not equally spaced) T-81 22 ft 27 ft 3/16 in., 1/4 in., 6, 3/4 in., A307 PWSMST A36 steel A36 stedL T-58 46 ft 24 ft 13/32, 9/32, and 5/16 in,. 52, 1-1/2 in., A325 SIRWP 1/4 in.; 5454-0 54S4-0 aluminum in 8-ft aluminum courses
=-
Overturning moment
/luid weight effective in resisting overturning O
--. 4 /, ,
.N
, , m - ,
y y Reaction due to Fluid weight compression in directly supported tank wall by the ground t l FIG. 5-13. Response to overturning moment in Palisades storage tanks. TABIZ 5-11. Field-erected tank static design accelerations. Design acceleration, g Horizontal Vertial l T-2 CST 0.20 0.133 T-81 PNSMST 0.025 0.025 T-58 SIRWf 0.23 0.17 f 1 82 . l ! . , _ g . _ _ - . . .
l l 1 TABLE 5-12. Response characteristics of field-erected tanks. Mode 1 is the fundamental (impulsive) mode; mode 2 is the convective mode. Spectral Frequency, Hz Damping, % acceleration, g T-2 CSr: Mode 1 11 7 0.40 Mode 2 0.33 0.5 0.18 T-81 PWSMST: Moce 1 11.7 7 0.37 Mode 2 0.37 0.5 0.19 T-58 SIRWP: Mode 1 20.3 7 0.42 Mode 2 0.25 0.5 0.22 5.4.4 Uncerground Piping In conjunction with the SEP reevaluatio1, a typical buried pipeline was analyzed for the 0.2-g SSE. The pipeline selected was the auxiliary feedwater line, which is f abricated from 6-in.-diameter ASTM A-376 seamless stainless steel. Tne line runs from the condensate-tank valve pit, under the turbine building, to the auxiliary building, as shown in Fig. 5-14. The elevation of most of the line is 578 f t 5 in. , which is approximately 11 f t 7 in, below the top of the turbine building base slab. The reanalysis considered the stresses induced in the pipe both by strains in the soil resulting from the propagation of elastic waves and by the effects of discontinuities and the relative displacements of attachment points. Stresses may be caused by primary (compression) waves, secondary (shear) waves, and surf ace waves with various angles of incidence to the pipe. St cesses may also result from relative end-point displacements caused by (Na motion of the structure at a penetration location. The phasing of the stresses due to soil strains caused by the various types of wave motion is 1 83 4 l
l i-l l N
/
fn i
/
AuxHiary , Containment ! build,ing m r ll Penetration Auxiliary feedwater line (El 578 ft 5 in.) Pipe bend Valve pit - h ~ 50 f t-- l FIG. 5-14. Schematic plan of auxiliary feeowater line. l l l normally not known, nor is the phasing of soil-induced stresses with those due to eno-point motion. Consequently, stresses were combined by the SRSS method. The analysis of the auxiliary feedwater line was conducted using the same median soil modulus and Poisson's ratio that were used for the SSI analysis of the containment and auxiliary buildings. A maximum soil strain of 2 x 10" in./in., corresponding to the 0.2-g SSE, was used, as discussed in Sec. 5. 5. 2. 3.5 EVALUATION OF CRITICAL STRUCTURES ! A reevaluation of the seismic capability of the critical structures was conducted using ?.oade developed in accord with SEP acceptance criteria. This reevaluation of the adequacy of the structures to withstand the 0.2-g SSE was t l based both on comparisons of tM se recalculated loads with the original design loads ano, where necessary, on further stress analysts. Where loads based on f l 84
the reevaluation guidelines are less than those used in the original design, the structure was in general judged to be adequate without additional evaluation. In cases where the recalculated loads exceed the original loads, j
.but where the resulting stresses are low compared to yield, the structures were again judged to be adequate. In general, damping values based on NUREG/CR-0098 were used (see Sec. 5.2.2) .
For those cases in which the seismic stresses are not low and where significantly low load ratios exist, conclusions were reached on the basis of ductility for Class 1 structures, as defined in Ref. 4. Accordingly, stresses above yield are considered acceptable provided the ability of the structure to perform its safety shutdown functions is not impaired. References 30 and 31 provide further discussion of rational ductility requirements for the inelastic response of critical structures. Load ratios were used to provide an initial screening of the expected responses of the structures and do not imply that inelastic responses would actually be expected. (Inelastic response would be expected only if the original design load was at or close to the elastic limit of the structural element. In f act, most design loads were well below the elastic limit.) Therefore, structures that do not exhibit load ratios less than 0.8 are considered acceptable. For structures with load ratios between 0.8 and 0.6, and for those for which original and analytical results were not available, more detailed investigations, including stress analyses of critical components, were conducted. No load ratios were below 0.6. A ductility-modified response-spectrum ar.alysis was not performed as part of this reevaluation, and all load ratios were based on linear analyses. 5.5.1 Containment Building As stamarized in Table 5-6, the loads developed in the containment building for the lower soil case are essentially tue same as those used in the original design. For the median and upper soil conditions, however, the seismic response loads determined in the present reanalysis exceed the original loads. The load ratio never f alls below the 0.8 to 0.6 range, but on the basis of the screening criteria adopted, some additional evaluation was necessary. A limited stress analysis of the containment building was conducted using 85
the loads oeveloped with current methods and with the damping values ; recommended in Ref. 4 for structures with stress levels at approximately one-half yield (see Sec. 5.4.1) . Concrete strengths (f ') of 4000 and 5000 psi were used for the concrete internals and containment vessel, respectively. The results indicate a minimum f actor of safety of nearly 1.5 tor the reactor internal structure (in E-W shear at El 590 f t) . The minimum f actor of safety for the cracking moment of the containment vessel is over 2.1; other factors of safety throughout the structure are even higher. The finding that all stresses were well below yield justifies the choice of camping values compatible with low stress levels. Details of equipment anchor capacity were not included as part of the structure evaluation--they were considered as part of the equipmant evaluation. However, it was concluded tha t the containment structure is capable of withstanding the 0.2-g SSE, based on SEP acceptance criteria. 5.5.2 Auxiliary Building i The original design loads for the auxiliary building are available only in the form of an envelope of maximum loads, with no indication as to whether they arose from the E-W or N-S response. (The load ratios presented in Table 5-9 are based on this load envelope.) Since the aesign'of the structure was apparently based on this envelope, the comparison of load ratios as a means of evaluation is considered conservative. The response tends to increase as the soil modulus increases. In no case does a load ratio fall below the 0.8 to 0.6 range. As in tne case of the containment building, however, some confirmatory stress analysis was conducted to verify the adequacy of the auxiliary building. Shear, moment, torsion, vertical seismic, and dead loads were consi ered. Dead loads were assumed uniformly distributed over the floor area. Various load combinations were considered: the combination of the total horizontal response load (for either principal direction) with 40% of the loads due to the remaining horizontal and vertical responses, and the total vertical load plus 40% of the two horizontal loads. Concrete strength was taken as 3000 psi, and 40-ksi yield was assumed for the A-15 intermediate-graoe reinforcing steel. In general, the loads decrease rapidly with increased elevation: All the 86
l l , ( ) most highly stressed members are located at El 590 ft. Also, although the torsional _ response is not the major contribution to the load, the presence of I moderate torsion causes elements located near the corners of the structure to be.the most highly stressed For primary response in the E-N direction, the two E-w shear walls in the northwest and southeast corners of the structure . ( (elements Nos. 22 and 39 in Fig. 5-15) were found to be the most highly ; i stressed. Similarly, for primary response in the N-S direction, the N-8 shear j walls on the east side and in the northwest corner of the building (elements' i
' l Nos.1 and 17 in Fig. 5-15) are the most highly stressed. Table 5-13 shows a ;
i . [ ' summary of the loads, stressen, and ultimate capacities of the most highly l ' I stressed elements. With the exception of element 17, the maximum shear i stress, including torsion, is less than the code value, including the $ factor ' for workmanship of 0.85. If the $ factor is neglected, the stresses in all elements would be less than code allowables. Maximum corresponding flexural stresses of approximately 160 psi occur, and no net flexural tension results. ]- All walls interior to the elements discussed above are less stressed, as are elements at higher elevations. Consequently, the auxiliary building is ! considered capable of withstanding the postulated 0.2-g SSE with no loss of l function and with possible light cracking in only one shear wall. 5.5.3 Building-to-Building Interaction Since the containment, auxiliary, and turbine buildings are supported on individual foundations, cons,ideratiots was given to the possibility of impact between the buildings. The turbine and auxiliary buildings are framed together at the lower elevations, but at El 624 f t, only slotted connections. j exist in' the N-S direction, and very light brackets resist E-W motion; the !
! structures are essentially uncoupled above El 610 f t. At El 610 ft, the maximum auxiliary building displacement normal to the turbine building (assuming the sof t-soil condition) is approximately 0.18 in., of which j approximately 0.16 in. is due to soil deformation. Since the slotted ! connections will not transfer large seismic loads, no damage that would impair the functional capacity of the auxiliary building is expected to result from the -interaction between the turbine and auxiliary buildings.
A 1-in. rattle space exists between the auxiliary building and the containment building. The locations of concern in any interaction-between 87
I I I l l l l 134 -l , 7///////////////////////////////////////// ////a =y ( Element No.1 -
//////sf/I @ 111, /Js'fs :
p__j g 25 - . -: H 13--i H9H HIP! 8 H Element 33 rfyf//
~
is siss ssi sA No.22 I mM m gag M vs1 i1 rsss1 m m H13-i - 40
//////////////A essss/s v/,s/ss, 22.5 %
<///////s>sss/siis////in
( 63.5 -l 1 l i ! -- 60 ! 1 -
> 1111 11111 1 1111111 1 11n f- Element
/ No.39 f-Element """ Walls resisting I
/ No.17 N S shear loads f / / s / / / 1 1,
-] j All dimensions in feet 26.5 i
x N
/
/
\ FIG. 5-15. Shear wall plan for the auxiliary building at El 590 ft. 88
l TABIE 5-13. Typical member loads and capacities at El 590 f t in the auxiliary building for tha postulated 0.2-g SSE. Axial stress on shear walls due to moment Maximum Direction Member and vertical load,a ksi shear load, Calculated capacity in shear, kips of number including loading (Fig. 5-15) for o max for %in torsion, kips concrete steel totalb E-W 39 0.121 0.048 468 407 402 688 E-W 22 0.162 0.007 732 454 456 744 N-S 1 0.123 0.046 4714 3380 3770 6080 N-S 17 0.157 0.013 954 502 509 8 20 aPositive numbers indicate compression.
. bTotal capacity was taken as 85% of the sta of the concrete and steel capacities, which were calculated
- assumingcf ' = 3000 psi and yf = 40 ksi (for A-15 intermediate-grade reinforcing steel).
i i
tnese two buildings are limited to elevations below 649 ft, since the relatively lightweight and flexible steel framing of the auxiliary building above El 649 ft is not expected to cause severe camage, even if it should strike the conteinment building. The sof t-soil assumption produces the maximum structural oeflection for both the auxiliary and the reactor buildings. At El 649 f t, the auxiliary building deflects about 0.23 in. in the direction of the containment building. The deformation at the base slab (El 590 f t) due to soil flexibility accounts for about 0.16 in of this value. The corresponding ceflections in the containment building are approximately 0.88 and 0.36 in., respectively. Although some of the natural trequencies of the auxiliary building are close to the fundamental of the containment building, the auxiliary building modes involve essentially only the steel superstr ucture. Containment building response frequencies are widely separated f rom all auxiliary building modes that involve significant displacements of the concrete shear wall structure. Consequently, the most appropriate estimate of relative displacement is obtained by combining the two structural displacements by the SRSS method. On this basis, the maximum relative displacement was calculated to be approximately 0.91 in., indicating that impact would not occur. Even if the displacements of the two struct ures were exactly out of phase, a maximum interference of approximately 0.11 in. (the absolute sum of relative displacements) would be expected for the ~ sof t-soil condition. Figure 5-16 illustrates these results graphically. Although some local concrete cracking and spalling could be anticipated, major structural damage sufficient to cause any less of functionality is not predicted for the sof t-soil condition, and no impact is expected in the median and upper soil cases, even on an absolute sum basis. Although no analysis of the turbine buildir.g was conducted, and although no through-soil coupling eff ects were considered in the analysis of the building-to-building interaction, no response moces that would cause loss of structural function of the auxiliary building are expected for the 0.2-g SSE. l 5.5.4 Field-Erected Tanks Tank shell and base plate integrity during the postulated 0.2-g SSE was evaluated for each of the three tanks. Initially, the assumption was made that the tanks behaved linearly, as shown in Fig. 5-17a. Tensile and 90
i l 765 g i i i
/
740 -
/ -
/
/
715 1.0-in, rattle / ~ space /
; --- Containment building 000 -
/ ~
o / --- Auxiliary building li / - SRSS combination C 665 -
/ -
G Absolute sum combination
)
640 - [ - I / 615 - - [ I "
/ l I 590 I I O 1 2 3 4 5 Disd cement (in-)
FIG. 5-16. Relative displacements of the containment and auxiliary buildings, asstaning soft-soil conditions. a compressive forces in the tank shell were evaluated, and attachment bolts and tank side-wall buckling were checked. If the number of hold-down bolts was determined to be insufficient to ensure linear behavior, then a nonlinear stress distribution, as shown in Fig. 5-17b, was assumed. In the latter case, the hold-down force resisting uplif t is provided by the weight of the tank fluid on the base plate, as well as by the yielded anchor bolts. The T-2 CST and T-58 SIRWT were both found to have adequate shell thickness and a sufficient number of anchor bolts to ensure a linear response. However, the T-81 PWSMST does not have sufficient anchor bolt I capacity to ensure linear raehavior. During a 0.2-g SSE, the T-81 tank is expectea to undergo substantial yielding and deformation of both the tank and the anchor oolts. Since the tank is constructed of A36 steel, with double fillet welds at the wall-base junction, ductile behavior of the tank is expected. The bolts have adequate shear capacity and the attachment brackets 91
M Hold down force from 1 pressure on sketch pla ie Foundation reaction (a) T, - - C, Tension
, i 4-Compression N
(b) Tm ax ---- -- C, Tension i i i Compression i i l FIG. 5-17. Linear (a) and nonlinear (b) tank wall force distributions resulting from seismic overturning moment M. T and C , signify longitudinal wall tension and compression, respe,ctively. 92
are more than adequate to. develop the anchor bolt ultimate strength. However, the ar hor bolt stresses for . the T-81 tank are well above yield, as well as above the ASME Code allowable for the faulted condition (70% of the ultimate I strength). E For the' A307 bolt material, the strain hardening across the thread stress area is suf ficient to develop scene yielding in the bolt shank before failure. Tank anchor bolts are not subjected to load reversals, so the bolts may .not f racture. Nevertheless, bolt failure is possible, and since tank buckling with subsequent loss of function is expected upon anchor bolt ' f ailure, increased anchor capacity should be installed. Stresses in the anchor bolts of the T-2 and T-58 tanks are acceptable. Stresses that might cause pullout of the anchor bolts and stresses in the concrete ring girders for all three tanks were evaluated and found to be i adequately low. The moment capacities of the ring girders were four.d to be sufficient, and the soil was found to have a large safety f actor against bearing capacity failure for all tanks. Frictional forces between the tanks and the underlying footings were high enough to preclude any relative displacement betwe 3n the tanks and ground. Hoop stresses were evaluated by combining absolutely the tank hydrostatic pressures with the SRSS of the pressures inuuced by the impulsive, convective, and vertical modes. All three tanks were found to have shell thicknesses sufficient to prevent yielding.
; Thus, with the exception of the possible f ailure of the T-81 anchor bolts, the critical elements of all three tanks are considered adequate to withstand the
} 0.2-g SSE. It is recommenced that modifications be implemented to increase l .the anchor capacity of the T-81 PWSMST. 5.5.5. Underground Piping ! In the evaluation of the adequacy of the buried pipelines at Palisades, the auxiliary feedwater line--a typical pipeline--was analyzed. To account f or possibly higher stresses in other lines with somewhat different configurations and stress allowables, conservative assumptions were made throughout the analysis. For instance, the assumed soil strain of 2 x 10" in./in. is higher than would be expected from a detailed 1
-geotechnic investigation of the Palisades soil conditions for the 0.2-g SSE. I Also, stresses were typically compared with ASME Code allowables. No stresses above . code-allowable ' values were computed for" the auxiliary feedwater '7e.
93
Maximum axial ano shear stresses of 27 and 15.7 ksi, respectively, were i calculated in the pipe oue to seismic wave propagation. These values do not account for the eff ects of discontinuities or end-point motion. The resulting prit.cipal stress of 35.2 kai may be compared with the ASME Code allowable of 44.9 kai for the f aulted condition. When a ourled pipe changes direction abruptly, as in the case of the auxiliary feedwater line (Fig. 5-14), axial strains induced in one leg will impose a ncrual loao on the transverse leg. This load must be resisted by the stiff ness of the pipe and the soil surrounding the transverse leg, which in turn creates shear and bending stresses at the elbow. These stresses were calculated assuming that the transverse leg acts as a beam on an e astic founuation. Tue coefficient of subgrade reaction was determined from Ref. 34. Based on the calculated deformation of the pipe (relative to the soil) of 0.16 in. at tais location and including the code stress-intensity f actor, a maximum bending stress of 29 kai results, together with a shear stress of 5.8 ksi. The principal stress of 29 ksi is well below the 4,.9-ksi allowable. The final location at which stresses in buried piping might be concentrated is the penetration where the pipe encers a building. Stresses here are caused by the building structure moving relative to the soil and imposing either lateral or axial displacements, or both, on the pipe. Tne auxiliary feeowater line was analyzed for stresses induced by the maximum outeral cuxiliary ouilding motion of 0.18 in. and a maximum lateral j cef ormation of 0.15 in. Maximum axial, bending, and shear stresses are 12, 2b.2, and 9 ksi, respectively. The principal stress of 40 ksi is less than the 44.9-ksi code allowable. It is therefore concluded that critical buried pipelines at Palisades are not expected to fail as a result of the postulated l l 0.2-g SSE. 5.6 SEISMIC INPUT MOTION FOR EQUIPMENT AND PIPING Seismic input motion for piping and equipment is typically cefined by
,36 means of in-structure (or floor) response spectra. Currently, floor respcnse spectra are usually generacco by means of time-history analyses, or by direct generation using random vibration techniques that use the response of the ouiloing structure as input. Befcre being used for design, these 94 l
- . - . . ..-_ ._. ._ _.m .
i spectra 'are' normally. smoothed and the peaks broadened _to account for modeling and material uncertainties. Unless a parametric study of the building is !. conducted, the peaks'of the floor response spectra are normally widened by
+154.'
. Horizontal' response spectra were generated for the original Palisades oesign analyses. of 'the containment and 'a6xiliary - building equipment and I ' piping, but no vertical . spectra were developed ~ for either structure. Spectra for10.54, 24, and 5% of critical _ damping for equipment were originally developed. The original spectra for the auxiliary building are envelopes of the maximum responses from the E-W and N-S directions.
As part of the SEP reevaluation, both horizontal and vertical
- . in-structure response spectra were generated for both' the containment and the auxiliary buildings, using' the new models and criteria described in Sec. 5.4.
f In accord with current-recommendations, spectra were generated for 34, 5%, and 74 equipment- damping. These values-reflect the somewhat higher damping i expected. Spectra were tenerated for each of the three soil cases ! considered. Smoothed and broadened envelopes encompassing the complete soil range were developed, as well as spectra for the individual soil cases. The spectra generated for the auxiliary building included the effects of torsion, although this contributes little to the response.- 5 Plots of' the the new in-structure response spectra are. presented -in i Appendix B. Where applicable, the corresponding original design spectra are [
.shown for comparison. 'In general, the new spectra exceed the original spectra
- f. at' both high _ and low frequencies, whereas the original spectra tend to be i slightly above the new spectra in the resonant range. Load ratios for f _ equipment in' the rigid ~ range are expected to be approximately 0.71 to 0.86 for 5 -the containment building and approximately 0.71 to 0.9 for the auxiliary toilding. For flexible equipment with frequencies of about' 4 Hz or less, load ratios as low ac 0.34 for the containment building and 0.23~for the auxiliary building can occur, depending on the soil modulus used. The effects of the
.new floor spectra and of these load ratios on equipment and piping are
~
I i , discussed in Chapter ' 6. d 95
I i CHAPTER 6: SEISMIC EVALUATION OF MECHANICAL AND ELECTRICAL EQUIPMENT AND FLUID AND ELECTRICAL DISTRIBUTION SYSTEMS
6.1 INTRODUCTION
6.
1.1 Purpose and Scope
This chapter reviews the reevaluation of selected mechanical and electrical equipment and fluid and electrical distribution systems at the l I Palisades Nuclear Power Plant. Based on that review, this chapter also l evaluates the ability of the reactor to shut down safely and to rsmain in a safe shutdown condition in the event of an SSE. The SEP review team purposely selected for reevaluation those components that were expected to have a high degree of seismic fragility; moreover, the review team believes that these components are tepresentative not only of those installed in the safe shutdown systems, but also of other seismic Category I systems, such as engineered safeguards. Thus, evaluation of these selected components establishes an estimated lower-bound seismic capability for the mechanical and electrical components and the distribution systems of the Palisades plant. Considered in terms of seismic design adequacy, nuclear power plant I equipment and distribution systems fall into two main categories--active and l passive. Within each main category, equipment and systems are further categorized as rigid or flexible. As discussed in R.G. 1.48 and Sec. 3.9.3. of the Standard Review Plan, aceive components are those that
- must perform a mechanical motion to accomplish a system safety function. For l
the purpose of this report, this definition is expanded to include electrical or mechanical components that are required for safe shutdown and that must move during or af ter a seismic event to perform their design safety function. Typically found in the active category are pumps, valves, motors and associated motor control centers, and switchgear. Seismic design adequacy of active components, which depends upon function as well as structural integrity, may be demonstrated by either analysis or 96
.. -_~ . -
j test. Testing is generally the preferred method, but because of size or
- weight-restrictions or difficulty in monitoring function, many active components are seismically evaluated by analysis. To ensure active component function by analysis, deformations.must be limited and predictable. ,
. l
'Therefore, total stresses in such components are normally limited to the elastic linear range of 0.67 to 0.8 times the yield stress of the material,
~
t and in no case would the total stress in a component be allowed to exceed the yield stress. Passive components considered in this report are those components that , are required for safe shutdown and for which the only safety functions are to i remain leak tight or to maintain structural integrity during or following the
-SSE. Typically found in the passive category are pressure vessels, heat exchangers, tanks, piping and other fluid distribution systens, transformers, and electrical distribution systems. t 4 'In determining seismic design adequacy by analysis, the most important 4
distinction between active and passive components is the stress level that the component is allowed to reach in response to the SSE excitation. For passive components, higher total stress liimits, ranging from yield to 0.7 times the ultimate strength of the material, are permitted by current design procedures and codes. f The designation of components and distribution systems as flexible or rigid is 'important in developing the magnitude of seismic input for component { evaluation. The seismic inertial acceleration of the equipment depends upon 4
' potential resonance with the supporting building structure, structure and equipment damping values, and equipment support elevations. Whether a component is designated as rigid or flexible may also depend on how it is
[ suppor ted. Many rigid components must be considered'and evaluated as flexible 4 because of their support flexibility. For the Palisades auxiliary building and containment internal structures, in-structure response spectra are such that equipment may be considered rigid for frequencies greater than 10 Hz. The maximum ~ floor acceleration is approximately 0.4 g ;(twice the SSE zero-period ground acceleration) . For f flexible components with fundamental-frequencies less than 20 Hz, the maximum ] seismic inertial' acceleration is approximately 2 g. l- - Af ter the components were categorized as active or passive and as rigid or flexible, a - representative sample f rom each group was evaluated to +- 97
l establish the seismic design margin or degree of adequacy of each group. In this way, seismic design margins for groups of similar components were f established without the need for detailed reevaluations of hundreds of individual components. Representative samples of components were selected for review by one of two methods: e Selection based on a walk-through inspection of the Palisades facility by a team comprising NRC staff, members of the SSRT, and the authors. Based on their experience, team members selected components in each group that appeared to have a high potential degree of seismic fragility. Particular' attention was paid to the components' support s tr uctures, e Categorization of the safe shutdown components into generic groups such as horizontal tanks, heat exchangers, and pumps; vertical tanks,
- heat exchangers, and pumps; motor control centers and motors.
l l
- The licensee was asked to provide seismic qualification data on the selected components.
I The rest of this chapter reviews the seismic capacity of the selected components and recommends, if necessary, additional analysis or hardware changes to qualify them for the 0.2-g SSE. Based on the detailed review of the seismic design adequacy of these representative components, conclusions are drawn as to the overall seismic design adequacy of seismic Category I equipment installed at Palisades. 6.1.2 Description of Selected Components Table 6-1 lists and describes the components selected by the team on the basis of its plant walk-through--components that are representative of the listed generic groups of safety-related components. Table 6-1 also gives the l basis for each selection. The review in this chapter emphasizes what are normally listed as auxiliary components. Such components are typically supplied by manufacturers who--unlike the NSSS vendors--may not have routinely designed and fabricated components for the nuclear power industry; particularly during the time this - 98 I l
TABLE 6-1. Mechanical and electrical components selected for seismic evaluation and the basis for selection. Item No. Description Reason for selection f Mechanical Components 1 Essential service water pump Has a long, vertical, unsupporte ' intake section which was originally statically analyzed for seismic erfects. 2 Auxiliary feedwater pumps Represents a horizontal component which is rigidly connected to a foundation mat. 3 Component cooling heat Unique component in that one heat exchangers exchanger is stacked on top of. the other and connected to it by two saddle supports. Concern was expressed about the heat exchangers' ability to withstand overturning effects in the transverse direction. 4 Component cooling surge tank A column-supported vertical tank. 5 Diesel generator oil tanks Anchor-bolt system for flexible in-structure flat-bottom tanks may be overstressed if tank and fluid contents were assumed rigid in the original analysis.
! 6 Boric acid storage tank A column-supported vertical tank with bracing.
7 Hydrazine tank - A tall, column-supported vertical. tank. Concern was expressed about overturning effects. 8 . Sodium hydroxide tank A horizontally mounted component supported by two saddles that do not appear to be seismically restrained. Concern was expressed about the saddles' ability to carry required seismic loads, particularly in the longitudinal direction. continued 99
TABLE 6-1 continued. Item No. Description Reason for selection 9 Safety injection tank Supported by a truss structure which is mounted at the top of the containment building. Concern was expressed about the increased acceleration values that the tanks will experience'at that elevation. 10 Motor-operated valves A general concern with respect to externally unsupported, motor-operated valves, particularly for lines 4 in. or less in diameter, is that the relatively large eccentric mass of the motor will cause excessive stress in the attached ' piping. 11 Control rod drive mechanism Particularly critical to ensure reactor coolant system integrity. 12 Pressurizer Same as item 11. 13 Steam generators Same as item 11. 14 Reactor coolant pumps Same as item 11. 15 Reactor vessel Same as item 11. Electrical Components 16 Battery racks Bracing required to develop lateral load capacity may not be sufficient to carry the seismic load. 17 Motor control centers Typical seismically qualified electrical equipment. Functional design adequacy may not have been demonstrated. In addition, anchorage to floor structure may not be adequate. 18 Switchgear Same as item 17. continued i 100
TABLE 6-1 continued. Item No. Description Reason for selection 19 Control room electrical Appear adequately' anchored at the panels base; however, there appear to be many components cantilevered from the front panel. The lack of front panel stiffness may permit significant seismic response of the panel, resulting in high , acceleration of the attached ; components. 20 Transformers Same as item 17. 21 Electrical cable raceways Cable tray support system does not appear to have positive lateral restraint and load-carrying capacity. plant was under construction. Therefore, if there is a reduction in seismic design adequacy, it would tend to be found in the auxiliary equipment, rather than in the major nuclear components. However, because of its importance to safety, the seismic design adequacy of the reactor coolant system components and support structures, to the extent information has been provided, is also evaluated in this report. In addition, portions of four piping systems were analyzed. The results of these analyses will be reported independently,38 but a preliminary summary is provided in Sec. 6.4. 6.2 SEISMIC INPUT AND ANALYTICAL PROCEDURES 6.2.1 Original Seismic Input and Behavior Criteria l Por seismic Category I mechanical equipment, all componeats and systen s originally cla ssified as Class 1 were designed in acccrdance with the criteria for load combinations and stresses listed in Table 4-3 under "Other Class 1 systems and eqt ipment." The manual Nuclear Reactors and Earthquakes, ' was used as the basic l design guide f or seismic analysis. Class 1 equipment and their supports were l 101
analyzed for accelerations at least equal to the acceleration of the floor on which they are supported. The seismic loads were applied to the equipment centers of gravity. The reactor protective system and nuclear instrumentation were specified to operate throughout a disturbance equivalent to a horizontal acceleration of 0.8 g. The seismic design criteria for the safety-related equipment controls and the emergency electric power systems were such that all controlling devices and systems would withstand the seismic disturbance without malfunction or improper action. Furthermore, it was specified that a seismic disturbance would not affect the operation of safety systems either momentarily or permanently. Some relays, breakers, emergency generators, and varioun other controlling devices similar to the type used in the Palisades plant have been shop tested and have been shown capable of withstanding seismic shock loads without malfunction. However, there is no information currently available as to the level of shock loads sustained, nor is there information to indicate that the specified criteria have been explicitly implemented in the design of the electrical components. For seismic Category I cable conduits, the flexibility of the cables and the supporting trays for above-ground cables was designed to accommodate differential seismic motions inside and outside the structure. . Cables leaving structures below ground are placed in plastic conduits which are subsequently encased in concrete for mechanical protection. Appendages (small masses elastically attached to large masses) were not considered dynamically coupled in the seismic analysis of large masses, because their inertial forces were assumed to be too small to affect the behavior of large masses. Their weights, however, were included in those of the large masses. Class 1 appendages to piping and equipment .(valve operators, for example) were analyzed statically to evaluate their effect on the piping and equipment. 6.2.2 Seismic Input for SEP Reassessment l i l l 1 Seismic input requirements for determining the seismic design adequacy of i mechanical and electrical equipment and distribution systems are normally based on floor or equipment response spectra for the various elevations at which the equipment is supported. These in-structure spectra, which are based l 102 l
on'R.G. 'l.60 spectra modified by the dynamic characteristics of the building,
- are shown in Appendix B. The in-structure spectra are based on the building models shown in Figs.' 5-3 and 5-7..
For mechanical and electrical equipment in general, a composite 74 equipment damping,.as' suggested in Sec. 5.3.1, is used in the evaluation for-the 0.2-g SSE. For piping evaluations, the equipment. damping associated with the SSE is limited.to 34. These damping ratios are also consistent with a l recent suusary of data presented -to define damping as a function of stress
- . level.40 ' For cable ' trays, recent tests seem to indicate that the damping ratios to be used in design depend greatly on the tray and support construction and the manner in which . the cables are placed. in the trays. -
t l Damping may be as high as 20% of critical damping.41 Horizontal seisNic - input loads have been assumed in this evaluation to be simultaneously applied . independent components. Depending on the geometry of the component being I evaluated, the resultant horizontal load varies from 1.0 to 1.4 times the
-individual component. load. Except where design adequacy is in question, we
! have censervatively applied the 1.4 factor to the check evaluations performed f in this reanalysis. 6.2.3 SEP Acceptance Criteria 4
- Seismic Category I components that are designed to remain leak tight or i to retain structural integrity in the event of an SSE are now typically designed to ASME Section TII Code,-Class 1, 2, or 3 stress limits for service
- condition D. The-stress limits for supports for ASME leak-tight components f are shown in Appendix F or Appendix XVII to the ASME Section III Code.
f -When qualified by analysis, active ASME Section III components that'must perform a mechanical motion to accomplish their safety functions typically . must meet. ASME Section III Code, Class 1, 2, or 3 stress limits for service condition B. (Recent increases to ASME Section III, Class 1, service level B limits have not been considered.) Supports for these components are also typically restricted to service condition B limits.
- For other equipment, which is not designed to ASME Section III Code requiremerts, and for which the design, material, fabrication, and examination requirements are typically less rigorous than ASME Section III Code requirements, the allowable stresses are limited to yield values for passive
- f. .
103
,_ - -. _ , _ - , - = , , - . , . _ . . ~ . - - . . _ - . _ _ , , _ - , - . - . - - - .
components and to the normal working stress (typically 0.5 to 0.67 times yield) for active components. The SEP acceptance criteria used to evaluate various equipment and distribution systems for the Palisades passive j components are given in Table 6-2. For active electrical components such as switches and relays, functional adequacy should be demonstrated by test. Experience in the design of such pressure-retaining components as vessels, pumps, and vr.lves (designed to meet ASME Section III Code requirements at 0.2 g) indicates that, except at the supports and nozzels, stresses induced by earthquakes seldom exc2ed 10% of the dead weight and pressure-induced stresses in the component body. Therefore, design adequacy of such equipment is seldom dictated by seismic design considerations. Seismically induced stresses in nonpressurized mechanical and electrical equipment, in fluid and electrical distribution systems, and in all component supporte nay be significant in determining design adequacy. However, because of the more restrictive stress and damping limits, the OBE rather than the SSE normally controls the design of piping systems. 6.3 EVALUATION OF SELECTED COMPONENTS 6.3.1 Mechanical Equipment 6.3.1.1 Essential Service Water Pump The essential service water (ESW) pump and motor unit is oriented vertically in the intake structure and supported at El 590 f t. As shown on Layne & Bowler, Inc., drawing 950X9 SH-1, the intake portion of the pump extends downward from the discharge head and pump base for 37 f t 10 in. The seismic analysis, as given by Layne & Bowler, Inc., was performed for simultaneous equivalent-static loads of 0.90 g acting in the horizontal direction and 0.14 g acting in the vertical direction. The pump and motor unit is located at grade; therefore, the seismic input is essentially the R.G.1.60 ground response spectrum. However, to be conservative, the response spectra for the auxiliary building base slab was , used in the analysis (see Figs. B-11 and B-12) . Overturning tensile and shear stresses in the pump base ancaor bolts were determined, as well as stresses at the attachment of the intake column pipe to the discharge head. 104
. .,. . . - . . . . . _ . - - . . -- - - .- - - = . .-
TABLE 6-2 SEP acceptance criteria for determining seismic design adequacy of passive mechanical and electrical equipment and distribution systems. , Components SEP acceptance criteria (SSE) Vessels, pumps and valves S 8dl 1 0 78uand 1.6Sy ASME III Class l-(Table F 1322.2.1) S agt 1 0.67Suand 1.33S y ASME III Class 2 (hc 3217) 8all 1.
- u d 1.25S y N III C ass 2 W 3321) 0,,7y 1 0.5S u and 1.25S y ASME III Class 3 (ND 3321) '
Piping S m all i 1* S uand 2.0Sy- ASME III Class 1 (Table F 1322.2.1) Sh 1 0.6Suand 1.5Sy ASME III Class 2 and Class 3_(NC 3611.2) Tanks No ASME III Class 1 o, 1 0.5Suand 1.25Sy ASME III Class 2 and Class 3 (NC 3821) Electric 'S,11 1 1.OS y equipment Cable trays Sall i 1.0Sy ASME supports Sall i 1.2S yand 0.7S u ASMd III, Appendices XVII and F, for ,. Classes 1, 2, and 3 Other supports Sall i 1.6S Normal AISC allowable stress increased by 1.6, consistent with NRC Standard Review Plan 3.8. Bolting Sall i 1.6S ASME III, Appendix XVII for bolting, where S is the allowable stress for design loads, f r 9
Decause the intake portion of the pump is oriented vertically as a cantilever beam, the dynamic characteristic of the intake suction pipe was determined. It was found to have a fundamental frequency of 1.0 Hz, assuming a weight distribution that includes the shaf t and contained water, as shown in Ref. 44. At this nstural frequency, the spectral acceleration for 7% damping is 0.32 g. The seismic accelerations were applied to the pump, and the resulting anchor bolt stresses were determined, considering simultaneous N-S and E-W loading. The eff ect of attached piping nozzle loads was not considered, since they were not available. (They are generally not significant in determining the overall adequacy of the pump body and support system.) Based on the ASME condition D stress limits, the analysis established a factor of safety of 2.04 for the assumed A307 anchor bolts. The stress calculated at the attachment of the discharge head to the intake column pipe was 10.8 ksi, which is well below the yield stress of 35.0 ksi, which is given in the original seismic design calculations for the column pipe. Therefore, we believe that the ESW pump will withstand a 0.2-g SSE seismic event without loss of structural integrity, provided the discharge head stresses are within code allowables. Insufficient information was provided to determine the stresses and material used in the discharge head; thus, allowable stresses there are unknown. Also, too few details were available to evaluate the functional adequacy of the pump in terms of motor impeller shaf t deformations or bearing f ailure. 6.3.1.2 Auxiliary Feedwater Pumps The auxiliary feedwater pumps are horizontal components supported on concrete pedestals and a mat foundation that is located at El 571 f t in the turbine building. The components consist of two pumps, one motor driven and one turbine driven. The motor-driven pump was supplied by the Bingham Pump Company and is shown on drawing FD-270663; the turbine-driven pump was supplied by the Elliot Company and is shown on drawing 602464. The original seismic design, considered an SSE Leismic load resulting from a 0.20-g horizontal acceleration and a 0.14-g vertical acceleration, acting. simultaneously. Since response spectra for' the turbine building are not available, the response spectra considered applicable for verifying scismic design adequacy 106
, .~. .. . . . . _ .
i l are'those calculated'for the auxiliary building at El 590-ft (see Pigs. B-ll, ll B-12, and B-19) . The corresponding. floor acceleration values for' the N-S, E-W, and vertical directions 'are 0.32, 0.34, and 0.28 g, respectively. The seismic. accelerations were applied -simultaneously to the pumps, and the resulting ^ mounting-bolt stresses were determined for three separate attachment locations: 'the base plate mounting, the pump mounting, and the turbine
; mounting.-' Based on ASME condition D stress limits for the assumed A307 bolts, the factors of safety are 11.6,-5.76, and 4.70, respectively.
In the analysis', . nozzle loads due to the attached piping were neglected, since they were not available. However, considering the relatively large safety factors, j nozzle loads would not be expected to contro) design. We believe that the-auxiliary feadwater pumps will withstand a 0.2-g SSE seismic event without loss of structural integrity; however, due to the lack of design detail, no attempt was-made to evaluate the functional adequacy of the pumps. 2 6.3.1.3 Component Cooling Heat Exchangers The component cooling heat exchangers (CCHXs) are two horizontal heat , exchangers located in the auxiliary building, one stacked on top of the other i and supported there by two saddles. The pair is supported on the floor by four saddles at E3 590 ft. Three of the four floor saddles are slotted in the longitudinal direction to permit thermal expansion. The heat exchangers are 4 shown ~on Industrial Process Engineers drawing F-5628-3. -The response spectra I for 7% damping (see Figs. B-ll and B-12) are considered applicable for 5 verifying seismic design adequacy. The' seismic qualification of the CCHXs was performed as described in Ref. 47. The seismic evaluation d6. ermined the dynamic response
-characteristiics of the exchangers and their saddle support system. The-
- evaluation indicated that the system is relatively rigid and has no response i frequencies below 33 Hz. As a result, horizontal seismic input accelerations in orthogonal directions are 0.34 and 0.32 g, respectively, corresponding to
; the 0.2-g SSE.
The seismic-accelerations were applied simultaneously to the heat exchangers, and the resulting anchor bolt and support saddle . stresses were . -determined. The analysis established factors of safety of 2.35 for the anchor bolts and 16.5 for the support saddle, . based on ASME III-l condition D stress limits . 107 i 2, , , - . - _ . .. -. ., --. -. . - ...
I In addition tc evaluating the CCHX saddle and anchor bolt support system, the seismic stresse., induced.in the tubes of the heat exchanger were Ldetermined, combined with other applicable loads, and compared to code allowables. The factors of safety determined.for the heat exchanger tube' , was 40.8; for ; the heat exchanger shell, it was 2.13. Both were controlled by hoop stresses .due to internal pressure rather than seismic stresses. It should be noted -that no evaluation .was made of nozzle loads-in the heat exchanger,'since they were determined from the. attached piping system analysis
; that was not available for evaluation. It has been generally found that such piping loads, which can be a limiting load to the nozzle, seldom have a significant effect on the heat exchanger support -loads.
2 In conclusion, we believe that the CCHXs will withstand a 0.2-g SSE
. seismic l event without loss of structural integrity or function. Our , conclusion is based on i
e Evaluation of the dynamic characteristics of the heat exchanger i support system and the supplemental analysis given in Ref._47. e Experience in reviewing similar saddle-supported heat exchangers. ]
' 6.3.1.4 Component Cooling . Surge Tank The component cooling surge tank is a column-supported component located ! -in the auxiliary building at El 649 f t. The surge tank is shown on Niles Steel Tank Company drawing 5935-M38-A. The response spectra for 7% damping , (Appendix B) .are considered applicable for verifying seismic design adequacy.
[ The ' seismic qualification of the surge tank was originally performed for a ! 0.30-g horizontal acceleration and a 0.l4-g vertical accelert. -
, applied 1 . simultaneously.
ll l - We .have reviewed the tank and support system to determine seismic design !. adequacy.48 The dynamic analysis considered the effective impulsive and i
-convective response of the contained fluid and determined fundamental response frequencies for the tank:- 0.71 Hz- for convective loading and 7.89 Hz for the tank and support system. For-the convective mode, a damping value of 0.5% was-used;'for the impulsive mode, 7% damping was used. TM analysis determined f
l ' gross dynamic characteristics of the tank und established minimum f actors of safety of 10.0 for compressive stresses in the tank legs.and 9.43 for-combined 108
stresses in the anchor bolts. As in the case of other canponents with attached piping, we did not evaluate nozzle capacities, since piping loads were not available. We believe that the component cooling surge tank will withstand the 0.2-g SSE without loss of structural integrity or function, based on e Check of the dynamic characteristics of the tank and an evaluation of support leg and anchor bolt stresses. e Experience in reviewing similar tanks. i 6.3.1.5 Diesel Generator Oil Storage Tanks The diesel generator oil storage tanks were not evaluated, since no drawings or design calculations were available. 6.3.1.6 Boric Acid Storage Tank The boric acid storage tank is a column-supported tank with cross bracing, as shown on Nooter Corporation drawing JN-D-31011. The seismic qualification of the tank and support system was performed as described in Ref. 49. We have reviewed the tank and support system and its anchors to determine seismic design adequacy. The tank, which is supported at El 590 f t in the auxiliary building, was evaluated for the corresponding floor
-response spectra from Appendix B. The dynamic analysis considered the effective impulsive and convective response of the contained fluid and determined the fundamental response frequencies for the tank: 0.56 Hz for convective loading and 18.8 Hz for the tank and support system horizontal mode (including impulsive loading) . For the convective mode, a damping value of 0.5% was used, and for the impulsive mode, 7% damping was usad. The analysis determined gross dynamic characteristics of the tank and established minimum factors of safety of 5.64 for compressive stresses in the support legs and 50.0 for combined stresses in the anchor bolts, in accordance with ASME III-2 condition D stress limits. Again, we did not evaluate nozzle capacities, since piping loads were not available. We believe that the boric acid storage i tank will withstand the 0.2-g SSE without loss of structural integrity or function, based on l 109 i
t
l I o Review of the stress analysis of the tank support supplied by the licensee. e Check of the dynamic characteristics of the tank and an evaluation of the tank and support system and anchor bolt stresses performed in connection with this report, o Experience in review of similar tanks. 6.3.1.7 Hydrazine Tank The hydrazine tank, as shown w Buff alo Tank Company drawings SK-1031-2 and SK-1031-4, is a tf.1, column-supported vertical vessel (12 ft long and 30_in. in diameter). The tank is supported at El 640 f t of the auxiliary building and was evaluated for the corresponding floor response spectra, shown in Figs. B-lS, B-17, and B-19. The dynamic analysis considered the effective impulsive and convective responses of the flui.d and tank system. The fundamental sloshing (convective) frequency was found to be 1.10 Hz, and the fundamental tank and support system frequency was found to be 27.1 Hz. For the convective and impulsive modes, the analysis used damping values of 0.5% and 7%, respectively. The analysis determined gross dynamic characteristics of the tank and established minimum factors of safety of 14.1 for compressive stresses in the tank legs and (assuming ASME III-2 condition D stress limits) 7.14 in the anchorage system,. We did not evaluate nozzle capacities, since piping loads were not available. We believe that the hydrazine tank will withstand the 0.2-g SSE without loss of structural integrity, based on e Check of the dynamic characteristics of the tanks and evaluation of support leg and anchor bolt stresses, o Experience in reviewing similar tanks. 6.3.1.8 Sodium Hydroxide Tank The sodium hydroxide tank is a horizontal vessel located on the auxiliary building and supported by two saddle supports at El 640 f t. The tank is shown on Buff alo Tank drawing SK-M-1054. The response spectra for 7% damping 110
. . .. ~ . .. ..- - . - .. . - . -. -.
l
' (see Figs. B-15,' B-17, and B-19) are considered applicable for verifying 4
seismic design adequacy. The seismic, qualification of the sodier ny Jroxide talot was performed as
- described in Ref. 53. This analysis was reviewd, and an independent evaluation of the dynamic response characteristics of the heat exchanger and its-saddle support system was made. The review indicates that the: system is
, relatively rigid and has no response frequencies below 17.8 Hz. As a result, horizontal seismic input accelerations-in orthogonal directions were { determined to be 0.49 and 0.42 g, corresponding to the 0.2-g SSE. The seismic accelerations were applied simultaneously to the sodium
- hydroxide ' tank, and the' resulting support saddle stresses and anchor bolt
} stresses were determined. The analysis established factors of safety of 3.52 j for the support saddles and (assuming ASME III-2 condition D stress limits) 15.9 for the anchor bolts. Therefore, we believe that the sodium hydroxide l j tank will w.thstand a 0.2-g SSE seismic event without loss of structural ! - integrity, based on
-e Review of the analysis in Ref. 53.
4 ! e Evaluation of the dynamic characteristics of the tank and support
- system and the supplemental analysis given in Ref. 54.
e Experience in reviewing similar saddle-supported tanks. l 6.3.1.9 Safety Injection Tank The safety injection nk is a tall vertical vessel, 32 f t 2 in. in
' length and 9 f t 0 in. in diameter. The tank is shown on Nooter Corporation ; drawing F-6171, and the support system is shown on Bechtel Corporation drawing C-246. The support system for the tanks consists of a series of trusses which
- are supported by the containment structure near the springline and connected together at the done centerline. The tanks are connected to the support trusses by means of a structural framework and vertical hanger members. The
; tanks were originally designed assuming a horizontal acceleration of 1.5 g and
, 'a vertical acceleration of 0.2 g, applied simultaneously to the center of-gravity of.the vessel.55,56
. The - tank, assumed to be full of water and with its truss and hanger l support system assumed rigid, was reevaluated dynamically as shown in Ref. 57 ,
[ 111
)
,_ . ._ ._ _- . __ m-_ . ._
l and found to have a. fundamental frequency of 25 Hz. Based on this frequency and the in-structure . response spectra determined at El 730 f t (Figs. B-3 and B-9), a resultant response spectrum fl.Ar acceleration of 0.6 g for 7% damping -
-was obtained.- The SRSS value for horizontal acceleration was then 0.85 g, which is less than the original. design value of 1.5 g. For the vertical
~
direction, an acceleration value of 0.34 g was obtained from the rigid end of the' vertical response spectrum. This represents a. load ratio of 1.34/1.20 = 1.117, an 11.7% increase in dead weight plus seismic load. With this revised seismic load, the tank and support trusses and structural framework will remain within ASME III-2 condition D stress limits, given that the tank and support system is rigid. However, the tank truss and structural framework support system may not be rigid. This system is highly complex and the determination of its frequency characteristics is beyond the scope of this report. If the support system is flexible, such that the tank is in the near-resonant range, the horizontal. response acceleration would be approximately 4.5 g and the vertical response 1.5 g. This would result in seismic loads on the support system three times that considered in the horizontal design and twice that considered vertically. As a result, ASME condition D stress limits would be exceeded. We recommend that a detailed reanalysis of the tank and its support system be performed by the licensee to determine the resultant dynamic characteristics and stresses in the system for the redefined seismic response spectra. l 6.3.1.10 Motor-Operated Valves l l . The motor-operated valves are shown' on Velan Engineering Company drawings P-33345, P-33345-4, and P33345-3 and Philadelphia Gear Corporation drawings ( l 02-405-0039 and 02-405-0085-4. The response spectra considered applicable for l 'the motor-operated valves are those for the auxiliary building and containment building internal structure at 3% damping (see Appendix B). It has been our experience that, for lines 4 in. in diameter and smaller, the eccentricity of motor-operated valves may cause additional significant piping stresses (in excess of 10% of code allowable) that should be considered in the computation of totel stresses.- The applicable stress levels are specified by Class 2, condition B, for active - valves and by condition D when only pressure _ boundary integrity is required. The stresses induced by. valve ! 112 l l
. ,_. __ .- _ _ , . - . . -.. _ _ _ _ ~_ .
i. r eccentricity increases as the line size decreases. Calculations performed on randomly selected motor-operated valves (2 in.,- 3 in., and 4 in. in diameter) installed in the Palisades Nuclear Power Plant l demonstrate that the stress levels reached are well in excess of the above-j mentioned lot, regardless of service condition. For- a typical ferritic piping material (S = 15,000 psi), the condition h B and D stress limits would be 18,000 and 36,000 pai, respectively. Preliminary calculations indicate that the stress levels shown in Table 6-3 would be reached in the pipe if a peak acceleration of 1.5 g were applied to the valves. Based on these values, it is recommended that the licensee evaluate the stresses induced from motor-operated _ valves in supporting pipe 4 j in. in diameter and maaller. - The licensee should show that stresses induced in the piping by. these valves are less than 10% of the pertinent service . l condition allowable stresses. Otherwise, the total stresses at motor-operated
- valve locations should be calculated to determine if they are within the established allowables.- Alternatively, we recommend that a requirement to support the valve operators externally be developed and implemented. In addition,' the licensee should provide an evaluation in the form of either test or analytical results which demonstrate the functional adequacy of the valves.
I 4 TABLE 6-3. Stress levels induced in supporting pipes by motor-operated valves.
- -Pipe diam., Stress, i in. psi t of condition B 4 of condition D J
4 8,200 454 234 3 15,300 85% 43% l 2 20,800 116% 58% i ) 6.3.1.11 Control Rod Drive Mechanism A susmary of the original stress analysis of the control rod drive . t i l mechaniem (CRDM) is given in Ref. 59. .The mechanism and seismic support are. I ! t' 113 l
shown en Combustion Engineering drawings 2966-E-2869, 2966-SE-2554, and 2966-SE-2557. A static seismic analysis of the drive mechanism was performed to determine if the seismic support allowables were exceeded. A' static horizontal SSE seismic load of 1.35 g was applied to the mechanism, and a stress evaluation of the various parts of the seismic support was performed. The stress evaluation was based upon a moment restraint placed 51 in, below the center of gravity of the mechanism. The response spectra for the CRDM, which correspond to the reactor vessel support elevation,-are given in Figs. B-1 and B-3. Since the fundamental frequency of the drive mechanism is 8.3 Hz, the peak acceleration in the two
. horizontal directions is 1.06 g for 24 damping. Therefore, the resultant horizontal acceleration is 1.49 g, and the ratio to the original design value of 1.35 g is'l.10. If the original CRDM seismic stresses are multiplied by 1.10, the resulting stress values are less than the allowable stress values, except for the tension stress in the plate bolts. For the plate bolts, the revised tension stree,s is 183 kai, as compared to an original allowable stress of 176 kai. The latter value was based on 110% of yield. Furthermore, there are two sources'ot additional margin not accounted for in the evaluation:
e Two percent damping of the CRDM was assumed, whereas Combustion Engineering testing has reportedly indicated that higher dampir.g could be justified. e Nonlinearities and friction in the seismic support hardware were neglected; these effects would tend to further reduce the predicted response. We believe that the CRDM will withstand the 0.2-g seismic load without loss of structural integrity. However, since some resultant stresses exceed yield, as well as current ASME condition B stress limits, the active function of the mechanism cannot be assured, based on the reviewed calculations. 6.3.1.12 Pressurizer The pressurizer is a vertical cylindrical vessel with a skirt-type support attached to the lower' head. The lower part of the skirt terminates in a bolting flange with sixteen 2-in bolts, which secure the vessel to its 114 l
. - , , . . , . . - - - , . - ~ .. , --- . - ,
~ . ._ m. . . . . ._
l foundation. , A summiary -of , the ' stress ~ analysis of the pressurizer is given . in Ref. 60. In 1967 a seismic analysis of the pressurizer shell and internal tubes, support skirt, shock' lugs, and pressurizer support bolts was
' per formed. The SSE evaluation _ assumed simultaneous horizontal and vertical
~ accelerations of 0.20 g. These accelerations were applied statically at the ~ center. of gravity of the pressurizer model. Since the pressurizer is at El 626 f t of the internal structure, the
- response spectra which correspond to El 616 f t and El 649 f t of' the internal structure are considered applicable (Figs. B-5 and B-7) . The fundamental i-frequency of the pressurizer is 18 Hz,61 which indicates a spectral acceleration of 0.32 g (74 damping) for the horizontal directioiss. For the
} vertical direction (Fig. B-3), the spectral acceleration is 0.34 g. l Therefore, the resultant horizontal acceleration is 0.45 g, 2.25 times the original design value of 0.20 g. For the vertical direction, the ratio of the revised acceleration to the original design acceleration value of 0.20 g is 1.70. 4 For the pressurizer heater-tube assemblies, the currently predicted I maximum bending stresses are 2.25 times the results of the analysis given in-Ref. 60, but the stresses are'small and well below the ASME III.-2 condition D stress values. For the support skirt, the; currently predicted axial stress is l 0.70 ksi, and the bending stress is 1.27 kai; again, both are small and well-within-the ASME III-l condition D limits. The pressurizer support lugs were . designed for loads due to pipe rupture plus the SSE. Since the-pipe rupture loads control fthe design and are much larger than the SSE loads, we believe that an increase in seismic loads will not affect the design of the shock-lugs. It should be noted that the design loads given are for pipe rupture , plus SSE; no individual loads were given. The original stresses in the support bolts due to overturning moment effects were multiplied by the ratio 1 of 2.25; the resulting stress was 18.6 ksi, which is less than the original allowable value of 55 ksi. Based upon review of the Combustion Engineering calculations and {. independent evaluation, we believe that the pressurizer support system will withstand the 0.2-g SSE seismic event without loss of structural integrity. Combination of SSE with LOCA loads was not evaluated. i 115
- s. . .
6.3.1.13 Steam Generators l l The steam generators are vertical cylindrical vessels, supported by the l l internal structure at El 615 f t 1 in. and laterally restrained at the operating deck (El 649 f t 0 in.) . A summary of the stress analysis of the steam generator is given in Ref. 62. The original seismic design specified 0.2 g in both horizontal and vertical directions, applied simultaneously. The response spectra for the steam generators, which correspond to El 649 f t of the internal structure, are given in Figs. B-3 and B-7. The fundamental f requency of the steam generator was not given but was assumed to be 10 Hz or greater; therefore, the corresponding spectral accelerations are 0.64 g for the -horizontal direction and 0.49 g for the vertical direction. The resultant value is 0.90 g for the horizontal direction, and the ratio of this acceleration to the original design value of 0.20 g is approximately ! 4.50. For the vertical direction, the ratio to the original design acceleration value of 0.20 g is 2.45. The steam generator supports were originally designed to withstand a load combination which includes pipe rupture loads in addition to seismic loads. A comparison of the maximum forces and moments for the support structures is given in Table 6-4. In addition, a comparison of the primary coolant nozzle TABLE 6-4. Forces and moments in the steam generator supports caused by loads due to the SSE and a main coolant pipe rupture. Support component SSE pipe break Support skirt: Horizontal force 0.3 x 10 lb 3.0 x 10 lb 6 Vertical force 1.4 x 10 lb 3.0 x 10 lb 6 l Moment about horizontal axis 5.0 x 10 f t-lb 12.0 x 10 f t-lb 0 Moment about vertical axis 0 12.0 x 10 ft-lb Upper support key Force per key 0.12 x 10 lb 0.3 x 10 lb Upper support snubber: 0 Force per snubber 0.1 x 10 lb 0.2 x 10 lb 116
'I l
l I loads for maximum _ seismic loads and pipe rupture loads is given in Table 6-5. l This table compares design loads, but a comparison of allowable forces or stresses for the various load combinations is not available. Likewise, no information concerning pipe nozzle loads, tubing lateral supports, or tubes and tube sheets was supplied. Because this information was not provided by i the licensee, we do not feel we can comment on the design adequacy of the steam generator and supports. TABLE 6-5. Forces and moments in the primary coolant inlet and outlet i , nozzle due to seismic and pipe rupture loads. Force, 106 lb Moment, 10 7 in.-lb Nozzle Fx Fy F, Mx M y Mz SSE load Inlet 12.0 10.4 10.8 12.0 10.2 12.0 I Outlet 10.12 10.02 11.2 11.2 10.4 11.6 Pipe rupture load s i-Inlet -3.46 +1.98 - - --
-4.85 Outlet -1.45 11.77 +1.35 +7.35 +7.35 +7.35 1
6.3.1.14 Reactor Coolant Pumps i The reactor coolant pumps are vertical components supported by the internal structure at El 608 f t 6 in. A summary of the stress analysis of the j reactor coolant pumps is given in Ref. 63. The origin.al seismic design specified 0.55 g in both horizontal and vertical directions, applied ! simultaneously. . The response spectra for the reactor coolant pumps, which correspond to El 616 ft of the' internal structure, are given in Figs. B-3 and B-5. The 3 fundamental frequency of the reactor coolant pump was not given but was assumed to be in the flexible range; therefore, the corresponding spectral 117
i acceleration is 0.9 g (74 damping) for the horizontal direction. The pump was ! assumed rigid in the vertical direction; the corresponding spectral
, acceleration is 0.34 g. The resultant horizontal acceleration is 1.27 g, 2.30 times the original design value of 0.55 g. The ratio of the revised vertical h
l acceleration to the original design acceleration value of 0.55 g is 0.62. The reactor coolant pump supports were originally designed to withstand a $ load combination which includes pipe rupture loads in addition to seismic d- loads. However, a comparison of seismic and pipe rupture design loads and a
- comparisen of allowable forces or stresses for the various load combinations was not provided. Therefore, based on the limited information provided by the licensee, we cannot comment on the design adequacy of the reactor coolant pump j and _ supports. ,
1
- 6.3.1.15 Reactor Vessel l The reactor vessel for the Palisades plant is supported at the nozzles
! (centerline El 618 f t 2-1/2 in.) by steel brackets, which are supported in turn by the primary shield wall. A summary of the stress analysis of the i reactor vessel is given in Ref. 64. The original seismic design specified a n
0.468-g horizontal acceleration and a 0.312-g vertical acceleration, acting simultaneously. _ The response spectra for the reactor vessel, corresponding to El 616 f t $ of the internal structure, are given in Figs. B-3 and B-5. Assuming the i reactor vessel to be rigid, the corresponding spectral acceleration is 0.28 g for the horizontal direction and 0.34 g for the vertical direction. The
; resultant horizontal accelegation is 0.395 g, whose ratio to the original 3 design value of 0.468 g is 0.84. For the vertical direction, the ratio of the revised acceleration to the original design acceleration value of 0.312 g is i 1.09. Based upon the above spectral acceleration ratios, it appears that the f reactor vessel will withstand a 0.2-g SSE without loss of structural integrity. However, due to the limited information provided by~ the licensee,
- we do not feel we' can conument on the actual design adequacy of the reactor 4 vessel and vessel internals. 3 4 118 L _ ___ _ - _ _ __. _ ___
6.3.2 Electrical Equipment The seismic qualification performed on the Palisades plant electrical equipnent, as provided by the licensee in Amendment No.15 to the FSAR,
- p. 7.7-1, is sumarized in Table 6-6. The qualification documentation listed in the third column was not provided for this evaluation.
6.3.2.1 Battery Racks The battery racks used for the Palisades plant were manufactured by Gould-National Batteries, Inc. They appear to be similar in design to the 125-V racks installed in the Dresden 2 and Ginna stations,1,66 except that additional diagonal bracing was added at the time of installation at the request of the architect / engineer. The floor response spectra for the auxiliary building (El 610 ft, Figs. B-13 and B-14) are assumed applicable to the battery racks. Given the rigidity of the racks, accelerations applicable to the racks are essentially the same as the floor accelerations. On this basis, we recomend that the wooden battens which now laterally restrain the batteries be strengthened or replaced so that friction between the batteries and their support rails no longer need be relied upon to carry the seismic loa 1. 6.3.2.2 Motor Control Centers The ac and de motior control centers (MCCs) are located in the auxiliary-building at El 607 ft. The ac MCCs were supplied by Cutler-Hammer, Inc., and are shown on drawing 94-D9801ED-837, sheets 1, 2, 3, 4, 9, and 10. The de l MCCs were supplied by Westinghouse and are shown on Westinghouse Electric Corporation drawing E13AC-950PB10, sheet 14. The original seismic design for the ac MCCs considered a 0.25-g horizontal acceleration and a 0.14-g vertical acceleration, acting simultaneously.67 For the de MCCs, the values were 0.283 g (horizontal) and 0.144 g (vertical), again acting simultaneously. The response spectra considered for seismic design adequacy were those for the auxiliary building at El 610 f t (see Pigs. B-13, B-14, and B-19) . The peak floor accelerations for the N-S, E-W, . and vertical directions are 0.38, 119
- . - _ - - . - . _ - - . - ~ . . .
E' TABLE 6-6. Original electrical and instrumentation seismic design L qualifications.
- l l
l Squipment Specified Design , Emergency 0.23 g horiz. 2.5 g for locomotive and marine service. generators 0.13 vert.
- 2400-V switch- 0.25 g horiz.- ,
gear: 0.14 vert. Breakers 3.0-g vibration test by supplier. Relays 5.0-g vibration test and dynamic analysis by supplier.
' Structure Bechtel analysis: structure is rigid.
. 480-V load 0.25 g horiz. Prototype unit shock tested by supplier [ . centers: 0.14 g vert. at 5 g max. l Transformers 6 g; shock tested at more than 6 g.
- . Breakers Relays Structure l Preferred ac bus, 0.28 g horiz. Dynamic analysis by supplier unit
[ Battery chargers, . remained operable at 0.75 g.
' inverters Batteries 0.30 g horiz. Designed for more than 0.3 g with cell i 0.14 g vert. impact spacers and braces.
Battery rack .0.30 g horiz. Bechtel analysis: structure braced and 0.14 g vert. rigid. 480-V MCC: 0.25 g horiz. Unit designed for 1.3 g for marine Breakers 0.14 g vert, service (momentary interruption only). S tarting . S tructure continued aAll seismic Class 1 equipment supported directly on the floor levels have been analyzed statically for the' floor acceleration, and the support structures, including the anchor bolt systems, have been designed to withstand the shear load and the equipment overturning moment. Where seismic Class 1 components are installed within structures, such as control panels or racks (which are supported from a concrete floor or wall), the structures have been analyzed and are rigid or restrained such that acceleration is not amplified above the specified floor-level acceleration. Component anchorages within a structure, such as instrumentation mounts, have been determined to be
- adequate, since (1) a conservative component mass was assumed, (2) a minimum standard ' anchor system was provided, and '(3) it was determined that an acceleration greater than 5.0 g, which is 'far above the design acceleration, would be required fbo reach yield stress in the anchorage system.
120 __ _ _ ._ ___ _ _ _ _ _ ~ _ . . . _ _ . . . .
TABLE 6-6 continued. Equipment Specified Design Main control 0.30 g horiz. Bechtel analysis: structure is rigid. boards Shutdown panel 0.20 g horiz. Bechtel analysis: structure is rigid. 0.13 g vert. Transmitters 0.30 g horiz. Prototype shock tested at more than 0.5 g 0.14 g vert, by supplier. Switches 0.30 g horiz. Prototype shock tested at 15 g by supplier. 0.14 g vert.. Cable trays -- Support system braced and rigid. 0.36, and 0.28 g, respectively. The peak spectral accelerations at 3% damping for the N-S and E-W directions are 1.65 and 1.57 g, respectively. The MCCs were considered flexible in the transverse direction and rigid longitudinally. Thus, typical ac and de MOCs units were analyzed for the peak spectral acceleration of 1.65 g in the transverse direction and for the floor acceleration value of _0.36 g in the longitudinal direction. The seismic accelerations were applied simultaneously to the MCds, and the resulting anchor bolt stresses were determined. The analysis established factora of safety of 20.0 for the ac MCC anchor bolts and 12.0 for the de MCC anchor bolts, when compared to ASME III-2 condition D stress limits. Therefore, we believe that the ac and de MCCs will withstand a 0.2-g SSE seismic event without failure of the control centers' anchorage system. However, no information was supplied concerning the design ) adequacy of the control cabinets or the functional adequacy of the cortained electrical components. Hence, additional analysis or test information is required before structural integrity and design adequacy can' be assured. 6.3.2.3 Switchgear The switchgear for the Palisades plant are located in the auxiliary building at El 590 ft and El 607 ft. They were supplied by the Allis Chalmers
. Mfg. Co. and are shown on Allis Chalmers drawings 18-463-546-417, 121
r 118-463-546-418,18-463-546-419,- 72-422-906, and 72-422-907. The original seismic design for the switchgear specified a 0.25-g horizontal acceleration and a 0.14-g vertical acceleration, acting simultaneously. ' Allis Chalmers' has stated thst, since the switchgear has withstood recorded input shipping shocks of 3 g (horiuontal) and 1 g (vertical) , the switchgear should withstand the scismic accelerations specified. This would be true if the support of the switchgear during shipment were similar to that of the installed switchgear. In general, however, this is not the casc. Also, the shock dur ing shipment gives no information as to functional adequacy during a seismic disturbance. The response spectra considered for seismic design adequacy are those for the auxiliary building at El 610 f t (Figs. B-13, B-14, and B-19) . The peak floor accelerations for the N-S, E-W, and vertical directions are 0.38, 0.36, and 0.28 g, respectively. The peak spectral accelerations at 3% damping for the N-S and E-W directions are 1.65 and 1.57 g, respectively. The switchgear cabinet support anchorage was analyzed for the peak spectral acceleration of 1.65 g in the transverse direction and the floor acceleration value of 0.36 g in the longitudinal direction. The seismic accelerations were simultaneously applied to the switchgear, and the resulting anchor bolt stresses were determined. The analysis established a factor of safety of 1.18 for the- assumed 7/8-in.-diameter A-307 bolts, based on ASME stress limits. If the bolt diameter is actually less than 7/8 in., the factor of safety would be less than one, and the design would be inadequate. Also, the number of anchor bolts per unit has been taken as four, whereas the drawings indicate the porsibility of six anchor bolts per unit. Therefore, we recommend that the licensee verify the size and number of anchor bolts used to secure the switchgear to the floor.- Furthermore, no analysis or test results have demonstrated the structural integrity of the switchgear racks or the functionality of the contained electrical components. Hence, additional analysis or tests are required before structural integrity and functional adequacy _can be assured. 6.3.2.4 Control Room Electrical Panels The control room electr Acal panels, which were supplied by the Harlo Corporation, are located in the control room at El 625 f t; their arrangement 122
and location are shown on Bechtel Corporation drawing M-183. The original seismic design specified a 0.3-g horirontal acceleration and a 0.14-g vertical acceleration. The method used to determine design adequacy assumed that the structure is rigid. Since response spectra for El 625 f t of the auxiliary building are not available, the floor acceleration values were taken as the average of the El 610 f t and El 640 f t values (Figs. B-13, B-14, B-15, B-17, and B-19) . The corresponding peak floor accelerations for the N-S, E-W, and vertical directions are 0.40, 0.39, and 0.28 g, respectively. Since these values are higher than the original spectral design values, it is recommended that the licensee verify the seismic design adequacy of the panels for a simultaneous seismic acceleration of 0.4 g in both horizontal directions and 0.3 g in the vertical direction. 6.3.2.5 Transformers The transformers for the Palisades plant, which were supplied by the ITE Circuit Breaker Company, are located in the auxiliary building at El 607 f t and are shown on ITE drawings 33-42924-P-02, 33-42924-P-01, and 73488-B31. The original seismic design specified a 0.25-g horizontal acceleration ar.d a 0.14-g vertical acceleration, acting simultaneously. The response spectra considered for seismic design adequacy are those for the auxiliary building at El 610 f t (Figs. B-13, B-14, and B-19) . The peak floor accelerations for the N-S, E-W, and vertical directions are 0.38, 0.36, and 0.28 g, respectively. The seismic accelerations were applied simultaneously to the transformer, and the resulting anchor bolt stresses were I determined. The analysis considered the center section to act separately from the end units. The resultant factor of safety, as measured against ASME III-2 l condition D stress limits for the assumed A307 anchor bolts, is 22.5. For the end units, the overturning effect produces uplif t, and since the end units are not tied down, we recommend that they be anchored. Insufficient information was supplied to evaluate either the structural adequacy of the framework which supports the transformers or the functional adequacy of the transformers. 123
6.3.2.6 Electrical Cable Raceways l Seismic loads were not considered in the original design of cable tray supports. Recent tests have indicated that damping ralues of 20% or more may be justified for cable trays and their supports; nonetheless, an evaluation of the existing cable tray system required for safe shutdown, including supporting documentation of the design assumptions used, is required before design adequacy can be assured. 6.4 PIPING The results of the Palisades piping analyses will be published separately. This brief summary is intended only as a preliminary overview. Portions of four major piping systems were analyzed e Residual heat removal (RHR) system. e Component ecoling system. e Auxiliary feedwater system (three portions were modeled, including the steam line to the P-8B turbine) . e - Regenerative heat exchanger (RHE) letdown system. Throughout, it was assumed that suitable stress analyses of the supports and substructure were performed for the original loads. The r6Jults of the analyses performed on the RHR piping indicate that stresses in this piping will be well within allowable limits during a seismic event equivalent to the postulated SSE. However, where compa:ison was possible, support loads were found to be generally higher than those determined in the original analysis. Further examination of the support members adajected to significantly higher loads is warranted. In addition, the anchor loads determined in the reanalysis are generally higher than the original loads. Further consideration should be given to the nozzles since the anchor moment loading has increased significantly. Piping stresses for the component cooling model are well within allowable limits for an SSE event. The support loads determined in the reanalysis are generally higher than the known original loads, but the increases do not appear large enough to warrant anticipation of failures. No conclusions were l 124 I
drawn for those supports where original loads were unknown. The anchor loads based on the SEP acceptance criteria are generally higher than the original loads. Further consideration should be given to the nozzles in cases where the loads have increased significantly. The reanalysis results for a portion of the auxiliary feedwater piping between pumps P-8A and P-8B show that ASME Code stress limits will be exceeded during an SSE event. The overstressed points occur near pump P-8A. These high stresses result from a deficiency in east-west lateral restraint. When relief valve (RV 0783) discharge loads are concidered in conjunction with the SSE loading, a large increase in the loads on support R227C is indicated. This support sh)uld be reevaluated for this increased load. Insufficient data were available for anchor load comparisons; thus, no conclusions were drawn concerning nozzle capabilities. Analysis of a model of a second portion of the auxiliary feedwater system showed no piping stresses above Code allowables for the SSE. Insufficient data were available for comparisons of support and anchor loading. No conclusions concerning support or nozzle structural adequacy were drawn. In; the third analysis of the auxiliary feedwater system, the piping between the steam line and the P-8B turbine was modeled. Results show that ASME Code stress limits will be exceeded during an SSE event. Additional lateral and vertical dynamic supports are needed in the vicininty of the piping upstream from valve PCV 0521A. Additianal vertical and lateral dynamic supports are also needed in the vicinity of valve MS 0522A. Support loads for the SSE csse are also generally higher than the known original loads. Results indicate that rod hangers Hll, H13, and H14 will probably buckle and become ine ff ective. Nonetheless, the increased support loads are not great enough to warrant anticipation of failure. No conclusions regarding support structural adequacy could be drawn in those cases where original loads were unknown. Insufficient data were available for anchor load comparison; therefore, no conclusions regarding anchor or nozzle structural adequacy were drawn.
' The results for the RHE letdown piping show that ASME Code stress limits will be exceeded during an SSE event. The high stresses are primarily due to a deficiency in axial and lateral restraint near the downstream vertical leg of the expansion loop nearest valve CV 2003. Insufficient data were available for support or anchor load cc:nparisons. No conclusions were drawn concerning support or nozzle structural adequacy.
l i 125 l
6.5- Supe 4ARY AND CONC WSIONS Table 6-7 susmarizes our findings on the sample of mechanical and electrical components and distribution systems that were evaluated to determine the seismic design adequacy of items required for the safe shutdown of the Palisades nuclear steam supply system. As discussed in Sec. 6.1, this sample includes components the review team selected, based on judgment and experience, as representative of the lower-bound seismic design capacity of Palisades. The components of the sample were also chosen to represent important generic groups. Based upon the design review and independent calculations for the SEP seismic load condition, we recommend design modifications or reanalysis of several mechanical and electrical components to ensure t. hat they can withstand the 0.2-g SSE without loss of structural integrity as required to perform safety functions. In general, no information was provided which demonstrated the functional a6equacy of mechanical and electrical equipment evaluated at the Palisades plant. The specific mechanical and electrical components which require additional evaluation and possible design modification are marked by I asterisks in Table 6-7. 1 J 4 4 i 126
- - . -. . - - =
TABLE 6-7. Summary of conclusions. 1 Item Description Conclusion and recommendation 1 Essential service water OK for structural integrity if discharge pump
- head stresses are within code allowables (no use of cast iron) . Aside from the
' anchor bolts, functional integrity was not
~
evaluated because of lack of design detail. 2 Auxiliary feedwater pumps
- OK for structural integrity. Functional integrity was not evaluated because of lack of design detail.
3 Component cooling heat OK. exchangers 4 Component cooling surge OK. tank 5 Diesel generator oil No evaluation was performed, since storage tanks
- no drawings or design calculations were t
available. 6 Boric acid storage tank OK. 7 Hydrazine tank OK. 8 Sodium hydroxide tank OK. 9 Safety injection tank
- OK if tank support structure is rigid.
Complex support structure should be evaluated for dynamic characteristics to ensure that rigidity assumption is correct. 10 Motor-operated valves
- Generic analysis of motor-operated valves on lines 4 in. in diameter or smaller should be performed to show that resulting stresses in the pipe are less than 10% of the applicable condition B (active) or condition D (passive) allowable stresses.
Otherwise, stresses induced by valve eccentricity should be introduced into piping analysis to verify design adequacy, or a procedure should be implemented whereby all such motor valves are externally supported. Also, verification of structural adequacy and function of the valves themselves were not demonstrated. continued 127
_ ~ _ . . - . . l l l TABLE 6-7 continued. i l Item Description Conclusion and recommendat!on 11 Control rod drive OK for structural integrity. Based on the mechanism
- calculations reviewed, active function l cannot be assured.
12 Pressurizer OK. l 13 Steam generators
- Insufficient information provided to verify l
design adequacy. l ! 14 Reactor coolant pumps
- Insufficient information provided to verify i design adequacy.
l l 15 Reactor vessel supports Insufficient information provided to verify l and internals
- design adequacy.
l 16 Battery racks
- Racks OK, except wooden lateral bracing should be replaced or strengthened to carry full seismic inertial loads.
17 Motor control centers
- Anchorage OK. No informati^n available to evaluate rack structural ade 'uacy or electrical component function. lity.
18 Switchgear* Anchorage OK if anchor bolts Lre 7/8 in. in diameter; otherwise, design modifications may be necessary. No information available to evaluate swityhaear tack _st rustural adequacro6ectrical component j ,-functionality. 19 Control room electrical Licensee to verify seismic design adequacy. panels
- 20 Transformers
- End units of transformers should be securely anchored. No information available to evaluate structural adequacy or electrical functionality.
21 Electrical cable raceways
- Cable tray support systems shcold be evaluated for seismic loads induced by l 0.2-g SSE.
l
- Components requiring additional evaluation and possible design modifications. See text.
I 128 l
i I l l REFERENCES
- 1. N. M. Newmark, W. J. Hall, R. P. Kennedy, J. D. Stevenson, and F. J.
Tokarz, Seismic Review of Dresden Nuclear Power Station-Unit 2 for the Systematic Evaluation Program, U.S. Nuclear Regulatory Commission, NUREG/CR-0891 (1979) .*
- 2. "Backfitting," U.S. Code of Federal Regulations, Title 10, Part 50.109.
- 3. U.S. Nuclear Regulatory Commission, Standard Review Plan for the Review of Safety Analysis Reports for Nuclear Power Plants, Office of Nuclear f Reactor Regulation, NUREG-75-087, LWR edition (1975) . **
- 4. . M. Newmark and W. J. Hall, Development of Criteria for Seismic Review of Selected Nuclear Power blants, U.S. Nuclear Regulatory Commission, NUREG/CR-0098 (1977) . **
- 5. T. A. Nelson, Seismic Analysis Methods for the Systematic Evaluation Program, Lawrence Livermore National Laboratory, Livermore, Calif. ,
UCRL-52528 (1978).
- 6. Consumers Power Company, Final Safety Analysis Report, Jackson,' Mich.,
included in NRC docket 50255-Al through A4 (1978, plus appendixes) . 7.--U.S.-Nuclear-Regulatory Commission, Seismic Design Classification, Regulatory Guide 1.29, Rev. 3 (1978) .
- 8. U.S. Nuclear Regulatory Commission, Damping Values for Seismic Design of Nuclear Power Plants, Regulatory Guide 1.61 (1973).
- 9. J. L. Alford, G. W. Housner, and R. R. Martel, Spectrum Analysis of Strong Motion Earthquakes, Califocnia Institute of Technology, Pasadena, Calif. (revised August 1964).
- 10. Consumers Power Company, Spent-Fuel Pool Modification Descriptio,n and Safety Analysis, Jackson, Mich. (November 1976) .
- 11. R. M. Marusich, Consumers Power Company, Jackson, Mich., letter to Howard Levin, U.S. Nuclear Regulatory Commission (August 28, 1979).
- 12. D. P. Hoffman, Consumers Power Company, Jackson, Mich. , letter to D. L.
Ziemann, U.S. Nuclear Regulatory Commission (June 8,1979) .
- 13. American Concrete Institute, Building Code Requirements for Reinforced Concrete, ACI 318-63 (1963) 129
L
-i
- 14. ' Consumers Power Company, Jackson, Mich. , Spent-Fuel Pool Modification, NRC docket No. 50255-926 (February 8,1977) . ,
- 15. " Seismic and Geologic Siting Criteria," Code of Federal Regulations, Title 10, Part 100, Appendix A.
- 16. U.S. Nuclear Regulatory Commission, Design Response Spectra for Seismic f .
I Design of Nuclear-Power Plants, Negulatory Guide 1.60 (1973). I
- 17. E. Kausel, R. V. Whitman, F. Elsabee, and J. P. Morray, " Dynamic Analysis !
of Embedded Structures,"- in -Trs.nsactions, Fourth International Conference s I on Structural Mechanics in Reactor- Technology, San Francisco, Paper K2/6, l Vol K(a) (1977). 18'. E. Kausel, and R. Ushijima, Vertical and Torsional Stiffness of Cylin& ical Footings, MIT Research Report R79-6, Dept. of Civil Engineering, Massachusetts Institute of Technology (February 1979) . !
- 19. J. E. Luco, " Impedance Functions for a Rigid *oundation on a Layered Medium," Nucl._ Engin. Design, Vol. 31 (1974).
- 20. T.'W.' Lambe, and R. V. Whitman, Soil Mechanics, John Wiley & Sons, New York (1969) .
- 21. F. E. Richart, J. R. Hall, and R. D. Woods, Vibration of Soils and Foundations, Prentice-Hall, Inc. , Englewood Cliffs, N.J. (1970).
- 22. H. B. Seed and I. M. Idriss, Soil Moduli and Damping Factors for Dynamic Response Analyses, University of California, Berkeley, Repcrt No. EERC l
70-10 (1970).
- 23. U.S. Nuclear Regulatory Countission, Combining Modal Responses and Spatial Components in Seismic Response Analysis, Regulatory Guide 1.92, Rev. 1 ;
(1976).
- 24. American Society M Mechanical Engineers, Boiler and' Pressure Vessel-l Code, Sec. III (1977).
I
- 25. American Concrete Institute, Code Require:.wnts for Nuclear Safety-Related Concrete Structures,- ACI-349 (1976) .
i ! - 26 . J. M. Roesset, R. V. Whitmari, and R. Dobry, " Modal Analysis for Structures with Foundation Interaction," J. Str. Div. , ASCE, Vol. 99, l No. ST3, 399-416 (March 1973). ;
- 27. A.~ S. Veletsos and J. Y. Yang, Dynamics of Fixed-Base Liquid Storage Tanks, Rice University, Houston (November 1976).-
- 28. A. S. Veletsos, " Seismic Effects in Flexible Liquid Storage' Tanks," in Proceedings, Fifth World Conference on Earthquake Engineering, Rome, I
pp. 630-639 (1974) . 130
- .~. . - _ . - - ., .- _ _ _ _ - - . , _ . - - _ _ . . ._ -__- __
- 29. R. P. Kennedy, Recommendations for Changes and Additions to Standard Review Plans and Regulatory Guides Dealing with Seismic Design Requirements for Structures, Lawrence Livermore Laboratory A-40 Program, EDAC 175-150 (June 1979) .
- 30. Newmark, N. M., "A Response Spectrum Approach for Inelastic Seismic Design of Nuclear Reactor Facilities," in Transactions, Third International Conference on Structural Mechanics in Reactor Technology, London, Paper K5/1 (1975) .
- 31. N. M. Newmark, " Seismic Design Criteria for Structures and Facilities, Trans-Alaska Pipeline System," in Proceedings, U.S. National Conference on Earthquake Engineering, Ann Arbor, Mich. , Earthquake Engineering Research Institute, pp. 94-103 (June 1975) .
32 M. A. Igbal and E. C. Goodling, Jr., " Seismic Design of Buried Piping," presented at 2nd ASCE Speciality Conference on Structural Design of Nuclear Plant Facilities, New Orleans, December 8-10, 1975.
- 33. G. C. K. Yeh, " Seismic Analysis of Long Buried Structures," in Proceedings, 2nd International Conference on Structural Mechanics in Reactor Technology, Berlin, Germany, Paper K4/8 (September 1973).
- 34. J. M. E. Audibert and K. J. Numan, " Coefficients of Subgrade Reaction of the Design of Buried Piping," presented at 2nd ASCE Speciality Conference on Structural Design of Nuclear Plant Facilities, New Orleans, December 8-10, 19/5.
- 35. K. K. Kapur and L. C. Shao, " Generation of Seismic Floor Response Spectra for Equipment Design," in Proceedings, Specialty Conference on Structural Design of Nuclear Facilities, ASCE, Vol. I (December 1973).
- 36. C. G. Duff, discussion of Ref. 35, in Proceedings, Specialty Conference on Structural Design of Nuclear Facilities, ASCE, Vol. III (December 1973).
- 37. U.S. Nuclear Regulatory Commission, Design Limits and Loading Combinations for Seismic Category I Fluid System Components, Regulatory Guide 1.48 (1973).
- 38. M. E. Nitzel, Summary of SEP Calculations for the Palisades Unit 1 Piping, EG&G, Idaho, Inc., to be published.
- 39. G. W. Housner and R. A. Williamson, Nuclear Reactors and Earthquakes,
,U.S. Atomic Energy Commission, TID-7024 (1963).
- 40. J. D. Stevenson, " Structural Damping Values as a Function of Dynamic 131
I Response Stress and Deformation Levels," paper K11/1 presented at 5th SMIRT Conference, Berlin, August 14-20, 1979. l 41 P. Y. Hatago and G. S. Reimer, " Dynamic Testing of Electrical Raceway l Support Systems for Economical Nuclear Power Plant Installation," paper
- F 79 166-0 presented at IEEE PES Winter Meeting, New York, February 1979.
l l 42. J. D. Stevenson, Evaluation of the Cost Effects on Nuclear Power Plant Construction Resulting f rom the Increase in Seismic Design Level, draf t prepared for U.S. Nuclear Regulatory Commission, Office of Nuclear Regulatory Research (May 1977).***
- 43. Layne & Bowler, Inc., Essential Service Water Pump Seismic Calculations, P.0, 5935-M-11-AC (May 1968).
D. J. Kirkner, Essential Service Water Pump Seismic Evaluation, Woodward-
~
44. l Clyde Consultants (February 1980) .
- 45. Bingham Pump Co., Auxiliary Feedwater Pump Seismic Calculations, P.O.
5935-M-17-Ac (September 1969).
- 46. D. J. Kirkner, Auxiliary Feedwater Pump Seismic Evaluation, Woodward-l Clyde Consultants (February 1980) .
! 47. P. J. Gallagher, Component Cooling Heat Exchanger Seismic Evaluation, i Woodward-Clyde Consultants (April 1980) .
- 48. P. J. Gallagher, Component Cooling Water Surge Tank Seismic Evaluation, Woodward-Clyde Consultants (April 1980),
i 49. Combustion Engineering, Inc., Boric Acid Storage Tank Seismic calculations, contract 1066 (October 1967) .
- 50. P. J. Gallagher, Boric At
- id Storage Tank Seismic Evaluation, Woodward-Clyde Consultants (April 1980) .
l
- 51. Buffalo Tank Division, Pethlehem Steel Corp., Hydraxine Tank Seismic Calculations, contract X-1031 NC (July 1973) .
- 52. P. J. Gallagher, Hydrazine Tank Seismic Evaluation, Woodward-Clyde Consultants (April 1980) .
- 53. Buff alo Tank Division, Bethlehem Steel Corp., Sodium Hydroxide Seismic Calculations, contract No. X-1032 NC (April 1973) .
- 54. P. J. Gallagher, Sodium Hydroxide Tank Seismic Evaluation, Woodward-Clyde Consultants (April 1980) .
- 55. Nooter Corp., Safety Injection Tank Seismic Calculations, job No. F6171 (March 1968).
- 56. Bechtel Corporation, safety Injection, Tank Support calculations, job No. 5735 (September 1967) .
132
57 P. J. Gallagher, Safety Injection Tank Frequency Analysis, Woodward-Clyde I Consultants (April 1980) . ,
- 58. F. A. Thomas, Motor Operated Valves Seismic Calculations, Woodward-Clyde Consultants (April 1980).
- 59. Combustion Engineering, Inc., Control. Rod Drive Mechanism Supports Seismic calculations, contract No.1966 (December 1967) .
- 60. Combustion Engineering, Inc., Analytical Report for Consumers Power Pressurizer, report No. CENC-lll4 (March 1969).
~
- 61. P. J. Gallagher, Pressurizer Frequency Analysis, Wcodward-Clyde Consultants (April 1980) .
- 62. Combustion Engineering, Inc., Analytical Report for Consumers Power Steam Generator, report No. CENC-1120 (May 1969) .
- 63. Byron Jackson Pump Div., Borg Warner Corp., Stress Report for 35x35x42 Pump Core (November 1969).
- 64. Combustion Engineerinet, Inc., Analytical Report for Consumers Power Reactor Vessel, report No. CENC-lll6 (October 1968) .
- 65. Gould National Batteries, Inc., Palisades Station Battery Racks Selection Chart, requisition No. 5935-E-12,
- 66. Rochester Gas and Electric Co., Ginna Station Battery Rooms A & B As Built Sketch of Battery Racks, plans and sections, SS-022180.
- 67. Bechtel Corporation, Palisades Plant Specification for 480 Volt Motor Control Centers, specification No. 5935-E-7 (December 1,1967) .
- 68. Bechtel Corporation, Palisades Plant Specification for DC Control and Distribution Centers, specification No. 5935-E-13 (June 7,1968) .
- 69. Bechtel Corporation, Palisades Plant Specification for 4160 and 2400 volt Switchgear, specification No. 5935-E-5 (August 29, 1967).
- 70. Bechtel Corporation, Palisades Plant Specification for Main Control Panels and Console, specification No. 5935-h -201 (November 29, 1967).
- 71. Bechtel Corporation, Palisades Plant Specification for 480 Volt Load Centers, specification No. 5935-E-6 (August 22, l':67) .
*Available for purchase from the NRC/GPO Sales Program, U.S. Nuclect Regulatory Commission, Washington, D.C., 20555, and the National Technical Information Service, Springfield, VA 22161 **Available for purchase from the National Technical Information Service, Springfield, VA 22161 o**Available in the NRC Public Document Room for inspection and copying for a fee.
133
. - _ ,. . . _ . _ - _ . _ _ _ _ . _ . . . . . . ~ . _ _ _ . . _
1 P 4 4 t l , APPENDIX As . EVALUATION OF THE EFhTIS. OF ASSUMING i J 204 MAXIMim MDDAL DAMPIN6 t. f' A.1 illrrRODUCTION j In the eva3uation of the seismic response of the Palisades containment [ i a.d auxiliary buildings, composite modal camping ratios, were computed by [ considering both the expected structure damping and the energy dissipation j within the soil. The latter . effect incluued buth geometric (or radiation) . camping ano soil. material damping. Calcul'ation of the composit'e modal damping' i j values was based on energy proportioning, as discussed in Chapter.5. I It was decided to limit the composite modal damping to no more than 20% of critical for both the ccntainment and auxiliary uuildings. 'Since both structures have calculated horizontal geometric damping ratios significantly [ aoove tne 20% limit, lecs energy was assumed to .be dissipated than woulo be . i ! l predicted theoretically for modes. involving significant-horizontal ~ soil respor.se. Furthermore, the use of modal camping often leads to a substantial ' } dif f erence between the location where energy is assumed to be dissipated and , t.ne location where it is actually dissipated. For instance, the assumption of moaal damping may lead to the prediction that less energy is dissipated in the soil and more in the structure enan is actually tne case for moues with large-4' amounts of soil damping and significant structural response. i I'n order to investigate tnese effec,ts, an'indepencent, conrirmatory l analysis was conducted for the containment builoing, using a discrete cashpot
) . to simulate the energy oissipation in the soil. This analysis was conuucted I
- for one case, using the full theoretical value computed for the horizontal i damping factor. A secono case was investigated using 75% of the theoretical
- horizontal geometric camping value. The 754 level was based on the' review of
} v..e set of; test results. conoucced for the San Onofre Nuclear Generating Station, Units.2 and 3.A These tests-involved concrete slabs embedded l -difterent amounts in the soil. The full theoretical value of rocking damping-
. was used in the analyses for both cases.
i p 135
. c. - ,-
A.2 MErHOD OF ANA1.WIS The analysis of the-containment butiding with a discrete cashpot simulating the energy cissipation that reJults fran horizontal translational
. soil modes was performed by direct integration of the equations of motion
~
using the a:Maputer program DRAIN-2D. For case 1, the full theoretical 444 horizontal geometric camping was combined with 54 soil material damping to give a horizontal energy dissipation corresponding to 39% of critical damping. For the second case, 26% of critical was used for the horizontal geometric damping. This was combined with the 54 soil material damping to give an overall value of 31% of critical for the horizontal damping. (These values were determined as described in Sec. 5.3.1.) The remainder of the two-oimensional analytical model was the same as was used in the modal analysis, described in Chapter 5. The same rocking damping was used, and the structural camping was again assumed to be 3% of critical for all mooes (as recommencea in Ref. A-3) for prestressed and well-reinforced concrete with stresses at no more than one-half the yield point. The analysis was conducted for the mecian soil case with a limited check for the upper-bound soil. Damping was incorporated into the model for the time-history analysis diff erently than for the modal analysis. For the time-history analysis using DRAIN-2D, the viscous camping matrix was assumea to be of tne form I ) [C ] = a[M] + S [K] + % , where ! [C] = system damping matrix, [M] = system mass matrix, 4 [K] = sial.ee stiff ness matrix, a,$ = damping oarameters, [Cy'=concentranodashpotmatrix. The dashpot was added along the diagonal of the system damping matrix at the cegree of treecem with which the cashpot is associateu. The input for the time-history analysis consisted of an artificial earthqaake whose response spectra are close to the smoothed spectra in l i 136
I R.G. 1.60 but do not necessarily envelop the R.G. 1.60 spectra at all trequencies (Fig. A-1). The direct integration of the equations of motion was performed using a technique which assumes constaat acceleration within the time step. The method is stable for all frequencies and does not introduce numerical damping, but small time steps are required. Integration time steps i j i i
; ,ii., i ; . i ..; i l
i i
;iii.
4 - 3 - 2 -
-8 _
e j _ 8
< 2 -
1 0 ' I ' 'l ' I ' ' I''l ' I ' ' I' O.1 0.2 0.5 1 2 5 to 20 50 100 Frequency (Hz) t FIG. A-1. Response spectrum (2% damping) used for the time-history analysis 1 of the cotstainment builcing, superposed on the correspunding smoothed spectrum ) from R.G. 1.60. 137
l l of 0.01 s wm o used in the analysis. This time step is expected to provide
. good accuracy for f requencies up to near 20 Hz, which is adequate for the region of interest for the reactor containment building.
l l A.3 BUILDING RESPONSE l Maximum shear and moment distributions throughout the structure, i including the containment vessel and concrete interr.al structures, as determined from the time-history analysis are shown in Figs. A-2 and A-3. For the purpose of comparison, the corresponding shear and moment distributions
- obtained from the response-spectrum modal analysis (using composite modal damping with a 20% upper limit) are shown in the same figures. This l comparison indicates that the latter analysis results in somewhat higher response throughout the structure, compared with either of the time-history cases. For case 1 (full theoretical geometric damping), base shear in the containment is approximately 10% greater using the 20% camping limit; shear remains about 4% greater at the higher elevations. Shear in the concrete l internals is about 5% to 13% greater for the 20% damping case. . Bending l
. moments are approximately 5% greater at the base of the containment vessel and
[ - about 11%, greater in the concrete internals. For case 2, using 75% of tne I theoretical damping, somewhat less reduction in response is indicated; however, both shear and moment response throughout the structure are somewhat less than computea using the composite modal damping limited to 20% of critical. These comparisons are r.ot expected to be exact, since the response spectrum produceu by the time history is not identical with the R.G.1.60 spectr um. Consequently, icentical responses woulo not be expected, even if the damping were treated in exactly the same manner. However, the results are considered representative in oetermining the effects of the 20% limit on modal damping. I From these results, it is apparent that for the Palisades analysis the use of composite modal damping limited to 20% of critical' produces slightly l conservative structural response results. Such an approach is theriefore i acceptaole. l I l l l l l
- 138 L
l I
. . ___m
765 n; . i i 740 - , - I l 715 - l - l
, 690 -
Response-spectrunk analysis - . 'E
.o I / Time history analysis (case 1)
I j , Time-history analysis (case 2)
"' 665 - #. -
)
i internal structure
--?
640 - l Containment l, vessel 615 - , 1; l. I I I I I I ' 590 U O 10 15 20 25 0 5 Shear (103 kip) Shear' distributi'on in the containment building as determined froic FIG. A-2. ! the response-spectrum analysis of Clapter 5 and two time-history analyses. s 1 139
- .. =. __ . ..
l l l I I I I I I I I I l i 740 l l - l 715 - l l t I 7 690 j l O~ l l 8 ! y Response spectrum analysis l 8 G - 665 +3- Time-history l analysis (case 1) l Time-history Internal l - 640 - analysis (case 2) structure
\
I , Containment ! vessel l l 615 - xx. I l I I I I I I ' 590 O 1 2 3 0 0.08 0.16 0.24 0.32 0.40 0.48 ! Moment (106 kip ft) FIG. A-3. Moment distribution in the containment building as determined from-the response-spectrum analysis of Chapter 5 r.. two time-history analyses. i 140
~ - . - , - ,e
A.4 IN-STRUCTURE RESPONSE SPECTRA l To evaluate the effects of composite mooal damping (as used in Chapters 5 and 6) on the response of equipment as well as on the response of the structure, in-structure response spectra were generated for several locations, using the respunse votained from the two time-history analysis cases uescribed above. In-structure response spectra for 34 equipment damping were generated for the median soil case at the containment building base slab (El 590 f t), at the top of the concrete internals (El 649 f t), and at El 730 f t in the containment vessel; and ror the upper-bound soil case at El 616 f t and El 649 ft in the internals. These spectra were smoothed and broadened as described in Sec. 5.6 and are shown in Figs. A-4 through A-8. For comparison, the corresponding spectra for 36 equipment camping, developed using the 204 maximum modal damping results, are snown in the same figures. At the base slab and containment vessel locations, the in-structure spectra produced by the two methods are similar in shape and magnituce, though the time-history results are, in general, somewhat lower than those developed using the 20% lis at on modal damping. This is particularly true in the high-frequency regions, wnere the decrease in the spectra corresponds to a similar cecrease in the structural response acceleration levels. Only very slight differences are evident wnen comparing spectra generated using the full theoretical geometric damping with those generated using 75% of the theoretical value. At the top of the concrete internal structure, some modification in the snape of the response spectrum is noteo. This occurs because of changes in the response contributions from the second and third horizontal modes. The second horizontal mode (5.8 Hz) is casically a soil translation mode, wnereas the third horizontal mode (12.9 Hz) is primarily shear oeformation of the concrete internals structure, with little soil displacement. Use of the discrete soil dashpot tends to add a significant amount of camping to the secono mode, while the damping of the third mode is occreased in comparison with the composite modal damping limited to 20% of critical. Thi-3 is reflected in the in-structure response spectra shown in Fig. A-5, where, for the time-history results, the response near 13 Hz is amplified while that near 6 Hz is attenuateo. 141
10 _ i i , , , , , , Reactor building - Base slab El 590 f t _ )
~ -
Response-spectrum analysis
~
Time history analysis (case 1) Time-history analysis (case 2) 3 _ _ C
.o E1 -
m 3 _ _ g _ _ u _ _ i' E- : i
/ -
~
1 O.1 , , , , , , , , I i i . e i i . il i e i e i i e 0.01 0.1 1 10 Period (s)
?
FIG. A-4. Comparison of base slab response spectra. All three assume 34 equipnent camping and median
- s. ail conditions. ,
'l
10 , , , , , _ internal structure ~ El 649 f t c Response spectrum analysis 5 - Time-history analysis (case 1) - 8 Time-history analysis (case 2) 51 m' _
~
--g -
.~
g - g - w __________. - - 0.1 . . . . . ..il i , , , . ,,il , , . , i i,, 0.01 0.1 1 to Period (s) FIC. A-5. Comparison of response spectra for the concrete internals (El 649 ft) . All three assume 34 equipment uamping and mecian soil conditions.
i - i . . i . iio g i a_ 10 _ i , . i , iii; i i i Containment vessel El 730 ft _ Response spectrum analysis ~ Time-history analysis (case 1) Time-history analysis (case 2) . g _ C
.9 -
E lii 1 -
~6 _ ' -
i 8 e --
' ' ' ' ' I ' ' ' ' ' ' ' '
. , . I ' ' '
0.1 . - 1 10 1 O.01 0.1 l ' Period (s) All three assume 34 equipment Comparison of response spectra for the containment vessel. FIG. A-6. camping stad mesian soil conditions. l I
10 , , , , , , internal structure _ El 649 ft _
~
en g
. ~
Upper-bound soil ,
.9 i
%u 3
__ MMian soil 's 1 -
$ g
~& A I ~
O s /
/
/
~ _
l 4 w
~
e' Response-spectrum analysis $ 2, ,
._________s ---Time-history analysis (case 1) _
0.1 i 1 i , i . - 1 1 i i i ' ' ' iil i ' ' ' ' O.01 0.1 1 10 Period (s) FIG. A-7. Ccaparison of response spectra for the concrete internals (El 649 f t), for upper-bound and mecian soil conditions. All tour assume 3% equipment camping.
3 , , . . . i
.g . > i i i a a i j '
~
Internal structure
- El 616 f t l
~
~
5 - Response-spectrum analysis
~
C o g- - Time-history analysis (case 1) - 1 -
~
- s O ~
~
/ -
/ .
w - e - _________s 0.1 ' ' ' ' ' ' ' ' ' ' ' ' ' 'l i i i i i iii 0.1 1 10 O.01 Period (s) Comparison of response spectra for the concrete internals (El 616 f t) . Both asstane 34 FIG. A-8. equipnent damping and upper-couna soil conoitions.
In the region of 11 Hz, the response spectra generated using the discrete dashpot approach exceed somewhat the corresponding spectra based on the composite modal camping metnod. This is true for the median soil case and for tie upper-bouno esse (Fig. A-7) . If the actual soil modulus is near or less tnan the median value used in the analysis, the envelope of spectra accounting for the soil range will cover' the increase in response obtained by using the aiscrete cashput method. In any event, as shown in Figs. A-4 through A-8, the uiscrepancies between the spectra generated by the two methods are limited to very narrow frequency ranges. Thus, it is concluded that the response spectra generateo using composite modal damping limited to 20% of critical are acequate for evaluations of piping and equipment. REFERENCES EUR APPENDIX A A-1. Woodward-McNeill and Associates, Soil-Structure Interaction Parameters: Development of Soil-Structure Interaction Parameters for Proposed Units 2 and 3, San Onofre Generating Station for Southern California Edison Co., Appenoix 4.7c (January al,1971) . A-2. G. H. Powell, DRAIN-2D Users Guide, University of California, Berkeley, Calif. (August 1975). A-3. N. M. Newmark and W. J. Hall, Development of Criteria for Seismic Review f of Selecten Nuclear Pcwer Plants, U.S. Nuclear Regulatory Commission, NUREG/CR-0098 (1978).** 1
. l **Available for pure' le from the National Technical Information Service ,
Springfield, VA 22131 147
l t APPENDIX B IN-STRUCTURE RESPONSE SPECTRA The following 20 figures depict the in-structure response spectra developed from the models shown in Figs. 5-3 and 5-7. The spectra are discussed briefly in Sec. 5.6. They are the basis of the reassessment of equipment and piping in Chapter 6. m ' 149
10 _ Reactor building - Base slab El 590 ft -
}
-- R.G.1.60 -
/ 0.005 --- Original design -
/__f0.02 l 3I 0.05
/l/
_ ~ s lI e lg4 0.03
.g 0.05 5 1 g
0.07 - 3 - - 8
/ -
u - g ~
< ~
m f r -
/
/ p\ \/// / \ -
H v/ \ o - lj / / ~~\ ~
/l
/ V e' \(g\
\
h -
/
9s - 0.1 ' ' ' ' ' ' l ' ' ' ' ' ' l ' ' ' ' ' ' ' ' O.01 0.1 1 10 Period (s) FIG. B-1. Comparison of horizontal in-structure response spectra for the containment cuilding base slab. Tne spectra recalculated on the basis of R.G.1.60 envelop the total range of soil properties. Labels indicate equipment damping ratios.
10 , , , , , Reactor building
~
[ Base slab El 590 ft
- Upper soil -
Median soil S Lower soil e
.o 5 1
_a> g - u - r Y -
'/ -
0.1 ' ' ' ' ' I > ' ' ' ' ==I = = i * ' t t. 0.01 0.1 1 10 Period (s) FIG. B-2. Horizontal in-structure response spectra for the contairunent building base slab, based on R.G. 1.00 spectra. The spectra illustrate the variation with the different soil conditions oiscussed in the text. The equipment damping ratio was 0.03 for all three spectra.
10 , _' Reactor building _ Base slab El 590 ft _ Tn w C
.9 3 1
~
$ ". 0.03 _
8 - 0.05 -
# - 0.07 -
w - 0.1 ' ' ' ' ' l ' ' ' ' ' I ' ' ' ' ' O.01 0.1 1 10 Period (s) FIG. B-3. Vertical it.-structure response spectra for the containment building base slab, based on R.G. 1.60 spectra. The spectra envelop the total range of soil properties. Labels indicate equipment camping ratios. These spectra are typical for all elevations in the contairment building.
10 , , , , , Reactor building ~ Base slab El 590 ft
~, o _ Upper soil e Median soil Lower soil 2 1 3 - -
g - g -
~
J . ~ 5.lI. // - I 0.1 > ' ' ' ' 'il - - - ' ' l ' ' ' ' ' O.01 0.1 1 10 Period (s) FIG. B-4. Vertical in-structure response spectra for the containment building base slab, based on R.G. 1.60 spectra. The spectra illustrate the sariation with the different soil conditions discussed in tne text. The equipment damping ratio was 0.03 for all three spectra. The spectra are typical for all elevations in the containment ouilding.
10 , ' ~
~
Reactor building [ Internal struct.El 616 ft
- rg R.G.1.60 -
g I I --- Original design _ _ g
,Ir- d (El 606 f t) r'yl l 0.03 0.05 e
c ili j sr3I I i 0.07
\
If il
/
S 1 l il ~
$ : ll / \ _
\\
/
k - p ^
~
$ // ['\ \ ~
g
^ ~ f
/
/
/ g 0.005 0.02 g g
#p d 1 0.05 ~
- - - - - -- - - - _ /,/ ,
_ ______s s N N 0.1 ' ' ' ' ' ' ' ' ' ' il > i i > >i O.01 '01
.I 1 10 Period (s)
FIG. B-5. Comparison of horizontal in-structur' esponse spectra for the internal structures (El 616 f t) . The spectra recalculated on the basis of R.G. 1.60 envelop the total range of soil properties. Labels inoicate equipment damping ratios.
l 10 , , , , , , , , l _ Reactor building - Internal struct.El 616 ft Upper soil Median soil .
- Lower soil Tn w
C
.9 '
3 1 3 - g - u - 4 - U v 1 0,1 i . . . . . . . ! . . . . . ...I . . . . . . . . 0.1 1 10 0.01 Period (s) FIG. B-6. Horizontal in-structure response spectra for the internal structures (El 616 f t), based on R.G. 1.60 spectra. The spectra illustrate the variation with the different soil conditions discussed in the text. The equipment damping ratio was 0.03 for all three spectra.
I 1 0 10 i - - - - . . . . . , , , , , ,,, _ is . , , , l Reactor building - l _ internal struct.El 649 ft 0.03 T0.05 0.07
~
2 8_ j Tu 1
\ _
8 U e 0.1 i i ' i i i>>l i i i > > >iI i i i i . .i. 0.01 0.1 1 10 Period (s) FIG. B-7. Horizontal in-structure response spectra for the internal structures (El 649 f t), based on R.G. 1.60 spectra. The spectra envelop the total range of soil properties. Labels indicate equipment camping ratios.
10 , , , , , Reactor building
~
[ Internal struct.El 649 ft
- Upper soil -
Median soil p L ower soil , 3 C
.9 3 1
_a> g - - g - g - w - 0.1 ' ' ' ' ' l ' ' ' ' ' l ' ' ' ' ' O.01 0.1 1 10 Period (s) FIG. B-8. Horizontal in-structure response spectra for the internal structures (El 649 f t), based on R.G. 1.60 spectra. The spectra illustrate the variation with'the different soil conditions discusseo in the text. The equipment damping ratio was 0.03 for all three spectra.
10 , , , , , , , ' '
- ,l ' ' '
'i; . , , , , ,,,
_' Reactor building _ Containment shell El 730 ft _ 0.03 0.05 - 0.07
'_E7i 8
z E 1 _e _ g _ y _ U
=
0.1 e i i > . , , , 1 , , , , , ' ' ' ' ' 0.01 o,; 3 10 Period (s) FIG. B-9. Horizontal in-structure response spectra for the containment shell (El 730 f t), based on R.G. 1.60 spectra. The spectra envelop the total range of soil properties. Labels indicate equipment camping ratios.
l 10 _
! Reactor building _
- Containment shell El 730 ft _
Upper soil
~ -
Median soil Lower soil , w 3 8_ E 1 _e g - - y - _ e g - _ 0.1 ' ' ' ' ' l ' ' ' ' ' l ' ' ' ' ' ' ' ' O.01 0.1 1 10 Period (s) FIG. B-10. Horizontal in-structure response spectra for the contairunent shell (El 730 f t), based on R.G. 1.60 spectra. The spectra illustrate the. variation with the different soil conditions discussed in the text. The equipment damping ratio was 0.03 for all three spectra.
10 _ i , , , ,
,,,j
~
_ Auxiliary building El 590 ft _
- E-W response .
r1 - Il l I R.G.1.60
~
/ 0.005 ---Original design -
fg
/f g 0.02
/j 0.05 -
3 [ ff' . l 0.03 c I 0.05 S / 'k,'. i g 0.07 2 m 1
- l \1 _
19
/ 1\ 1L __ -
)
Q \
,/ \ ____
/ /
5o . / L_____T \ _ j,
\
- _ _ _ _ _ _p/ ./ / \
_____/ \
. \ _
0.1 ' ' ' ' ' l ' ' ' ' 'l ' ' ' ' ' ' ' ' O.01 0.1 1 10 Period (s) FIG. B-ll. Comparison of E-W horizontal in-structure response spectra for the auxiliary building base slab. The spectra recalculated on tne basis of R.G. 1.60 envelop the total range of soil properties. Labels inoicate equipment damping ratios.
10 _
^
Auxiliary buildingEI 590 ft _ N-S response . r1 - I \ '
~
/ I R.G.1.60
- 0. 5 --- Original design -
/g 3
/l 1 0.05 -
II I 0.03 Tn f c j tiI 0.05
.9 \\1 0.07
% 1
- / iii -
/
ji - lli g - ggL___ -
# : ,/ \1 \
. /
, p_____ -
#' L-_ _ -
v - _jy ,
\ -
.___._/ i
\ -
0.1 ' ' ' ' ' ' 'l ' ' ' ' ' ' ' ' O.01 '.l 01 1 10 Period (s) FIG. B-12. Counparison of N-S horizontal in-structure response spectra for the auxiliary building base slao. Tne spectra recalculated on the basis of R.G.1.60 envelop the total range of soil properties. Labels indicate equipment damping ratios.
10 , l Auxiliary building El 610 ft
-- E-W response _-
2 e
.o' E 1
_o_ _ o 8 -
~
Z _ 0.03 _ y. 0.1 ' ' * - > >>if i . . . . ,,,1 , , , , , ,,, 0.01 0.1 1 jo Period (s) FIG. B-13. East-west horizontal in-structure ' response spectrum for the auxiliary building (El 610 ft), oased on R.G. l.60 spectra. The spectrum envelops the total range-of soil properties. The equipment camping ratio is 0.03.
10 ,
. . . . . ...g - i - - -
'>>i * ' ' ' '
Auxiliary building El 610 ft I N-S response - t
^
w cn C
.9 lis 1 t . -
8 . -
} 0.03 _
g - - 1
' ' ' ' ' I ' ' ' ' ' ' ' ' ' ' '
0'O.01 0.1 1 10 Period (s) FIG. B-14. North-south horizontal in-structure response spectrum for the auxiliary building (El 610 f t), based on R.G. 1.60 spectra. The spectrtan envelops the total range of soil properties. The equipment damping ratio is 0.03.
10 ,
, . . . . .1 . . . . . .; _
~
[ Auxiliary building El 640 ft g _ E-W response gg .
- I \
0 l 1 7 1 0.02 005 ~ I R.G.1.60 0.05 j,,n 1 i
--- Original design
~
f l 0\ 1
/ 't 'il s // iii e // i I \ 0.03 S J 1
7 g 0.05 T N T -
// s 0.07 -
O N
// g
/ s _
/ N N
/ '
/ ] -
/ 1 p..
.__.3 s .
_ _ - - - _ - _f,
\ r s _
u_ _ __ _ _ ss \
\ _
0.1 ' ' ' ' ' ' ' ' ' ' ' l ' ' ' ' '
.e.01 ' '01.I 1 10 Period (s)
FIG. B-15. Comparison of E-W horizontal in-structure response spectra for the auxiliary building (El e40 f t) . The spectra recalculated on the basis of R.G. 1.60 envelop the total range of soil properties. Labels inoicate equipment damping ratios.
l 10 , , , , , , , ,
~
[ Auxiliary building El 640 ft _
- E-W response _
i l - - l Upper soil 3 Median soil c s Lower soil 2 .1 - - 3 - - g - - o
. ~
e _ 0.1 ' ' ' ' ' ' ' ' l i i ' ' 'l ' ' ' ' ' O.01 0.1 1 10 Period (s) FIG. B-16. East-west horizontal in-structure response spectra for the auxiliary building (El 640 f t), based on R.G. 1.60 spectra. The spectra illustrate the variation with the different soil conditions discusseo in the text. The equipnent damping ratio was 0.03 for all three spectra.
10 ,
,,g
~
Auxiliary building El 640 ft - [ N-S response O [ ll I \ -
. I jr 1 f 005 0 0.02 R.G.1.60
- gi 0.05 ji , --- Original design -
T gl
/ il
/ / \' 'd 3 /// ,3 e // i 8 // \
S 1
// I -'
0.03 s -
// s -
g - ff s 0.05 -
\ 0.07
\g s -
w
\
E -
, s -
- N
- v __- 3 s _
=======- t. _ _ __ _ _4 ' -
s\
\\
s g - 0.1 ' ' ' ' ' 'l ' ' ' ' ' ' l ' ' ' ' ' ' ' ' O.01 0.1 1 1A Period (s) FIG. B-17. Comparison of N-S horizontal in-structure respons e spectra for the auxiliary build ng (El 640 ft). The spectra recalculated on the basis of R.G. 1.60 envelop the total range of soil pr oerties. Labels indicate equipment damping ratios. i l
l i 10 _ [ Auxiliary building El 640 ft - N S response Upper soil S Median soil o Lower soil c 2 1 8 o g sj - . 0.1 i i > > i. I i i . . . ...I , , , , ,,, 0.01 0.1 1 10 Period (s) FIG. B-18. North-south horizontal ita-structure response spectra for the auxiliary building (El 640 f t), Dased on R.G. 1.60 spectra. The spectra illustrate the variation with the different soil conaltions discussed in the text. The equipment damping ratio was 0.03 for all three spectra.
~
10 . . . . . . . , . . .. . . ...g . . . . . .>> Auxiliary building El 640 ft _ _ Vertical response - l t- - Tn 8_ 1 E. _ _ 3 - 0.03 - 0.05 - 0.07
}
0.1 ' ' ' ' ' I ' ' ' ' ' l ' ' ' ' ' O.01 0.1 1 10 Period (s) FIG. B-19. Vertical in-structure response spectra for the auxiliary building (El 640 f t), based on R.G. 1.60 spectra. The spectra envelop the total range of soil properties. Labels indicate equipment damping ratios. These spectra are typical for all elevations in the auxiliary building.
10 , , , , ,
~
Auxiliary building El 640 ft - [ Vertical response - Upper soil ~ M En f edian soil c Lower soit
~
y'1 - , 7 - g - E e
// $ 0.1 - ' ' ' ' ' l ' ' ' ' ' l ' ' ' ' '
y 0.01 0.1 1 10 i 5 Period (s) 5 3 5 FIG. B-20. Vertical in-structure response spectra for the auxiliary building (El 640 f t), based on f R.G. 1.60 spectra. Tne spectra illustrate. the variation with the different soil conditions discussed in j the text. The equipnent damping ratio was 0.03 for all three spectra. The spectra are typical for all 3 elevations in the auxiliary building. 5 h 5 7
N.3 acRu 335 " (7 77) U.S. NUCLEAR REGULATORY COMMISSION NUREG/CR-1833 BIBLIOGRAPHIC DATA SHEET UCRL-53015 4 TITLE AND SUBTtTLE (Add Volume No.. of apprepnasal 2. (Leave blushi Seismic Review of the Palisades Nuclear Power Plant Unit 1 as Part of the Systematic Evaluation Program 3. RECIPIENT'S ACCESSION NO.
- 7. AUTHORISI 5. DATE REPORT COMPLETED M ON T H l YEAR T. A. Nelson, R. C. Murray, D. A. Wesley, J. D. Stevenson December 1980
[ PERFORMING ORGANIZATION NAME AND MAILING ADDRESS (include 240 Codel DATE REPORT ISSUED MONTH l YEAR Lawrence Livermore National Laboratory January 1981 P. O. Box 808 s to ,v, y ,n , Livermore, California 94550 8 (Leave warski
- 12. SPONSORING ORGANIZATION N AME AND M AILING ADDRESS (lactude 2,p Codel Office of Nuclear Reactor Regulation 10 PROJE CT,TASKlWORK UNIT NO.
U. S. Nuclear Regulatory Commission Washington, D.C. 20555 11 CONTRACT NO Fin Nos. A-0233 & A-0415
- 13. TYPE OF REPORT PE RIOD COVE RE D (l'icig5,ve daars/
Technical NA ,
- 15. SUPPLEMENTARY NOTES 14 ILeaw otm*/
hla
- 16. ABSTR ACT 000 words or less)
A limited seismic reassessment of Unit 1 of the Palisades Nuclear Power Plant was performed by the Lawrence Livermore National Laboratory for the U. S. Nuclear Regulatory Commission as part of the Systematic Evaluation Program. The reassessment focused generally on the reactor coolant pressure boundary and on those systems and components necessary to shut down the reactor safety and to maintain it in a safe shutdown condition following a postulated earthquake characterized by a peak horizontal ground acceleration of 0.29 Unlike a comprehensive design analysis, the reassessment was limited to structures and components deemed representative of generic classes. Conclusions and recomendations about the ability of selected structures and equipment to withstand the postulated earthquake are presented.
- 17. KE Y WORDS ANC DOCUMENT AN ALYSIS 1 74 DE SC RIP TO RS 17b. IDENTIMERS OPEN-ENDED TERUS
- 18. AVAILAP.LITY ST ATEMENT 19 SE CURITY CLASS (Tn,s reporre 21 NO OF PA;d 5 UNCLASSIFIED UNLIMITED 2 g G Q f Q SITmspost 22 PRICE NIC FORM 335 67-77) t
_ - _ _ _ _ _ _ _ _ _ _ . _ _ _ - _ _ . _ - - _ - -}}