ML17275B162

Nonproprietary Version of Chugging Loads-Revised Definition & Application Methodology for Mark II Containments (Based on 4TCO Test Results).

ML17275B162
Person / Time
Site:	Columbia
Issue date:	07/21/1981
From:	Bedrosian B, Ettouney M, Verderber J BURNS & ROE CO.
To:
Shared Package
ML17275B161	List: ML17275B160 ML17275B162
References
NUDOCS 8108040527
Download: ML17275B162 (286)

Text

Chugging Loads-Revised Definition and Application Methodology for Mark I I Containments (Based on 4TCO Test Results)TECHNICAL REPORT (5 (5 Q 0 flQ (5 9 (5 p 99 (5 0 p Bf 0 (5-'.'>>(5(pi>>'.'f08040527 810724 PDR ADOCK 05000397 PDR Burns and Roe, Inc.Engineers 8 Constructors Woodbury, New York

~~II LIST OF FIGURES FIGURE NO.5-4 5-5 5-6 5-7a 5-7b 5-7c 5-7d 5-7e 5-8a 5-8b 5-8c 5-8d 5-8e 5-9 DESCRIPTION Fluid-Structure Boundary-WNP-2 Suppression Pool Reactor Building Model Envelopes of Calculated Responses for WNP-2, and Measured Responses at JAERI-Containment at Vent Exit Elevation Reactor Building Response-Asymmetric Loading: Containment Vessel at Mat Reactor Building Response-Asymmetric Loading: RPV Support Reactor Building Response-Asymmetric Loading: Containment Vessel at Stabilizer Truss Level Reactor Building Response-Asymmetric Zoading: Containment Vessel at.Mid-Submergence Depth Reactor Building Response-Asymmetric Loading: Outside Building Wall Elevation 521'eactor Building Response-Nearly Symmetric Zoading: Containment Vessel at Mat Reactor Building Response-Nearly Symmetric Loading: RPV Support Reactor Building Response-Nearly Symmetric Loading: Containment Vessel at Stabilizer Truss Level Reactor Building Response-Nearly Symmetric Loading: Containment Vessel at Mid-Submergence Depth Reactor Building Response-Nearly Symmetric Zoading: Outside Building Wall Elevation 521'ibration Tolerance Observations PAGE NO.126 127 128'i 29 130 131 132.133 134 135 136 137 138 139 I

Su~arur Tests were conducted during 1975/76 by General Electric Company (GE)in their 4T test facility for the domestic Mark ZZ utilities for the purpose of evaluating the containment pool dynamic effects resulting from a postulated loss-of-coolant accident (LOCA).Based on chugging data recorded during these tests, an empirical load definition was deve-.loped This load definition was based on direct application of pressure traces measured on the boundary of the 4T test I facility to the wetted perimeter of Mark ZZ containments and, as a result, could not account for differences between the 4T test f acility and the Mark ZZ containments with respect.to vent length (vent acoustics), suppression.

pool geometry (pool acoustics) and flexibility of suppression pool structural boundaries.

Zn order to account for these differences, it became necessary'to develop a chugging load definition at the"source" (i.e., at vent exit).Such an improved.chugging load definition was'eveloped together with the application methodology for Mark ZZ contain-'ents for specific application to Washington PubLic Power Supply: System-Nuclear Project No.2 (WNP-2).This defini-tion was based on conclusions reached after evaluation of tests conducted to assess effects of steam-condensation pheno-mena in Mark ZZ type (over/under) pressure suppression systems.Two main conclusions from these tests were:.

I I I a)chugging effects are mainly due to the sudden (impulsive) collapse of the steam-water interface which occurs near the vent exit during the chugging regime and, in view of this, chugging could be repre-sented by an impulsive load applied there;and,"b)bulk fluid motions during chugging being relatively small, a linear formulation (small displacements/

velocities) is adequate for predicting the dynamic pressures induced in the far field (away from vent exit)and the dynamic response of the pool boundary structures.

A single vent design load specification was derived to bound, statistically, the 4T test data supplied by GE as representative of Hark II conditions during LOCA.The appli-cation methodology for NNP-2 containment was also developed, properly accounting for all important plant specific parameters:

length of downcomer vents (vent acoustics), 3-D multi-vent suppression pool geometry with a sloped bottom (pool acoustics) and the flexibility of the suppression pool structural boundary.Two loading conditions were developed for, and considered'n the design of, the multi-vent configuration of WNP-2: a nearly symmetrical loading and an asymmetric loading.

I Additional condensation tests were performed during 1979-1980 by GE for the U.S.Mark ZZ Owners Group, in a modified configuration of the 4T test facility, known as the."4TCO" test facility.Selected and" conservatively represen-tative (most severe)4TCO chugging data supplied by GE were evaluated/analyzed with the objectives:

a)to examine in light of the 4TCO data, the adequacy of the existing improved chugging load definition; and, b)to revise, where necessary, this (improved) load definition and the application methodology for the Mark lX containment of NNP-2.Analysis of the 4TCO chugging data, as well as of the chugging data which became available from other tests during the same time period, resulted in the following main findings: a)the impulsive nature of chugging (sudden collapse of the steam-water interface) was confirmed; b)it was determined that the 4TCO data included some stronger/larger amplitude chugs which exhibited characteristics (frequency content, spatial dis-tribution) different from those of the 4T chugs;c)the random nature of chugging was confirmed; I I I d)the strength/amplitude of chugging, although random, appeared to be dependent on system conditions, i.e., stronger chugs appeared to cluster within limited time windows corresponding to specific system con-ditions.As a consequence, the following revisions to the single vent design load specification were implemented:

the"source" load was defined as an impulsive pressure gradient (acceleration) applied over the steam-water interface at vent exit;this resulted in better matching of the characteristics exhibited by the stronger 4TCO chugs;to account for the random nature of the chug strength/amplitude each strongest ("key")chug was averaged (in terms of Fourier amplitude spectrum)with the largest neighboring

("companion")

chug to obtain an"average" or"mean" chug for each time win-dow for which 4TCO chugging data were supplied.Zt is significant to note that the single vent design"source" load developed for WNP-2 in fact envelopes the 4TCO data at almost all locations of the 4TCO tank wetted boundary and throughout the frequency range of interest;it also enve-lopes the 4T data.

I I I The two loading conditions originally developed for'NP-2 were basically retained (a nearly symmetrical loading and an asymmetric loading)in a manner compatible with the revised single vent design load specification.

In ordex to account for physical realities observed during steam-condensation tests in multi-vent configuxations (JAERI, CREARE), vent-desynchronization is specified for both these loading con-ditions adopting the approach used in the Long Term Improved Generic Chugging Load Definition developed by GE for Hark II Owners Group, in a-manner compatible with the two loading con-ditions for WNP-2.To verify the adequacy of chug strength averaging and of vent desynchronization, the dynamic pressures calculated on the wetted wetwell wall of WNP-2 were compared with wall pressures recorded during large-amplitude chugs in the 7-vent full scale tests conducted by the Japan Atomic Energy Research Institute (JAERI)in a test facility representative of the Mark ZZ geometry.The calculated pressures were found to bound the JAERI data.

I I I I 1.0 Introduction and Back round The original chuggingload def inition was developed using chugging data recorded during the 4T tests conducted by General Electric Company (GE)for the domestic Mark II utili-ties in the 4T (single vent/unit cell)test facility during late 1975 and early 1976,[1].This load definition was based on direct application of pressure traces measured on the boun-dary of the 4T test facility to the wetted perimeter of Hark II containments,[2].It soon became apparent that this method of application of 4T data to Mark II containments did not account for differences between the 4T test facility and the Mark II containments with respect to vent length (vent acoustics),.

single vent versus multi-vent suppression pool geometry and flexibility of suppression pool structural boun-daries.In order to account for these differences, it became necessary to develop a chugging load definition at the"source", i.,e., at vent exits.Because of schedule constraints, such an improved chugging load definition was developed, together with the application methodology to Hark II containments, for specific application to Washington Public Power Supply System-Nuclear Project No.design load the 4T test containment 2 (WNP2)during 1978/79, f 3, 4].A single vent specif ication was derived to bound, statistically, data supplied as being representative of Mark II conditions expected during a postulated ZOCA.

I I I l I I I I The single vent design load was defined as a pressure source at the vent exit and since its definition was indepen-dent of the 4T test facility characteristics which were dif-'erent from those of WNP-2 (vent length, suppression pool geometry and flexibility of suppression pool structural boundary)it was assumed to be directly transferable to vent exits in the WNP-2 containment.

Two loading conditions were developed and considered in the design of WNP 2: a nearly symmetrical loading and an asymmetric loading.The applica-tion methodology for WNP-2 containment accounted for the plant specific parameters governing the response: length of down-comer vents, 3-D multi-vent suppression pool geometry with a sloped bottom, and the flexibility of suppression pool struc-tural boundary (steel containment, the concrete pedestal and the foundation mat).Zn 1979-1980, additional condensation tests were per-formed by GE for the U.S.Nark XX Owners Group in a modified configuration of the 4T test facility, known as the"4TCO" test facility (5].The original, 4T test facility included a drywell located adjacent to the wetwell, a confi'guration which required a vent with three bends and a total length of about 90 feet.Zn the 4TCO facility, the drywell vessel was mounted

above the wetwell to represent'he over/under pressure suppression configuration with straight vertical vent, approximately 45'ong, representative of Mark ZZ plants.Although the 4TCO tests were planned and performed with the objective of gathering test data to be used for confirmation of the DFFR Condensation Oscillation (C.O.)load definition, the data were recorded for the entire transient including chugging, thus providing an additional data base for chugging as well.Selected chugging data obtained from regions of the 4TCO tests during which the most severe chugging effects were recorded were made available by General Electric Company as being conservatively representative for Mark lZ plants during the chugging regime.The 4TCO chugging data supplied, (6], are evaluated and results and conclusions piesented in this report.The conclusions of this evaluation together with the conclusions reached following the evaluation of multi-vent test data by GE and presented in Reference 7 report (regarding the random nature of chug strength and chug initiation time from vent-to-vent during a pool chug in a multi-vent configuration) are used in this report: (a)to examine in light of the 4TCO data, the ade-quacy of the (improved) chugging load definition developed previously, using 4T data, for applica-tion to Vii1P-2[3g 4]I and (b)to revise, where necessary, this (improved) chugging load definition and the application methodology for the Mark II containment of WNP-2.The 4TCO chugging data supplied by General Electric and the multi-vent tests evaluated by General Electric are identified in Chapter 2..The evaluation of 4TCO chugging data including the analytical studies performed in the process of data evaluation and the characteristics of single vent 4TCO chugs derived from data evaluation/analysis are described in Chapter 3.The eva-luation shows that revision in the improved chugging load definition is necessary.

The revised single vent load definition and the revised application methodology for the Mark II containment of WNP-2 based on theconclusions reached following the eva-luation of'TCO test data (presented in Chapter 3)and on the conclusions reached from the evaluation of multi-vent:

test data (presented by General Electric in Reference 7)are pre-sented in Chapter 4.The results of application of the revised chugging load definition to the WNP-2 plant (i.e., reactor building/containment structure responses) are presented in Chapter 5.

2.0 The New Chu in Data 2..1 The Sin le Vent 4TCO Test Data The 4TCO test facility, test variables, test matrix, test instrumentation and test results are described in detail in Reference 5.The-test facility is shown in Figure 2-1.The wetwell pressure transducer locations are shown in Figure 2 2~The pressure time histories recorded at the bottom center (channe1 28)were scanned by General Electric Company to identify significant chugs.Two hundred ninety-seven chugs were identified (See Table 4-2 of Reference 7).*Table 2-1 provides a summary of the 4TCO chug data compiled from infor-mation provided by General Electric[8]in November 1980.Table 2-1 identifies seven regions from six tests which recorded the largest chugs (based on the bottom center pressure (BCP)mean square power (msp)and peak over pressure*information from Reference 8 is provided in Tables 2-1 and 2-2 of this report for identification of chug numbers and time window numbers used in this report and to establish their correspondence with information subsequently published in Reference 7.

I 1 I (POP))of all 4TCO tests.General Electric Company selected 7 key chugs (one for each of the seven regions)because the power" spectral density (PSD)envelope of these chugs closely approximated the PSD envelope of the entire sample of chugs[7, 8].Several chugs (called neighboring chugs)occurring before or after each of the seven key chugs together with the key chug define a region, or a time window making seven regions[8].The region numbers and the number of chugs in each region are also identified in Table 2-1., A total of 35 chugs in seven regions or time windows were selected by General Electric Company as the chugging data base[8].The 4TCO chugging data base identification parameters are shown in Table 2-2.The 4TCO bottom center pressure time histories for the thirty-five.chugs were recorded on magnetic tapes at.4939 millisecond interval and supplied to Burns and Roe[9].In addition, the data from all the 28 replay channels from all tests were digitized at 1 millisecond interval and supplied on magnetic tapes to the Mark II Owners Group[6].The data for the 35 chugs of the chugging data base were obtained from these tapes for the evaluation presented in Chapter 3.2.2 Multi-vent Test Data Multi-vent test data from two test programs (the CREARE subscale tests and the JAERI full scale tests)have I

recently become available.

General Electric Company has eva-luated these multi-vent test data for the Mark II Owners Group and has incorporated the significant findings of multi-vent effects (the random nature of chug strength and chug ini-tiation time from vent-to-vent during a pool chug in a multi-vent configuration) in the Generic Chugging Load Definition Report[7].The results of these data evaluations and conclu-sions reached will also be adopted for the chugging load defi-nition for WNP-2..

BURNS AND ROE g.lNC~PROPRZETARY 3.0 4TCO Chu in Data Evaluation and Anal tical Studies 3.,1 Xntroduction 3.2 4TCO Chu in Data Evaluation 3.2.1 Waveform Characteristics of Boundary Pressures BVRNS AND ROE I INC~PROPRXETARY BURNS AND ROE, lHC., PROPRIETARY 3.2.2 Spatial Dist:ribuhion of Boundary Pressures BURNS AND ROEI XNC.PROPRXETARY I)

CHUGGING LOADS-REVISED DEFXNITION AND APPLICATION METHODOLOGY FOR MARK II CONTAINMENTS (based on 4TCO test results)TECHNICAL REPORT Prepared By BURNS AND ROE'NC.for'pplication to WASHINGTON PUBLIC POWER SUPPLY SYSTEM NUCLEAR PROJECT NO.2 Prepared By: M.M.Ettouney Senior Civil Engineer Approved By: B.Bedrosian l Assistant Chief Civil Engineer F.J Patti Chief Nuclear Engineer Submitted By: J Verderber ect Engineering Manager!Date:

I I 4'(I BURNS AND ROE g INC.DISCLAIMER OF RESPONSIBILITY Neither the Burns and Roe, Xnc.nor its affiliates or related entities nor any of the contributors to this document make any warranty or representation (expressed or implied)with respect to the accuracy, completeness, or usefullness of'he information contained in this document, or that the use of such information may not infringe privately owned rights;nor do they assume any reponsibility for liability or damage of any kind which may result from the use of any of the infor-mation contained in this document., 1-BURNS AND ROE, INC.PROPRIETARv NOTICE This document contains proprietary information of Burns and Roe, Inc.and it is not to be reproduced or furnished to third parties nor the information contained therein utilized, in whole or in part, without the prior express written per-mission of Burns and Roe, Inc.,

I I GENERAL ELECTRIC COMPANY DISCLAIMER OF RESPONSIBILITY This document contains proprietary information of General Electric Company pursuant to contracts with certain utilities owning plants utilizing Mark ZI Containments.

Except as otherwise provided in such contracts, the General Electric Company does not: A.Make any warranty or representation, expressed or implied, with respect to the accuracy, completeness, or usefulness of the proprietary information con-tained in this document, or that the use of any proprietary information disclosed in this document may not infringe privately owned rights;B.Assume any responsibility for liability or damage which may result from the use of any proprietary information disclosed in this document;or C.Imply that a plant designed in accordance with the proprietary information found in this document will be licensed by the United States Nuclear-Regulatory Commission or that it will comply with Federal, State or Local regulations.

I I I I I GENERAZ EZECTRXC COMPANY PROPRIETARY NOTICE This document contains proprietary information of the General Electric Company and it is not to be reproduced or furnished to third parties nor the information contained therein uti-lized, in whole.or in part, without the prior express written permission of General Electric Company.

I I TABLE OF CONTENTS Item Disclaimers Proprietary Notices Table of Contents, List-of Tables Zist;of Figures S'ummary 1.0 Introduction and Background 2.0 The New Chugging Data 2.1 The Single Vent 4TCO Test Data 2.2 Multi-vent Test Data 3.0 4TCO Chugging Data Evaluation and Analytical Studies Paca e.No.V111 10 10 11 13 3.1 Xntroduction 3.2 4TCO Chugging Data Evaluation 3.,2.1 Naveform Characteristics of Boundary Pressures 3.2.2 Spatial Distribution of Boundary Pressures 3.2.3 Summary of" Characteristics of the 4TCO Chugs 3.3 Analytical Studies and Correlation with Test Data 13 13 13 15 19 3.3.1 3.3.2 3.3.3 Finite Element Model of the 4TCO System Response Sensitivity to Source Parameters and Correlation with Test Data Response Sensitivity to System Parameters and Correlation with Test Data 21 22 l I I TABLE OF CONTENTS Item Pacae Na.3.4 Conclusions 4.0 Revised Chugging Load Definition 4.1 Introduction 33 35 35 4.2 4.3 Summary Review of the (Improved)

Chugging Load Definition Based on 4T Test Data Revisions Reauired in the (Improved)

Chugging Load Definition to Account for the New Chugging Data 4.3.1 Revision Zn Source Impulse Based on 4TCO Data 36 39 39 4.3.2 Revision Zn Source Strength Based on 40 4TCO and Multi-vent Data 4.3.3 Revision Zn Application Methodology For Mark ZI Containments Based on Multi-vent Test Data 4.4 Single Vent.Design Zoad Specification 4.,4..1 Required Average Spectrum 4.,4.2 Design Impulsive Sources 4.4.3 Summary of Single Vent Design Load Specification 43 43 43 47 4.5'4~6 Application of Single Vent Load Specification 48 to Multi-vent Mark IZ Containments 4.5.1 Spatial Variation of Chug Strengths 48 4.5.2 Desynchronization of Chugs Summary of Comperative Review Between'NNP-2'1 and Mark ZI Generic Chugging Load Definitions 4.6.1 Computational Methodologies 51 I I TABLE OP CONTENTS Item~pa e No.4.6.2 WNP-2 Plant Unique Characteristics 4.6..3 Application Methodologies for Mark II Containments 5.0 WNP-2 Reactor Building Response to Chugging Loads 5.1 Introduction 5.2 Theoretical

Background

5.2.1 Treatment of Multiple Vents 5..3 WNP-2 Response to Chugg ing Loads 5.3.1 Containment Wall Design Pressures and Comparison with Test Data 5.3.2 Structural Response 5.4 Discussion of Calculated Structural Response to Chugging Loads 6.0 List of References 53 55 57 57 60 63 63 66 Tables Figures Appendix A-Analogy Between Acoustic and Structural Boundary Conditions Al-A6 I I jIST OF TABLES TABIE NO.DESCRIPTION PAGE NO.2-1 Summary of.4TCO Chug Data 70 2-2 4TCO Chugging Data Base Identif ication Parameters 71'-1 Identification of the Companion Chug Used for Averaging with Key Chug 72 4-2 Single Vent Design Source Definition 73 5-1 Chug Start Times for Random Phasing 7'4 5-2 Maximum Rigid Hall Pressures on WNP-2 Containment at Vent Exit Elevation (Node 15)75 5-3 4TCO Pressure Maximums and Average at Channel 20 76 5-4 JAERI Peak Positive Chugging Pressure Amplitudes 77 5-5 Maximum Computed Accelerations for NNP-2 Reactor Building 78 I

LIST OF FIGURES FIGURE NO.DESCRIPTION PAGE NO.2-1 Test Configuration for Nark El Condensation Oscillation (4TCO)Tests 79 2-2 4TCO Tests-Wetwell Pressure Transducer Locations 80 3-1 Comparison of Pressures and~Fourier Amplitude Spectra of Key Chug and A Neighboring Chug-Time Window No.1 81 3-2 Comparison of Pressures and Fourier Amplitude Spectra of Key Chug and A Neighboring Chug-Time Window No.2 82 3-3 Comparison of Pressures and Fourier Amplitude Spectra of Key Chug and A Neighboring Chug-Time.Window No.3 83 3-4 Comparison of Pressures and Fourier Amplitude Spectra of Key Chug and A Neighboring Chug-Time Window No., 4 84 3-5 3-6 Comparison of Pressures and Fourier Amplitude Spectra of'Key Chug and A Neighboring Chug-Time Window No.5 Comparison of Pressures and Fourier Amplitude Spectra of Key Chug and A Neighboring Chug-Time Window No..6 85 86 3-7 Comparison of Pressures and Fourier Amplitude Spectra of Key Chug and A Neighboring Chug-Time Window No.7 87 3-8 Comparison of Pressures Measured at Channel 28 and Channel 26 During Time Window No.6, Chug 52 88 I

LIST OF FIGURES FIGURE NO.DESCRIPTION PAGE NO-3-9 3-10 Comparison of Pressures Measured at Channel 24 and Channel 20 During Time Window No.6, Chug 52 Phase Relationship Between Pressures Measured at Channel 20 and Channel 28 Versus Frequency-Time Window No.1 89 90 3-11 Phase Relationship Between Pressures Measured at Channel 20 and Channel 28 Versus Frequency-Time Window No.,'l 91 3-12 3-13 Ratios of Fourier Amplitudes of Pressures Measured at Channel 28/Channel 20-Vertical Distribution of Peak Pressures-Six Chugs, Time Window No.1 92 93 3-14 3-15 3-16 Vertical Distribution of Peak Fourier Amplitudes of Pressures-Two Chugs-Time Window No.1 Ratio of Fourier Amplitudes of Pressures Measured at Channel 20 and Channel 21 Comparison of 4TCO and 4T Data-Pressures Measured at Bottom Center 94 95 96 3-17a 3-17b Vent-Pool Model (Fluid Elements)Structural Finite Element Model of 4TCO Tank 97 98 3-18 3-19 Schematic Presentation of Pressure Source at Vent Exit in 4TCO System Fourier Amplitude Spectrum of Pressure Calculated at Channel 28 with Pressure Source at Vent Exit 99 100 LIST OF FIGURES FIGURE NO.3-20 3-21 DESCRIPTION Schematic Presentation of Acceleration Source at Three Locations in 4TCO System Fourier Amplitude Spectrum of Pressure Calculated at PAGE NO.101 102 3-22 Fourier Amplitude Spectrum of Pressure Calculated at Channel 28 with Acceleration Source Located 6'bove Bottom 103 3-23 Fo uri er Amp 1 i tude Spectrum o f Pressure Calculated at Channel 28 with Acceleration Souxce Located 3'bove Bottom 104 3-24 3-25 Comparison of Vertical Distxibution of Normalized Maximum Pxessure Calculated with Pressure and Acceleration Sources Vertical Distribution of Fourier Amplitudes of Pressures Calculated with Acceleration Sour'ce at Vent Exit 105 106 3-26 3-27 3-28 Fourier Amplitude Spectrum of Px'essure at Channel 28 with 1418 fps, Decreased from 1600 fps, Figure 3-21 Fourier Amplitude Spectrum of Pressure at Channel 28 with 2400 fps, Decreased from 4800 fps, Figure 3-21 Phase Relationship Between Pxessures Calculated at Channel 20 and Channel 28 Versus Frequency 107 108 109 3-29 Cw Versus Resonant Frequency'-

Analytical Curve 6 Its Applz.cation 110 4 I.I ZIST OF FIGURES FIGURE NO.DESCRIPTION PAGE NO 4-1 Comparison of Fourier Spectra of Pressures of Key Chug and Companion Chug Measured at Channel 28-Time Window No.2 4-2 Comparison of Fourier Spectra of Pressures of Key Chug and Companion Chug Measured at Channel 28-'ime Window No.3 112 4-3 4-4 4-7 4-8 4-9 Design Spectrum and Required Average Spectrum-Channel 28 Design Spectrum and Required Average Spectrum-Channel 26 Design Spectrum and Required Average Spectrum-Channel 24 Design Spectrum and Required Average Spectrum-Channel 20 Design Spectrum and Required Envelope Spectrum-Channel 28 Design Spectrum and Required Envelope Spectrum-Channel 26 Design Spectrum and Required Envelope Spectrum-Channel 24 113 114 115 116 117 118 119 4-10 Design Spectrum and Required Envelope 120 Spectrum-Channel 20 4-11 4-12 5-1 5-2 Source Strength Distribution-Asymmetric Zoading Case Source Strength Distribution-Nearly Symmetric Loading Case General Corss-Section of WNP-2 Reactor Building Wetwell Plan View at Elevation of Downcomer Exits 121 122 123 5-3 Finite Element Model of WNP-2 Suppression Pool with a Radial Row of Three Vents 124 I I BURNS AN D ROE t INC~PROPRIETARY 4 I I BURNS AND ROE, INC.PROPRIETARY BURNS AND ROE'NC.PROPRIETARY 3.2 3 Summary of Characteristics of the 4TCO Chugs Ke Chu s

BURNS AND ROE'NC.PROPRlETARY Nei hborin Chu s and the Ori inal 4T Chu s I

BURNS AND ROE I INC~PROPRIETARY 3.3 Anal tical Studies and Correlation with Test Data 3 3 1 Finite Element'odel oC the 4TCO System-2 1.-

BURNS AND ROE, INC.PROPRIETARY 3.3.Z Response Sensitivity to Source Parameters

.and.Corelation with Test Data BURNS AND ROE, ZNC.PROPRlETARY Waveform Characteristics of Boundar Pressures Pressure Source<<23>>

BURNS AND ROE, INC.PROPRXETARY Acceleration/Pressure Gradient Source avocation of Acceleration Source I I BURNS AND ROE, INC.PROPRXETARY Saatial Distribution of Boundar Pressures and Correlation with Test Data BURNS AND ROE'NC.PROPRZETARY I I I I I BURNS AND ROE'NC.PROPRIETARY Conclusion 3.3.,3 Response.Sensitivity to System Parameters and Correlation with Test Data<<27 l I I BURNS AND ROE, INC.PROPRIETARY Sensitivity oC the Response Frequencies to C, C s w l I BURNS AND ROE~INC~PROPRI ETARY

BURNS AND ROE, IN'ROPRIETARY Sensitivity of the FSI Mode Frequency to C W

BURNS AND ROE~INC~PROPRI ETARY

BURNS AND ROE, INC.PROPRIETARY Damping in the 4TCO System (D , D)s w<<32>>

BURNS AND ROE g INC~PROPRIETARY 3.4 Conclusions

<<3 3>>

I I I l BURNS AND ROE'NC.PROPRIETARY I I I I BURNS AND ROE, ENC.PROPRlETARY 4.0 Revised Chu in Zoad Definition 4.1 introduction I~l I BURNS AND ROE, ENC.PROPRXETARY 4.2 Summary Review of the (Zm x'oved)Chu in Toad Definition Based on 4T Test Data l

BURNS AND ROE, XNC..PROPRXETARY

<<3 7>>

BURNS AND ROE, INC.PROP RIETARv I

BURNS AND ROE, INC.PROPRZETARY 4.3 Definition to Account for the New Chu in Data 4.3.1 Revision in Source Impulse Based on 4TCO Data

BURNS AND ROE'NC.PROPRIETARY 4.3 2 Revision in Source Strength Based on 4TCO and Nulti-vent Data I

BURNS AND ROE g XNC.PROPRIETARY

BURNS AND ROE'NC.PROPRlETARY

BURNS AND ROE~INC..PROPRIETARY 4.3..3 Revision in Application Methodology for Mark II Containments Based on Multi-vent Test Data 4.4'in le Vent Desi n Goad Specification 4.4.1 Required Average Spectrum I I I I BURNS AND ROE, ZHC.PROPRXETARY

~44~

BURNS AND ROE'HC.PROPRXETARZ 4.4.Z Design Xmpulsive Sources I

BURNS AND ROE, INC~PROPRIETARY BURNS, AND ROE g INC.PROPRXETARY' 4.3'ummary of Single Vent Design Koad Specification BURNS AND ROE g XNC.PROPRXETARY 4.5 Application of Single Vent Zoad Specification to u x-vent i ar ontaxnments 4.5.1 Spatial Variation of Chug Strengths C Asymmetric foading Case Nearly Symmetric Zoading Case

BURNS'ND ROE, INC.PROPRIETARY 4.5.2 Desynchronization of Chugs BURNS AND ROE~lNC.PROPRXETARY BURNS AND ROE g INC.PROPRIETARY 4.6 Summar of Com arative Review Between'WNP-2'nd Nark II Generic Chu in Eoad Definitzons 4 6.1 Computational Methodologies

'WNP-2'ethodolo The.computation methodology used for the source extraction from the 4TCO data and for its application to the multi-vent WNP-2 containment are similar to those used in Reference 3.Namely, a fully coupled model representing the vent steam/suppression pool water/4TCO tank structure was used for source extraction, thus resulting in an impulsive source free of the test facility characteristics.

Similarly, the computational methodology

'for application of the"source"

design load to the multi-vent geometry of WNP-2 containment utilized a fully coupled model which directly accounted for.all important plant specific parameters:

length of downcomer vents (vent acoustics), 3-D multi-vent suppression pool geometry with a sloped bottom (pool acoustics) and the flexi-bility of the suppression pool structural boundaries.

Generic Methodolo Subsequent to the development of the above methodo-logy[3J', General Electric Company presented an improved chugging methodology

[18]and more recently the generic chugging load definition based on the 4TCO and multi-vent test data:[7].The methodology developed by GE[18,.7j is based on principles which are similar to those of, the WNP-2 methodo-logy., Zt recognizes the impulsive nature of chugging, acoustic nature of the steam response in the vent, acoustic nature of the water response in the pool, and it recognizes the need to address in the load definition the presence of vent response characteritics and of the fluid-structure interaction (PSZ)effects in the dynamic pressures measured in the test facility.However, the computational methodologies used for source extraction from test data and for its applica-tion to the multi-vent Mark ZZ containments are based on two assumptions

[181 (not required in the'WNP-2'omputation methodology):

(a)The vent is not acoustically coupled to the pool, and (b)The principal effect of the fluid-structure interaction (PSI)is to reduce the frequency of the tank ringout.With the above two assumptions, the hydrodynamic model of the test facility used for source definition is reduced to solving an acoustic.wave equation in a flat bottom axisymmetric rigid tank.To account.for the vent par-ticipation, vent harmonic (sine wave)response is added as a.forcing function to the impulsive pressure source., Prior to application of these sources to Hark II plants two modifications are required to account for the dif-ferences, if any, in the vent response characteristics and in the PSZ response characteristics between the test facility and the Hark II plant as described in References 7 and 18.The hydrodynamic model of the Mark II containment used in calculating the chugging boundary pressures due to a sour'ce consists of solution of'he acoustic wave equation in a rigid, flat bottom annular tank 4.6.2 WNP-2 Plant Unique Characteristics

There are two characteristics of the WNP-2 plant, which are, not in common with the other domestic Mark XX plants: containment shell structure built of stiffened steel plate, the sloped bottom pool geometry.The evaluation of the 4TCO data presented.in Chapter 3 shows that the boundary pressure traces from key chugs con-tain significant participation of the FSI response.-

The magnitude of its participation vary from chug to chug and i;s dependent on the system conditions Since the WNP-2 contain-h ment shell structure is built of steel (as is the 4TCO tank wall)~its response to impulsive chugging sources will include significant participation of the FSX mode (as in the 4TCO tank)which may be axisymmetric as well as non-axisymmetric..

To obtain realistic responses of the WNP-2 containment to chugging Loads, it is essential to use the methodology which directly accounts for the FSX.effects of the test facility in-the source extraction methodology (thus resulting in sources which are free of the test facility characteristics) and which directly accounts for the FSX'effects (axisymmetric as well as non-axisymmetric) in the containment response calculations.

<<54>>

As stated before<the generic methodology is appli-cable to a containment with a flat bottom pool geometry., Zts use for NNP-2 containment would require idealization of the sloped bottom floor to.a flat bottom floor.Such an idealiza-tion could be practically based on only one (the fundamental) pool acoustics mode.Since the Mark ZZ pool acoustic response to chugging loads involves participation of many=modes of vibrations of the coupled vent/pool/structure system, such an idealization would result in over-simplification of the problem., For the above reasons, the'WNP-2'evised chugging load definition is developed, implemented and presented in.this report.'4.6.3 Application Methodologies for Mark ZX Containments Elements of the two load application methodologies for Mark ZX containments are similar.Based on multi-vent test data both methodologies:

recognize random variation of chug strength from vent-to-vent and use an averaging technique (although averaging is used in'WNP-2'efini-tion, it is shown that design sources bound all unaveraged 4TCO data at Channel 28),-55>>

BURNS AND ROE p INC., PROPRlETARY recognize random variation of chug initiation times'rom vent-to-vent (the'NNP-2'ethodology conservatively assumes that the three vents in one radial.row are in-phase, see.Chapter 5).

5..0 WNP-2 Reactor Buildin Resonse To Chu in Zoad 5.1 Zntroduction The application of the chugging load methodology of the previous chapters-is presented in this chapter.The theoreti-cal background of the structural analysis is presented.

The structural and suppression pool models are discussed.

The results of the analysis, and their comparison with JAERZ test results are presented..

5.2 Theoretical Back round The analytical.

methods that.were used.in the applica-tion of this, chugging, load methodology to the NNP-2 contain-ment is similar to that of Reference 3, Section 5.1.It was shown that the total hydrodynamic pressures, P2 (~)on the fluid-structure boundary can be expressed by PZ~Pj~+Ma (fU U (A3 (5.1)where 4)~Rigid wall pressures Ma (AJ~Hydrodynamic added mass matrix~Forcing frequency~~UP)~Accelerations of the fluid-structure interface

The structural equation of motion can then be expressed (5-2)where US (Q)=Structural displacements

~Appropriate transformation matrix K~~Dynamic stiffness matrix~2 (Ms+.T Ma TT)+iQ Qs+Ks (5 3)Ms=Structural mass matrix~-Structural damping.matrix~Structural stiffness matrix For any specified case of loading, the rigid wall pressures Pi (JlJ can be obtained,.

equation 5.2 can then be solved to obtain the required'tructural displacements.

For more details, refer.to Reference 3.5.2.1 Treatment of Multiple Vents A cross sectional view of the WNP-2 reactor building is shown in Figure 5-1 and a plan view, of the wetwell at the I I I I elevation of vent exits is shown in Figure 5-2.There are 102 downcomers (18 downcomers of 28" diameter and 84 downcomers of 24"'iameter) located in thirty-four radial lines arranged in.an axisymmetric manner.A.three dimensional finite element model of the WNP-2 suppression pool that has a set of three vents in one radial row is shown in Figure 5-.3 and its structural boundary in Figure 5-4.The analysis is performed for any given chugging loading case using this model.and assuming the source loads at the three vent exits to be of a unit strength and occurring in-phase..

The structural and pool responses are evaluated then for, this set of three vents..Zet.the response measure of interest located.at angle 0 and time t from the reference radial vent row be represented by the vector X (&,t).lf the chugging load intensity at the vent exits corresponding to the ith radial row is assumed to be E i, the total building response, U (0,t), can be obtained as: U (<'t)-34 Li-X (e-ei, t y,)i=1 (5.4)where i~The angular position of the it radial row.measured from the reference row as shown in Figure 5-2.

I I BURNS AND ROBED ZiVC~PROPRIETARY Pi Random chug start time for the ith radial row, Table 5-1.The assumption that the chugs occur in-phase at the three vents in each row is more conservative than the case where all 102 vents are assumed desynchronized since the variance of the chug start times assigned to 34 radial rows is smaller than th'e variance of the chug start times assigned to 102 vents.5.3 WNP-2 Response to Chugging Loads The single vent load definition and the'assoc'iated multi-vent application.

methodology of Chapter 4 and the theoretical approaches of Section 5.2.are used to obtain the suppression pool boundary pressures and structural responses to chugging loads.The axisymmetric finite element model of the reactor building is shown in Figure 5-5.This model is a more refined version of the model used in Reference 3.It was shown that refined modeling techniques give more realistic results (Reference,11).

BURNS AND ROE'NC..PROPRIETARY 5.3 T'ontainment Wall Design Pressures and Comparison With Test Data, To obtain.the maximum poo1 boundary pressures for comparison with test data the source strengths at the WNP-2 vents were assumed equal and the sources along the 34 radial rows were assumed desynchronized with chug start times as given in Table 5-1'.The maximum rigid wa11 pressure value calculated at the containment wall at the vent exit elevation at 0'zimuth for each design source is shown in Table 5-2.I The maximum rigid wall pressure measured, at the vent exi;t elevat'ion in the 4TCO'ystem (Channel 20)during each of the seven time-windows (Table 2-2)and its average with maxi-mum pressure of the companion chug (Table 4..1)is shown in Table 5-3.

The maximum modified*pressures measured in different JAERI tests at the vent.exit elevation at the containment.

wall are shown in Table 5-4..Its comparison with the design wall pressures of Table 5-2 shows that the design wall pressures are higher than the JAERI results., To provide additional comparison with JAERI data, Fourier amplitude spectrum of the containment wall pressures at the vent exit elevation at 0'zimuth was obtained for each of the seven design sources The envelope of the seven spectra is shown in Figure 5-6 and compared with the envelope of Fourier amplitude (modified averaged)spectra of measured pressures in JAERI tests[7J.The envelope of design pressures completely envelopes.

the JAERZ data by a significant margin at all frequencies The above comparisons demonstrate that the chugging load definition including the averaging method and the desynchronization procedure used in the application methodo-logy for the Nark II containments is conservative.

The JAERI multi-vent test facility shown in Figure 3-1 of Reference 7 includes 7 vents in a 20'ector.The WNP-2 vent configuration shown in Figure 5-2 shows six vents in a 21.8'ector..

In order to account for this difference in the number of vents between the two systems, JAERI test results are divided by a factor of 1.17 based on a previous study[19], which compared peak boundary pressures resulting from in-phase equal strength source application to seven vents versus six vents in a 20'egment..

5.3.2 Structural Response Structural responses were calculated, using the axi-symmetric finite element model of the reactor building shown in Figure 5-5 subjected to the pool boundary pressures calcu-lated for each of the seven design sources Response spectra at several locations were calculated.

The-envelope spectrum curves (with the peaks spread by+15%)corresponding to 0.58, 1%,.2%and.4S of the critical damping values are plotted'or selected locations (foundation mat.at primary containment vessel, RPV pedesta3.at vessel support.elevations, containment vessel at.stabilizer.truss level, containment vessel at mid-submergence depth'nd reactor building at elevation 521'in.Figures 5-7 and 5-8.f or the asymmetric and the nearly sym-metric loading cases,.respectively 5.4 Discussion of the Ca3.culated Structural Res onse to Chu in'oads The calculated NNP-2 reactor building responses to chugging loads show a pattern similar to that of the cal'cu-l lated responses to SRV 3.oads, Reference 11..The reactor building can be'ivided again into three zones: The wetwell zone, including the containment struc-'ure boundary, where the hydrodynamic pressures are app3.ied, and where the structural responses are the largest.

ii., The drywell.zone, including the, containment struc-ture boundary, where the high responses of zone (i)have been attenuated through the RPV pedestal and the containment shell..Although smaller, they're still of a finite magnitude for the low damped steel containment structure..

iii.The third zone consists of the biological shield and the reactor building walls and floors outside pri-mary containment The structural response accelera-tions calculated for WNP-2 are negligibly small Et is noted.here.that negligibly small responses were also calculated for NNP-2.when sub j ected to SRV loads[11I'nd this.low level of predicted response was verified by'easurements taken during Caorso and Tokai-?.SRV tests.This is due to the fact that the Load path from zone (i)where the hydrodynamic pressures are applied, to this zone is through the mat and the soil and as shown in References 12 and 13 the soil compliances reduce rapidly as the fre<<fluency of excitation increases; this explains the above mentioned large reductions in the structural responses in this zone.Table 5>>5 shows the maximum computed structural re-sponse accelerations in the three zones of the NNP-2 reactor

building.Figure 5-9 shows the vibration tolerance obser-vations as documented in References 14, 15, and 16.Examina-tion of the maximum computed structural response accelerations at locations outside the containment structure of HNP-2 reac-tor building (zone iii)indicates that they fall.consistently near the curve of Figure.5-9 labeled BEGIN TO PERCEIVE".

These findings lead to the conclusion that evaluation of safety related piping and~equipment for chugging responses need only be carried out in zones (i)and (ii).It is noted that this conclusion is consistent with the results of WNP-2 reactor building analysis for SRV loads (Reference 11)..

6 0 LIST OF REFERENCES"Mark XX-Pressure Suppression Test Program Phase XI and XXI Tests," NEDO-13468,/NEDE-13468-P, including-Errata (1).2"Mark IX Containment Dynamic Forcing Function Information Report (DFFR),"'EDO-21061/

NEDE-21061-P, Rev.3,, June 1978, General Electric Company.3.-Chugging Loads-Improved Def inition and Application.

Methodology to Hark XI Containments,"'echnical Report (Proprietary), Prepared by Burns and Roe, Xnc , for Application to Washington Public Power Supply System, Nuclear-Pr'oject No.2, June 1'5g 1979.4.Non-proprietary Version of 3..4T Condensation Oscillation Test Program Final Test Report," NEDE-24811-P (Proprietary), Hay 1980, General Electric Company.6."Transmittal of Computer Tapes Containing 4T C.O.Test Data,"'eneral Electric Company Letter HKXX 1814-E, dated July 15, 1980 (Including the Two

Ref erenced Letters dated 1/2/80, 4/29/80)to Hark XX Consultants.

7."Generic Chugging Load Definition Report," General Electric.Document NEDE-24302 (Proprietary), April 1981 8."4T C.O.Chugging Data Base Information," General Electric Company Letter MKXI-1970-E (including seven Attachment), dated November 18, 1980 to Burns and Roe, Inc.L'etter dated July 24, 1980 from Creare Xnc.to Burns and Roe, Inc 10."Random Data: Analysis and Measurement Procedures,.

J..S Bendat.,-A..G., Piersol, Wiley-Interscience, 1971"SRV Loads-Improved Definition and Application Methodology for Hark II Cont:ainments," Technical Report (Proprietary)

Prepared by Burns and Roe, Inc.for Application to HPPSS WNP-2, July 29, 1980.1'2."Dynamic Response of Structures in Layered Soils,"'..

Chang-Liang, Department of Civil Engineering Report Number R74-10, Massachusetts, 1974.13"Dynamic Stiffness Functions of Strip and Rectangular Footings on Layered Media," G.Gazetas, Massachusetts Institute of Technology, Dept.of Civil Engineering, M.S.Thesis, 1975.

14."Data Averaged From 7 Sources," Report 1, D.E.Goldman, Naval Medical Research Znstitute, March 1948.15."Foundation Vibrations," R.E.Richart, Journal of ASCE, Vol.86, No SM-4, August 1960.16."Vibration of Zsolated Foundations for Boiler Feedpumps," H., A..Franklin, Presented at the 1979 Annual Convention, American Concrete Znstitute, Milwaukee, Wisconsin, March 1979.17."Xmproved Structural Analysis Methods For Prediction of, Containment

-Response to Suppression Poo1 Hydrodynamic Roads", Technical Report, Revis ion 2, Prepared.by Burns and Roe, Znc.for General-Electric Company, January 1981'I 8"Mark ZZ Zmproved Chugging Methodology", General Electric Document NEDE'4822 (Proprietary), May 1980'l9."Fukushima Dai-Ni: Unit 3, Containment Response to the LOCA Steam Condensation Loads" Technical Report (Proprietary), prepared by Burns and Roe, Xnc., for Toshiba Corporation April 1981..

f GENEBAG ELZCTRlC.COMPANY PBOPBZZTABZ TABZZ 2 SUMMA'P 4TCO CHUG CATA l l 1 I I I QEHEKLL, ELECTRIC COMPANY PROPRIETARY Table 2-2 4TCO CHUGGING DATA BASE IDENTIPICATION PAEQMETERS

-73.-

I BURNS AND ROE, INC.PROPRIETARY TABLE 4-1 IDENTIFICATION OF THE COMPANION CHUG USED FOR AVERAGING WITH KEY CHUG I I BURNS AND ROE~ZHC~PROP RZETARY TABLE 4-2 SXNGLE VENT DESZGN SOURCE DEFXNXTZON

<<73>>

~I I I TABLE 5-3.CHUG START TINES PQR RANDOM PHASING Radial Row 2, 6 10 12 13 15 16.17 Chug Start Time (Sec)0 022826 0.021631 0.014281 0.037928 0.026944 0.025257 0.0092329 0.,035325 0.036614 0.,017459 0 030120 0.018409 0.034114 ,0.041211 0'34711 0.024805 0.031562 Radial Row 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 Chug Start Time (Sec)0.012603 0.035364 0.026404 0.039039 0.021682 0 024879 0.044889 0.024589 0.0057321 0.023880 0.033905 0.015641 0.013154 0.022173 0.026808 0.046634 0.027705 Vari.ance~0.881708 x 10 (sec.)-4 2 Note: The three vents in each radial row are assumed to chug in-phase.<<74

BURNS AND ROE, INC.PROP RIETARY TABLE 5-2 i~EMUi4 RIGID WALL PRESSURES ON WNP-2 CONTAINMENT AT VENT EXIT ELEVATION (NODE 15)

I, GENERAL ELECTRIC COMPANY PROPRIETARY

Ãi Ql A z CI Vl&I X g fQ g H Z Cl OÃmE Ll lh O O

BUPSS hiVD ROE, Zi<C.P.OPRZETARY GENERAL ELECTRZC COMPANY PROPRZETARY TABLE 5 4 JAERZ PEAK POSZTZVZ CHUGGZ.IG P.RES SURE RMLZTUDES~77~

TABLE 5-5 MAXIMUM COMPUTED ACCELERATIONS FOR WNP-2 REACTOR BUILDING ZONE I Xnside and on the boundary of the containment struc-ture below El.510'OCATION Containment Wall (Quencher Elevation)

RESPONSE DIRECTION Horizontal MAXIMUM ACCELERATION (g)1.63 XX Inside and on the boundary of the containment struc-ture above El.510'ontainment Wall El.520'lorizontal RPV Support Vertical Stabilizer Truss Horizontal 0.18 0.060 0.038 IXX Outside the con-tainment structure Outside Building JIorizontal Wall El.521'ertical 0.002 0.029

VACUUM&REAKER IIL1 It ORYWELL I I 1 2 It 52.5 ft VENT SRACE 454't 4T TANK YBIT (OCWNCOMER)

V E."IT aRACS Si EAM QENERATOR Il.S It I'I.OW MEASURINQ VENTI'Rl W7LSHZNGTOH PVBLXC POWER SUPPLY SYSTEM NUCLEAR PK4ECT HQ Test Configuration for Park II Conclensation Oscillation (4TCO)Tests>>79-

C H.18&19 0'0'25 315 CH.17~-45o 20.0'H.22&'23 I~-~, 225 315 CH.21 4'5 CH.20 12.0'H.24 C H.26&27 Oc4'25 315 CH.25~'5o 00 0.0o 6.0'.0'.0'h.

28 (Bottom Center)WASHZNGTON PUBLZC POWER SUPPLY SYSTEM NUCLEAR PROJECT NO 2 4TCO Tests Wetwell Pressure Transducer Locations PZGURE 2~2 GENERAL ELECTRIC COMPANY PROPRIETARY WASHZNQTON PUBLXC POWER SUPPLY SY~COMPARISON OF PRESSURES&FOURIER PZGUBE AMPLITUDE SPECTRA OF i<EY CHUG 6 A NUCEZAR PROJECT NO 2 NEIGHBORING CHUG-TIME WINDOW NO.1

GENERAL ELECTRIC COMPANY PROPRIETARY WASHINGTON PUBLIC POWER SQPPXY SY~NUCLEAR PROJECT NO~2 COMPARISON OF PRESSURES&FOURIER AMPLITUDE SPECTRA OF KEY CHUG 6 A NEIGHBORING CHUG-TIME WINDOW NO.2 FIGURE 3>>2 GENERAL ELECTRIC COMPANY PROPRIETARY t t WASHZNGTON PUBLZC POWER SUPPXiY SYSTRC NU~PROJECT HO 2 COMPARISON OF PRESSURES&FOURIER AMPLITUDE SPECTRA OF KEY CHUG 6 A NEIGHBORING CHUG-TIME WINDOW ViO.3-83-PZGURE 3<<3

GENERAL ELECTRIC COMPANY PROPRIETARY t t WASHINGTON PUBLIC POWER SUPPLY SYSTEM%NUCXZAR PROD~HO 2 COMPARISON OF PRESSURES&FOURIER AMPLITUDE SPECTRA OF KEY CHUG S(A@NEIGHBORING CHUG-TIME WINDOW NO.4-84-PIGURE 3-4 I I I I GENERAL ELECTRIC COMPANY PROPRIETARY

~~~@PmXZC ZewZz~~zzz zz~COMPARISON OF PRESSURES&FOURIER PIQUE AMPLITUDE SPECTRA OF KEY CHUG&A 3-5 NEIGHBORING, CHUG-TIME WINDOW NO.5 GENERAL ELECTRIC COMPANY PROPRIETARY

%LSHZNQTON PUBLIC PowER SUPPLY SYSTzH COMPARISON OF PRESSURES&FOURIER AMPLITUDE SPECTRA OF KEY CHUG 6 A 3-6 NEIGHBORING CHUG-TIME WINDOW NO.6 I I I l I GENERAL ELECTRIC COMPANY PROPRIETARY WASHINGTON PUBLXC POWER SUPPLY SY~NUCZZAR PEKL7EZT HO COMPARISON OF PRESSURES&FOURIER AMPLITUDE SPECTRA OF KEY CHUG&A NEIGHBORING CHUG-TIME WINDOW NO.7-87-FIGURE 3~7 I I I I GENERAL ELECTRIC CO%?ANY PROPRIETARY WLSHMGTON pUBLzc pggzz gypsy gyp~COMPARISON OF PRESSURES MEASURED AT CHZ2iNEL 28 AND C~~EL 26 RJCLEAR PRCLTEZT NO 2 DURING TIHZ WINDOW NO.6,CHUG"'

I I GENERAL ELECTRICCOMPANY PROPRIETARY WASHZNGTON PUBLZC POWER SUPPLY SYSTEM NUCLEAR PROJECT NO 2 COMPARISON OF PRESSURES MEASURED AT CHANNEL 24 AND CHANNEL 20 DURING TIME WINDOW NO.6 I CHUG n 2 I I)

BURNS AND ROE i INC~PROPRIETARY M7LSHZNGTON PUBLZC~SUPPLY SYSTM NUCLEAR PROJECT HO 2 PHASE RELATIONSHIP BETWEEN PRESS.PZUURE MEASURED AT CHANNEL 20 AND CHANNEL 3>0 28 VERSUS FREQUENCY-TIME WINDOW NO.l l I BURNS AND ROE g ZNC~PROPRIETARY

~HZNGTON PUBLIC PowER sgPPLY SY~PHASE RELATIONSHIP BETWEEN PRESS.PIGURE MEASURED'AT CHANNEL 20 AND CHANNEL 3-gg 28 VERSUS FREQUENCY-TIME WINDOW NO.l I.I l BURNS AND ROE, INC.P ROP RIETARY~~~@~Q~$QPPT~NU~PROJE~HQ 2 RATIOS OF FOURIER AMPLITUDES OF PRESSURES MEASURED AT CHANNEL 28/CHANNEL 20 a

GENERAL ELECTRIC COMPANY PROPRIETARY WASHZNGTON PUBLZC POWER SUPPLY SYSTZ?4 i%JCLEAR PROJECT NOi 2 VERTICAL DISTRIBUTION OP PEAK PRESSURES-SIX CHUGS, TIME WINDOW No.1 PZGURE 3-l3 I I GE'FERAL ELECTRIC COMPANY PROPRIETARY

%LSHXNQTON'UBLIC POWER SUPPL" SYFZBC NUCLEAR.PR03ECT NO 2 VERTICAL DISTRIBUTION OF PEAK FOURIER AMPLITUDES OF PRESSURES TWO CHUGS-TIME WINDOW NO.1-94-PIGURE 3-14 5 l 1 BURHS ALOD ROE I XNC~PROPRZETARY WASHINGTON PUGLXC PGWER SUPPLY SYSTEM%NUCLEAR PROJECT NO 2 RATIO OF FOURZER AMPLITUDES OF PRESSURES MEASURED AT CH.20/CH.21 95>>FIGURE 3-15 l

GENERAL ELECTRIC COMPANY PROPRIETARY WASHINGTON PUBLIC POWER SUPPL SYS~COMPARISON OF 4TCO S 4T DATA PRESSURES MEASURED AT BOTTOM NUCLEAR PBCQECT NO 2 CENTER-96-FIGURE 3-16

588.0 54RO 543;0.505.0 466.0 424.0 Modified 4T Tank 382.0 328:0~276.0-262.0 216.0 195.0 180.0..144.0 127.0 108.0 60.0 72.0 p pll 36.0 24.0 0.0 12.0 18.0 26.'0 34.0 42.0 0.0" 12.0 MLSHZHGTON PUBLZC PC%ZR SUPPLY SYSTEM NUCLEAR PROJECT No~2-97-VENT-POOL MODEL (FLUlD ELEMENTS)PZGURE 3-3.7a

4~J~e~~~~~~~r~

l BURNS AND ROE, INC.P ROP RIETARY WASHXNGTON PUBLIC POWER SUPPLY SYSTRC NUCLZAR PROJECT NO 2 SCHEMATIC PRESENTATION OF PRESSURE SOURCE AT VENT'XIT IN 4TCO SYSTEN-99-PEGURE 3-18 I

ak~R n sm g ggC MNQ UI OIH C~o~9~KRC ea x a ta Q (A RAA td hJ C HQO a n Ul tO I O V)Q.II)Ul S hl IL Q.n n o n n UI 9 Hz 27 Hz~~45 Hz-63 Hz CIJ=4000/CS=-!600 hl M I-8 20.Oo IJD.QO 6D.QD 00.DQ f.REQUFHC'5 (Hl)!00.QD!20.Qo!0,00!60.QD

BURNS AND ROE, INC.PROPRIETARY WASHXHGTON PUBLIC PC5KR SUPPLY, SYSTEM NUCLEAR PROJECT NO 2 SCHEMATIC PRESENTATION OF ACCELERATION SOURCE AT THREE LOCATIONS'IN 4TCO SYSTEM

n CV 3&Hz CU=I1800/CS=1600 n CO n OgQ td M Q~Ol H HOR Q OH CII C 6 QVC Cpa PS a CII HQO<OS%AC g 0+M oo Q HH o CD ILI o Co Co LLJ K o n-&Hz-24 H I"40 Hz 50 Hz I-60 Hz I 20.OD IIO.00 GO.OD 00.00 FREQUENCY (t)Z)!00.00.'20.QO IgO.00!GO.JO

~~~5 u'4 k~O(O X Ol Q M CA H td C Cd HQ(R>V n~Vi C g cpu w a cn&OR ai C h)g Ze O o fV m CL W D M M ld K n CJ 20.00 CtJ=~BOO/CS=i600 tlO QQ 60 03 gQ.QD)00.OQ 120.03)40.OQ)GO 00 FRFQUEHCY (HZ)

A n n Ul P4 CIJ=4'000/C5=!600 QOQ AC+Cd g W R c&C H an<XC 0 M g tI7~N OOg RO RbfO 0 V K R td 0cn g\M"p CL V)e Q.4J lL M CO n n CI V)0 a a n n 0.00$0.00 tlO.00 GO.QQ 00.QD FREQUENCY (l)Z)l00.DO 120.QO l 40.QO l GO.QO

BURNS AND ROE i INC~PROPRIETARY m~~~c~mme ms~COMPARISON OF VERTICAL DISTRIBUTION zZmaz'F NORMALIZED MAX PRESS CALCULATED 3-24 W~C PSOJEC"'O.

2 i W/PRESS.&ACCELERATION SOURCES-105-

BURNS AND ROE, INC.PROPRIETARY

%LSHXNGTQN PUBLXC POWER SUPPLY SYSTEM NU~~PRGJFCT HO 2 VERT.DISTR.OF FOURIER AMPLITUDES OF PRESS.CALCULATED WITH ACCEL.SOURCE AT VENT EXIT FIGURE 3-25

n n~35 Hz I g=u800/CS=1410 cn a 0 CO W M V)QJ K a 7 Hz I~21 Hz I 40 Hz I a a a 2o.oo ilo.oo ao.00 on.on FREQUENCY (HZ)l 00.00 l'c'.n.dn l 40.DD l GD.QD

Ch+I n n (V C IJ.=2 l}00/C S.=}6 00~~UWN nRc/MAL 0 C hf lA g Q OAH R." 8 C)OgM thgK N P3 AQ C w II g Q M~AQ F05 I o M n IJj K M CO LJJ K Q.n a Cl CI n n-O trz 2$Hz 29 Hz D.OD 20.00 IlD.00 60.00 BD.00 FREQUEHC't'H2) l00.DD l20.00 l lO.00 1GO.DD

BU RNS AND ROE I ENC~P ROP R I ETARY WLSHZNGTOH PUBLZC POWER SUPPLY SYSTEM NUCLEAR PROJECT HO 2 PHASE RELAT1ONSHXP BETWEEN PRESSURES CALCULATED AT CH.20 AND CH.28 VS.FREQUENCY-109-FZGURE 3-28

BURNS AND ROE I INC~PROPRIETARY WASHZNGTON PUBLZC POWER SUPPXZ SYSTEM NUCLEAR PBOJECT NO+2 C.VS.RESONANT FREQUENCY ANALYTICAL CURVE&ITS APPL.-110-

o o Chug No.2 (Key Chug}I a CO o I+>/l'I 0~1 0 o o 20~QQ'IO.QQ 60.QO 80.QC I 00~00 I 20.00 I 40.QQ I 60.CQ FRCOUENCY (HZ)o o Chug No.3 (Companion Chug)I I2 o D CA 0 hJ o o III VI UJ IX 0 o o ZQ.CC 40.QO 60.QQ 90.QO I 00.00 I'Q.00 I 40.CQ I 60.QQ FREOUKNC'({H7.)WASHRf~~N PUBLIC PCSIER SUPPLY SYSTEMS HUCZZAR PROJI&2 HO 2 COMPARXSON OF FOURIER SPECTRA OF PRESS OF KEY CHUG&COMPANION CHUG MEASURED AT CH.2S-TIME HINDOOS NO.2-111-I I I I a a Chug No.3 (Key Chug)cs Oa V)0~a QJ a lZ a D V)Vl li!.0 00 ZO.QQ'40.QO'Q.QQ 8Q.QQ lQO.00!ZQ.QQ'.!O.QD!GD.DD FREOUENC'i (HZ)Chug No.4 (Companion Chug)a O I as O~V)Q c as VE a!i!a!Z 0 DO ZQ.CD CO.DQ GO.CQ 8D.QQ:DO.QQ!ZQ.QO'.AO.DQ!GQ.QQ FRFQUE!(C" (HZ)WLSEKNGTCH PURL):!" PO~SUPPLY SYSTRt i&JCZZAR PRC4cX.~HOa 2 COMPARZSON OF FOURIER SPECTRA OF PRESS'F KEY CHUG&COMPANZON CHUG MEASURED AT CH.28, TZME NZNDON NO.3 PZGURE 4-2-112-

BURNS AND ROE, INC.PROPRIETARY I i t WASHXNGTON PUBLIC POWER SUPPLY SYSTEM NUCZZAR PROJECT NO~2~DESIGN SPECTRUM AND REQUIRED AVERAGE SPECTRUM-CHAVv.NEL'8 PZGURE 4-3.

.BURNS AND ROE, INC.PROPRIETARY WASHZNGTON PUBLZC POWER SUPPLY SYSTEM NUCLZAR PROJECT HO~2~DESIGN SPECTRUM AND REQUIRED AVERAGE SPECTRUM-CHANNEL 26 PZGURE 4-4-114-I I I I BURNS AND ROE, INC.PROPRIETARY WASHINGTON PUBLIC POWER SUPPLY SYSTEH NUCLEAR PROJECT HO 2~DESIGN SPECTRUM AND REQUIRED AVERAGE SPECTRUM-CHANNEL 24-115-BURNS AND ROE g I NC~PROP R I ETARY WASHINGTON PUBLIC POWER SUPPLY SYSTEM NUCLEAR'RMECT NO 2~DESIGN SPECTRUM AND REQUIRED AVERAGE SPECTRUM-CHANNEL 20 FIGURE 4-6

BURNS AND ROE, INC.PROPRIETARX i t WASHINGTON PUBLIC POWER SUPPLY SYSTEH NUCLEAR PROJECT NO 2.DESIGN SPECTRUM AND REQUIRED ENVELOPE SPECTRUM-CHANNEL 28 FIGURE 4-7-3.17-I I l BURNS AND ROE I INC~PROPRIETARY WASHZNGTON PUBLZC POWER SUPPLY SYSTEM NUCZZAR PROJECT NO 2~DESIGN SPECTRUM AND REQUIRED ENVELOPE SPECTRUM-CHANNEL 2 6 PZGURE 4-8-118-BURNS AND ROE I INC~PROPRIETARY WASHINGTON PUBLZC POWER SUPPLY SYSTEM NUCLEAR PROJECT No 2.DESIGN SPECTRUM AND REQUIRED ENVELOPE SPECTRUM-CHANNEL 24 PZGURE 4-9-119-l I I BURNS AND ROE, INC.PROPRIETARY wAsHINGT0N Paar'nna svpprv scrag DESIGN SPECTRUM AND REQUIRED ENVELOPE SPECTRUM-CHANNEL 20>>120-

S c.CT:ON B-B 0 0 0 0 0 gl I I 0.Ot 2~~i~f MASHZH~iN PUBLIC PO~SUPPLY PiaTKi HQ~PMJECT HO SOURCE STRENGTH D ISTRIBUTION ASYMMETRIC LOADING CASE-121-

If j I ll ghkk i L.ii, gal.S lk'JAX SECTION B-B 0 0 0 0 0 tJl 0 Qw (U I Z 0 I-O UJ M/../9~MN WASHINGTON PUBLIC PONER SUPPLY SYSTEM NUCLEAR PROJECT NO<2.SOURCE STRENGTH DISTRIBUTION NEARLY SYMMETRIC LOADING CASE-122-FIGURE 4 12 I

le TCiN-BRlCCs K CRAAIC SRICGK~G!RCKR 5TCM~Kc¹/~Roa'v CRAblK Rl'iiW'QIR~aR.I E'CC-la~:~~~~~e.~~-,aa-Z Paal~R" AC i QR.=1'1 1 i l CR STEEL.PRlQAR~:.I HEN~V~MSK1 B)4L M&~l SHlPWALK P~cl'lg~~~~~~~+i I I~ia.w<=-1 RlRCl<al L',MNc O'P.YYZ-i'l I"-SHlc.~, Ylkl f SUP..=~~14N baal I': WA, L-VE'i p~~~g&~Q'I I='~a~SLlPPP+!I.I~~~~~<I.I cc cvvc~~~~'.'.HO.HULK~'uQ~~i COL.Uhl H 5~~i4R PR~ws~l l FQQslC~7lOH i%As~~'~V'~0 l hl&LE'R:81R>~~~~0~~~~~~~~~~4 r~~~WLSHZHGXOH PUBLIC PCWER SUPPLY SYS~HUCLZAR PROJECT HO~2 GEaKRAL CROSS-SECTXON OP WNP-2 REACTOR BUZLDXNG-123-I LQ 2 GCha~CCa'-

25=QM 5~C=D l2.064 I37.65 4 I05.88o>>29 II 47o'8 90.04 4~'4I2'3.534 52.94o 42.35 48.24o 3I.76 o l58.824 l69.4I 2I.I 8 o I0994 I800o 0.0 270.0~~nSK J~v.PU3X PCnZR M S"~WETWELL PLAN VIEW AT ELEVATlON OF DOWNCOMER EXlTS 5-2>>124-

~~~uMVRY i-I~~~~)~

WASHZNGTON PUBLZC PQWER SUPPLY S~~NUCLEAR PROJECT NQ 2~FLUID STRUCTURE BOUNDARY-WNP-2 SUPPRESSION POOL PZGURE 5-4-.12 6-f 605.88'67.38 547'21 500'88.8'77.6'70'63.25 456.5'49.75'43'27.75'22.25 SOIL/418,25 STRUOTlBE 414 25 I NTERFACE 410,25 NOD'ES WASHINGTON PUBLIC POWER SUPPLY SYSTEM NU~~PROJECT NO>>2~REACTOR BUZLDXNG MODEL-127-FIGURE 5-5

BURilS AND ROE, ZNC.PROPRIETARY GE?iEEVJ ELECTRIC CO.'IPANY PROPRZ~TARZ WASHINGTON PUBLIC POWER SUPPLY SYSTEM NUCLEAR PROJECT NQ~2 ENVELOPES OF CALCULATED RESPONSES FOR FIGURE WNP-2 AND MEASURED RESPONSES AT JAERI CONTAINMENT AT VENT EXIT ELEVATION THETA TRANSLATIOl4 O lt I g4 Ill 4l~Lk~'0 cga.00 IO.04'%0.00 40.00 IO.04 l00.00 INl.00 I 0.00 I CO.00 FRFOURNCT IN'L)VERTICAL TRANSLATION D A I g4 IC le lal~IJ D EJ g4 40.40 W4.00 M.04 l4.04 I 00.04 I IO.00 I%4.00 I CO.00 FROOUQICT INZ)HORIZONTAl TRANSLATICN O Q~V I<Z4 IC lal III~CJ~(g 4 b.OO IO.<Q'l0.04 40.00 lO.OO 104.00 l)0.04 I%0.00 ICO.OO FRE'OURNCT UI\)Note: Multiply all acceleration values by 1.18.WASHTMGTCN PUBLZC PQWER SUP LX SZS~NU~PM4ECT HO 2 REACTOR BUILDING RESPONSE-ASYMMETRIC LOADING: CONTAIN),I, NT VESSEL AT MAT PZGQ3E 5 7 a-129-I I I I I I 2~<<THETA TRANSLATION 2<<Og IJ~I~0 I<<4~4 IC<<<<I 4 IJ EJ<<0 b.04 to.04<<Io.00 Oo.00 Io.00 IOO.00 I to.00 Ilo.00 100.04 F Rt:OtjENCY N2)V.ERTICAL TRANSLATION I<<04 ill l<<k<<Oo'b.oo-to.oo<<~f Rf QUEttC'f<ltL I HORIZONTAL TRANSLATION lh<<I<<0<<40<<<<<<~<<<<k 4 EJ EJ<<0 b.04 t0.00 14.04 44.00 00.00 I04.04 IO0.04 I 0.04 IO0.00 F RCQUtttC'l Ilt'C I Note: multiply all acceleration values by 1.18.WASEKBCZQN PUBLZC POWER SUPPXIY SYSTZA HU~~PRCQZCT HO~2 REACTOR BUILDING RZSPONSZ-ASYK4ETRIC LOADING: RPV SUPPORT FIGURE S-jb-130-

THETA TRANSLATION "1 5 ar zo.ao uo.oa eo.oo so.oo Loo.oo tza.ao.uo.ao ice.oa FRCQUKtuCY (HZ)HORIZONTAL TRANSLATION

%.oa zo.aa'ua.aa 80.Oo 80.Oa 100.Oa L20.Qa Lua.aa boa.00 FPEQUEtuCY

<HZ>Note: Multiply all acceleration values by 1.18.NASHZHQTON PUBLZC POWER SUPP'Y~A NUCLEAR PBQZECT'O~

2 REACTOR BUILDING RESPONSE-ASYMMETRIC LOADING: CONTAINMZNT VESSEL AT STABILIZER TRUSS.LEVEL PZ GORE 5-7 c-131-I I I I THETA TRANSLAT(ON l44 wQ 4 g4 I40 EJ~Ek~0 El 10.00 00.00 l0.00 IOIOO IOO.04 l10.00 I10.04 IOO.OO FHE'OUQICV IIII>VERTICAL TRANSLATlON a ll4<ga IK 4I lal~ll lg4 h.04 10.00 IO.00 IO.00 00.00 IOO.04 l10.04 IlO.04 IOO 04 FtICOUf NCT (1IZ1 HORIZONTAL TRANSLATlON O 0 I 4l wQ 0't'b.04 10.04 IO.OO IO.OO 00.00 I04.04 I10.04 IIO.OO I00.00 f REQUCIICI NL1 Note: Multiply all acceleration values hy 3..18.WASHINGTON

?UBLZC PCWER SUPPLY SY~NUCLKQt?BOJECT HO 2 REACTOR BUZLDZNG RESPONSE-ASYMMETRZ LOADZNG-CONTAZNMENT VESSEL AT MZD-SUBMERGENCE DEPTH PEGVRE 5-7 d-132-

Theta Translation Ou~4?~O4 O I44 4I 4 Iu 4 IJ I?'II.uo Io.oo co.oo co.oo co.uo Ioo.oo I to.oo i co.uo I co.oo f RCOUEIICY (III I I~\Vertical Translation 4 Qu?O4 gu tu lu u tu tu4 to II'b.oo, co.oo co.uo cu.oo co.oo Ioo.oo Ico.uo I o.uo ccu.oo f RROUfIICY IIQ)4 Horizontal Translation t Ou 4 O 4 O I IC ceo 4J~~ay 4 4 O'b.uo Io.oo co.uo co.oo co.oo duo.oo cto.ou cco.oo cco.oo f RCOUCIICY UIK)Note: Multiply all acceleration values by 1.18.RLSEDiQTOH PQBLZC PCRER SUPPLY SiE'8 HU~PBCCEZT HO REACTOR BUZLDZNG RESPONSE ASYHMETRXC LOADZNG: OUTSXDE BUZLDZNG gALL ELEVATZON 521'133-THETA TRANSLATION ua?~t l 4l 4J IJa Co')I.04 IO.00 10.44 IO.00 10.04 100.04 IIO.04 I IO.00 100.04 FRCOUKIIC)

UIE)4 l4 ll VERTICAL TRANSLATION a 4la IK 4l 4l tJ 4lo go b.00 10 00'10.00 00.00 10.00 100.04 I 4.00 1%0.00 100.04 FRCOUCIICY IIIL)HORIZONTAL TRANSLATION a Qa?.Cl a I ga la i@a 4J tJ o i/4 tl.04)0.04'14.00 44.00 10.00 100.04 140.00 I 0.04 I<0.04 FRCOURIICT IIIL)iVote: Multiply all accelexation values by 1.18.MASHISGTCN PUBLIC PGWER SUPPLY SYS~NU~PRAT-C2 HQ 2 REACTOR BUILDING RESPONSE-KUGKY SYMMETRIC LOADING: CONTAINMENT VZSSEL AT MAT-ZGUBE 5-8a'134-

0 THETA TRANSLATION C)g IJ ce~0 CC Wg wa CJ D D)I.04)O.00'10.00 CO.00 ll.04 100,04 I CO.04 1%4.00 I M.00 FRCOUFNCT IIIt)VERTICAI TRANSLATION D Ia D I CC D W Wa CJ CI a COD)0.00 10,00 la.00 IO.00 100.00 I)0.00 I 0.00 Ill.00 FRROUENCY INZ)HORIZONTAL TRANSLATION COD D D I CC~a W>>wa CJ II.04)0.04 10.44 10.40 IO.04 100.04 I)0.04 I 0.04 104.04 FRKOUKtlCY (NR)iVote: Multiply aU.acceleration values by 1.18.)t)ASKPifQTON PUBIC PCQER SUPPLY SY~NU~PBQJE~D No 2 REACTOR BUILDING RESPONSE-NEARLY SYMMETRIC LOADING: RPV SUPPORT PX GUNK 5-8 b-135-

THETA TRANSLATiON R Oe CI 44 W ua QO~>>.QQ 20.00'l:l.00.<Q.QQ BQ.QQ IQO.QQ l20~JQ l40.OQ!OQ~OJ ePsOUf ICY tHZ'ORiZONTAL TRANSLATiON Cl p)gO~I laJ~~r.~00 20.00 40.00""0.00 80.00 I QQ.00 I 20.00 I'40.00 l 60.00 FRfQUfNCY (HZ]Note: Multiply all acceleration values by 1.18.MAKKHGTON PUBLZC PCWER SUPPLY SYS~iVUCXZiQL MAC HO 2 REACTOR BUZLDXNG RESPONSE-NEARLY SYMMETRZC LOAD XNG: CONTATNVZNT VESSEL AT STABILIZER TRUSS LEVEL rzavaz 5-8 c-136-I Theta Translation O Q4 Z O~gO ac~a lJ~Cg D.00 t0.00 t0.00 40.00 40.04 IM.00 I'CD.00'.10.09 I CD.00 FREOUENCt (hatt Vertical Translation O'?o I ggEh EC~a~N CJ Ef 0.04 t4,04 10,04 CCI.05 Ql.Oo l00.04'R0, 04 lWO.00 ICl.05 FRFOUENCT IHZI A a'orizontal Translation

'X tl I gR EC~n EJ~n QA 0.04'lO.00 lO.C4 sO.M 104.OQ I'CS,04 I%0.00 I CO.00 FREOUCHI:T IIIt)Note: Multiply all acceleration values by 1.18.WASHZHGTON PUBIC POWER'UPPLY SY~NUCLEAR PROJECT HO 2 REACTOR BUiLDING RESPONSE-K~LY FIGURE SYMMETRIC LOADING: CONTAINMENT g-8 d VESSEL AT MID-SUBMERGENCE DEPTH-137-

Theta Translation l O4~4 EJ S Eo 4 I Ea l 4 EO E44 IJ EJ Ea 4 4~I'b.ao lo.oo Ea.oo co.ao to.ao too.w toa.oa t O.oa too.aa FRZt)VZttCY lllZ)Vertical Translation EJ4 4 EJ 4 Ea 4 EK ill ill EJ~EJ 4 0,4 ta,aa.Ea.00 40.00 Eo.00 taa.00 I to.00 IEO.00 I to,oa FRft)VCttCY ItlZl Horizontal Translation W EO4 Et I Ea Ea E4 4 l4 4 EJ EJ Eo'b.oo to.oo Ea.ao EO.oo to.Oo too.ao t to.oo EEO.Oo t Eo.oa FRROVCttCY It)Z)Note: Multiply all acceleration values by 1.18.WZttsiKS~~

PQBLZC PCWER SUPPLY SYS~NU~PROJEX.~HO z REACTOR BUIIEDING RESPONSE-NEARLY SYMMETRIC LOADING: OUTSIDE BUILDING HALL ELEVATION S21'GVBE 5-8e-138-RQF"RKPJC-I.a i i i I~A<MAIN8 PCtlNQA-,ICONS I I I I AV3RAQc LIlIIT CP Hllblkt4 TCI-RkiIIC-CS Cl C iik iii CJ CJ i i I I I i i I I i I I II>>I~i I~I I i I I I I I I l 1 lllll!I.i l lllll!I I I l.33cIi'I ic 38 I I I I I IIII!I I I I!Q.QI~i I I i i I i I I 1 i I I I I II I.I I I I i I I I I I I II i I I I I I III I I l l ll I I l I I ll!i I Iilil I I llii Q.CQ1 3 IQ PRSCUS'ICY CP VIGRATICiV,~3Q IQQ WSBZZ~PQ3X~~~~~~~HCC~I.~c HQ-139-

APPENDIX A ANALOGY BETWEEN ACOUSTIC AND STRUCTURAL BOUNDARY CONDITIONS A.l Introduction Response of the analytical model of the 4T system to two types of sources, the pressure source and the acceler-ation source applied under the vent is presented in Section It is shown that the two sources excite different modes of the 4T system.The difference in the excited modes is a result of different boundary conditions at the pool vent interface which appear in the analytical solutions of the two problems.It is important to recognize this basic difference between the two types of sources because the 4TCO key chugs which are simulated using the acceleration source contain the response of the system which cannot be simulated using the pressure source.It is shown in this Appendix that the acceleration/

pressure sources of vibration in acoustic fluid vibration problems are analogous to the force/displacement (or acceler-ation)sources in structural vibration problems.To emphasize the difference in the boundary conditions and their importance in changing the characteristics of the response, a simple structural vibration problem is first presented in this Appendix, then the anology is shown between the two problems.-Al-

A.2 Forced Lon itudinal Vibrations of a Prasmatic Bar The equation of motion of'ongitudinal vibrations of a prismatic bar due to external force p(x)f(t)and the formulation of its solution is shown in Figure A-l.The eigenmodes and eigenvalues (frequencies) for two cases of boundary conditions are shown in Figure A-2.It is seen that the fundamental mode frequency for Case 1 (fixed-free) is one half of the fundamental mode frequency for Case 2 (fixed-fixed)

.It is shown in Figure A-3 that vibration of the bar due to a force applied at the free end is completely defined by the eigenmodes and frequencies corresponding to the Case 1 whereas the vibration of the bar due to a displacement motion prescribed at the free end is defined by the eigenmodes and frequencies corresponding to the Case 2.In other words, the characteristics of response of the bar due to an external force is completely different from the characteristics of response of the same bar due to an imposed (displacement) motion.The imposed displacement motion requires a change in the boundary condition at the location of imposed motion which alters the response characteristics of the system.-A2-

A.3 A'nalog Between Acoustic Fluid Vibrations and Structural Vi rations The equation of motion for the longitudinal vibrations of a prismatic bar is analogous to the equation of motion of acoustic fluid.Bp Bu p (x t)+~~~0 C.Qt Qx where p=u and Q Q-(note'6)du.du x X X E AE.'.Pressu=e Source in Acous ic Fluid Problems=Zmposed D'splacement m Structural Vib at on Problems ard Accelerat'on (or p ess" e g ad'ent)=-Appl ied.force~~Source in Acoustic r lu'id Problems Structural Vibration P=oblems A.4 Conclusion From the above analogy, it is evident that the response characteristics of acoustic fluid system excited by an acceleration source (which is analogous to externally applied force in structural vibrations) will be different from those excited by pressure source (which is analogous I to imposed motion in structural vibrations)

.Since key chugs are properlv simulated using accel-eration source, for chugging load definition to bound the key chug data, an acceleration source must be used.A.5 References.

A-l"Vibration Problems in Engineering", Timoshenko, S., D.Van Nostrand Company, Inc.-A3-I I I u (x,o)l 8 u pA~u ec So lution f zom Q u(x.t)"~B.u(x t)n~l, u(x,t)I 2 I'I (3 u!I AE-dx, I.Q x p (x)f (t)dx FOR CE S ACTI N G ON ELEMENT dx AT (x, f)(SEE SKETCH OF Q u(x, t)+AP~=p(x)z((:)gt X (x)P (t)Ecuatior.of Motion (A-1)Solut'or"or Homogeneous Boundary Con"'t'ons (A-2)BAR BELOW)Refe ence A-1 may ce su~+rimed as oU.ows: X(x)=Eicenmodes oz the eicenvalue ecuation.'{(x)+Z x (x)~0 2 wim appropriate homoce..eous boundary c8ndition9 (e.g., Case 1 and 2 in Figure A-2)p (t)=Solution oz the ecuation-o mot'on p (t)+U II'o f (t)/Ap wi"5 appzopz'ate initial conditions whe e,, a=E/p 2=-Q3~recuency oz vibration of natural mode n"-Kna g p=p(x)X (x)dx X (x)o o ()=A cosG3 t+B SinG3 n" r.n,"'n.+~t f (tl)S~zU (t-tl)dtl~Ap g 1 n 1)1.(A-3)where con tents A, B a e obtained rom initial cond'tions and aze set erual to"ezo'-tne system at zest at.t~0.x=O iÃ.I-APPLIED FORCE p(x)'(t)P=MASS DEN SITY E=MODULU S OF, ELASTI CITY A=AREA OF CROSS SECTION WASHINGTON PUBLIC POWER SUPPLY SYSTEM NUCLEAR PROJECT NO~2 LONGZTUDZNAL TRANSZENT VZBRATZONS OF A PRZSHATZC BAR rImRE A-1 4

Fixed Case 1 Free Fixed Case 2 Fixed x=o Boundary>Conditions:

U(o)=o,-.(g)=o Bu Qx U(o)=o, U(z)=o Eigenmodes:

Z (x)=Sin~07t x n=1,3,5 x (x)=Sin-n n=1(2(3 11=2 Eiaenva1ues:

nW K n 2g, n7l K n WASHZNGTON PUBLZC POWER SUPPLY SYSTEN EIGENMODES A'ND EIGENVALUES FOR TWO CASES OF BOUNDARY CONDITIONS NUCLEAR PK4ECT NO 2-A5-FZGURE A-2

Problem l Applied Force at Free L'nd Problem 2 Imposed Displacememt at Free End x=0 A, L",P-e P(t)R Ai F(P I x=0 o(t)K)Pl n-l u (x,t)=-Z><(u)j (t)pAX n=l,3, S., G3 where, u (x~t)p ul(x, t)+u 2 (x, t)u>(x, t)~~D(t)H OO 0>Q OA H t3 OO R NH gK WO an x (x)~sin-2-j)-n I(x n Pu (t>I ri(t~>S(.uCJ (t t>.)dtj 0 n l'=8/p where, x (x)sin-n ITx n g lu<t>f ri<t()SsuCJ<t-tZ>ritZ 0 nT(a'3 n g D(t)d D dt Note that the response characteristics (x,4))of the same bar are dif ferent in the above two problems.The series solution of Problem l i9 based on the boundary conditions of Case l n'Figure A-2).The series part of the solution (U2 (x.t))of Problem 2 is based on the boundary conditions of Case 2 (Figure A-2).

)I I I I I