ML20039E513
| ML20039E513 | |
| Person / Time | |
|---|---|
| Site: | West Valley Demonstration Project |
| Issue date: | 12/31/1981 |
| From: | Johnson N, Walls J SCIENCE APPLICATIONS INTERNATIONAL CORP. (FORMERLY |
| To: | NRC OFFICE OF NUCLEAR MATERIAL SAFETY & SAFEGUARDS (NMSS) |
| References | |
| CON-FIN-B-6955 NUREG-CR-2236, NUREG-CR-2236-V01, NUREG-CR-2236-V1, SAI-148-026, SAI-148-26, NUDOCS 8201070398 | |
| Download: ML20039E513 (71) | |
Text
. . .
N 4
. NUREG/CR-2236 sal-148-026 Vol.1 4
i
- Seismic Resistance Capacity Evaluation of Spent Fuel Storage Racks and Fuel at West Valley,
- New York Main Report Prepared by N. E. Johnson, J. C. Walls Science Applications, Inc. p
' S acEWED U Nuctear Regulatory f= 4. $82' hg ys -
commission -
i
\p
's, q .
l 1
~-
NOTICE This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, or any of -
their employees, makes any warraty, expressed or implied, or assumes any legal liability or responsibility for any third party's use, or the results of such use, of any information, apparatus product or process disclosed in this report, or represents that its use by such third party would not infringe privately owned nghts.
l l
l l
l l
l l
l l Available from l
l GPO Sales Program Division of Technical Information and Document Control U. S. Nuclear Regulatory Commission Washington, D. C. 20555 Printed copy price: $4.75 and
. National Technical Information Service
.. Springfield, Virginia 22161
_ , . , - _ _ . _ - ~ . _ _ . .
NUREG/CR-2236 sal-148-026 Vol.1 Seismic Resistance Capacity Evaluation of Spent Fuel Storage Racks and Fuel at West Valley, New York Main Report Manuscript Completed: January 1981 Date Published: December 1981 Prepared by N. E. Johnson, J. C. Walls Science Applications, Inc.
800 Oak Ridge Turnpike Oak Ridge, TN 37830 Prepared for Division of Fuel Cycle and Material Safety Offico of Nuclear Material Safety and Safeguards U.S. Nuclear Regulatory Commission Wcshington, D.C. 20555 NRC FIN B6955 9
i
ABSTRACT This report documents an evaluation of the seismic resistance capacity of the Fuel Receiving Station (FRS) spent fuel storage racks and canisters at the Nuclear Fuel Services Reprocessing Plant at West Valley, New York. The primary objective of this work was to determine the threshold ground acceleration above which potential yielding, permanent deformation or collapse, and/or excessive deformations would occur to the storage rack structure and canisters. Examination of the failure threshold levels show that the splice joint in the top rail of the storage rack will be the first item to reach yield stress. This event occurs at 0.198 The results also show that th9 top rail splice joint would probably be the first component to break at a threshold of 0.22g. A probable scenario of events following the top rail splice joint failure is that loads will redistribute in the storage rack causing the anchor bolts connecting the rack columns to the pool floor to fail. This will subsequently cause the rack to deflect in an East-West direction until it impacts the pool wall or another rack. Although this event could cause a few canisters to become dislodged from the rack and fall to the pool floor causing crushing of the canisters, such an occurrence is considered to be unlikely.
iii l
1 l
SWSIARY This report documents an evalua. ion of the seismic resistance capacity of the Fuel Receiving Station (FRS) spent fuel storage racks and canisters at the Nuclear Fuel Services Reprocessing Plant at West Valley, New York. This work was performed by Science Applications, Inc. (SAI) for the Nuclear Regulatory Commission (NRC) Division of Fuel Cycle and Material Safety. This report has been prepared in two volumes. This volume contains ins text of the report and Volume 2 includes appendices containing technical supporting data generated during the preparation of the report. The primary objective of this work was to determine the threshold ground acceleration above which potential yielding, permanent deformation or collapse, and/or excessive deformations would occur to the storage rack structure and canisters.
In general, the analysis procedure used in this task was as follows.
Finite element mathematical models of the racks were formulated using the PAFEC computer program as the basic analysis tool. In addition, a program developed by SAI to perform seismic response spectra analyses was used. This program, entitled BLDRES, uses the results from a PAFEC modal analysis to determine the deflections and internal member loads of a structure which has been excited by a seismic ground acceleration spectra. The internal member loads obtained from the BLDRES program were then compared to values of loads which would cause structural
" failure" at critical points in the structure. These allowable values for the rack structure were determined from an evaluation of its structural design. The comparison of allowable load values to actual values for a given ground acceleration enable the calculation of threshold ground accelerations for initiation of various failure modes.
An examination of the failure threshold levels show that the splice joints in the top rail of the storage rack will be the first item to reach yield stress. This event occurs at 0.19g. The results also show that this splice joint would break at a threshold of 0.22g. It should be noted that these thresholds for yielding and ultimate stress are values which apply to the racks supporting the heaviest loaded canisters I in the present pool configuration (i.e. , those which contain fuel from the Wisconsin Electric Power Company. Canisters which contain lighter fuel assemblies will have higher failure thresholds.
A probable scenario of events following the top rail splice joint failure is that the redistribution of load will be such that the floor / column welds will yield and/or fail thus causing very large deflections of the storage racks in the East-West direction. Although this event could cause a few canisters to become dislodged from the rack and fall to the pool floor, such a failure is not expected to occur.
v
i l
TABLE OF CORTENTS - YOLUBE 1 Section Eagg ABSTRACT iii 4
j
SUMMARY
Y TABLE OF CONTENTE vii LIST OF FIGURES ix LIST OF TABLES xi i
PREFACE xiii 1.0 INTRODUCTICN 1 2.0 TECHNICAL APPROACH 3 2.1 General Discussion 3 2.2 Special Analysis Considerations 4 2.2.1 Response Spectra Analysis Methodology 5 2.2.2 Hydrodynamic Effects 5 2.2 3 Earthquake Directionality Effects 7 23 Assumptions 7 2.4 Computer Software 8 2.4.1 Finite Element Analysis 8 2.4.2 Ground Response Spectra Analysis 10 2.5 Canister Sliding Analysis 12 3.0 FACILITY DESCRIPTION 15 3.1 Pool Arrangement 15 32 Structural Configuration 15 3 2,.1 Storage Racks 15 3 2.2 Canisters 23 4.0 MATERIAL BEHAVIOR 25
. 4.1 Material Strength 25 l 4.2 Friction Coefficients 25 vii r
--, . , . , - - - - . , - , , . n.,- --,. , - . - . - , - - ... , _, . - . , , , - . . .
l l
1 TABLE OF CONTENTS - VOLUBE 1 (cont'd)
\
l Section Page 5.0 STRUCTURAL DESIGN REVIEW 27 5.1 Storage Rack Strength 27 l 5.2 Canister Strength 29 6.0 SEISMIC RESPONSE ANALYSIS 31 a 6.1 Pool Sloshing 31 6.2 Single Rack Response 32 3
6.2.1 Finite Element Models 32 6.2.2 Modal and Seismic Response Analysis Results 35 6.3 Multiple Rack Response 35 631 Finite Element Models 35 6.3 2 Modal and Seismic Response Analysis Results 40 t
6.4 Single Canister Response 53 70 SEISMIC RESISTANCE CAPACITY EVALUATION 57
- REFERENCES 61 i
i l,
l t-a I viii
LIST OF FIGURES - VOLINE 1 Finure Title f.a&R J
2.1 The Response Spectra Analysis Method 6 2.2 NRC Re6ulatory Guide 1.60 Design Response Spectra 9 23 The BLDl!ES Protfram 11 24 Detailed Drawings of Fuel Canister and Support System 13 3.1 Plan View of Repreessing Facility 16 32 Plan View of FRS Pool 17 33 Elevation Views of FRS Fool 18 3.4 Existing Canister Arrangement in FRS Pool 20 35 Layout of Typical Storage Rack 22 4.1 Typical Stress-Strain Curve for 6061-T6 Aluminum Alloy 26 6.1 Finite Element Model R1 33 6.2 Model R1S - Mode Number 1 36 6.3 Model R1SLT - Mode Number 1 37 6.4 Finite Element Model RS 39 6.5 Finite Element Model R6 41 6.6 Finite Element Model R7 42 6.7 Finite Element Model R8 43 6.8 Model R5B - Mode Number 1 44 6.9 Model R5BLT - Mode Number 1 45 6.10 Model R6 - Mode Number 1 46 6.11 Model R6LT - Mode Number 1 47 6.12 Model R7A - Mode Number 1 48 6.13 Model R7 ALT - Mode Number 1 49 6.14 Model R8A - Mode Number 1 50 6.15 Model R8 ALT - Mode Number 1 51 6.16 Canister Shear Force vs. Distance From North Wall 54 6.17 Maximum Canister Shear vs. Percent of Canisters Effective 55 j
ix l
1 LIST OF TABLES - VOLUME 1 Table Title Pane 3.1 Description of Fuel Assemblies Stored in FRS Pool 19 3.2 List of Drawings for Storage Racks and Canisters 21 5.1 Yield Moment Capacity of Structural Connections 28 in Storage Rack 5.2 Ulticate Moment Capacity of Structural Connections 28 in Storage Rack 6.1 Canister Weight Summary 34 6.2 Summary of Results for Single Rack Models 38 6.3 Summary of Reaults for Five Rack Models 52 71 Yield Stress Ground Acceleration Thresholds 58 7.2 Ultimate Stress Ground Acceleration Thresholds 58 7.3 Probable Failure Scenario 60 xi
I
- PREFACg This report was prepared by Science Applications, Inc. (SAI), Oak Ridge, Tennessee, under Basic Ordering Agreement (BOA), No.
NRC-02-80-035, No. 0005 as issued by the U. S. Nuclear Regulatery Commission (NRC), Division of Fuel Cycle and Material Safety, Washington, D. C.
The authors would like to acknowledge Mr. R. W. Starostecki who is SAI's General Program Manager for this BOA and Mr. C. J. Haughney who is the NRC technical monitor for the task. Their guidance and review efforts have been invaluable during the course of this work. In addition, a number of suggestions made by Dr. A. T. Clark of the NRC and Dr. W. Coffman, an NRC consultant, during their technical review of this work have been extremely helpful. Acknowledgement is given to Mr.
J. P. Duckworth and his staff of Nuclear Fuel Services, Inc. for conducting an excellent tour and providing data during our visit to the West Valley facility. Acknowledgement is also given to Mr. R. C. Dong of Lawrence Livermore National Laboratory and Mr. D. R. Peterson of SAI/La Jolla who provided valuable technical advice during the formulative stages of this project. Special acknowledgement is given to Mrs. P. P. Delaney of SAI who provided administrative support and to Mr. G. N. Lagerberg of SAI who provided technical support during the project.
l xiii
- 1. INTRODUCTION This report documents an evaluation of the seismic resistance capacity of the Fuel Receiving Station (FRS) spent fuel storage racks and canisters at the Nuclear Fuel Services Reprocessing Plant at West Valley, New York. This work was performed by Science Aoplications, Inc. (SAI) for the Nuclear Regulatory Commission (NRC), Division of Fuel Cycle and Material Safety. This report has been prepared in two volumes. This volume contains the text of the report and Volume 2 includes appendices which contain all technical supporting data generated during the preparation of the report. The primary chjective of this work was to determine the threshold ground acceleration above which potential yielding, permanent deformation or collapse, and/or excessive deformations could occur to the storage rack structure and canisters.
The work involved the review of a previously published structural and seismic analysis of the FRS (Ref. 1), a review of structural drawings of the storage racks and canisters used to store the spent fuel assemblies, and an on-site inspection of the facility which was made to obtain weight and other design data on the present storage configuration at West Valley, A series of finite element mathematical models of the storage racks and canisters were utilized to perform the appropriate seismic response analyses. The determination of seismic resistance capacity was dcne using assumptions which were thought to be realistic to reflect actual "as-built" conditions. That is, this determination was made with the understanding that analyses done to evaluate an actual structure may differ from those done in support of an original design with the aim of removing known conservatism wherever possible. The task did not include the determination of the probability of occurrence of any expected scismic event of a particular magnitude.
Section 2.0 contains a description of the technical approach used to perform the seismic resistance capacity evaluation. Included in this section is a general discussion of the step-by-step process used to perform the evaluation, a description of the various assumptions made to expedite the evaluation, and a description of the various analysis techniques used to perform the work. Section 3.0 contains a description of the facility, including the pool, storage racks, canisters, and fuel assemblies stored in the canisters. Information about the material used for the fabrication of the storage racks and canisters is presented in Section 4.0 along with discussions of the friction coefficients used in the capacity evaluation process.
Results of a structural design review of the storage rack and canisters are presented in Section 5.0. This review was made to evaluate the rack's structural design characteristics as they might affect its 1
seismic response capacity. Section 6.0 contains details of the seismic response analysis performed on single racks and groups of racks while Section 7 0 presents the final seismic resistance capacity evaluations of the system in terms of threshold grottnd accelerations. Extensive tabulations of data developed during this evaluation are included in appendicies which are contained in Volume 2 of this report.
4 2
- 2. TECHNICAL APPROACH This section contains a description of the methodology, assumptions and analytical techniques used in the seismic resistance capacity evaluation of the storage racks and canisters at the West Valley FRS pool. Section 2.1 contains a general discussion of the step-by-step procedure used in performing the evaluation. Section 2.2 discusses certain aspects of the approach used to perform the storage rack response analysis that required special consideration. Section 23 contains a description of the various assumptions made in conducting the study and Section 2.4 contains a description of the analysis techniquea which were used in performing the evaluation. Finally, Section 2.5 discusses the potential restraint provided by the canisters between adjacent racks.
2.1 GENERAL DISCUSSION The seismic analysis of the spent fuel storage racks was actually conducted in two (2) steps. Step I of the study was further sub-divided into three (3) phases. Phase I involved the gathering of data required to perform the study. This data included reports generated during previous structural evaluations of the West Valley facility; design drawings for the FRS pool, storage racks and canisters; a description of the fuel assemblies currently stored in the FRS pool; and miscellaneous other information required during the study. A portion of this data was obtained by the authors during a visit to the West Valley site.
Phase II consisted of a detailed review of the structural design of the storage racks and their connections to the FRS pool walls and floor.
The results of this review are presented in Section 5.0.
Phase III included a number of simplified scoping analyses which were performed in order to identify some of the structural dynamic characteristics of the FRS pool / storage rack / canister system and to establish the analysis method to be used in the subsequent detailed seismic response capacity evaluation. The scoping analyses included an evaluation of the sloshing characteristics of the FRS pool and preliminary finite element analyses of the storage rack / canister system.
The potential use of both the time history and the ground response spectra analysis methods were conside *d during the Phase III effort.
As a result of this consideration, the response spectra analysis method was selected for subsequent seismic analyses of the storage rack / canister system. This method requires that the ground motion be defined as a response spectrum. As described in Reference 2, a response spectrum is represented by a curve of maximum response motion 3
1 l
l
of an infinite series of single degree-of-freedom spring-mass systems as a function of the natural frequency of the systems. Since every structure can be defined dynamically by its various natural frequencies and their corresponding mode shapes and generalized masses, it is possible to use the ground response spectra analysis method to obtain an approximate solution for the response of the structure with very little analysis cost.
The time history analysis method, on the other hand, requires that ground motion be defined as an amplitude (e.g., acceleration) function of time. Although the use of this method may produce " accurate" results, it exhibits two distinct disadvantages. First, even though any given earthquake as measured at a particular location can be represented as a well defined amplitude function of time, no two earthquakes are ever exactly the same. Consequently, it is generally necessary to perform several time history analyses in order to provide an envelope of expected structural response. Second, the use of the time history analysis method requires an integration of the prescribed equations of motion which is generally a very costly procedure.
Step II of the study involved the use of the response spectra analysis method in the performing of a detailed seismic response capacity
- evaluation and the determination of ground acceleration failure
] thresholds. The results of the Step II effort are documented in Sections 6.0 and 7 0 of this report.
2.2 SPECIAL ANALYSIS CONSIDERATIONS The determination of the seismic resistance capacity of any structure is, to a great extent, a qualitative and subjective analysis based on experience gained by evaluating damage to existing structures during earthquakes. The capability of a particular ground motion to do damage to a structure is a function of the:
- 1. Intensity of ground shaking,
- 2. Duration of ground shaking,
, 3 Frequency characteristics of ground shaking, and
- 4. Capability of ground motion to put energy into the structure and
- the ability of the structure to absorb this energy.
Many other factors also enter into the determination of the seismic i
resistance capacity of the storage rack / canister structural system.
The primary objective of the evaluation performed during this study was to attempt to consider the most important aspects of this problem in a reasonable period of time and in a realistic manner.
Included in the following subsections is a description of the methodology used to perform the seismic resistance capacity evaluation and a discussion of the effects of some of the special factors which were considered during the evaluation.
4
2.2.1 Resoonse SDectra Annivsis Methodology The methodology described below bas generally used to determine the seismic resistance capacity of the storage racks in terms of threshold damage ground accelerations. A general flow diagram illustrating this methodology is shown in Fig. 2.1.
- 1. Finite element models of the individual storage racks and multiple storage racks were developed to represent existing configurations of spent fuel assembly storage in the FRS pool at West Valley. These models were formulated using the PAFEC computer program (Ref. 3) as the basic analysis tool. During the development of these models, special consideration was given to the evaluation of the effects of water on the dynamic characteristics (e.g., mass and damping) of the canisters. These effects are further discussed below in Section 2.2.2.
- 2. Using the PAFEC finite element computer program as described in Section 2.4.1, a modal analysis was performed on each model to determine natural frequencies, normal modes, generalized masses and related parameters.
3 The BLDRES program described in Section 2.4.2 was then executed using the modal analysis results generated by the PAFEC program.
' The BLDRES program uses the PAFEC modal analysis results to determine the deflections and internal member loads of a structure which has been excited by a seismic ground acceleration spectra. The BLDRES program was executed assuming an input ground acceleration of 0.10g for each model to simplify subsequent normalization.
- 4. Individual structural member stresses were obtained by combining results from earthquakes in each orthogonal direction using criteria discussed below in Section 2.2 3
- 5. The internal member loads obtained from the BLDRES program were then compared to values of loads which would cause structural
" failure" at critical points in the structure. These allowable load values for for the storage rack structure were determined from an evaluation of the storage reck structural design and are discussed in Section 5.1. The comparison of allowable load values to actual values for a given ground acceleration (e.g.,
0.1g) enable the calculation of threshold ground accelerations for the initiation of various failure modes.
2.2.2 Hydrodynnmic Effects The fact that the storage racks are submerged in a water environment requires a careful study of the hydrodynamic interaction of the water and structure during a seismic event. The motion of the water in the pool could induce additional forces into the canisters and racks.
Also, for a structure vibrating under water, the concepts of added mass and damping must be considered.
5
i I
i a
t GEOMETRY. M ASS
- AND STtFFNESS
! CHARACTERISTICS OF STRUCTURE I
i 1 PAFEC GEOMETRY FINITE ELEMENT - AND 4 ANALYSIS MODE SHAPE PROGRAM PLOTS l
k
.I EARTHOUAKE DAMPlNG j DIRECTION AND CHARACTERISTICS j MAGNITUDE OF STRUCTURE I
2 i
A BLDRES RESPONSE
= ANALYSIS % RESPONSE S' ECTR A 4 COM81 NATION RESULTS ANALYbfS METHOD PROGRAM I
l 1f DISPLACEMENT
! AND STRESS 2 RESPONSE RESULTS l
i i Figure 2.1. The Response Spectra Analynis 'Pethod.
t i
6
, ~ - - . . - - - . ., ,. , , - , - . . ,-r,_-_..._,. _,.m . , ,.r,_, , ., _ . _ .-, _ _ , m--,-_.-.
The effective mass of a canister embedded in water is the sum of its body mass, the mass of the fuel assembly including the mass of any trapped water, and the added mass from its interaction with the surrounding water. Rigorous computation of added mass is impractical i
because of the complex structure-water interaction. Rather, the methods commonly used to perform this computation are based on engineering judgment derived from analytical and experimental work on single structures in an infinite medium (Ref. 4). Depending on the shape of the response spectra of applicable seismic and dynamic loads and the stiffness and mass of the rack, inaccuracies or uncertainties in estimating the added mass can affect the predicted response significantly. Therefore, the added mass was varied within the limits of possible mass magnitude and its effects on the computed structual response were evaluated.
Added damping is not as thoroughly investigated in the literature as is added mass. Establishing a fixed value of damping for a general multiple member structure is very difficult, if not impossible, since damping can be significantly influenced by member arrangement, spacing, and relative motions among the members. Some measured values of damping for structures similar to those evaluated in this study (Ref. 4) are in the range of 8 to 12 per cent of critical.
2.2 3 Earthauake.Directionality Effects Even though earthquake forces act in three principal directions of a structure simultaneously, the effects of these forces in the three principal directions are unlikely to reach their maximum simultaneously. It is an accepted seismic design practice that ,
structural responses during earthquakes be combined considering effects in three orthogonal directions (e.g., one (1) vertical and two (2) horizontal), each calculated based on the full spectrum value specified for the site. Any of several methods are used to combine the responses computed from these differing directions (Ref. 5). One such method, the square root of the sum of the squares (SRSS) method, was used to combine earthquakes in these three (3) principal directions in this report.
23 ASSUMPTIONS A number of simplifying assumptions were made in order to expedite the evaluation process discussed in this report.
As previously stated, the primary objective of this evaluation was to determine the threshold ground acceleration above which " failure" of the structure could occur. The input seismic ground acceleration spectra used during this evaluation was assumed to be based on the generic ground motion spectra defined by the Nuclear Regulatory Commission 'NRC) and presented in NRC Regulatory Guide 1.60 (Ref. 6). l A plot of these spectra for both horizontal and vertical earthquakes 7
l l
1 are shown in Fig. 2.2.
Another basic assumption involved the determination of the thresholds above which a " failure" was presumed to occur. It should be noted that the evaluation process was based on the use of linear, elastic analysis techniques. Since a failure of the facility would probably involve inelastic deformations, it can be seen that the analysis techniques used during the evaluation would only result in the determination of approximate damage thresholds.
With the above in mind, yield failure was defined as that stress above which any component or connection of the structure would exhibit permanent deformation when aubjected to a ground acceleration. As discussed more fully in Section 4.1, this stress was defined to be 40,000 psi for the 6061-T6 aluminum used to fabricate the rack and canisters. Complete collapse or ultimate failure of the structure was assumed to occur when the stress on any component or connection of the rack structure exceeded the average ultimate strength of the aluminum material used in the rack. This stress was defined to be 45,000 psi (See Section 4.1) for the 6061-T6 aluminum used to fabricate the rack ,
and canisters.
The amount of damping used in the response spectra analysis of the storage racks must consider the amount of structural damping, added damping from the water environment and additional damping from potential canister impact. For all analyses presented in this report the damping is assumed to be ten (10) per cent of critical. As discussed in Section 2.2.2, this damping value is consistent with values measured in structures similar to those evaluated in this study.
Accordingly, it is believed that this is a realistic value for the structural arrangement under consideration.
2.4 COMPUTER SOFWARE This section includes a description cf the finite element analysis techniques and computer sof tware used in the work documented in this report. Included is a description of the PAFEC finite element analysis computer program and a special purpose program, BLDRE3, written by SAI to perform seismic response analyses.
2.4.1 Finite Element Analysis As previously mentioned, the PAFEC (Erogram for Automatic Einite Element Calculations) program has been used to perform the finite element analyses required for this study. PAFEC is a general purpose, three-dimensional, linear and non-linear finite element analysis program.
PAFEC employs many data preparation aids including a free format input with engineeering key weeds, a series of powerful mesh generation facilities, and extensive plotting options. Capabilities include 8
~66mw ^^
~
- A,. ~ n ~ x^ 6~ -
n/ bb,C n uh^,6
^ .
A,
^69 N,,.6.. :m^6^
! 4, -
L ,. x ,, s x- s / s
- ' Eib(x,O3
_ ^ ~
%Q d(6pfs 3'O' '
A po%s
.N$3
- i[. sXM$(66b l~ 23D (XXY'h:.TOOB c-OS(' R "kDS 6A66Ai.1 'o 6A66A c. e, 6 '
3s^t,pc. A A
(6^^7
, s x ^/ c 8z A, ,
(tu
.C O S/X>^i,
- n . _ ..
<(^OS('-S('C'5
._.C.
(a). e oaiso= vat oman aaweses prevaa - sentro to e oairo= rat Ih)- """**"*""'""""~'**'8""4"""'^'
OROUNO aCCiLEAaf108e 08h0U880 ACCf LERaT8088 Figure 2.2. NRC Regulatory Guide 1.60 Design Response Spectra.
l l elastic stress, large deflections, modes and frequencies and . dynamic response, creep, . plasticity, and heat transfer analysis. The PAFEC l element library contains more than eighty (80) elements. Models may be composed of springs, masses, beams, membranes, 3-D solids, thick and j thin plates and shella, and axisymmetric solids with non-symmetric l loads. Most involve linear to cubic isoparametric elements, but hybrid j and semi-loof elements are also employed. PAFEC has restart l capabilities and control options for maximum versatility and economy.
L In addition to the extensive passive graphics options availible in PAFEC, the Pigs (EAFEC Interactive graphics suite) program is available to interactively view and modify thv PAFEC data base. PIGS requires Tektronix 4000 series terminals with cursor control and can be used to study both pre- and post-solution results.
It should be noted that only a portion of the capabilities available in PAFEC were used to perform the analyses documented in this report. For example, the finite elemerit models used in these analyses consisted of only beam and mass elements.
The PAFEC program was used to determine natural frequenciesr and normal mode shapes for the various finite element models under consideration.
As discussed more completely in the next section, this PAFEC output data was used as input to a program developed by SAI which computed the desired response to a predetermined seismic ground response spectrum.
2.4.2 Ground ResDonse SDectra Annivsis The responae of a structure to a prescribed ground motion (e.g.,
earthquake) can be determined by using finite element models and any one of several methods of solution. One such method, which requires that the ground motion be defined as a response spectrum, is commonly referred to as the ground response spectra analysis method. As described in Reference 2, a response spectrum is represented by a curve of maximum response motion of an infinite series of single degree-of-freedom spring-mass systems as a function of the natural frequency of the systems. Since every structure can be defined dynamically by its various natural frequencies and their corresponding modo shapes and generalized masses, it is possible to use the ground response spectra analysis method to obtain an approximate solution for the response of a structure with very little analysis cost.
The DLDRES (RuiLD.ing Efaponse) program has been developed by SAI to determine the response of a structure to a prescribed ground motion using the ground response spectra analysis method. A simplified flow chart for this program is shown in Fig. 23 As seen from this flow chart, the program uses modal analysis data (i.e., frequencies, mode shapes, modal stressses, etc.) generated by PAFEC together with-definitions of the prescribed earthquake magnitude, spectral shape and direction, the damping in the structure and the method by which modal response data is to be combined to produce total response to compute i
! 10 l
i l _ _ _ . _
DEFINE INPUT DATA STEPI MODAL ANALYSIS DATA FROM CALCULATE GROUND MOTION DEFORMA-FINITE ELEMENT ANALYSIS TlON RESPONSE FOR nth MODE AT ALL
- FREQUENCY (fn) NODES ON STRUCTURE
- MODE SH APE (t n) i on = +n l'an So
- MODAL STRESS (dn)
- GENERAllZED MASS (Mn*)
CALCULATE TOTAL GROUND MOTION
- PARTICIPATION FACTOR (l'an) 5 DEFORMATION RESPONSE AT ALL 1
- MODES TO BE USED NODES ON STRUCTURE
- STRUCTURAL DAMPING e DESCRIPTION OF SEISMIC EVENT U - RESPONSE SPECTRA SHAPE (Sol
- EARTHOUAKE DIRECTf0N STEP 11 CALCULATE GROUND PAOTION STRES3 DEFINE RESPONSE '
COMBINATION METHOD EESPONSE FOR NTH MODE IN EAct! ELE I MENT ON'S,TRUCTURE ,
. SRSS n = #n l'an So
,i
. ABSOLUTE
. NRC REG.GU!OE 1.92 m CALCUL ATE TOTAL GROUND MOTION 4 "~
STRESS RESPONSE IN EACH ELEMENT -
CN STRUCTURE i
c i -
Figure 2.3. The BLDhES Pregram. ,. ,
i s h I 4
s v -
k 1
l %
e
\
e .
4 g
~
a . \
4 ; l 1
) /,
}
r v f the desired deformations at each node and internal loads (or stresses) ;
- in each member of the finite element model of the structure.
'2 5 CANISTER SLIDING ANALYSIS
,s Ab impbet, ant consideration in determining the seismic response of the g- storage racks and canisters in the FRS pool is the amount of resistance th'a - canisters provide to the dynamic response of the storage rack
' +
-structure.
The' Misters are supported on the racks by a support ring, which is
- ' notched [ to slide over the lips of the rails of adjacent racks (Fig.
~
2 '. 4 ) . lThe support ring has a two (2) inch high by one quarter (1/4) j inch thick cylindrical lip which contacts the support rail over an arc length of. approximately 12.7 inches. In its nominal position, the lip of_ thr rack has a one cparter (1/4) inch clearance on all sides of the
,, suppcrt ring notch. Therefore, there are no structural connections 4
botacen the canisters and the storage racks.
~
Consequently, except for gravity oc . vertical forces, no lateral force can be developed between the canisters and rack except for that developed by friction between the two ' parts. If sufficient friction is available, the canisters could provide additional stiffness between adjacent racks and hence influence their dynamic response, c
The canister restraint between adjacent racks has been addressed in this report by developing finite element models of multiple racks with the canister restraint between racks represented by equivalent bar members. The dynamic response of the multiple rack system is then obtained by the procedure given above in Section 2.2.1. Forces obtained in ,the, canister equivalent bars are then evaluated to
, determine whether they are within the bounds of force which could be
{ -
developed . by consideration of known ranges of the ceafficient of friction between the two surfaces.
The coefficient of static friction between dry surfaces of aluminum has a wide range' end has been reported to be as high as 1.9 The effect of water submergence on the coefficient .of friction is not well documented but a reduction is probable.
The consideration of canister restraint and sliding is included in the seismic response analysis of the storage rack and canister system given in Section 6.0.
f a n
, , 12 e-
, t 7, g
'M
i i I L
,a m.. 1 p
s, w
't..
s .
30' 6.5 :
.- 20 diam -. (approximate) I 1 l g_ ... B M ^7 12 .tii.d.
.. 8
, Enlarged section h 7.5 7.5
-____-- I p- - --
7 anister C
\ upport S ring l <
4 ' i' I
""~
O R Typ. ; Fuel storage l4 6 _L;. . ; ,/ rack beam f
000 12 f t.1 1n. 1 - - - - - - - -
h25 or i i i 16 f t.1 tn. E - y - - - --,I u ~.,)
/ 0.101 wat!
/ '(aluminum) . . ..
0 0 Dimensions in inches (except where noted)
- Canister
- - Figure 2.4. Detailed Drawings of Fuel Canister and Support System.
f 13 i
l 3 FACILITY DESCRIPTION This section contains a description of the Storage Racks and Canisters located in the Fuel Receiving Station (FRS) Pool. The current arrangement of fuel canisters in the pool are described in Section 3 1, and the structural description of the storage racks and canisters are given in Section 3 2.
3.1 POOL ARRANGEMENT The FRS pool is located in the Northeast corner of the Nuclear Fuel Services Reprocessing Plant as shown in Fig. 31 and consists of three cells. These cells, as shown in Fig. 3 2, include the water treatment, cask unloading and fuel storage cells. The fuel storage cell contains the storage racks and is filled with water to a depth of twenty-eight feet. A cross-section thru the. FRS pool showing the storage rack area is shown in Fig. 3 3 The fuel storage pool contains forty-two storage.
rows with space for twenty-one canisters in each row giving a total capacity cf 882 canisters. As shown in Fig. 32, the arrangement of the canist ;rs on the racks is such that one canister is supported by two adjacent racks. Therefore, there are forty-three storage racks.
Presently, fuel assemblies from five different sources are stored in canisters in the pool. Table 3.1 lists the f'2el source, weight and size for the assemblies. The present location of the various types of fuel in the pool storage rows are shown in Fig. 3 4. It can be seen from this figure that several empty rows exist as well as rows of empty canisters.
3.2 STRUCTURAL CONFIGURATION A structural description of the storage racks and the canisters are given in following subsections. A list of drawings used to describe these components is given in Table 3 2.
3.2.1 storare Racks A sketch of a single storage rack frame is shown in Fig. 3.5. The rack is constructed from 6061-T6 aluminum alloy structural members. The top rail on which the canisters rest is an extruded beam which features edge lips over which the canisters slide when they are placed on the racks. The rail member is made in three pieces with splices occuring as shown in Fig. 3.5. The complete structural details of the storage rack are shown on Bechtel Corporation Drawing 1A-M-7 The rack is anchored to the North wall by four 3/4 inch diameter bolts on the top rail member and by two 3/4 inch diameter bolts on the lower channel section member. The six vertical columns of the rack frame are welded to transversely oriented base beams at the bottom cf the pool. The 15 l
1
,w-y y.. --n_. g 3 --
4 2
U
\
J M 7 anipulator service module repair shop
'7 N
- t_
1 '
/ t , , ,/, , , / ,',, ,. .
l
', Fuel receiving station .
Offices e /- '//// *//_
J 1
1 Process building
(
s 3
.r e
Transfonner Utility roorn - - ,
lodine andidac!C recovery cell C 3 4
) % \j
' J Figure 3.1. Plan View of Reprocessing Facility.
l 4
t 16 I
Rack beam _
r i.u r i.7s a m
i I l 1 Canisters % -
I I 1 i l 1 1 I Arrangement of canisters on racks Fuel storage cell N 75 /
-25.5- Water 4 -- re tment
\s (T e ,
() \ #
15 14.5 36 :fiack3 ' g g
-- 24 i _ \ I2 4 0.75 -
3.5 l
1.5- 106 26 -
unloading Gate cell Dimensions in feet l \
\ \
l I Figure 3.2. Plan View of FRS Pool.
17 I
l
75 WATER SURFACE N
/ ////
RACK AREA H //// / .
EAST WEST ELEVATION 40 _
1{ ,
r -
11 1 2"
17
/ ////
RACK AREA
~ //// / s NORTH SOUTH ELEVATION NOTE - ALL DIMENSIONS IN FEET Figure 3.3. Elevation Views of FRS Pool.
18
i.
i
] Table 3 1. Description of Fuel Assemblies Stored in FRS Pool, i
4 Source. Weight Size Length (lbs) (inch square) (inch) l 1,
- Consumers Pouer j Big Rock Point 465 6.5 82. and 84.
Wisconsin Power j (WEPCO) l 1390 7 763 159.5
) Commonwealth j Edison-Dresden I 300 4.4 134.0 l Rochester Gas and i
Electric (RG&E) 1300 8.0 160.2 j Jersey Central 615 (*30) 5.263 173 0
.)
i i
l i
i i
3 1
19 l
4 4
,, ,--.,-w,,,,yv ,. c--,,...,v.,--.v. --,,,,y.r,.,--. -----,.---y,-,.-.,-r,,- , , -- - - - - - , ,-,,- --%,I
42 40 35 30 25 20 15 to 5 1 A A A A A A C B B B B C C C C C D D D D D D D D D D X X X X X X X E E E E E A A A A A A A C B E .B B C C C C C C D D D D D D D D D D D X X X X X X X E E E E B A A A A A A C B B B B C C C C C C D D D D D D D D D D D X X X X X X X E E E E C A A A A A A C B B B B C C C C C D D D D D D D D D D D X X X X X X X E E E E D A A A A A A C B B B B C C C C C D D D D D D D D D D D X X X X X X X E E E E E A A A A A A C B B B B C C C C C D D D D D D D D D D D X X X X X X X E E E E F A A A A A A C B B B B C C C C C D D D D D D D D D D D X X X X X X X E E E G A A A A A A C B B B B C C C C C D D D D D D D D D D D X X X X X X X E E E H A A A A A A C B B B B C C C C C D D D D D D D D D D D X X X X X X X E E E J A A A A A A C B B . B C C C C D D D D D D D D D D D X X X X X X X E E E K A A A A A A C S B B B C C C C D D D D D D D D D D D X X X X X X X E E E L A A A A A A C B B B B C C C C D D D D D D D D D D D X X X X X X X E E E M A A A A A A C B B B B C C C C D D D D D D D D D D D X X X X X X X E E E N A A A A A A C B B B B C C C C D D D D D D D D D D D X X X X X X X E E E P A A A A A A C B B B B C C C C D D D D D D D D D D D X X X X X X X E E E R A A A A A A C B B B B C C C C D D D D D D D D D D X X X X X X X E E E S A A A A A B B B B B C C C C D D D D D D D D D D X X X X X X E E E T o A A A A A C B B B B C C C C D D D D D V D D D D X X X X X X E E E U A A A A A B B B B C C C C D D D D D D D D D D X X X X X X E E E V A A A A A B B B B C C C C D D D D D D D D D D X X X X X X E E E W A A A A A B B B B C C C C D D D D D D D D D D X X X X X X E E E X i t i, Note e The canister arrangement shown on this flyure was existing as of 8-19-80 LEGEND A - RG & E D - JCPL b - CONSUMERS POWER E - CON. ED.
C - WEPCO X - EMPTY CANISTER Figure 3.4. Existing Canister Arrangement in FRS Pool.
Table 3 2. List of Drawings for Storage Racks and Canisters.
Drawing Number Title 1A-M-7 Fuel Storage Rack 1M-1, Fuel Receiving Storage 1A-Q-8 Anchor Bolt Setting Plan and Sections, Fuel Receiving and Storage Building 1A-Q-3 Foundation Plans, Fuel Receiving and Storage Building 1B-T-6 Fuel Pool, Fuel Canister IV-6, Fuel Receiving and Storage Area 1B-T-1052 12-1/2" x 16'-1" Fuel Storage Canister (IV-6A)
)
I l
4 l
}
I 21 1 l
1
8 ea. % in. diam. bolts (typical)
Detail
.. -..t.n
.i.-y,----- splice 4 ea. 3/4 in. diam ;
' ll ' anchor bolts li a,
2 ea.
I 3/4-in. diam g b t \
_L 5.5 0 0 6 - ,-
1 in. [___ i , co */S Exwded 7[3.6 1/4-in.
plate end stop /
[
/ ;
// _
- . s -
2 ea. 5/8-in. diam anchor bolts 6W F9.8 columns \
l
/~,i '
- l at each column All members and fasteners are aluminum m--,3 Note: The data on this 6WF9.18 sketch was obtained frorn base beams Bechtel Corp. Drawing No.
lA-M-7.
Figure 3.5. Layout of Typical Storare Rack.
22
l beams are anchored to the bottom of the pool by two 5/8 inch diameter bolts at each' column location.
3.2.2 canisters A sketch of a typical canister is shown in Fig. 2.4. The canisters are all made from 6061-T6 aluminum alloy. The shell wall is 0.101 inches thick and has an inside diameter of 12 5 inches. The canisters are manufactured in two configurations one of which is 12 ft. 1 in. long and the other of which is 16 ft. 1 in. long. These two configurations are required to accomodate the various types of fuel assemblies. The support ring, which supports the canister on the top rail of the storage racks, is twenty inches outside diameter with notches provided for sliding over the lips of the rails. A top lifting ring is also twenty inches diameter. Details of the canisters are given either on Bechtel Corporation Drawing 1B-T-6 or Nuclear Fuel Service Drawing 1B-T-1052.
t l
7 23
- 4. MATERIAL BEHAVIOR All material used in the storage racks and canisters is 6061-T6 aluminum alloy. All finite element nodels used in the seismic response analysis of Sectiog 6.0 assumed that the modulus of elasticity for this material was 10x10 psi and that Poisson's ratio was 0 3 A mass density for the material was assumed to be 0.098 pounds per cubic inch. ,
4.1 MATERIAL STRENGTH For the determination of member and joint structural capacities, yield strength was taken as 40,000 psi and ultimate strength was taken -as 45,000 psi. These values were obtained from Ref. 7 and are typical or average values for the material. A typical stress-strain curve is shown in Fig. 4.1.
When 6061-T6 aluminum is welded the resulting weld and heat affected zone has considerably less strength than the parent metal. For weld capacity calculations in this report a value for yield stress of 17,000 psi and a value for ultimate stress of 27,000 psi were assumed. These values are typical values and were obtained from Ref. 7 4.2 FRICTION COEFFICIENTS As discussed in Section 2.5, the shear resistance to the racks provided by the canisters resting on the top rail of the racks is entirely dependent on the coefficient of friction between the canister and the top rail member. Both of these materials are 6061-T6 aluminum. Values of the static coefficient of friction between dry aluminum surfaces has
, been reported to be in the range of 0.5 to 1 9 The reason for such 2
high values appears to be caused by galling between the surfaces. The effect of water submergence of the coefficient of friction is not well
, documented. It might te expected that the water environment would have the tendency to lower the coefficient of friction but no data was found to substantiate this.
l l
I l
l l
25 l
l
. . _ ~ -~ .- _ ._ _ _,
l l
l l
l i
50
+o '
p.
p%
i.
o 7- 4 c g to S & k,0.00/ in/4 Figure 4.1 Typical Stress-Strain Curve for 6061-T6 Aluminum Alloy.
l l
l l
t 26 i
l I
l . - . _ . . . - - - - . . - . - . - . . - . . - . . . , . . . . -
- 5. STRUCTURAL DESIGN REVIEW This section contains the results of a structural design review of the storage racks and canisters. The review was made in order to evaluate the rack's structural design characteristics as they might. affect its seismic response capacity. ~I 5.1 STORAGE RACK STRENGTH The structural design of a typical storage rack was reviewed in order to determine the member and joint capabilities in terms of the magnitude of force and/or moment that could be resisted before the occurence of failure. Failure, for the purpose of this design review, is defined as the force and/or moment which causes stress levels in the material to reach ultimate or yield strength levels. The ultimate strength and yield strength for the 6061-T6 aluminum members and anchor bolts was taken as 45,000 psi and 40,000 psi, respectively (Ref. 7). For welds, the ultimate stress and yield stress were taken as 27,000 psi and 17,000 pai, respectively (Ref. 7). These stress levels are average values of ultimate strength and thus represent realistic conditions upon which to base actual failures in the storage racks.
An examination of the storage rack's design led to the identification of several potentially weak structural connections. These were the North wall anchor connection to the top extruded rail member, the splice connection of the top extruded rail member and the floor anchor attachment to the vertical members.
i Detailed calculations on the capability of these connections are presented in Appendix I. A summary of the yield and ultimate moment capacity of these joints is given in Tables 5.1 and 5.2, respectively.
The significance of these joint moment capacities can be seen if comparisons are made with the ultimate moment capacity of the members themselves. For example, the top rail extruded member has an ultimate moment capacity of 1,178,600 inch-pounds for out-of-plane bending (See Appendix I for calculations). Comparing this value to the moment required for anchor bolt failure at the rail / north wall connection from Table 5.2, it is seen that the bolts fail at only sixteen (16) percent of the member capacity. Also, the connection used to splice the rail fails at only about six (6) percent of the member capacity. These
( weaker points in the structure become points of moment limitation for
! dynamic response of the structure under a seismic event. These joint l capabilities will be utilized to determine thresholds of failure in terms of ground acceleration in Section 7 0.
l 27 l
Table 5.1. Yield Homent Capacity of Structural Connections in Storage Rack.
Yield Moment Connection Capacity Description of Failure (inch-lbs) s.
Top Rail at North Wall 168.560 Anchor Bolt Tensile Failure 186,300 Weld Failure Rail Splice 64,200 Bolt Shear Failure (Out-of-Plane Bending)
Column at Floor 56,420 Anchor Bolt Tensile Failure (In-Plane Bending)
Column at Floor 225,400 Weld Failure (Out-of-Plane Bending)
Table 5.2. Ultimate Moment Capacity of Structural Connections in Storage Rack.
4
" Ultimate Moment Connection Capacity Description of Failure (inch-lbs) 1 Top Rail at North Wall 189,600 Anchor Bolt Tensile Failure 295,900 Weld Failure Rail Splice 72,230 Bolt Shear Failure (Out-of-Plane Bending)
Column at Floor 63,470 Anchor Bolt Tensile Failure (In-Plane Bending)
Column at Floor 393,800 Weld Failure (Out-of-Plane Bending) i 28
5.2 CANISTER STRENGTH The question of canister strength must consider the potential loadings on the canister during a seismic event. As discussed in Section 3.2.2, the canisters are basically a long cylindrical shell with a support ring which supports the canisters on the storage racks and a-top lifting ring for handling. The spacing of the canisters are such that the top rings of adjacent canisters are nominally only one inch apart.
Therefore, as long as the canisters remain on the racks, contact between adjacent canisters could only occur at the top ring. Consideration of the close spacing, weight of the fuel and canister combinations, and maximum potential accelerations lead to the conclusion that any forces of impact between canisters at the top ring will be relatively low and will not cause damage to the canister.
The only feasible scenario which could lead to canister failure appears to be an event leading to the canisters being knocked off the racks and falling to the floor of the pool. Sliding of the canisters along the storage rack top rail is prevented at the south end of the rail by an end plate. The canister support ring would normally impact this end plate and prevent the canister from falling off the south end of the rack.
Calculations performed in Appendix I show that a very large force would be required for failure of the end plate. The possibility that such a large force would exist is highly unlikely during an earthquake of the type that might be expected to occur at West Valley (e.g., about 0.2g ground acceleration). However, if it were assumed that canisters could fall off the ends of the racks either as a result of end plate failure or some other unexpected behavior, then they would be limited in movement in the South direction by the South wall of tha rol which is only four feet from the ends of the rack. The canisters could then hit the south wall and fall in an East-West direction to the bottom of the pool. This could cause large localized loads on the shell of the canister and possibly cause crushing at the points of impact.
29
i
- 6. SEISMIC RESPOBSE ANALYSIS The seismic response analysis of the storage racks and canisters presently existing in the FRS pool at West Valley is presented in this section. The analysis methodology and assumptions used to perform this analysis were previously discussed in Section 2.0. Consideration of water sloshing in the pool and its effects upon the behavior - of the
- storage racks is given in Section 6.1. The seismic response of a single rack is discussed in Section 6.2. Section 6 3 contains the j analysis results obtained from consideration of multiple rack response.
The effects of canister constraint between adjacent racks is also considered in this section. Finally, the dynamic response of a single canister is discussed in Section 6.4.
6.1 POOL SLOSHING l The motion of the water in the pool during an earthquake was evaluated in order to determine the significance of hydrodynamic effects.
4 Calculations for the sloshing natural frequencies and for the maximum vertical displacement of the water surface are given in Appendix II.
These calculations are based on formulas taken from References 8 and 9 Calculations were made for a depth of water of twenty-eight (28) feet and also for a depth of water of eleven (11) feet. The latter depth was used because, with the storage racks and canisters present, the depth of water over the canisters is only eleven (11) feet. The first mode natural frequency for sloshing in the long (i.e., East-West) direction, with a depth of water equal to twenty-eight (28) feet, is 0.17 Hz. For a depth of water equal to eleven (11) feet this first mode natural frequency drops to 0.12 Hz.
Calculations to determine the maximum vertical displacement of the water surface were completed for both depths of pool water.
Consideration of the interaction of sloshing in both the long and short directions of the pool was also taken. Maximum vertical displacements were found to be 3 5 feet and 4.5 feet for the eleven (11) foot and twenty-eight (28) foot pool depth, respectively, for an earthquake of 0.2g ground acceleration.
This amplitude of slosh is such that some water could splash out of the pool for this magnitude earthquake. However, it is felt that very little water would be splashed out and that the canisters and racks would still be covered to a sufficient depth. Also, it will be assumed that no additional hydrodynamic pressures are induced on the canisters as a result of surface sloshing since the canisters and . .'ks should always be sufficiently far enough below the surface of the water to minimize such an effect.
31
l 6.2 SINGLE RACK RESPONSE l As described in Section 3.0 and shown in Fig. 3.4, the pool has l forty-three (43) storage rack frames, which provide storage rows for j
the canisters which contain the various fuel assemblies. Considering l the present arrangement of canisters in the pool, it was assumed that a first approximation to the dynamic behavior could be obtained by analysis of a single rack frame.
The single rack response approach considers only a single storage rack frame to be effective in providing support to the canister mass. That is, no loads can be transferred from the storage rack frame being analyzed to adjacent storage rack frames.
6.2.1. Finite Element Models A finite element model of a single storage rack was formulated using the structural members defined on the facility design drawings. This model, designated Model R1 and shown in Fig. 6.1, is a planar model consisting of bar elements with restraints at thq anchor points in the north wall and at the bottom of the pool. The' details of the model formulation are given in Appendix III.
The primary variable for a seismic analysis of this model is the value of mass to consider for the canister plus fuel assembly plus effective mass of the water. As noted in Section 3 1 and Table 31, there are presently six (6) different types of fuel assembly stored in the pool.
Also, empty canisters are present in the pool. With consideration of the canister and fuel weight in water, a range of weights is obtained as shown in Table 6.1. This table gives potential weights of canister and fuel for both the sixteen (16) foot and twelve (12) foot canister lengths as well as weights with and without the weight of " trapped" water inside the canister. Since holes exist in the canisters to provide for water circulation, the amount of truly trapped water is difficult to determine. In addition, the amount of added mass due to the hydredynamic effects of water-structure interaction is difficult to assess because the response of a single canister may be such that only a portion of the canister may be moving in the water. (See Section 6 3). For these reasons, it was assumed that the upper bound of canister mass would be that represented by the Wisconsin Electric Power Co. (WEPCO) fuel assemblics with trapped water included. From Table 6.1 this value is seen to be 1,627 2 pounds. It is seen that without trapped water the weight drops to 1,119 3 pounds for this assembly. This should represent the possible range of mass for this canister / fuel configuration in the pool.
In order to differentiate between the two values of mass used on Model R1, the model with a canister weight of 1,627.2 pounds was designated Model RIS and the model with a canister weight of 1,119 3 pounds was designated Model RISLT.
l l
l l 32 l
l
i i
?
?
4 r
+
3 o
4 lot.
i
- ]
, 4 6' ' 5' t
,, g I
- 80 1
< c ~ tal A*'M #Au kT ,4
'[
l14 5'..
11 7 /Il0 107 30
, / fo3 J
z A
'4 W '$os l %
- lly / ' op s
'y / "'I L
' \
/ IC l8 f
YA R ute 5,3* Finite gy'nent po,ge, qy, 33
, , , - - - - - - -. , , . , - r
Table 6.1. Canister Weight Summary.
16 Foot Canister 12 Foot Canister Weight Weight Weight Weight With Without With Without Fuel Type Trapped Trapped Trapped Trapped Water Water Water Water (lbs) (lbs) (lbs) (1bs)
Empty Canister 931 3 76.3 705.5 62.9 Rochester Gas and 1,611.6 1,066.3 1,385.8 1,052 9 Electric (RG & E)
Consumers Power 1,139 9 413 1 914.1 399 7 (Big Rock Point)
Wisconsin Electric 1,627 2 1,119 3 - -
Power (WEPCO)
Jersey Central 1,200 3 518 3 - -
(JCPL)
Commonwealth Edison (Dresden I) 1,156.0 488.3 930.2 474.9 (2 Rods /Can)
Note - All weights assume can is submerged in water.
34 l
l l
-,_.~_,_,m.---- . - - - ~ - - - - - - - ~ ~ - - - - - - - ' - ' ~ ' " - - * " - ~ ' ~ ~ ' ^ '
l 6.2.2. Modal and Seismic Resoonse Analysis Results The PAFEC computer program was used to perform a modal analysis of the two single storage rack Models R1S and R1SLT. A computer plot of the first mode shape for each of these models is shown in Figs. 6.2 and 6 3, respectively. Additional modal analysis results for each of these models are presented in Appendix III. These analysis results were used as input to the BLDRES program in order to determine the deformations of and internal loads in the structural members of the rack when subjected to a seismic ground acceleration in each of the X, Y, and Z directions. Some of the results from this analysis are surcarized in Table 6.2 and presented in more detail in Appendix III.
63 MULTIPLE RACK RESPONSE In order to assess the possible effect of canister constraint between adjacent racks, a study was made to evaluate the seismic response cf a group of racks and canisters. As discussed more fully in the next paragraph, a five (5) storage rack system was selected for this analysis, based on the present configuration of canisters stored in the pool and the weight variation of fuel stored in the canisters.
From the canister arrangement shown in Fig. 34 it is seen that the heaviest canisters are the ones which contain fuel from Wisconsin Electric Power Co. (WEPCO). Also, because of essentially empty rarks on each side of the group of five (5) racks supporting the WEPCO fuel, it was decided that this group would be a logical choice for studies of meltiple rack response.
The purpose of this multiple rack analysis was to determine the conditions under which the canisters could modify the seismic response from that predicted by the single rack response analysis of Section 6.2. As discussed in Section 2.5, canister constraint between adjacent racks is essentially dependant on the coeficient of friction between the aluminum surfaces of the canister support ring and support rack top rail.
The approach taken to evaluate canister restraint between racks is described in the finite element model descriptions and modal response results given below.
6 3.1. Finite Element Models The first finite element model formulated for the multiple rack response study was the five (5) rack system shown in Fig. 6.4. This model, designated Model RS, consists of five (5) storage rack frames identical to those used for the single rack finite element model. For this model the effect of canister restraint between racks is simulated by equivalent bar elements between the racks in the plane of the top support rails. The properities of these egivalent bar elements were obtained by the determination of axial and tangential stiffness 35
r
- r 3 i.
E n o
e3s sEsu s
en e
- e. Sa e. e.
s e n5
. r ,e
.. e m z7 s -
e s
s
_ e v.
s
- .m u
s _ r r ee - ms e , _ sm t
n e.
s e es i
v sa ew s sa e
s r a ne as s
n mg--
e e ,
s"7 o.
t 1-e r. 09 3
e vH g.T N y.
e n,re s iA L
P
_ N .
_ t i 1
- r s
e s - e i
- b ms m e
__ u t _
N H
m
- e a -- e d N s.
p o s '- h a ep M e
r y
, - S h
a -
ef l , - d S d e S t
R i
a -_ n n r d l L
E R I e
g r o n e R r
W e e f r l l
L e of e H d ' . d s
t n e d A
A d
i ' .
U D M
o t M _ -
s l I A i - 3. e R
S r R a - - - n. .
g u 2 L p _ - - ,.
E e 6 W o - r F- T a e f . i l-e
- e r
N.
- - u T
s L
E V
w a
i F
g f -
e -
w 1 gg s
t t
e r jX a.
L@
_m fifts Wst VM1Ev.sv-fusL sTestatt RAtst-west Res.ilvrPCs.8tAssis itstisees 7 g
u me b
ttft FTUBS I e ..eSep v.
Lerend E . .u.s . .s.
Undeformed Shape
--- Deformed Shape O
s-fliddle Rail y revens virumi l - g ,-
_ _ _ . z w a - -- ,
X l - - T - ~ ~ _ ' ~ ~
l
_ _ ,I , #
m __
N Top Rail o
14EE SWE lef91ER t meexv. t,rrs wart RSWff UERP%
Innvas us, e 3e 1 as weer se. ..m e 4 r.ge jeje see, I; v, ,, Enm etas . sassaw n sattus, se se Figure 6.3. Model R1SLT - Mode Number 1.
Table 6.2. Suneary of Results for Single Rack Models.
Maximum Moment Canister First Mode Maximum at Model Weight Frequency Deflection Wall Anchor (1bs) (llz) (inch) (in-lb)
R1S 1,627 2 1.019 1.49 70,158 R13LT 1,119 3 1.223 1.19 54,817 Note - These results assume ten (10) per cent modal damping and earthquake excitation in the Z direction.
38 l
I A
- *E f i 7s ~
},o- /Q,
~
/f@%;1W'/ d
% >< > /
0~, m y' y1 >h1%,
/
Y Jk )p sA t /
- /
/
- rv
.. l" S RACK FRAMES- EAc.H
)( ,< SAME As Mopet R1 LUMPED CA41 STER /Fust MASS (TYPICAL).
NOTE- Att " Dims.ustous iu feer.
Figure 6.4. Finite Element Model R5.
1 39 l
properties of the support ring of the canisters and equating these to equivalent bar properties for use in the finite element model. The details of these calculations are given in Appendix IV. In the model a
, single bar between racks. represents three and one half (3 1/2) canisters since there are twenty-one (21) canisters per rack and only six (6) bars between racks. The details of the model formulation are given in Appendix IV. The lumped masses representing the canister / fuel weight were located at grid points midway between racks with the center of mass located thirty-two (32) inches below the plane of the top rail of the rack.
As was done with the single rack model, two values of mass were used on the five (5) rack model. The model with a canister weight of 1,627.2 pounds was designated Model R5B and the model with a canister weight of 1,119.3 pounds was designated Model R5BLT.
Three (3) additional models of the five (5) rack system were mado.
These models are modifications of Model R5 with the only change being the number of effective canister bar elements between the racks. Model R6, shown in Fig. 6.5, has half of the bar elements connecting the racks removed. The ones removed are those nearest the north wall anchor points of the racks. For Model R7A, shown in Fig. 6.6, an additional row of bars connecting the racks at the X = 24 ft. position on the rack are removed, and for Model R8 A, shown in Fig. 6.7, the rows of bars connecting the racks at both the X = 24 ft. and X = 30 ft.
postions are removed. For Models R6, R7A and R8A, the mass which was lumped at a point between the racks, for the bars that were eliminated, is distributed equally to points on the racks. Details of the finite element models for Models R6, R7A, and R8A are given in Appendices V, a VI and VII. As with Model RS, the additional Models R6, R7A, and R8A 4
were analyzed with the two versions of canister mass.
6.3 2 Modal and Seismic ResDonse Analysis Results The PAFEC computer program was used to perform modal analyses on all the versions of the five rack models described above. A computer plot of the first mode shape for each of these models is shown in Figs. 6.8 thru 6.15. Additional modal analysis results for esch of these models j are presented in Appendices IV thru VII. These analysis results were used as input to the BLDRES program in order to determine the deformations of and internal loads in the structural members of the
, rack when subjected to a seismic ground acceleration. The results from this analysis are summarized in Table 6.3 and presented in more detail in Appendices IV thru VII.
t It should be noted that analysis results from Models R5B and R5BLT are j only valid as long as it can be shown that the canisters are effective in transmitting loads between racks. Furthermore, since the canisters are only supported on the top rail of the storage rack by their dead weight, it is seen that the only method by which loads can be transmitted from one rack to another is through friction shear forces
[
40
hY Nt.75
.,, 9
/
Q$
Y
- ,- + rs< / /
o' ( >
'g s y P ~<>? / '
2 Ds ygv / ~~
, / / ,,
/
X vj
, ,, 6 Rack FRAMES - EACH
, SAME As MODEL Ri LUMPED GAMISTER / FUEL k4 Ass (TYPLCAL).
KloTE: ALL tiuEusioMS 14 FEET, Figure 6.5. Finite Element Model R6.
41
aY N / l.73
- 3 (sY' / f Y f
~
// O /
3' F s g / / #
/
S a
k y ro/yg[&/ J / f( / ~
/,.
en p 6 RACK FRAMES - EACH o n 54AE As MoDEL Rd.
Luueep GuisTER/Fuet Mass (TYPICAL')
OTE ALL DIMEMsto45 14 EEET.
Figure 6.6. Finite Element Model R7.
42
rI 1
nY Z 'D N
.+ ,> $ W((
Y
</
b Y f
~,
Y ~<ssY f
)V 'y '
$ Jk3v. / ""
< ,/ / "'
y g 6 RACK FRAME 5 - EACH SAME As hAopst Ri
)( ,
er LUMPED 6.Au tsrER./ FUEL Mass (Tveic.at)
Wore : Att tim e.a ste es io vest-Figure 6.7 Finite Element Model R8.
43
torts unst vaury.ny-rusL eronacs RACN-18ENL RSELt 4 f ts01sDal
,7 l uC 3 _-
1.I\ 1 Legend E were enes x . e.sens Undeformed Shape
[ * ,',",[.
Deformed Shape Z
16 W
X
-e-- v teuses WIEWER r ___ _--
_ . .
i
-#~.i _n_ _ _ ---
- = ==
=-,,-..-
-- ,- - - = - ,
y_
e_ . . q _ _ _ _ _ _ .
.:::= -
_ _ _ _ ~ _ ~_ ; _. t- -- ~~~- - - -
"_~;.,**~~ .- _ .z-w*=---l.~ ~ ~. ; -
~
- ~ ~ -
_- l-
-[ ,,,L l_- 1 Middle Rail (Typical)
Top Rail (Typical) nee owe sesore e rivesme . r, ego sentr etsmer _ -
e qe ge ge 3e m. en=im so. ,
ei e-r ve.as.n e' 84 'i I& li Tr gnsw MtflPLT pf to sota . e = * **
Figure 6.8. Model R5B - Mode Number 1.
~
. t
. e .
R .
_. h .
F 00 i
[
ese s s n . 1 1 e
.. M e
i s. m, s. w e e e e nt e
m .. - s. .Me e t e.
- v:
' n e s r t
e v I f
. n u re uu.s a rTC s f m
- e r WE B
S H t
M W
n.
B t r l m t S D s
u ac e oW NF t s Ws
.n __
t.
M E__. .
e , _
S"7-0e 41, __
e1 w
v _.
e ,a_
Z te,eetr _ .
,l 7 cr c[ p P 0 t . 1 1r, a l
_ _ _ n r
- l I e
b m
- - - u N
e
__ e p
d o
. a ep M C ,.
. ) .
- h e l 2 _ S a e
w a h -
e, c .- -
d S e T
- - ~ --
i d
, p - n m d y L 8
e r e B e T
( - _
g o mr 5 s - ~_ - - e f R
n )
l l
'~ -
L d
e o f
t a '
i n e l e
n c
R a
' - %sy - U D e i
p - d w
e y e ~
o a - - M T
'^
l m ( d - - -
e d , .. '- -
c n
e a l i
i M - ~- . _ . ' r- __
J
- r.
9 t a - -
n s
R J _
e n 6 a
a p
o ' - .
J _
r e
o.
4 v T Tf _
,Y]- F_ r u
n.
t e .
- r _ . _ . _-
, i g
n _ . F
'~r:- .~
a v _- - .
_ r -
a .
~- _
ef.l1p# '
. w _
s t
f a
r X
_ 3u
Ill 1l l'I<{l i l' 1 1
) i nE r
r a s e
wEse m s swe, o SE e.
a.
e e ,
g s..r. .
n, e
e s
a s
e
, Se e
s g w m
y ,
r.
wy aem re e
suw a t s
s e
t n sss e
s mm a ss os s
.a a -
e e e s
a v
e n
oE s
e e ee g_7 e G e, eva e
Z q r int o rls s s w s
i se a
_ _ _ m .
__ i 1
. - r e
- - e
- b
- ,2, m u
2 N e
) __ Z . 2 _
~~__ p e l a ep d q_. , -
a _
c h
S a o i h M
)
l a
c
(
T y
p
~
~
'N_Q-
._ - '- . ~
~
~
d g
n e
d S e
m d r e o mr 6 R
i p
l F' T-p. a,
. e f L e of
_ T y
R i
a
~
\_4 -
~
d n eD l
e
( ~ __ - _ 1 d
_ - 1 e - - _ o l
i l
d
_ M a d . .
._ R i
M : -
- s .
_ : a. 0 e p - s,
. o : _ - 1 T _ _. . : __
e 6
_ J,y_ : r
. - l.)
e s e r
- : _ e u e i g
r
=
m F
_ _~
x m
I sins wat vausy.sv.fWEL STORM RAtN-NEDGL 200 8tfstisDet 7
' ' ~ ~
. 0=.
Legend o m nEse . e.asse Undeformed Shape , " ,l",,,
--- De formed Shape Z
h z.--_
X
~ __ f tauses vergun
_ _-;;=.-
_ .s
___.._ - _.1 , - -
g i
- - _ _ .i- c 4,. - -. _ ' c = == - rr
~~
.~--'-
_l. .e-'f ,'3' ~ [
. . -__ ._:= g
. ;;=- ,
N _~-- -
.l_- ~ '_~~ 1, , # ; ; ; ; ; = ' F 7__ J -
L Middle Rail (Typical)
Top Rail (Typical) lEEE Stet 1880E51 t FTWacut? - 4410 genft ames _ _
e . Se qe 3s_ se go pes sne se om. t
= s - rw.u.= i se is 1.7.17, seas - eessee .*
nununw IMfinna Figure 6.11. Model R6LT - Mode Number 1.
3 t
- m .
'- E s es os vue e i e ae 7' s.
e e a
- s. w e
r J
n1 r
e 4
4 svr s
e I
e u
a e
s e
s s.
w a
m *
- wyc s see u Y an r. =.
r i
ra e n et e
v e vw e wu at n so m
n
_ mg e
e sf,
,r
_ e1 r
_ere
,3 s gr e-r e Z g r ntw
-d ; m *E y [ F e- t t .
_ _ _ _. _ es 1 r
- , e
~ - - b
- - m
~ -
u
~ N e
e s
2 ~- ~.. - 2 e d s, ) ' _ ~ -- - -' _ p a ep o l !s,1 :.l h M
, a ~ - -- - - S a s
e c - -
s i , = = e -
c .
h p d S y ~~ ~ - ~ - d n
e m d 7 A
u T ~~ , e n ( ~ ~
g r e R t ) ~ ' .- , .-
e f o mr e l l , l m a i - = .- - L e of e a- c a - - ~ - -
d d
n R n e e
a i
y p
e
- ~- ; ; ; - U D M
o m - ~ _ -_ _ -
T l _
e -
m a
( d d T __~ ~~
r
- e s
r l
i i
M ~ _ : __ _~ _ - - , 2 e a e. 1 R ~ _ 2 _- _~ _ - - r e
e ~ _ 0 _- . 6 v . - r p
- _ 2 _- __~ e wg e
- e. o ~ .
. - s r :
T r y
r
- n. _: .-
. - u u
v a - _ : _ _ -_ -
- - _ - m i
F t
s e -(. .- .- -
s t
r a
t X
e*
m~E
't
=
essu s ,
t f
n e
e 1s t t
7' sao
- e.
- s. ,
e e ,
, tv.
9 t
9*
.. 5 n 9 e vt P
f m
u r
e : ,
N.
T Wx t
t F
R E
B C
T I
fe r.
N
=
r f e B C. e 1 s ve 5wF B wu g
g e vI R E s u
n e.
t P M e~
jTe e;
4J -
e ;r es r
1F.e e
Z t n ,
- r
- t F EF rp Es sr .
e e - 1 t
ruin
_ r
_ _ f i e
. b
_ - - - m
- - - -. u
- - - - N e
s
, - ~ .. - ,. d e
e o
t
- ~. ~- ~- - M t
i
)
l
~ -- ~- - e i
v e
s a
c gs - s .sss.
p a ep T i , , q h p = - S a L t.
a y -_ -- - ~ - d S h o A 7
i n (
T , , _ - d e R
) , , , , , n m d t
r e r e l
~=
l g o mr l e a i = = = _P e e e c a ~ : ; ; _ L f
e of d w i R s p ; ; ; d o e
sw T
(
y l
d e ' _- _
- n e U D M
T 37>1.-
r }
a n
d . . . 4. I 7 o e
l i
i H _. . . _ .. ~ .
- ,. 3 t
s a 1
~ _ _. -- ~ - - e.
. R - e m
u p . _.. - - .. - - -
- t 6
r-t o .
. - .- - e
- r T
- n. - ._ .
- - u t
r g u - _ - .- - - ,
i a
v -
. F t
e
[. . . - .- . .
w e
t t
i t
x c@ _
l , 11 i T
m E t
M r '
eo "se se a e.
T 8 9 'f a J 14
=- - P m%
u t e
s e
8?I
- n- f R
E ,m
'N "
7 m?r F
E te s
_- f eU F
M e.
T D W s.
" wR xT D Wa MF L E oer 4
M y
9 e
]d 8 e g7 9
1 Te09 t
re7 9 q f nat
,?n ,
z$lF r # 9 P
F t
_ wi .
M 1
__ __ _ r
- e
- b
, - r u
- N 2 - ~- e 2 2
~- d s
s ) - e o e l '_ Z p M
t e
a !
!I I. 'a , a ep c h t
s
- - - S a -
e i - -
e p - - - h s y - = -
- d d
e S A a T S s ( ~% n c d R
- e r o ce o
N_W*-
) g t l ~,? l -
r a
l
- e f r l e c l
a ~
- _
L e o e s
m- i p
F ~
d n e f d o
o n y e _
'- , l' D M a T l
_ _- 2 7-
'-,F+
s ( d -
r a d ni - , .
l i
_ -:_~ ~ ~ _ s 4 t
s -
e i
s i
a M ~
- ~ - m 1 R - ~ - -
. s. .
. s a - ~ - - e 6 v p - - - - - t e c - - . - e r
T e 1. , ,-t- T-l-n- e r
a.
v r
L L
a
.l
- ~.
m e
i g
i F
u v -
- - - - s r - - -
a ~
- ~ - - -
w s
t r
e t k
, c:
l l 1l il
r 7
i n "
E e e i
= m n se e.m s.
= o w.
g '
e e w
r i
v ni r ,*
. .- e e
e e *=
?r e m r m
u n
e v v. a s
o =
l I
r y
e y
we or .
==
r r= a. r e
e v=
v er u n i===
u n
g mg 3eT
.. : E' gJ L
e*W g
sV D
.'~?
3a
'8,,.
Z e ,F e ln i 1
-d gr f
- c_ _
t t
I r
e
- . b
. m
.. - . u e N
. - e 2 ;- : d
- ~:_
s o
e - e M er
_ ' _ 2 p t - a ep -
s n )
I '
~1 1. _ ' . _ h i
v l
a -
. S a T e
c _ -
h L r i
._ , s.r d d S e A p - n 8 u y - , m d R e r e
- b. W-' . -
n -
t
)
l T
( - l g
e f o mr l r a ,
e , L e of e c l - 3 -s d d m- - -
a i
p i
a - = - ' - n e U D o
M c y R .
A n T 0 -
e ( e . ! - j-i_ _- _ i -
l e .
u e
l i
d d -
s.
r 5
- . _ - - 1 t
e R
a i M -
e.
e n -
- r r 6 u p - ._ ae e
, o - - - -
e t. l t_ r T ,
m.
r e
- - w u
u g
n a -
i s
i F
v _ - _
r -
a w
{_ .
s t
r X
w-l l(
Table 6 3 Summary of Results for Five Rack Models.
Maximum Moment Maximum Canister First Mode Maximum at Canister Model Weight Frequency Wall Anchor Deflection Bar Shear (1Ls) (Hz) (inch) (in-lb) (1bs)
R5B 1,627.2 1 749 0 79 97,742 9,774 l
l R5BLT 1,119 3 2.098 0.63 76,856 7,719 R6 1,627.2 1.174 1.06 79,823 3,747 R6LT 1,119 3 1.410 0.84 61,459 2,846 R7A 1,627.2 1.142 1.14 71,103 3,040 R7 ALT 1,119 3 1.373 0 91 53,509 2,114 R8A 1,627.2 1.138 1.21 65,835 1,888 R8 ALT 1,119 3 1.366 0 97 51,544 1,431 Note - These results assume ten (10) per ceu i, modal damping and earthquake excitation in the Z direction.
52
I between the rail and the canister. The shear forces in the canisters for Model R5 are plotted in Fig. 6.16 as a function of distance from the north wall. Also plotted on this figure is the maximum frictional force that could be developed for the canister mass under consideration for coefficient of friction values of 1.0 and 2.0. It is seen that the shear forces from the north wall outward, to about the center of the rack, exceed forces that could be developed by friction. When this _
happens the canisters begin to move or slide around on the rack. ]
Therefore, the results from Model R5 are not realistic. This ,
conclusion led to the development of Models R6 and, subsequently, to Models R7A and R8A. The maximum values of shear in the bar elements between racks are plotted in Fig. 6.17 for each of the models considered. It can be seen that, for any of the models to be valid, an unrealistically high coefficient of friction would have to exist or else that the canisters begin to slide and loose their effectiveness .
for stiffness. An exception to this could possibly be Models R7A and _
R7 ALT. For those models to be valid the coefficient of friction would have to be 1.55 for Model R7 A and 1.08 for Model R7 ALT for the case of a 0.1g earthquake. These coefficient of friction values may be feasible.
Based on these analyses of the five rack system it 13 generally concluded that the canisters will begin to slide and/or move around on the racks at low ground accelerations and, therefore, will not be effective in providing stiffness between storage rack frames at higher '
ground accelerations. This leads to the conclusion that tbc single rack model can be used to predict the response of the system and that racks will tend to respond individually, n
6.4 SINGLE CANISTER RESPONSE An individual canister with its contained fuel assembly is supported on a the storage rack in a manner that makes its response to motion of the rack similar to that of a rigid pendulum. If the canister is assumed to act as a rigid pendulum, hinged at its support point on the racks, ,
its first mode natural frequency, considering an added mass of water, is approximately 0.20 Hz. Because of this low first mode frequency, the canister motion is likely to be such that the motion will be the same as the rack at the support point but very low at the lower end of the canister near the bottom of the pool.
This behavior will affect the amount of added water mass to be >
considered in the dynamic response of the storage rack. _
The design and spacing of the canisters are such that the top lifting ring of twenty (20) inches in diameter and spacers at the bottom end of the canister prevent the cans from moving more than one (1) inch before impact with an adjacent canister. This restricted movement capability should prevent any damage to the canisters from adjacent canister impact as sufficient energy could not be generated to produce high enough forces to damage the canisters.
53
I* i i i i i MODEL RSS LE4hdD R650LT5 EbR MoDELS War 4 ADPED WATEP.,
2s* ~
RE50LT3 FOR MoDEL5 -
, Moott RGBLT wirHour AppEp WATER v.
(Aul5TER SLIDiW4
'3 N 7HRE5 HOLD (C57) FOR s
zooo - N s cc6FFici&+1r CF _
N g FP_le.Teod = g.
- N e -
s 6--
[ \
\
isoo -
s _
N s 4: \
v More5 %
a' .
\
? (i) RE.5045 PRESE4TEO 04 .
.\. 'I MR-looo Ul5 Fl4UR6 A S S U M E. \ K= 7.0 019 4:: E 0 0 H D g \ MopQ R7A 7 A<.46Lec4T1od s s
- s (7) d ut 5 9 2.5 W lu. S u D 6 s IF " die $ HEAR l C 5 T. F O R , \ .N .
Fogca Pet udesTee & " '- \
g N -
EX4Ecos T11E Gdl5TER \
SLIDIM4 Tlif45 HOLD. gpg g [ _' }
MoDEL R'7 ALT o i I I l i i O 6 12 18 24- So %
VISTAucE FROM MoRTH WALL (FEET)
F l y,u re 6.16.
Caninter Shear Force vs. Distance From North Wall.
54
sooo . , .
i MOTE 5
. (.0 REsuLTS PRE 564T&D on TNis F% ORE A550ue. o.(9 4Roudo Ac.4Ete2.Ariod.
2" -
(O dAut511E25 WILL SLIDE IF THE MAytMuu -
n THEAR FOR4E PER CARLSTER M4EtDS y '
THE CAdl5TER SLIDiM4 THRESHot.O.(CSF).
2 %
3 0
~\ N
' N 2 coo -
d
\ N RE50LTS T-oR MODELS wig Appep wArge,,
8- N
$ \
g -
N
\
V isoo -
s -
N. N
.u \
d \
Of \
@ ~ REWLT5 FOR mod 6L5 A c s F Fo R /.4,* 7 0 '
- WiiHodT MPED VlATEE ' N N
k looo - -
e cse Fea wi.s . . ._
9 -
5
's '
s' 3 CSF F.oA M .
1,0
. =
y~
Soo - -
4 e -
- E D L si at J d d 5 y a a R .'
, i . l . t , i O j 10 0 So 40 40 20 C (
Pelt 4hlT OF G415TERS THAT Do Nor SLIDE Figure 6.17. Maximum Canister Shear vs. Percent of Canisters Effective.
55
- 7. SEISMIC RESISTANCE CAPACITY EVALUATION The seismic resistance capacity evaluation of the storage racks at West Valley was based on the present configuration of canister storage, the rack strength analysis discussed in Section 5.0, and the seismic response analysis discussed in Section 6.0. Some of the results obtained during this evaluation are presented in this section.
From the results obtained from the multiple rack response analysis reported in Section 6 3, it was seen that the coefficient of friction between the canisters and storage rack has an effect on the type of structural response of the system to a seismic event. The actual coefficient of friction is not known, especially for the submerged environment of the FRS pool. The range of values are likely to be anywhere from 0.5 to 1 9. Results from the multiple rack seismic response analysis show that, if the c.afficient of fricticn is below approximately 1.0, then the canisters slide and the response can be predicted accurately by the single rack model (i.e., Models R13 or R1SLT). However, if the coefficient of friction is in the range of 1.0 to 1.6, then some of the canisters may not slide and the response might possibly be better represented by Models R7A or R7 ALT, which have about twenty-five (25) percent of the canisters effective for Icad transmittal between adjacent racks.
In order to include this possible range of coefficient of fr1 tion 9 in the seismic resistance capacity evaluation of the storage racks, ground acceleration thresholds which cause yield stress and ultimate stress for three critical points in the storage rack were calculated using results from Models R13, R1SLT, R7A and R7 ALT. These results are tabulated in Table 7.1 for the yield stress threshold event and in Table 7.2 for the ultimate stress threshold event. Detailed calculations for these thresholds are presented in Appendix VIII.
An examination of the ground acceleration thresholds reveal that the bolts in the top rail splice will be the first items to reach yield stress loads (0.19g), with the North wall anchor bolts second (0.21g).
It should be noted that, at ground acceleration levels of 0.19g and above, the coefficient of friction necessary to make results from Models R7A and R7 ALT valid would approach unrealistically high values, thus making results from the single rack Models R13 and R1SLT even more likely to represent the actual behavior of the system. The magnitude of these high values can be seen by multiplying the shear forces / canister shown in Figs. 6.16 and 6.17 for Models R7A and R7 ALT by a factor of 1 9 (i.e. , the ratio between 0.19g and the 0.1g normalization ground acceleration), which in turn causes the coefficient necessary to prevent sliding to also increase by a factor of 1.9 57
Table 7.1. Yield Stress Ground Acceleration Thresholds.
Ground Acceleration Threshold (g)
Location on Storage Rack Model Model Model Model R1S R1SLT R7A R7 ALT Bolts - Top Rail Splice 0.19 0.25 0.22 0.28 Anchor Bolts - North Wall 0.21 0.28 0.21 0.28 Weld - Column / Floor Beam 0.22 0.28 0.29 0 37 3
Table 7 2. Ultimate Stress Ground Acceleration Thresholds.
I Ground Acceleration Threshold (g)
Location on Storage Rack Model Model Model Model R1S R1SLT R7A R7 ALT Bolts - Top Rail Splice 0.22 0 31 0.24 0 32 Anchor Bolts - North Wall 0 24 0 31 0.24 0 32 Weld - Column / Floor Beam 0 39 0.49 0.51 0.65 58
As previcusly discussed, the only difference between Model R1S and Model R1SLT is that the added mass of water contained in the canisters is included in Model R1S. Even though higher thresholds can be achieved if the added mass of the contained water is not included, as can be seen from Tables 7.1 and 7 2, it is thought to be realistic to include it.
When yielding begins in the splice joint of the top rail, the distribution of internal member forces will change. Bending moments at the floor reactions will increase while tha North wall anchorage moments will decrease. This is caused by the reduction in the moment carrying capability of the splice joints and the resultant increase in the relative stiffness of the load path from the top rail to the floor as compared with that along the top rail to the North wall. In addition, it should also be noted that when yielding begins nonlinear effects become important and the finite element models used for the response predictions become invalid. However, if it is assumed that the results can be linearly extrapolated to ultimate stress levels, the ground acceleration thresholds would be those given by 7able 7.2.
These results, which reflect conservatively low thresholds, show that the bolts in the top rail splice would break at a threshold of 0.22g.
Of course, it should be noted that these ground acceleration thresholds for yield and ultimate stress are values which apply only to the racks supporting the heavier loaded canisters in the pool configuration.
Canisters which contain lighter fuel assemblies as noted in F'g. 3.1 and Table 3.1 will have higher failure ground acceleration thresholds.
A probable scenario of events following the initiation of yielding in the splice joints is given in Table 7.3. It is seen from this table that the yielding thresholds are relatively close for the splice joint, North wall anchor bolts and the column / floor beam. As discussed above, it is felt that the redistribution of load after splice joint yield will be such that the floor / column welds will yield and cause very large deflections of the storage racks in the East-West direction. This response will be somewhat dampened by the distribution of load to the more lightly loaded storage racks adjacent to the racks supporting the WEPC0 canisters. Eventually, permanent deformation of the racks in the East-West direction will occur. Because of the empty rows in the pool, this deformation could conceivably reach a maximum value of about two feet. Even with this occurrence, it is unlikely that the canisters themselves would become damaged or fall to the floor.
In conclusion, based on the conservative assumptions made in this analysis, it can be safely stated that no da::; age will occur to the spent fuel stored in the FRS pool from the failure of the storage rack system unless the threshold ground acceleration from an earthquake significantly exceeds 0.19g.
59
Table 7.3 Probable Failure Scenario.
Maximum Ground Acceleration Description of Event (g) 0.19 Bolts.in top rail splice joint yield.
0.21 North wall anchor bolts yield.
- 0.22 Column / floor beam welds yield and bolts in top rail splice joint may break.
l 0.24 North wall anchor bolts and/or column / floor beam welds break depending on the effect of load redistribution.
60
REFERENCES
- 1. Dong, R. G., and Ma, S. M.; " Structural Analysis of The Fuel Receiving Station Pool at The Nuclear Fuel Service Reprocessing Plant, West Valley, New York," Lawrence Livermore Laboratory Report No. UCRL-52575, May 5, 1978.
- 2. Clough, R. W. and Penzien, J. Dynnmics of Structures, McGraw-Hill Book Company, New York, 1975.
3 PAFEC 75 Data Precaration, PAFEC Ltd., Nottingham, England, November, 1978.
- 4. Dong, R. G., " Effective Mass and Damping of Submerged Structures,"
Lawrence Livermore Laboratory Report No. UCRL-52342, April 1, 1978.
- 5. Structural Analysis and Desien of Nuclear Plant Facilities, ASCE Manual on Engineering Practice No. 58, American Society of Civil Engineers, New York, New York, 1980.
- 6. U. S. Nuclear Regulatory Commission, ' Design Response Spectra for Seismic Design of Nuclear Power Plants," Regulatory Guide 1.60, December, 1973 7 Aluminum Standards and Data -
1974, Sixth Edition, The Aluminum Association, Washington, D. C., March, 1979
- 8. Dong, R. G., and Tokarz, F. J., " Seismic Analysis of Large Pools,"
g Lawrence Livermore Laboratory Report No. UCRL-52167, November 17, 1976.
- 9. Johnson, C. M., et. al., "F1t J-Structure Interaction for Seismic Input," Proceedings of the Second ASCE Conference on Civil Engineering and Nuclear Power, Vol. VI: Design and Analysis of Nuclear Facilities for Earthquakes, September 15-17, 1980, Knoxville, Tennessee.
- 10. Fisher, J. H. and Striuk, J. H. A., Opide to Desian Criteria for Bolted and Riveted Joints, John Wiley and Sons, New York, 1974.
61 i
1 l
- 11. Bruhn, E. F., Analysis and Design of Flight Vehicle Structures, Tri-State Offset Company, Cincinnati, Ohio, 1965.
b 62
NRC a oRu 335 1. REPORT NUMBE R (Assped by DOC)
U S. NUCLE AR REGUL ATORY COMMISSION
,,,,, NUREG/CR-2236, Vol. 1 BIBLIOGRAPHIC DATA SHEET SAI-148-026 4 TlTLE AND SUBTlTLE (Add Volume No, s! war @rian} 2. (Leave blek)
Seismic Resistance Capacity Evaluation of Spent Fuel Storage Racks and Fuel at West Valley, New York 3. RECIPIENT'S ACCESSION NO.
Main Report
- 7. AUTHOR (S) 5. D ATE REPORT COMPLE TED M ON TH l YE AR N.E. Johnson, J.C. Walls January 1981
- 9. PERF OHMING ORGANIZATION N AME AND MAILING ADDRESS (Include 2,p Codel DATE REPORT ISSUED MONTH l YEAR Science Applications, Inc. December 1981 800 Oak Ridge Turnpike s it,., u,n a i Oak Ridge, TN 37830 B (Leave Nank)
- 12. SPONSOHING OHGANIZATION N AME AND M AILING ADDRESS (Include Ira Codel
- 10. PROJECT / TASK / WORK UNIT NO.
Division of Fuel Cycle and Material Safety Office of Nuclear Material Safety and Safeguards 11. CONT RACT NO U.S. Nuclear Regulatory Commission Washington, DC 20555 FIN B6955
- 13. TYPE OF HE POH T PE RIOD COVE RE D (inclus,ve dams /
Technical 15 SUPPLEMENTARY NOTES 14 (Leme Wa"*/
16 ABSTH ACT (200 words or less1 This report documents an evaluation of the seismic resistance capacity of the Fuel Receiving Station (FRS) spent fuel storage racks and canisters at the Nuclear Fuel Services Reprocessing Plant at West Valley, New York. The primary objective of this work was to determine the threshold ground acceleration above which potential yielding, permanent deformation or collapse, and/or excessive deformations would occur to the storage rack structure and canisters. Examination of the failure threshold levels show that the splice joint in the top rail of the storage rack will be the first item to reach yield stress. This event occurs at 0.19 9. The results also show that the top rail splice joint would probably be the first component to break at a threshold of 0.22 . 9A probable scenario of events following the top rail splice joint failure is that loads will redis-tribute in the storage rack causing the anchor bolts connecting the rack columns to the pool floor to fail. This will subsequently cause the rack to deflect in an East-West direction until it impacts the pool wall or another rack. Although this event could cause a few canisters to become dislodged from the rack and fall to the pool floor caus-ing crushing of the canisters, such an occurrence is considered to be unlikely. Volume 1 contains the text of the report and Volume 2 includes appendices containing technical supporting data generated during the preparation of the report.
17 KE Y WOR DS .v.D DOCUME NT AN ALYSIS 17a DE SC RIP T OHS 17b IDE N TIF IE RS! OPE N EN DE D TE RMS IB AV AIL ABILITY ST ATEMENT 19 CU TY CLAS h.s reporr/ 21 NO OF PAGES Unlimited 20 gp,T* ## ""
ss '$8"""# i N RC F OHM 335(777)
UNITED STATES NUCLEAM REGULATORY COMMISSION F ]
W ASHiteGTON O. C. 20565 POST AG E AN D P E E S P4 8D u.s. MucLE An nKoutAvomy OFFICI AL BUSINESS PEN ALTY FOR PRIVATE USE,5300 COMMISSION L
L J 1
l l
f I
_ _ _ _ _ .