ML20091C015
| ML20091C015 | |
| Person / Time | |
|---|---|
| Site: | Cook  |
| Issue date: | 07/10/1991 |
| From: | Wang Y HOLTEC INTERNATIONAL |
| To: | |
| Shared Package | |
| ML17329A102 | List: |
| References | |
| HI-90488, HI-90488-R06, HI-90488-R6, NUDOCS 9108050009 | |
| Download: ML20091C015 (262) | |
Text
l ~f dgguQdNg gagsp+t#gegemahamwtp4emw.ghn h
hf kkh $
ra$lllkh p
$gqNy A se
$ssnon$e$ss$$$
yn$y+RER,ORTEQR@
NSI E esed%+
nw e n % am+ wa.a a STORAGE +nDENSIEICATIONR44%
S#jhg
! II '
dessessem P_mwith.sw.~hy;:nd A. n e. fMau . .
1%uh h _ ~ _ X -
ERRE
-w ~w -
LICENSING REPORT FOR STORAGE DENSIFICATION OF D.C. COOK SPENT FUEL POOL .
INDLANA MICHIOAN POWER COMPANY -
by Holtec International AEPSC Contract No. C-7926
> Holtec Project 00480 E
i HOLTEC INTERNATIONAL I REVIEW AND CERTIFICATION LOG DOCUMENT NAME: LICENSING REPORT FOR STORAGE DENSIFICATION I OF D.C. COOK SPENT FUEL POOL I HOLTEC DOCUMENT I.D. NO.
HOLTEC PROJECT NO.
CUSTOMER / CLIENT:
HI-90488 00TED AMERICAN ELECTRIC POWER
~~
(INDIANA MICHICAN POWER CO.)
REVISION BLOCK ISSUE NO. AUTHOR REVIEWER Q.A. MANAGER APPROVED *
& DATE & DATE & DATE BY & DATE 3 ORIGINAL PC b~ c'N #" *' ~ #
'~
^p
- si% Q b , ufi t,f ec 75' Y ?N fah F' ? Q 'r' '"I ' */9 0 REVISION 1 4, ,j Q[j , '
I b YN7YN "'",4 REVISION 2 St U '
%3Ml >It*/1/ w/k, ,% Am 6 M>rN REVISION 3 D#7 bM N/N Y N.M
<g Y w SI> sh i n . rlu/W W //M -!.A e f REVISION 4 y l 7/ a REVISION 5 g [ .,
hh,jp REVISION 6 h~lO%A!3 fDY. S V Y ' '? l
% p M So W - NCei ^
O' Yi c )c 1fICffI rr 1/10 /9/ A!! 'l
- l NOTE: Signatures anc printed names are required in the review block.
I
- Must be Project Manager or his Designee.
I This document conforms to specification and the applicable sections of the governing codes.
the requirement of the design This document bears the ink stamp of the professional engineer who
.I is certifying this document.
fCY Professional Engineer SEAL
. , y ;. I'#;O;3.
,yy,.jyp--_.,
.m ru.s 3,g..A,3 _
1 Q/ g ..m
,,,,f.,;,.m_~.-i--e* * 'Q u 9$1 KR45H.'!A PAL SIIIGH 1:
~~;"
, ...,2j },i},
mm . w e
- ' *" j
l s
SUMMARY
OF REVISIONS
-g Revision 1 contains the following number of pages of text fincludina tables, but excludina ficures):
l Title Page 1 l Review and Certification Log 1 Summary of Revisions Page 1 Table of Contents 4 List of Figures 3 Section 1 8 Section 2 17 Section 3 8 Section 4 (later)
Section 5 23 Section 6 48 Section 7 5 Section 8 13 Section 9 (later)
Section 10 11 Revision 2 contains the following number of pages of text (includina tables and ficures):
Title Page 1 Review and Certification Log 1 Summary of Revisions Page 1 Table of Contents 4~
List of Figures 3 Section 1 8 Section 2 18 Section 3 14 Section 4 Not included in Revision 2 Section 5 33 Section 6 78 Section 7 5 Section 8 17 Section 9 Not included in Revision 2 Section 10 Not included in Revision 2 0
SUMMARY
OF REVISIOLLS Revision 3 contains the following number of pages of text (includina tables and ficuresi:
Title Page 1 1 Review and Certification Log 1 Summary of Revisions Page 2 Table of Contents 4 List of Figures 3 Section 1 9 Section 2 19 i 3ection 3 Section 4 14 34 Appendix A to Section 4 9 Section 5 33 Section 6 76 Section 7 5 Section 8 17 Section 9 11 Section 10 5 Section 11 4 Revision 4 contains the same number of pages as Revision 3 with I these excentions:
List of Tables (added to Rev. 4): 3 Sections revised in Rev. 4 now contain the following number of pages:
Section 1 9 Section 2 18 Section 4 35 Section 5 34 Section 9 11 Individual pages revised and transmitted in Revision 4 are:
Pages 3-1, 3-3, 3-4, 6-6, 6-15, 6-18, 6-28, 6-30, 7-4, 7-5, 7-6, 8,7 and 10-2.
I I
b -
r b
j~
SUMMARY
OF REVISIONS Holtec Report HI-90488 Revision 5 The following is revised in Revision 5:
Pages 4-8 and 4-9 Section 9 Appendix A Page v of Table of Contents Revision 6 The followino naces are revised in Revision 6:
~
List of Figures Table of Contents (page v) 2-1 4-15, 4-16 l 5-2, 5-3, 5-5, 5-6, 5-8, 5-9, 5-10, 5-12, 5-15, 5-16, 5-17, 5-18, 5 5-19, 5-20, 5-7', 5-24 through 5-38 6-3<. 6-4, 6-35 7-2, 7-4 8-6, 8-7, 8-13 10-4 11-3 i
I
TABLE OF CONTENTS l
5
1.0 INTRODUCTION
1-1 2.0 MODULE DATA 2-1 2.1 Synopsis of New Modules 2-1 2.2 Mixed Zone Two Region Storage (MZTR) 2-1 2.3 Material Considerations 2-4 2.3.1 Introduction 2-4 2.3.2 Structural Materials 2-4 2.3.3 Poison Material 2-4 2.3.4 Compatibility with Coolant 2-7 2.4 Existing Rack Modules and Proposed Reracking 2-7 Operation 3.0 CONSTRUCTION OF RACK MODULES 3-1 3.1 Fabrication Objective 3-1 3.2 Mixed Zone Two Region Storage 3-2 3.3 Anatomy of Rack Modules 3-2 3.4 Codes, Standards and Practicos 3-5 for the D.C. Cook Spent Fuel Pool Racks 3.5 Materials of Construction 3-9 I 4.0 CRITICALITY SAFETY ANALYSES 4-1 4.1 Design Basis 4-1 4.2 Summary of Criticality Analyses 4-4 4.2.1 Normal Operating Conditions 4-4 4.2.2 Abnormal and Accident Conditions 4-6 4.3 Reference Fuel Storage Cells 4-8 4.3.1 Reference Fuel Assembly 4-8 1 4.3.2 High Density 7uel Storage Cells 4-9 i
J L
TABLE OF CONTENTS (continued) 4.4 Analytical Methodology 4-10 g 4.4.1 Reference Design Calculations 4-10
[ 4.4.2 Fuel Burnup Calculations and 4-12 j Uncertainties
!- 4.4.3 Effect of Axial Burnup 4-13 Distribution 4.5 Criticality Analyses and Tolerances 4-15 I 4.5.1 Nominal Design 4-15 E 4.5.2 Uncertainties due to 4-15 Manufacturing Tolerances 4.5.2.1 Boron Loading Tolerances 4-15 4-16 4.5.2.2 Boral Width Tolerance 4.5.2.3 Tolerance in Cell Lattice Spacing 4-16 4.5.2.4 Stainless Steel Thickness 4-16 l
E 4.5.2.5 Tolerances Fuel Enrichment and Density 4-16 Tolerances 4.5.3 Water-gap Spacing Between Modules 4-17 I 4.5.4 Eccentric Fuel Positiening 4-17 4.6 Abnormal and Accident Conditions 4-17 4.6.1 Temperature and Water Density 4-17 1 Effects 4.6.2 Dropped Fuel Assembly 4-18 4.6.3 Lateral Rack Movement 4-18 1 4.6.4 Abnormal Location of a Fuel 4-19 Assembly 4.7 Existing Spent Fuel 4-19 4.8 References 4-21 5.0 THERMAL-HYDRAULIC CONSIDERATIONS 5.1 Introduction 5-1 5.2 Spent Fuel Cooling System Description 5-2 5.2.1 System Functions 5-2 5.2.2 System Description 5-3 5.2.3 Performance Requirements 5-4 5.3 Decay Heat Load Calculations 5-4 il
I TABLE OF CONTENTS (continued) 5.4 Discharge Scenarios 5-5 5.5 Bulk Pool Temperatures 5-6 ,
lI 5.6 Local Pool Water Temperature 5-11 l
I 5.6.1 Basis 5-11 l 5.6.2 Model Description 5-12 5.7 Cladding Temperature 5-13 l
8 5.8 Blocked Cell Analysis 5-16 l 5.9 References for Section 5 5-16 i 6.0 RACK STRUCTURAL CONSIDERATIONS 6-1 6.1 Introduction 6-1 6.2 Analysis Outline 6-2 6.3 Artificial Slab Motions 6-3
) 6.4 Outline of Single Rack 3-D Analysis 6-5
(
l 6.5 Dynamic Model for the Single Rack 6-7 Analysis
- E 6.5.1 Assumptions 6-9 l5 6.5.2 Model Description 6-11 l 6.5.3 Fluid Coupling 6-12
{ 6.5.4 Damping 6-13
- 6.5.5 Impact 6-13 6.6 Assembly of the Dynamic Model 6-14
.I 6.7 Time Integration of the Equations 6-17 of Motion lg 6.7.1 Time History Analysis Using 6-17 l3 Multi-Degree of Freedom Rack Model 6.7.2 Evaluation of Potential for 6-19
( Inter-Rack Impact i
6.8 Structural Acceptance Criteria 6-19 6.9 Material Properties 6-21 I
I ...
111
I TABLE OF CONTENTS (continued) 6.10 Stress Limits for Various Conditions 6-22 6.10.1 Normal and Upset Conditions 6-22 (Level A or Level B)
I 6.10.2 Level D Service Limits 6-25 6.11 Results for the Analysis of Spent Fuel 6-25 I Racks Using a Single Rack Model and 3-D Seismic Motion I 6.12 Impact Analyses 6.12.1 Impact Loading between Fuel Assembly and Cell Wall 6-28 6-28 6.12.2 Impacts between Adjacent Racks 6-28 I 6.13 Weld Stresses 6-31 6.13.1 Baseplate to Rack Welds and 6-29 Cell-to-Cell Welds 6.13.2 Heating of an Isolated Cell 6-30 6.14 Whole Pool Multi-Rack Analysis 6-30 6.14.1 Multi-Rack Model 6-32 6.14.2 Results of Multi-Rack Analysis 6-34 6.15 Bearing Pad Analysis 6-36 6.16 References for Section 6 6-37 7.0 ACCIDENT ANALYSIS AND MISCELLANEOUS 7-1 STRUCTURAL EVALUATIONS 7.1 Introduction 7-1 7.2 Refueling Accidents 7-1 7.2.1 Dropped Fuel Assembly 7-1 7.3 Local Buckling of Fuel Cell Walls 7-2 7.4 Analysis of Welding Joints in Rack 7-3 due to Isolated Hot Cell I 7.5 Crane Uplift Load of 3000 lbs'. 7-4 7.6 References for Section 7 7-4 l
8 8 iV l
TABLE OF CONTE!1TS 8.0 STATIC AND DYNAMIC AllALYSES OF FUEL POOL STRUCI'URE 8.1 Introduction 8-1 8.2 General Features of the Model 8-3 8.3 Loading Conditions 8-6 8.4 Results of Analyses 8-10 8.5 Pool Liner 8-11 8.6 Conclusions 8-11 8.7 References for Section 8 8-12 9.0 RADIOLOGICAL EVALUATIOt1 9-1 9.1 Fuel Handling Accident 9-1 9.1.1 Assumptions and Source 9-1 Term Calculations I 9.1.2 Results 9-4 9.2 Solid Radwaste 9-5 9.3 Gaseous Releases 9-5 9.4 Personnel Exposures 9-5 9.5 Anticipated Exposure during Reracking 9-6 9.6 References for Section 9 9-8 10.0 IN-SERVICE SURVEILLANCE PROGRAM 10-1 1 10.1 Purpose 10-1 10.2 Coupon Surveillance 10-2 10.2.1 Description of Test Coupons 10-2 10.2.2 Benchmark Data 10-3 10.2.3 Coupon Reference Data 10-3 10.2.4 Accelerated Surveillance 10-4 10.2.5 Post-Irradiation Tests 10-4 10.2.6 Acceptance Criteria 10-4 10.3 References for Section 10 10-5 i
V
~
J TABLE OF CONTENTS '
L-
[ 11.0 COST / BENEFIT ANALYSIS 11-1 11.1 Introduction 11-1 11.2 Project Cost Assessment 11-1 11.3 Resource Conunitment 11-3 11.4 Environment Assessment 11-3 I
l ~
I I
I I
I 5
vi
LIST OF TARIJJ l
Table 1.1.1 Discharge Schedule Table 1.1.2 Available Storage in the Donald C. Cook Pool l Table 1.1.3 Rack Module Data, Existing and Proposed Racks Table 2.1.1 Module Data Table 2.1.2 Common Module Data Table 2.1.3 Module Data Table 2.3.1 Boral Experience List (Domestic and Foreign)
Table 2.3.2 1100 Alloy Aluminum Physical and Mechanical Properties Table 2.3.3 Chemical Composition (by weight) - Aluminum (1100 Alloy)
Table 2.3.4 Boron Carbide Chemical Composition, Weight %
Boron Carbide Physical Properties Table 4.1 Summary of Criticality Safety Analyses Normal Storage Configurations Table 4.2 Summary of Criticality Safety Analyses Interim Checkerboard Loading Table 4.3 Reactivity Effects of Abnormal and Accident I conditions Table 4.4 Design Basis Fuel Assembly Specifications Table 4.5 Reactivity Effects of Manufacturing Tolerances Table 4.6 Effect of Temperature and Void on Calculated
- Reactivity of Storage Rack Table 5.4.1 Fuel Specific Power and Pool Capacity Data Table 5.4.2 Data for Scenarios 1 through 3 vii
LIST OF TABLES (continued)
Table 5.4.3 Data for Scenarios 1 through 3 Table 5.5.1 Pool Bulk Temperature and Heat Generation Rate Data Table 5.5.2 Time-to-Boil for Various Discharge Scenarios Table 5.6.1 Peaking Factor Data Table 5.6.2 Data for Local Temperatures Table 5.7.1 Local and Cladding Temperature Output Data for the Maximum Pool Water Condition (Case 1)
Table 6.3.1 Correlation Coefficient Table 6.5.1 Degrees of Freedom Table 6.6.1 Numbering System for Gap Elements and Friction Elements Table 6.6.2 Typical Input Data for Rack Analyses (lb-inch units)
Table 6.9.1 Rack Material Data (200*F) l Support Material Data (200*F) l Table 6.11.1 Stress Factors and Rack-to-Fuel Impact Load Table 6.11.2 Rack Displacements and Support Loads Table 6.14.1 Rack Numbering and weight Information Table 6.14.2 Maximum Displacements from WPMR Run MP1 Table 6.14.3 Maximum Displacements from WPMR Run MP2 Table 6.14.4 Maximum Displacements from WPMR Run MP3 Table 6.14.5 Maximum Rack Displacements and Foot Load Table 8.4.1 Safety Factors for Bending of Pool Structure Regions i
i I
l viii
> LIST OF TABLES (continued)
Table 9.1 Inventories and Constants of Significant Fission Product Radionuclides 1
Table 9.2 Data and Assumptions for the Evaluation of the Fuel Handling Accident Table 9.3 Typical Concentrations of Radionuclides in the Spent Fuel Pool Water Table 9.4 Preliminary Estimate of Person-Rem Exposures During Reracklng Table 11.1 Donald C. Cook Nuclear Plant Worst Case Spent Fuel Inventory i
I I
I I
I I
I l
I i l ix l .
I I LIST OF FIGURES Figure 2.1.1 Cook Spent Fuel Pool Layout (upper bound cell count 3616 cells)
Figure 3.3.1 Seam Welding Precision Formed Channels Figure 3.3.2 Composite Box Assembly Figure 3.3.3 Array of Cells for Non-Flux Trap Modules l Figure 3.3.4 Adjustable Support Leg Figure 3.3.5 Elevation View of Rack Module Figure 4.1 Normal Storage Pattern (Mixed Three Zone)
Figure 4.2 Interim Storage Pattern (Checkerboard)
Fi gure 4. 3 Acceptable Burnup Domain in Regions 2& 3 Figure 4.4 Fuel Storage Cell Cross Section Figure 4.5 KENO Calculational Model I Figure 4.6 Equivalent Enrichment for Spent Fuel ar.
Various Burnups for Initial Enrichment of 4.95%
Figure 4.7 Effect of_ Water-Gap Spacing Between Modules on System Reactivity I Figure 4.8 Acceptable Burnup Domain in Regions Showing Existing Spent Fuel Assemblies 2 & 3 Figure 5.5.1 Pool Bulk Temperature Model Figure 5.5.2 Donald C. Cook SFP Normal Discharge, One Cocling Train, Case la Figure 5.5.3 Donald C. Cook SFP Normal Discharge, One Cooling Train, Case Ib Figure 5.5.4 Donald C. Cook SFP Normal Discharge, Two Cooling Trains, Case 2 Figure 5.5.5 Donald C. Cook SFP Full Core Offload Two Cooling Trains, Case 3 Figure 5.5.6 Donald C. Cook SFP Full Core Offload One Coc ing Train, Case 4 Figure 5.5.7 Cook SFP Loss of Cooling Scenario, Case la E X
m
~
LIST OF FIGURES (continued) i Figure 5.5.8 Cook SFP Loss of Cooling Scenario, Case ib Figure 5.5.9 Cook SFP Loss of Cooling Scenario, Case 2 Figure 5.5.10 Cook SFP Loss of Cooling Scenario, Case 3 Figure 5.5.11 Cook SFP Loss of Cooling Scenario, Case 4 Figure 5.6.1 Idealization of Rack Assembly Figure 5.6.2 Thermal Chimney Flow Model Figure 5.6.3 Convection Currents in the Pool Figure 6.2.1 Pictorial View of Rack Structure Figure 6.3.1 DBE - N-S Acceleration Time History Figure 6.3.2 DBE - E-W Acceleration Time History Figure 6.3.3 DBE - Vertical Acceleration Time History LIST OF FIGURES (continued)
Figure 6.3.4 Horizontal Design Spectrum and N-S Time History Spectrum (5% damping)
I Figure 6.3.5 Horizontal Design Spectrum History Spectrum (5% damping) and E-W Time Figure 6.3.6 Vertical Design and Time History Derived I Spectra (5% damping)
Figure 6.3.7 OBE - N-S Acceleratior Time History Figure 6.3.8 OBE - E-W Acceleration Time History Figure 6.3.9 OBE - Vertical Acceleration Time History Figure 6.3.10 Horizontal Design Spectrum and Time History Derived N-S Spectrum (2% damping)
Figure 6.3.11 Horizontal Design Spectrum and E-W Time History Derived Spectrum (2% damping)
Figure 6.3.12 Vertical Design and Time History Derived Spectra (2% damping) xi
m LIST OF FIGURES i (continued)
Figure 6.5.1 Schematic Model for DYNARACK Figure 6.5.2 Rack-to-Rack Impact Springs Figure 6.5.3 Impact Spring Arrangement at Node i Figure 6.5.4 Degrees of Freedom Modelling Rack Motion Figure 6.5.5 Rack Degrees of Freedom for X-Z Plane Bending Figure 6.5.6 Rack Degrees of Freedom for Y-Z Plane Bending Figure 6.6.1 2-D View of Rack Module Figure 6.14.1 Rack and Foot Pedestal Numbering for Cook Multi-Rack Model 1 Figure 6.14.2 Cook Pool Multi-Rack Seismic Analysis, Run MP2 Rack 16 to Rack 17 South Corner Dynamic Gap at I Rack Top Figure 6.14.3 Cook Pool Multi-Rack Seismic Analysis, Run MP2 i Rack 16 to Rack 17 South Corner Dynamic Gap at Rack Top i Figure 6.14.4 Cook Pool Multi-Rack Seismic Analysis, Run MP3 Rack 12 to Rack 18 West Corner Dynamic Gap at Rack Top Figure 6.14.5 Cook Pool Multi-Rack Seismic Analysis, Run MP3 Rack 12 to Rack 18 East Corner Dynamic Gap at Rack Top _
Figure 7.3.1 Loading on Rack Wall Figure 7.4.1 Welded Joint in Rack i Figure 8.2.1 Isometric View of Cook Spent Fuel Pool k Figure 8.2.2 Overall Finite Model of Cook Pool P Top View Figure 8.2.3 Overall Finite Model of Cook Pool Bottom View Figure 8.3.1 Pedestal Load vs. Time xii
1.O INTRODUCTION Donald C. Cook is a twin unit pressurized water nuclear power reactor installation owned and operated by Indiana Michigan Power Company. Donald C. Cook received its construction permit from the AEC in March, 1969, and its operating License in October, 1974 for Unit 1 and December 1977 for Unit 2. The two reactors went into i commercial operati.on in August, 1975 (Unit 1) and Julv, 3978 gunit 2), respectively. Th* Donald C. Cook fuel storace system is made up of a fuel pool 58'-3 1/8" long x 39'-1 9/16" wide with an integral cask laydown area. The pool creaently contains 1367 spent fuel storage assemblies and 36 miscelleteous hardware items.
Thus, out of the total installed storage capaci.y of 2050 storage cells, 1403 storage cells are presently occupied. Since the full core has 193 fuel assemblies for both Donald C. Cook reactors, maintaining full core offload capability from one reactor implies I thr,t 1857 storage cells (2050 minus 193) are available for normal offload storage. Table 1.1.1 provides the data on previous and projected fuel assembly discharge in the Donald C. Cook spent fuel pool. Table 1.1.2, constructed from Table 1.1.1 data, indicates that Donald C. Cook will lose full core discharge capability (for one reactor) in 1995. This projected loss of full core discharge capability prompted the present undertaking to increase spent fuel stcrage capability in the Donald C. Cook pool.
I I
i 1-1
u a O purpose of this licensing submittal la to rcrack the Donald C.
Cook pool and equip it with new poisoned hig -
.sity storage I racks containing 3613 storage cells. The reracking also entails relocation of the thimble plug tool, spent fuel handling tool, Rod Cluster Control Assembly (RCCA) change tool, and Burnable Poison e Rod Assembly (BPRA) tool brackets to the South wall adjacent to a the cask pit.
Twenty three free-standing poisoned rack modules positioned with a I
1 prescribed and geometrically controlled gap between them will contain a total of 3613 storage cells (including 3 triangle cells located at the SW, NW and NE corners of the pool). Out of these cells, the peripheral cells located in each rack module are flux-trap cells *, and the interior ones are of the so-called non-flux trap type. The storage cells suitable for storing fresh fuel (up to 5% enrichmant) are uniquely identified (:;ee Section 4.0, Figures 4.1 and 4.2), and are surrounded by non-flux trap cells which have a burnup restriction on the fuel which they can store.
F.onsistent with the cc7cep' of two regiert storage, the placement of fuel with a given burnup in the allowable location is 4
administratively controlled. No credit is taken for uoluble boion i in normal refueling and full core offload rtorage conditions.
A flux trap construction implies that there is a water gap between adjacent storage cells such that the neutrons emanating from a fuel assembly are thermalized before reaching an adjacent fuel assembly.
1-2
I l It is noted that the proposed reracking effort will increase the I number of licensed storage locations to 3613 and, as indicated in Table 1.1.2, will extend the date of loss of full core discharge capability through the year 2008. Table 1.1.3 presents key comparison data for existing and proposed rack modules for Donald ;
C. Cook.
The now spent fuel storage racks are free-standing and self supporting. The principal construction materials for the new racks are SA240-Type 304 stainless steel sheet ar.d plate stock, I and SA564-630 (precipitation hardened stain 1 css steel) for the
- adjust- able support spindles. The only non-stainless material utilized in the rack is the neutron absorber material which is boron carbide and aluIrinum-composite sandwich available under the l patented product name "Boral".
The new racks are designed and analyzed in accordance with Section III, Division 1, Subsection NF of the ASME Boiler and Pressure Vessel (B&PV) Code. The material procurement, analysis, and I fabrication of the rack modules confonn to 10CFR 50 Appendix B requirements.
This Licensing Report documents the design and analyses performed to demonstrate that the new spent fuel racks satisfy all governing requirements of the applicable codes and standards, in particular, "OT Position for Review and Acceptance of Spent Fuel Storage and Handling Applicati7ns", USNRC (1978) and 1979 Addendum thereto.
1-3
- o. _ __ _ _ _ _
m The safety assessment of the proposed rack modules involved L demonstration of its thermal-hydraulic, criticality and structural adequacy. Hydrothermal adequacy requires that fuel cladding will net fail due to excessive thermal stress, and that the steady l state pool bulk temperature prescribed for the spent fuel pool to satisfy the pool structural will remain within the limits strength constraints. Demonstration of structural adequacy primarily involves analysis showing that the free-standing rack modules will not impact with each other or with the pool walls I under the postulated Design Basis Earthquake (DBE) and Operating Basis Earthquake (OBE) events, and that the primary stresses in the rack module structure will remain below the ASME BlPV Code allowables. The structural qualification also includes analytical demonstration that the subcriticality of the stored fuel will be maintained under accident scenarios such as fuel assembly drop, accidental misplacement of fuel outside a rack, etc.
The criticality safety analysis shows thct the neutron I multiplication factor for the stored fuel array is bounded by the USNRC limit of 0.95 (OT Position Paper) under assumptions of 95%
probability and 95% confidence. Consequences of the inadvertent placement of a fuel assembly are also evaluated as part of the criticality analysis. The criticality analysis also sets the requirements on the length of the B-10 screen and the areal S-10 density.
This Licensing Report contains documentation of the analyses performed to demonstrate the large margins of safety with respect g
to all USNRC specifled criteria. This report also contains the B results of the analysis performed to demonstrate the integrity of the fuel pool reinforced concrete structure, and an appraisal of 1-4 1
l
s L
radiological considerations. A cost / benefit analysis demonstrating reracking as the most cost effective approach to increase the on-site storage capacity of the Donald C. Cook 11uclear Plant has also
( been performed and synopsized in this report.
{ All computer programs utilized in performing the analyses
- ' documented in this licensing report are identified in the 8 appropriate sections. All computer codes are benchmarked and verified in accordance with Holtec International's nuclear Quality I Program.
The analyses presented herein clearly demonstrate that the rack module arrays possess wide mergins of safety from all three -
thermal-hydraulic, cr!.ticality , and structural - vantage points.
The 13 o Significant Bazard Consideration evaluation submitted to the Commission along with this Licensing Report is based on the descriptions and analyses synopsized in the subsequent sections of this report.
~
I 1
I I
I 1-5 1
u
] Table 1.1.1 DISCHARGE SCHEDULE Number of Cumulative Month / Assemblies Inventory I Cycle lA*
Igar 12/1976 D_ischarced 65 In the Pool G5 5 2A 4/1978 64 129 5 3A 4/1979 64 193 1B** 10/1979 80 273 4A 5/1980 65 338 I 2B SA 6A 5/1981 5/1981 7/1982 92 64 64 430 494 558 3B 11/1982 72 630 I 7A 4B 7/1983 3/1984 80 92 710 802 BA 4/1985 80 882 I 5B 9A 2/1986 6/1987 88 80 970 1050 6B 5/1988 80 1130 10A 3/1989 80 1210 1 7B 6/1990 77 1282 11A 10/1990 80 1362 8B 11/1991 76 1438 I 12A 9B 2/1992 3/1993 80 80 1518 1598 13A 6/1993 80 1678 10B 7/1994 80 1758 14A 10/1994 80 1838 11B 11/1995 80 1918 15A 4/1996 80 1998 1 12B 3/1997 80 2078 16A 8/1997 80 2158 13B 7/1998 80 2238 A - Reactor Unit 1 B - Reactor Unit 2 1-6 1
L a
k r
I Table 1.1.1 (continued)
DISCHARGE SCHEDULE Number of Cumulative Month / Assemblies Inventory cvele Xggr Discharced In the Pool 17A* 12/1998 80 2318 14B** 1/2001 80 2398 18A 4/2000 80 2478 ISB 5/2001 80 2558 19A 8/2001 80 2638 16B 9/2002 80 2718 l 20A 17B 12/2002 1/2004 80 80 2798 2878 21A 6/2004 80 2958 18B 5/2005 80 3038 22A 10/2005 80 3118 19B 9/2006 80 3198 23A 2/2007 80 3278 1 20B 1/2008 80 3358 24A 7/2000 80 3438 21B 7/2009 80 3518 1
I A - Reactor Unit 1 I B - Reactor Unit 2 I
i i,
i 1-7
. Table 1.1.2 AVAILABLE STORAGE IN THE DONALD C. COOK POOL NUMBER OF STORAGE LOCATIONS AVAILABLE l Month /
With Present Licensed Capacity After Reracking Cycle liar (2050 Locations) f3616 Locations) 7B 6/1990 768 2334 11A 10/1990 683 2254 8B 11/1991 612 2178 I 12A 9B 2/1992 3/1993 532 452 2098 2018 13A 6/1993 372 1938 I 10B 14A 11B 7/1994 10/1994 11/1995 292*
212 132**
1858 1778 1698 15A 4/1996 52*** 1618 I 12B 16A 3/1997 8/1997 1538 1458 13B 7/1998 1378 I 17A 14B 12/1998 1/2000 1298 1218 18A 4/2000 1138 I 15B 19A 16B 5/2001 8/2001 9/2002 1058 978 898 20A 12/2002 818 I 17B 21A 1/2004 6/2004 738 658 18B 5/2005 578 19B 9/2006 418 1 23A 2/2007 338*
20B 1/2008 258 24A 7/2008 178**
1 21B 7/2009 98 25A 11/2009 18***
I
- Date of loss of full core offload capability from both reactors.
Date of loss of full core offload capability for one reactor.
Date of loss of normal discharge capability.
1-8
I
!I
.J
!I Table 1.1.3 RACK HODUIf ATA, EXISTING AND PROPOSED RACKS 5
ITEM EXISTING RACKS PROPOSED RACKS Nulabor of cells 2050 3616*
Number of modules 20 23 i
. Neutron Absorber Boral Boral I
(Nom.) cell pitch, inch 10.5" 8.97" (Nom.) cell opening
] size, inch 8.884 1 0.124 8.75" 0.04 I
Include three triangular corner storage cells.
-I 1-9
i
, 2.0 MODULE DATb 2.1 Synopsis of _11ew Modulqa .
The Donald C. Cook spent fuel pool consists of a 39'-1 9/16" x 58'-3 1/8" rectangular pit with a 10'-4" x 10'-6" space designated for cask handling operations. The pool is connected to the fuel transfer canal through a weir gate on the West wall. This gate is normally closed.
At the present time, the Donald C. Cook pool contains medium density racks with a 10.5" nominal assembly center-to-center pitch. There is a total of 2050 storage cells in the pool. There are two sizes of modules, 10x10 and 10x11. The 10x10 module weighs 33,800 lb. and the 10x11 module weighs 37,200 lb.
Figure 2.1.1 shows the module layout for the Donald C. Cook p ; 11 after the proposed reracking campaign. As shown in Figure 2 - 1.1 and tabulated in Table 2.1.1, there are twenty-thre.. Lacks I- containing a total of 3613 storage cells with a 8.97" nerinal '
pitch.
The essential cell data for all storage cells is given in Table 2.1.2. The physical size and weight data on the modules may be found in Table 2.1.3. In summary, the present reracking application will increase the licensed storage capacity of the Donald C. Cook pool from 2050 to 3613 cells.
2.2 Mixed Zone Three Recion Storagg (MSTR):
The high density spent fuel storage racks in the Donald C. Cook pool will provide storage locations for up to 3613 fuel assemblies and will be designed to maintain the stored fuel, l having an initial enrichment of up to 5 wt% U-235, in a safe, coolable, and suberitical configuration during normal dischargo and full core offload storages and postulated accident conditions.
I 2-1
I l
I All rack modules for Donald C. Cook spent fuel pool are of the so-called " free. standing" type such that the modules are not attached I
l to the pool floor nor do they require any lateral braces or ;
restraints. These rack modules will be placed in the pool in their designated locations using a specific s11y designed lifting device, and the support legs remotely leveled (using a telescopic removable handling tool) by an operator on the fuel handling bridge. The leveling operations are done when the support legs are I lifted off the floor. Except for the crane, no additional lifting aquipment is needed while leveling is being performed.
As described in detail in Section 3, all modules in the Donald C.
Cook pool are of "non-flux trap" construction. However, the module If baseplates extend out by 7/8" (nominal), such that the nominal gap between the adjacent walls of two neighboring racks is 2" (nom.).
Thus, although there is a single screen of neutron absorber panel
] between two fuel assemblies stored in the same rack, there are two
- poison panels with a water flux trap (2" wide) between them for fuel assemblies located in cells in two f acing modules. Out of i these flux trap locations, and peripheral cell locations (cells adjacent to pool walls) a certain number of storage cells are designated for storing fresh fuel. In addition, as described in Section 4, a certain number of interior cells in each rack are designated for storing fresh fuel of 5% wt. U-235 (max.)
enrichment. In this manner, a sufficient number of locations without any burnup restriction (F.egion I cells) are identified to
- enable unrestricted full core offload of the Donald C. Cook reactor in the spent fuel pool. These so-called Region I cells are identified in Section 4 of this report. The remaining storage cells have enrichment /burnup restrictions. Appropriate
~
restrictions on the enrichment /burnup of the stored fuel in Region II and Region III cells are presented in Section 4.
t t
I 2-2
l Each rack module is supported by at inast four legs which are remotely adjustable. Thus, the racks can be made vertical and the l I top of the racks can easily be made co-planar with nach other.
The rack module support legs are engineered to accommodate l
variations of the pool floor. The support legs also provide an under rack plenum for natural circulation of water through the storage cells. The placement of the racks in the spent fuel pool has been designed to preclude any support legs from being located over existing obstructions on the pool floor.
The Donald C. Cook racks are subjected to mandated seismic loadings per the plant UFSAR. The Design Basis Earthquake (DBE) and Operating Basis Earthquake (OBE) seismic response spectra are provided and synthetic time histories are generated. These acceleration time histories are applied as inertia loads (see Section 6.3).
Under these seismic events, the rack modules have four designated I locations of potential impacts (i) Support leg to bearing pad (ii) Storage cell to fuel assembly contact surfaces (iii) Baseplate edges (iv) Rack top corners The support leg to pool slab bearing pad impact would occur I whenever the rack support foot lifts off the pool floor during a seismic event. The " rattling" of the fuel assemblies in the storage cell is a natural phenomenon associated with seismic conditions. The baseplate and rack top corners impacts would occur if the rack modules tend to slide or tilt towards each other during the postulated DBE or OBE seismic events. Section 6 of this report presents the analysis methodology and results for all three locations of impact, and establishes the structural integrity of racks under the load combinations specified for plant I the conditions required by the NRC.
I .
I 2-3
A bearing pad, made of austenitic stainless steel, is interposed between the support foot and the liner such that the loads transmitted to the slab by the rack module under steady state as well as seismic conditions are diffused into the pool slab, and allowable local concrete surface pressures are not exceeded.
l Section 8 of this report presents the detailed pool structure analysis.
2.3 Material Considerations 2.3.1 Introduction Safe storage of nuclear fuel in the Donald C. Cook spent fuel pool requires that the materials utilized in the fabrication of racks be of proven durability and be compatible with the pool water environment. This section provides the necessary inf ormation on l
this subject.
I 2.3.2 Struc3 ural Materials The following structural materials are utilized in the fabrication of the spent fuel racks:
- a. ASME SA240-304 for all sheet metal stock.
- b. Internally threaded support legs: ASME SA240-304.
I c. Externallf threaded support spindle: ASME SA564-630 precipitation hardened stainless steel.
- d. Weld material - per the following ASME specification:
SFA 5.9 ER308.
2.3.3 Poison Material In addition to the structural and non-structural stainless material, the racks employ Boral, a patented product of AAR Brooks
& Perkins, as the thermal neutron absorber material. A brief description of Boral, and its fuel pool experience list follows.
g Boral is a thermal neutron absorbing material composed of boron I carbide and 1100 alloy aluminum. Boron carbide is a compound l
I 2-4
\ ___ _ _--_ _ _-_____ _ ___- -____- _ - ___ ___ _ __ - - _ _ __ _ _ _ - __ __ . .. . . . .
l I having a high beron centent in a physically wtabit and chemical inert form. The 1100 alloy aluminum is a '.ight weight metal with I high tensile strength which is protectu: f u .n corrosion by a highly resistant oxide film. The two maiegir.'s, boron carbide and aluminum, are chemically compatib..e and ideally suited for long-t- use in the radiation, thermal and chemir:al environment of a spent fuel pool.
Boral's use in the spent fuel pool as the neutron absorbing material can be attributed to the fol2; wing reasons:
(i) The content and placement of boron carbide provides a very high removal cross section for thermal neutrons.
(ii) Boron carbide, in the form of fine particles, is homogenously dispersed throughout the central layer I of tne Boral.
(iii) The boron carbide and aluminum materials in Boral I do not degrade as a result of long-term exposure to gamma radiation.
(iv)
I The thermal neutron absorbing central layer of Boral is clad with permanently bonded surfaces of aluminum.
(v) Boral is stable, strong, durable, and corrosion re sis trant .
The passivation process of Boral in an aqueous environment results in the generation of hydrogen gar. If the generation rate of hydrogen is too rapid, then swelling of Boral may occur.
Laboratory studies by Boral's supplier indicate that the rate of hydrogen generation is a strong function of the so-called impurities in the chemical composition of the boron carbide powder, namely sodium hydroxide and boron oxida. AAR Brooks &
Perkins has instituted a strict program of monitoring of the chemistry of boron carbide used in the manuf acturing of Boral to ensure that no swelling of the panels will occur. Furthermore, l
2-5
I !
i randomly selected coupons of Boral panels fron production runs are !
subjected to swelling test checks to preclude any possibility of swelling of Boral.
Boral is manufactured by AAR Brooks & Perkins under the control and surveillance of a computer-aided Quality Assurance /Qualir.y l Control Program that conforms to the requirements of 10CFR50 i Appendix B, " Quality Assurance Criteria for Nuclear Power Plants and Fuel Reprocessing Plants". As indicated in Table 2.3.1, Boral j has been licensed by the USNRC for use in numerous BWR and PWR spent fuel st rage racks and has been extensively used in overseas
- E 5 nuclear installations.
] Boral Material Characteristics Aluminum: Aluminum is a silvery-white, ductile metallic element 1
that is abundant in the earth's crust. The 1100 alloy aluminum is used extensively in heat exchangers, pressure and storage tanks, chemical equipment, reflectors and sheet metal work.
It has high resistance to corrosion in industrial and marine atmospheres. Aluminum has atomic number of 13, atomic weight of 26.98, specific gravity of 2.69 and valence of 3. The physical /
mechanical properties and chemical composition of the 1100 alloy aluminum are listed in Tables'2.3.2 and 2.3.3.
I The excellent corrosion resistance of the 1100 alloy aluminum is provided by the protective oxide film that develops on its surface from exposure to the atmosphere or water. This film prevents the I loss of metal from general corrosion or pitting corrosion and the film remains stable between a pH range of 4.5 to 8.5.
1 I
I 2-6
r L
Doron Carbide: The boron carbide contained in Boral is a fine F
s granulated powder that conforms to ASTM C-750-80 nuclear grade Type III. The particles range in size between 60 and 200 mesh and the material conforms to the chemical composition and properties l listed in Table 2.3.4.
2.3.4 Comoatibility with Coolant All materials used in the construction of the Donald C. Cook racks i have un established history of in-pool usage. Their pl.ysical, chemical and radiological compatibility with the pool environment is an established fact at this time. As noted in Table 2.3.1, Boral has been used in both vented and unvented configurations in fuel pools with equal success. Consistent with the recent practice, the Donald C. Cook rack construction allowo full venting of the Boral space. Austenitic stainless steel (304) is widely used in nuclear power plants.
2.4 Existino Rac': Modules and Pronosed Reracking operation The Donald C. Cook fuel pool currently has medium density rack modules containing a total of 2050 storage cells in twenty modules. At the time of the proposed reracking operation, approximately 1678 cells (botween 6/1993 and 7/1994) out of 2050 locations will be occupied with spent fuel. There is sufficient number of open (unoccupied) cells in the pool to permit relocation of all fuel such that the existing modules can be emptied and removed from the pool, and new modules installed in a programmed manner.
A remotely engagable lift rig, which is designed to meet the criteria of NUREG-0612 " Control of Heavy Loads of Nuclear Power Plants", will be u s,ed to lift the empty modules. Auxiliary Building Crancs will be used for this purpose. A module change-out 2-7
scheme and procedure will be developed which ensures that all L modulee being handled are empty when the module is moving at a height which is more than 12" above the pool floor.
The Auxiliary Building has two overhead cranos which ride on rails that traverse the entire fuel handling area of the building. Each crane has a main hook rated .at 150 tons. These hooks are single I failure proof (SFP) (up to 60 tons). In addition there is an auxiliary hoist on the East Crane rated at 20 tons.
Pursuant to the defense-in-depth approach of LUREG-0612, the following additional measures of safety will be undertaken for the reracking operation.
(i) The crane and hoist will be given a preventive maintenance checkup and inspection within 3 months of the beginning of the raracking operation.
(ii) The crane hook will be used to lift no more than 50% of its single failure proof capacity of 60 tons I at any time during the reracking operation.
maximum weight of any module and its associated handling tool is 24 tons).
(The (iii) The old fuel racks will be lifted no more than 6" above the pool floor and held in that elevation for approximately 10 minutes before beginning the l vertical lift.
(iv) The rate of vertical lift will not exceed 6' per minute.
(v) The rate of horizontal movement will not exceed 6' per minute.
E (vi) Preliminary safe load paths have been developed.
The "old" cr "new" racks will not be carried over any region of the pool containing fuel.
(vii) The rack upending or laying down will be carried out in an area which is not overlapping to any safety related component.
(viii) All crew members involved in the reracking operation will be given training in the use of the lifting and upending equipment. The training l
l 2-8
I seminar will utilize videotacos of the actual lifting and upending rigs on typical modules to ig be installed in the pool. Every crew member Vill be 3 required to pass a written examination in the use of lifting and upending apparatus administered by the rack designer.
(ix) Referring to Figure 2.1.1, it is noted that the fuel handling bridge crane cannot access storage I cells facing the east wall and several locations in the southwest corner. Therefore, it will be necessary to load the inaccessible cells with fuel I when the rack is staged a certain (approximately 20 inches) from the pool wall.
Having loaded these cells, distance the module will be lifted approximately 4 inches above the pool liner, I and laterally transported to its final designated locations. A fuel shuf fling and rack installation sequence has been developed to ensure that all I heavy load handling criteria of NUREG-O t!12 are s at.i s fied . The rack handling rig is designed with consideration of the rack module weight along with the contained fuel assembly mass.
The fuel racks 'will be brought directly into the Auxiliary Building through the access door which is at ground level (609' elevation). This direct access to the building greatly facilitates the rack removal and installation effort.
I Tne "old" racks will be decontaminated to the extent practical on-site and approved fer shipping per the requirements of 10 CFR71 and 49 CFR 171-178, be housed in shipping containers, and transported to a processing facility for volume reduction. Non-decontaminatable portions of the racks will be shipped to a licensed radioactive waste burial site or returned to site for storage if disposal access is unavailable. The volume reduction is expected to reduce the overall volume of the racks to about 1/10th of their original value.
I All phases of the reracking activity will be conducted in
. accordance with written procedures which will be reviewed and
. approved by I&M.
I I 2-9
L F
l I
I I
Table 2.1.1 MODULE DATA I Module I.D. Quantity Array Cell Size Total Cell Count for this Module Tyne A** 13x14 910 I B C
5 4
4 12x14 13x12 672 624 D 2 12x12 288 I E F
G 4
2 1
13x11 12x11 12xiO 572 264 120 H* 1 13x14 -
(8x2) 151 Total 23 3616 I
I Non-rectangular module.
Three of the A modules have one triangle cell to accommodate pool corner curvature.
2-10 ..
I l
I I
I Taole 2.1.2 COMMON MODULE DATA Storage cell inside dimension: 8.75" i 0.04" Storage cell height (above the baseplate): 168 1/16" Baseplate thickness: 0.75" (nominal)
Support leg height: 5.25" (nominal)
Support leg type Remotely adjustable legs Number of support legs: 4 (minimum)
Remote lifting and handling provision: Yes Poison material: Boral Poison length: 144" Poison width: 7.5" Cell Pitch: 8.97" (nominal)
I I
I I
I I 2-11
r L
F L
r Table 2.1.3 MODULE DATA Dimensions (inch)*
l Shipping Weight 3 Module I.D. East-West North-South (kips)
A 117-3/16 126-3/16 25.7 B 108-1/8 126-3/16 23.7 C 117-3/16 108-1/8 22.5 D 108-1/8 108-1/8 20.9 E 117-3/16 99-1/16 20.8 1 F 108-1/8 99-1/16 19.3 a 108-1/8 90-1/8 17.7 H 117-3/16 126-3/16 23.9 I
I All dimencions are bounding rectangular envelopes rounded to the nearest one sixteenth of an inch.
I I
I l
l 1
l 2-12 l .. ,
Table 2.3.1 BORAL EXPERIENCE LIST (Demestic and Foreign)
Pressurized Water Reactors Vented Construc- Mfg.
Plant Utility tion Year I Bellefont 1, 2 Donald C. Cook 1, 2 Tennessee Valley Authority Indiana & Michigan Electric No No 1981 1979 Indian Point 3 NY Power Authority Yes 1987 Maine Yankee Maine Yankoe Atomic Power Yes 1977 Salem 1, 2 Public Service Elec & Gas No 1980 Seabrook New Hampshire Yankee No I
Sequoyah 1,2 Tennessee Valley Authority No 1979 Yankee Rowe Yankee Atomic Power Yes 1964/1983 Zion 1,2 Commonwealth Edison Co. Yes 1980 I Byron 1,2 Braidwood 1,2 Yankee Rowe Commonwealth Edison Co.
Commonwealth Edison Co.
Yankee Atomic Electric Yes Yes Yes 1988 1988 1988 Three Mile
'I Island I GPU Nuclear Yes 1990 4
Bolling Water Reactors Browns Ferry 1,2,3 Tennessee Valley Authority Yes 1980 Brunswick 1,2 Carolina Power & Light Yes 1981
'g Clinton Illinois Power Yes 1981 g Cooper Nebraska Public Power Yes 1979 1
Dresden 2,3 Commonwealth Edison Co. Yes 1981 Duane Arnold Iowa Elec. Light & Power No 1979 J.A. Fitzpatrick NY Power Authority No 1978 E.I. Hatch 1,2 Georgia Power Yes 1981 Hope Creek Public Service Elec & Cas Yes 1985 E Humboldt Bay Pacific Gas & Electric Yes 1986 5 Lacrosse Dairyland Power Yes 1976 Limerick 1,2 Philadelphia Electric No 1980 g Monticello Northern States Power Yes 1978 g Peachbottom 2,3 Philadelphia Electric No 1980 Perry, 1,2 Cleveland Elec. Illuminating No 1979 Pilgrim Boston Edison No 1978 I Shoreham Susquehanna 1,2 Vermont Yankee Long Island Lighting Pennsylvania Power & Light Yes No 1979 Vermont Yankee Atomic Power Yes 1978/1986 Hope Creek Public Service Elec & Gas Yes 1989 i
I
!I I 2-13
I il Table 2.3.1 (continued)
!I Foreign Installations Using Boral France i 12 PWR Plants Electricito de France I
South Africa l
l Kooberg 1,2 ESCOM i
! Switzerland
! Beznau 1,2 Nordostschweizerische Kraftwerke AG Gosgen Kernkraftwerk Gosgen-Daniken AG Taiwan Chin-Shan 1,2 Taiwan Power Company Kuosheng 1,2 Taiwan Power Company lI I
I I
I a 2-14
I I
I Table 2.3.2 1100 ALLOY ALUMINUM PHYSICAL AND MECHANICAL PROPERTIES i
I Density 0.098 lb/cu. in.
2.713 gm/cc Melting Range 1190-1215 deg. F 643-657 deg. C Thermal Conductivity 128 BTU /hr/sq ft/deg. F/ft (77 deg. F) 0.53 cal /sec/sq cm/deg. C/cm Coef. of Thermal 13.1 x 10-6/deg. F Expansion 23.6 x 10- /deg. C I- (68-212 deg. F)
Specific heat 0.22 BTU /lb/deg. F (221 deg. F) 0.23 cal /gm/deg. C Modulus of 10x106 pai Elasticity Tensile Strength 13,000 psi annealed (75 deg. F) 18,000 psi as rolled Yield Strength 5,000 psi annealed (75 deg. F) 17,000 psi as rolled Elongation 35-45% annealed (75 deg. F) 9-20% as rolled Hardness (Brinell) 23 annealed 32 as rolled 4 Annealing Temperature 650 deg. F 343 deg. C 2-15
4 I
i iI 1
i f
j Table 2.3.3 i CHEMICAL COMPOSITIOli (by weight) - ALUMItiUM (1100 Alloy)
(
99.00% min. Aluminum i
1.00% max. Silicone and Iron
!g 0.05-0.20% max. Copper l 5 .05% max. Manganese
! .10% max. Zine d
.15% max. others each
- I
!I lI
!I I
!I 4
- I 2-16
I I
Table 2. 3.4 I jlQRQU CARBIDE CHEMICAL COMPOSITION. Weicht %
I Total boron 70.0 min.
I B10 isotopic content in natural boron 18.0 Boric oxide 3.0 max.
I Iron 2.0 max.
Total boron plus 94.0 min.
total carbon BORON CARBIDE PHYSICAL PROPERTIES Chemical formula BC 4
Boron content (weight) 78.28%
Carbon content (weight) 21.72%
Crystal Structure rombohedral Density 2.51 gm./cc-0.0907 lb/cu. in.
Helting Point 24500 C (44420 7)
Boiling Point 35000 C (63320 7)
I Microscopic thermal-neutron cross-section 600 barn I
I I 2-17
~~M .~M M M M M M M M M -
M M M M M 58*-3 1/8* _,
q 1081/W s._ 117 ?/t r 7{t 1173/1C a ;__lal/Ir ; illlitr___ . 11Z21e 25/1#
ol a -
i l
I A1 81 '
B2 A2 i A3 A4 h ; Z-8 13 x 14 12x14 12 x 14 13 x 14 13 x 14 13 x 14 2 I
C1 D1 D2 C2 C3 c4 o 8 13 x 12 12 x 12 12 x 12 13x12 13 x 12 13 x 12 I _} . } m _ . .
.{
m *
.O g E1 F1 F2 E2 E3 E4 h 13x1. 12 x 11 12 x 11 13 x 11 13 x 11 13 x 11 4 :)*;e N
. ==.1 J $,\ NN MN" i N' M, i b A5 B3 B4 H G NN\
E 1M'M b b'
.- . sCdSR EAN'k te 8 13 x 14 12 x 14 12 x 14 13x14-8x2 12x10 < 'N N\\ b
.\ \ -
l,-
[ x \x } \ . \x; 3 - N' N gp -.
N c 1r
- i. N N p '
y N '
~3, ~
r} 453f4L 1(f.4= ~l Typical flack to flack Gap: 7 Total Storage: 3616 ceDs (loclude 3 Lhigular corner cets)
FIGURE 2.1.1 COOK SPENT FUEL POOL LAYOUT
I 3.0 CONSTRUCTION OE RACK MODULEE The object of this section is to provide a description of rack module construction for the DonaJd C. Cook spent fuel pool to enable an independent appraisal of the adequacy of the design.
Similar rack structure designs have recently been used in previous licensing efforts for Kuosheng Unita 1& 2 (Taiwan Power Company);
I J.A. Fit:: Patrick (New York Power Authority); Indian Point 2 (Consolidated Edison Company of New York, Inc.); Three Mile Island Unit 1 (GPU Nuclear); and Hope Creek 1 (Public Service Electric &
Cas Company). A list of applicable codes and standards is also presented.
3.1 Fabrication Objective The requirements in manufacturing the high density storage racks I for the Donald C. Cook fuel pool may be stated in four
- interrelated points:
I (1) The rack module will be fabricated in such a manner that I there is nQ weld splatter on the storage cell surfaces which would come in contact with the fuel assembly.
(2) The storage locations will be constructed so that I redundant flow paths for the coolant are available.
(3) The fabrication process involves operational sequences I which permit immediate verification by the inspection staff.
R (4) The storage cells are connected to each other by y austenitic stainless steel corner welds which leads to a honeycomb lattice construction. The extent of welding is selected to "detune" the racks from the seismic input motion of the Operating Basis Earthquake (OBE) and Design Basis Earthquake (DBE).
I
.I 3-1
'I
4 I
I 3.2 tibteLApng__Two Region Storng I
All rack modules designed and fabricatea for the Donald C. Cook spent fuel pool are of the so-called "non-flux trap" type. In the non-flux trap modules, a single screen of the poison panel is interposed between two fuel assemblies. The po3 son material utilized in this project is Boral, which does not require lateral support to prevent slumping due to the inherent s tif f ntis s .
However, accurate dimensional control of the poison location is essential for nuclear criticality and thermal-hydrsulic considerations. The design and fabrication approach to realize this objective is presented in the next sub-section.
3.3 Anatomy of Rack Modules
+
As stated earlier, the storage cell locations have a single poison panel between adjacent austenitic stainless steel surfaces. The significant componente of the rack module are (1) the storage box subassembly (2) the baseplate, (3) the thermal neutron absorber material, (4) picture frame sheathing, and (5) support legs.
(1) i I The rack module manufacturing begins with fabrication of the box. The " boxes" are fabricated from two precision formed channels by see2n welding in a machine equipped with copper chill bars and pneumatic clamps to minimize I distortion due to welding heat input.
shows the box.
Figure 3.3.1 I The minimum weld penetration will be 80% of the box metal gage which is 0.075" (14 gage). The boxes are manufactured to 8.75" I.D. (nominal inside dimension). !
I The design objective calls for installing Boral with minimal surface loading. This is accomplished by die forming a " picture frame sheathing" as shown in Figure I
I !
l 3-2 I
I
I 3.3.2. This sheathing is 0.035" thick and is made to precise dimensions such that the offset is .010" to
.005" greater than the poison material thickness.
As shown in Figure 3.3.1, each box has leur lateral 1" diameter holes punched near its bottom edge to provide I autillary flow holer. The edges of the sheathing and the box are welded togeher to form a smooth edge. The bcx, with integrally e -.n6eeted sheathing, is referred to as the " composite box'.
The "compocite boxes" are arranged in a checkerboard array to form an assemblage of storage cell locations I (Figure 3.3.3). The inter-box welding and pitch adjustmant are accomplished by small longitudinal connectors. Further details are given later in this section.
This assemblage of box assemblies is welded edge-to-edge I as shown in Figure structure dependirq 3.3.3, resulting in a honeycomb with axial, flexural and torsional rigidity on the extent of intercell welding provided.
It can be seen from Figure 3.3.3 that the edges of each I interior resulting box are connected to the contiguous boxes in a well defined path to resist shear.
I' (2) Baseplate _: The baseplate provides a continuwas horizontal surface for supporting the fuel assemblies.
The baseplate is attached to the cell assemblage by
- B fillet welds. The baseplate in each storage cell has a S" diameter flow hole. The baseplate is 3/4" thick to withstand accident fuel assembly drop loads postulated and d?.scussed in Section 7 of this report.
(f 1 thermal neutron absorber material: As mentioned in I tLo preceding section, neutron absorber material.
Boral is used as the thermal (4) Picture Frame Sheathina: As described earlier, the
'I sheathing serves as the locator and retainer of the poison material. Figure 3.3.2 shows a schematic of the sheathing.
l I
l 3-3 I
P I
I The poison material is placed in the customized flat depression region of the sheathing, which is next laid on a side of the " box". The precision of the shape of I the sheathing obtained by die forming guarantees that the poison sheet installed in it will not be subject to surface compression. The flanges of the sheathing (on I all four sides) are attached to the box using skip welds. The sheathing serves to locate and position the poison sheet accurately, and to preclude its movement under seismic conditions.
(5) Suecort Leas: All support legs are the adjustable type (Figure 3.3.4). The top portion is made of austenitic I steel material. The bottom part is made of SA564-630 stainless steel to avoid galling problems.
I Each support leg is equipped with a readily accessible socket to enable remote leveling of the rack af ter its placement in the pool. Lateral holes in the support leg provide the requisite coolant flow path.
An elevation cross-section of the rack module shown in k; .
Figure 3.3.5 shows two box cells, and a developed cell in elevation. The Boral panels and their leculon are also indicated in this figure. The boral panels are positioned such that the entire enriched fuel portion of
[].J the fuel assembly is enveloped by the thermal neutron absorber material.
The joint between the composite box arrays and the I baseplate is made by single fillet velds which provide a minimum of 7" of connectivity between each cell wall and the baseplate surface.
As shown in Figure 3.3.4, the support leg is gusseted to provide an increased section for load transfer between I the support legs and the cellular structure above the baseplate. Use of the gussets also minimizes heat input induced distortions of the support / baseplate contact region.
I I
I 3-4 I
I .
i I
I 3.4 Codes, Standards, and Practices for the Donald C. Cook Seent Fuel Pool Racks l The fabrication of the rack modules for the Donald C. Cook spent fuel pool is performed under a strict quality assurance system suitable for manufacturing and complying with the provisions of 10CFR50 Appendix B.
The following codes, standards and prcetices will be used as applicable for the design, construction, and assembly of the st.ent fuel storage racks. Additional specific references related to I detailed analyses are given in each section.
- a. Codes and Standards for Desion and Testino (1) AISC Manual of Steel Construction, 8th Edition, 1980.
(2) ANSI N210-1976, " Design Objectives for Light Water Reactor Spent Fuel Storage Facilities at Nuclear Power Stations'.
(3) American Society of Mechanical Engineers (ASME),
Boiler and Pressure Vessel Code,Section III, I Subsection NF, 1989.
(4) ASNT-TC-1A, June, 1980 American Society for I Nondestructive Testing (Recommended Practice for Personnel Qualifications).
(5) ASME Section V - Nondestructive Examination (6) ASME Section IX -
Welding and Brazing Qualifications I (7) Building Code Requirements for Reinforced Concrete, ACI318-89/ACI318R-89.
I I 3-5 I
I
I (8) Code Requirements for Nuclear Safety Related I Concrete Structures, ACI 349-85 and Commentary ACI 349R-85 (9) Reinforced Concrete Design for Thermal Effects on Nuclear Power Plant Structures, ACI 349.1R-80 .
(10) ACI Detailing Manual - 1980 (11) ASME NQA-2, Part 2.7 " Quality Assurance Requirements of Computer Software for Nuclear Facility Applications (draft).
(12) ANSI /ASME, Qualification and Duties of Personnel Engaged in ASME Boiler and Pressure Vessel Code I Section III, Div. 1, Certifying Activ.ities, N626 1977.
- b. Material Codes (1) American Society for Testing and Materials (ASTM)
Standards - A-240.
(2) American Society of Mechanical Engineers (ASME),
Boiler and Pressure Vessel Code,Section II - Parts A and C, 1989.
- c. Weldina Codes
-g ASHE Boiler and Pressure Vessel Code, Section IX-3 Welding and Brazing Qualifications (1986) or later issue accepted by USNRC.
- d. Quality Assurance, Cleanliness, Packacino, Shineino, Receivina, S_t o r a c e , and Handlina Recuirements j (1) ANSI N45.2.2 -
Packaging, Shipping, Receiving, 3 Storage and Handling of Items for Nuclear Power Plants.
(2) ANSI 4 5 . 't .1 - Cleaning of Fluid Systems and Associated Components during Construction Phase of Nuclear Power Plantn.
I I.
I 3-6
'I I
l I
I (3) ASME Boiler and Pressure Vessel, Section V, Nondestructive Examination, 1983 Edition, including I Summer and Winter Addenda, 1983.
ANSI N16.1-75 Nuclear Criticality (4) -
Safety operations with Fissionable Materials Outside Reactors.
(5) ANSI - N16.9-75 Validation of Calculation Methods I for Nuclear Criticality Safety.
N45.2.11, 1974 Quality (6) ANSI -
Assurance I Requirements Plants.
for the Design of Nuclear Power
.g (7) ANSI 14.6-1978, "Special Lifting Devices for g Shipping Containers weighing 10,000 lbs. or more for Nuclear Materials".
I (8) ANSI N45.2.6, Testing Personnel.
Qualification of Inspection and (9) ANSI N45.2.8, Installation, Inspection.
(10) ANSI N45.2.9, Records.
(11) ANSI N45.2.10, Definitions.
(12) ANSI N45.2.12, QA Audits.
(13) ANSI N45.2.13, Procurement.
(14) ANSI 45.2.23, QA Audit Personnel.
- e. Other References (In the references below, RG is NRC Regulatory Guide)
(1) RG 1.13 - Spent Fuel Storage Facility Design Basis, Rev. 2 (proposed).
I (2) RG 1.123 -
(endorses ANSI N45.2.13) Quality Assurance Requirements for Control of Procurement of Items and Services for Nuclear Power Plants.
(3) RG 1.124 - Service Limits and Loading Combinations for Class 1 Linear Type Component Supports, Rev. 1.
I 3-7
)
I
F
~
(4) RG 1.25 -
Assumptions Used for Evaluating the Potential Radiological Consequences of a Fuel i Handling Accident in the Fuel Bandling and Storage Facility of Boilj ng and Pressurized Water Reactors.
(5) RG 1.28 - (endorses ANSI N45.2) - Quality Assurance I Program Requirements, June, 1972.
(6) RG 1.29 - Seismic Design Classification, Rev. 3.
(7) RG 1.31 -
Control of Ferrite Content in Stainless Steel Weld Metal, Rev. 3.
(8) RG 1.38 - (endorses ANSI N45.2.2) Quality Assurance Requirements for Pr.ekaging, Shipping, Receiving, Storage and Handling of Items for Water-Cooled I Nuclear Power Plants, March, 1973.
(9) RG 1.44 -
Control of the Use of Sensitized Stainless Steel.
(10) RG 1.58 - (endorses ANSI N45.2.2) Qualification of Nuclear Power Plant Inspection, Examination, ano I Testing Personnel, Rev. 1, September, 1980.
(11) RG 1,64 -
(endorses ANSI N45.2.11) Quality Assurance Requirements for the Design of Nuclear I Power Plants, October, 1973.
(12) RG 1.71 - Welder Qualifications for Areas of I Limited Accessibility.
(13) RG 1.74 -
(endorses ANSI N45.2.10) Quality Assurance Terms and Definitions, February, 1974.
(14) RG 1.85 -
Materials Code Case Acceptability ASME Section III, Division 1.
(15) RG 1.88 -
(endorses ANSI N45.2.9) Collection, Storage and Maintenance of Nuclear Power Plant Quality Assurance Records, Rev. 2, October, 1976.
(16) RG 1.92 -
Combining Modal Responses and Spatial Components in Seismic Response Analysis.
3-8
)
I
[
(17).RG 3.41 -
Validation of Calculation Methods for Nuclear Criticality Safety.
~
(18) General Design Criteria for Nuclear Power Plants, Code of Federal Regulations, Title 10, Part 50, Appendix A (CDC Nos. 1, 2, 61, 62, and 63).
(19) NUREG-0800, Standard Review Plan, Sections 3.2.1, 3.2.2, 3.7.1, 3.7.2, 3.7.3, 3.8.4.
(20) "OT Position for Review and Acceptance of Spent Fuel Storage and Handling Applications," dated i April 14, 1978, and the modifications to this document of January 18, 1979. (Note: OT stands for Office of Technology).
!21) NUREG-0612, " Control of Heavy Loads at Nuclear Power Plants".
Guide 8.8, "Information Relative to I (22) Regulatory Ensuring that Occupational Radiation Exposure at Nuclear Power Plants will be as Low as Reasonably Achievable (ALARA).
(23) 10CFR50 Appendix B, Quality Assurance Criteria for Nuclear Power Plants and Fuel Reprocessing Plants (24) 10CFR21 - Reporting of Defects and Non-Compliance 2.5 Materials of Construction.
Storage Cell: ASME SA240-304 Baseplate: ASME SA240-304 Support Leg (female): ASME SA240-304 Support Leg (male): Ferritic stainless steel (anti-galling material) ASME SA564-630 Poison: Boral 3-9
,m ,__ - _. - m... _ a . ._1. .._.m.. . _.. . m. _ . . . _ _
e
!I 4
!I lI I
Lat eral I
flow i
holes
/
lI s
I s.7s-
- I v i
/
ig oe lI
. 1e s._
/ - 's" I
I ,1em,. 3.3.1 se m metersc eesc1szcm scemeo CHANNELS it g me I
. - . . .e.__._ E - -u%= a ..,a.. >
I
!I
!I
, \
'I
.I !m T O in 2 5
. 3 5
!l $
'a I
1
~
..I
' '=
I
.I
\ o I O ig
- 3-11 I
I I
I 1. <
?. is r r-- = 9 _m_,.n ., .=__ 1_ +;. x_ _ r, : r I
I \
4 >
3 I <
m
.- _ __ _ _ )
a.
n
,- rpm q -
a -4,. y v--- pKC w_ ,
6 > <
m > >
I h I r, a.
y T r
Ii-- A c V I g 7 ~ T\
p
> . > < s I <
// -A ,z..
e
',, /,
i m_
\e-y y . , -
b d > >
l I
> s L \ < s I, .\ .t J, f T sw ~/
,%s t I %-
I 4 > W > s. -
I c
\' / V 'I l
\' / / g- 7 I Figure 3.3.3 ARRAY OF CELLS FOR NON-FLUX TRAP MODULES I
3-12
.I
I I t.
Baseplate in I LNN\\\\\\\i i
k \ \ \ \ \ \ \ \ \ \ \ \ \w
/ ~[ ,
I / /
'N
/)!
/ -r W
![/) Gusse;
___q g p_ __ _ _
I p__q O p___
I I
I .
m 1 I C) ()' '
I x. n I 8 CJ j
I Figure 3.3.4 ADJUSTABLE SUPPORT LEG I
2.n g
Box Cells I.
I Cell Pitch Boral panel
=
I '
l Developed cell sk . .
t I
?
f l h Sheathing
~
( i #
L /
i I
C i i l' .
l )
f
/p . .' .
I / s I
/ -
._ / .
v
~
< / ; /
l i
'i l E '
3 :
11 i ll
-E - --- -
- A )!
7 Y/// //// '/// //f/// h h
Baseplate One Inch Lateral Flow Hole _(Typical)
Figure 3.3.5 ELEVATION VIEW OF RACK MODULE I
3-14
I 4.0 CRITICALITY SAFETY ANALYSES 4.1 Desian Basis The high density spent fuel storage racks for Donald C. Cook Nuclear Plant are designed to assure that the effective neutron multiplication f actor (k,,,) is ecual to or less than 0.95 with the racks fully loaded with fuel of the highest anticipated reactivity, and flooded with unborated water at the temperature within the operating range corresponding to the highest reactivity. The maximum calculated reactivity includes a margin for uncertainty in reactivity calculations including mechanical tolerances.
I uncertainties are statistically combined, such that the final k,,,
All will be equal to or less than 0.95 with a 95% probability at a 95%
confidence level.
Applicable codes, standards, and regulations or pertinent sections thereof, include the following:
o General Design Criteria 62, Prevention of Criticality in Fuel Storage and Handling.
o USNRC Standard Review Plan, NUREG-0800, Section 9.1.2, Spent Fuel Storage, Rt..v. 3 - July 1981 o USNRC letter of April 14, 1978, to all Power Reactor Licensees -
OT Position for Review and Acceptance of Spent Fuel Storage and Handling Applications, including modification letter dated January 18, 1979.
I c USNRC Regulatory Guide 1.13, Spent Fuel Storage Facility Design Basis, Rev. 2 (proposed), December 1981.
o ANSI ANS-8.17-1984, Criticality Safety Criteria for the Handling, Storage and Transportation of LWR Fuel Outside Reactors.
4-1 I
I To assure the true reactivity will always be less than the calculated reactivity, the following conservative assumptions were made:
o Moderator is assumed to be unborated water at a I temperature within the operating range that results in the highest reactivity (determined to be 20 *C).
I o The effective multiplication factor of an infinite radial array of fuel assemblies was used (see section 4.4.1) except for the boundary storage cells where leakage is inherent, o Neutron absorption in minor structural members is neglected, i.e., spacer grids are analytically replaced by water.
I The design basis fuel assembly is a 15 x 15 (Standard) Westinghouse containing UO2 at a maximum initial enrichment of 4.95 t 0.05 wt%
I U-235 by weight. For fuel assemblies with natural UO2 blankets, the enrichment is that of the central enriched zone. Calculations confirmed that this reference design fuel assembly was the most reactive of the assembly types expected to be stored in the racks.
Three separate storage regions are provided in the spent fuel storage pool, with independent criteria defining the hignest potential reactivity in each of the two regions as follovs:
o Region 1 is designed to accommodate new fuel with a maximum enrichment of 4.95 0.05 wt% U-235, or spent fuel regardless of the discharge fuel burnup.
o Region 2 is designed to accommodate fuel of 4.95% initial enrichment burned to at least 50,000 MWD /MtU (assembly average), or fuel of other enrichments witn a burnup yielding an equivalent reactivity.
o Region 3 is designed to accommodate fuel of 4.95% initial enrichment burned to at least 38,000 MWD /MtU (assembly "I average), or fuel of other enrichments with a burnup yielding an equivalent reactivity.
4-2
!I ll The water in the spent fuel storage pool normally contains soluble boron which would result in large subcriticality margins under
- actual operating conditions. However, the NRC guidelines, based upon the accident condition in which all soluble poison is assumed
- to have been lost, specify that the limiting k,,, of 0.95 for normal I stcrage be evaluated in the absence of soluble boron. The double I
, contingency principle of ANSI N-16.1-1975 and of the April 1978 I
NRC letter allows credit for soluble boron under other abnormal or accident conditions since only a single independent accident need be considered at one time. Consequences of abnormal and accident conditions have also been evaluated, where " abnormal" refers to
! conditions which may reasonably be expected to occur during the lifetime of the plant and " accident" refers to conditions which are not expected to occur but nevertheless must be protected i against.
- I
- I
!I 1
I I
1 .
I 4-3 I
L 4.2 Summary of Criticality Analyses 4.2.1 Normal Ocoratino Conditionq
_l The design basis layout of storage cells for the three regions is shown in Figure 4.1. In this configuration, the fresh fuel cells (Region 1) are located alternately along the rack periphery (where neutron leakage reduces reactivity) or along the boundary betwee..
I two storage modules (where the water gap provides a flux-trap whicu reduces reactivity). High burnup fuel in Region 2 affords a low-rativity barrier between fresh fuel assemblies and Region 3 fuel of intermediate burnup. There are at the present time, an adequate number of spent fuel assemblies to nearly fill and " block off" the Region 2 barrier locations (see Sec' ion 4.7). Thus, the administrative controls required are comparable to a conventional two-region storage rack design.
I Prior to approaching the reactor end-of-lif e, not all storage cells are needed for spent fuel. Therefore. an alternative configuration may be used in which the internal cells are loaded in a
(
j{}
checkerboard pattern of fresh fuel (or fuel of any burnup) with empty cells, as indicated in Figure 4,2. This configuration is intended primarily to facilitate a full core unload when needed, prior to the time the racks are begirning to fill up.
Figure 4.3 define the acceptable burnup domains and illustrates the limiting burnup for fuel c-! variouc initial enrichments for both Region 2 (upper curve) or Region 3 (lower curve) , both of which assume that the fresh fuel (Region 1) is enriched to 4.95% U-235.
I Criticality analyses show that the most reactive configuration occurs along the boundary between modules with the reactivity of 4-4 l
l
i l
I i 1
- the edge configuration being slightly lower. The bounding criticality analyses are summari::ed in Table 4.1 for the design basis storage condition (which assumes the single accident i
condition of the loss of all soluble boron) and in Table 4.2 for the interim checkerboard loading arrangement. The calculated I maximum reactivity of 0.940 (same for both the normal storage condition and the interim checkerboard arrangement) is within the I regulatory limit of a k,,, of 0.95. This maximum reactivity includes calculational uncertainties and manufacturing tolerances (95% probability at the 95% confidence level), an allowance for uncertainty in depletion calculations and the evaluated effect of the axial distribution in burnup. Fresh fuel of less than 4.95%
enrichment would result in lower reactivities. A; cooling time increasec in long-term storage, decay of Pu-241 results in a continuous decrease in reactivity, which provides an increasing suberiticality margin with time. No credit is taken for this decrease in reactivity other than te indicate conservatism in the calculations.
The burnup criteria identified above (Figure 4-3) for acceptable storage in Region 2 and Region 3 can be implemented in appropriate administrative procedures to assure verified burnup as specified in the proposed Regulatory Guide 1.13, Revision 2. Administrative procedures will also be employed to confirm and assure the presence of soluble poison in the pool water during I fuel handling operations, as a further margin of safety and as a precaution in the event of fuel misp3acement during fuel handling operations.
I The thick base-plate on the rack modules extend beyond the storage cells and provide assurance that the necessary water-gap between modules is maintained.
4-5 I
l For convenience, the minimum (limiting) burnup data in Figure 4.3 for unrestricted storage may be described as a function of the initial enrichment, E, in weight percent U-235 by fitted polynomial expressions as follows; i
For Recion 2 Storace Minimum Burnup in MWD /MTU =
2 3
- 22,670 + 22,220 E - 2,260 E + 149 E I fpr Recion 3 Storace Minimum Burnup in MWD /MTU =
- 26,745 + 18,746 E - 1,631 E2 + 98. 4 E3 1
4.2.2 Abnorinal and Accident conditions Although credit for the soluble poison normally present in the I spent fuel pool water is permitted under abnormal or accident conditions, most abnormal or accident conditions will not result in exceeding the limiting reactivity (k,g of 0.95) even in the absence of soluble poison The effects on reactivity of credible abnormal and accident conditions are discussed in Section 4.7 and summarized in Table 4.3. of these abnormal or accident conditions, only one has the potential for a more than negligible positive reactivity effect.
s
- -6
I l
The inadvertent misplacement of a fresh fuel assembly has the l potential for exceeding the limiting reactivity, should there be a concurrent and independent accident condition resulting in the loss of all soluble poison. Administrative procedures to assure the presence of soluble poison during fuel handling operations will preclude the poss3bility of the simultaneous occurrence of the two independent accident conditions. The largest reactivity increase I (+ 0.065 Sk) would occur if a new fuel assembly of 4.95% enrichment were to be positioned in a Regic:1 2 location with the remainder of the rack fully loaded with fuel of the highest permissible
! reactivity. Under this accident condition, credit for the presence of soluble poison is permitted by NRC guidelines , and calculations indicate that 550 ppm soluble boron would be adequate to reduce the k,,, to the calculated k,,, (0.940) and approximately 450 ppm would be sufficient to assure that the limiting k ,, , of 0.95 is not exceeded.
1
'I I
I Double contingency principle of ANSI N16.1-1975, as specified in the April 14, 1978 NRC letter (Section 1.2) and implied in the proposed revision to Reg. Guide 1.13 (Section 1.4, Appendix A).
4-7
I 4.3 Reference Fuel Storace Cells I
4.3.1 Peference Fuel Assembly.
]
I The design basis fuel assembly, described in Figure 4.4, is a 15 x 15 array of fuel rods with 21 rods replaced by 20 control rod guide tubes and 1 instrument thimble. Tab]e 4.4 summarizes the I fuel assembly design specifications and the expected range of significant manufacturing tolerances. As shown below, initial cell calculations with CASMO-3 indicated that the H 15 x 15 fuel exhibited a slightly higher reactivity in the storage rack cell than either the H 17 x 17 standard or optimized (OFA) fuel or the ANF fuel assembly designs.
Burnup Cell
, Fuel tvoe Enri.chment MWD /KcU Lg I E 15 x 15 H 15 x 15 2.5 2.5 0
10 1.0261 '
0.9210 H 17 x 17 OFA 2.5 0 1.0205 i E 17 x 17 OFA 2.5 10 0.9144 4
W 17 x 17 Stnd 2.5 0 1.0217 5 17 x 17 Stnd 2.5 10 0.9188
! ANF 15 x 15 2.5 0 1.0148 ANF 17 x 17 2.5 0 1.0126 H 15 x 15 4.95 0 1.1941 ',
H 15 x 15 4.95 40 0.9204 H 17 x 17 OFA 4.95 0 1.1933 E 17 x 17 OFA 4.95 40 0.9149 t
H 17 x 17 Stnd 4.95 0 1.1980 ANF 15 x 15 4.95 0 1.1857 ANF 17 x 17 4.95 0 1.1883 Highest values II I
4-8
r l Based upon the calculations listed above, the Westinghouse 15 x 15 rod design was used as the basis for the criticality calculations.
4.3.2 ILich Densit" Fuel Storace Cells 1
The nominal spent fuel storage cell used for the criticality analyses of the Donald C. Cook spent fuel storage cells is shown in Figure 4.4. Each storage cell is composed of Boral absorber panels positioned between a 8.75-inch I.D., 0.075-inch thick inner stainless steel box, and a 0.035-inch outer stainless steel sheath which "orms the wall of the adjacent cell. The fuel assemblies are normally located in the center of each storage cell on a nominal lattice spacing of C.97 i O.04 inches. The Boral absorber has a thickness of 0.101 t 0.005 inch and a nominal B-lO areal density of 0.0345 g/cm 2.
I l
l .
4-9
4.4 Analvtical Methodolocv 4.4.1 Reference Desian Calculations In the fuel rack analyses, the primary criticality analyses of the high density spent fuel storage racks were performed with the KENO-Sa computer code package *, using the 27-group SCALE cross-section library and the NITAWL subroutine for U-238 resonance shielding effects (Nordheir integral treatment). Depletion analyses and determination of equivalent enrichments were made with the two-dimensional transport theory code, CASMO-3 . Benchmark calculations, presented in Appendix A, indicate a bias of 0.0000 with an uncertainty of 0.0024 for CASMO i and O.0090 1 0.0021 (95%/95%) for NITAWL-KENO-Sa. In tracking long-term (30-year) reactivity effects of spent fuel stored in Region 2 of the fuel storage rack, previous CASMO calculations confirmed a continuous reduction in reactiv ;y with time (af ter Xe decay) due primarily to m Pu-241 decay and Am-241 growth.
KENO-Sa Monte Carlo calculations inherently inudde statistical uncertainty due to the random nature of neutron tracking. To minimize the statistical uncertainty of the KENO-calculated reactivity, a minimum of 500,000 neutron histories in 1000 generations of 500 neutrons each, are accumulated in each calculation. For the design calculation for the racks, 1,250,000 histories were used to confirm convergence of the KENO-Sa I calculation.
Figure 4.5 represents the basic geometric modal used in the KENO-Sa calculations. This model effectively describes a repeating array of 10 storage cells in the X-direction separated by a 2-inch water lF
" SCALE" is an acronym for Standardized Computer Analysis for Licensing Evaluation, a standard cross-section set developed by ORNL for the USNRC.
4-10
gap between modules and an infinite array of cells in the Y-
[ direction (periodic boundary conditions). In the axial (2) direction, the full length 144-inch fuel assembly was described with a 30-cm water reflector. A similiar model was used for l
I calculations of the rack peripheral cells where the calculations were made with both water and concrete reflectors (a concrete reflector gave a slightly higher reactivity by 0.004 6k).
Larger models, encompassing an entire storage module (half of an 11 x 11 array, run for 1,250,000 neutron histories to assure convergence) confirmed results cotained with the smaller infinite array model. The larger model was also used to confirm the reactivity calculation for the checkerboard arrangement with fresh I fuel and empty cells in Region 3 and in the investigation of the consequences of potential accident conditions with a misplaced fresh fuel assembly. In addition, the corner intersection was explicitly modeled and, as expected, gave a lower reactivity than the reference design calculation.
In the CASMO-3 geometric model (cell), each fuel rod and its cladding were described explicitly and reflecting boundary conditions (zero neutron current) were used in the axial direction and at the centerline of the Boral and steel plates between storage I cells. These boundary conditions have the effect of creating an infinite array of storage cells in all directions and provide a I conservative estimate of the uncertainties in reactivity attributed to manufacturing tolerances.
Because NITAWL-KENO-Sa does not have burnup capability, burned f uel un represented by fuel of equivalent enrichment as determined by tA@G -3 calculations in the storage cell (i.e. an enrichment which Fialdu the same reactivity in the storace cell as the burned fuel) .
4'1
t Tigure 4.6 shows this equivalent enrichment for fuel of 4.95%
initial enrichment at various discharge burnups, evaluated in the storage cell.
4.4.2 Fuel Burnue Calculations and Uncertainties CASMO-3 was used for burnup calculations i.n the hot operating l condition. CASMO-3 has been extensively benchmarked (Appendix A and Refs. 2 and 7) against critical experiments (including plutonium-bearing fuel). In addition to burnup calculations, CASMO-3 was used for evaluating the small reactivity increments (by differential calculations) associated with manufacturing tolerances.
I Since there are no critical experiment data with spent fuel for determining the uncertainty in burnup-dependent reactivity i calculations, an allowance for uncertainty in reactivity" was assigned based upon the assumption of 5% uncertainty in burnup.
I This is approximately equivalent to 5% of the total reactivity decrement. At the design basis burnups of 38 and 50 IGD /KgU, the uncertainties in burnup are t 1.9 and 2.5 MWD /KgU respectively.
To evaluate the reactivity consequences of the uncertainties in burnup, independent calculations were made with fuel of 36,100 and 47,500 MWD /MtU burnup in Regions 2 and 3, and the incremental change from the reference burnups assumed to represent the net uncertainties in reactivity. These calculations resulted in an g
incremental reactivity uncertainty of 0.0047 6k in Region 2 B (isolation barrier at 50 MWD /KgU burnup) and 0.0019 for Region 3 (at 38 MWD /KgU burnup). .In the racks, the fresh unburned fuel in Region 1 strongly dominate the reactivity which tends to minimize the reactivity consequences of uncertainties in burnup. The 0nly that portion of the uncertainty due to burnup. Other uncertainties are accounted for elsewhere.
4-12
r L
allowance for uncertainty in burnup calculations is a conservative l estimate, particularly in view of the substantial reactivity decrease with time as the spent fuel ages.
l 4.4.3 Effect of Axial Burnuo Distribution Initially, fuel loaded into the reactor will burn with a slightly i skewed cosine power distribution. As burnup progresses, the burn distribution will tend to flatten, becoming more highly burned in the central regions than in the upper and lower ends, as may be seen in the curves compiled in Ref. 4. At high burnup, the more reactive fuel near the ends of the fuel assembly (less than average burnup) occurs in regions of lower reactivity worth due to neutron leakage. Consequently, it would be expected tha : over most of the burnup history, distributed burnup fuel assembli s would exhibit a slightly lower reactivity than that calculat c ' for the average i burnup. As burnup prograsses, the distribution, to some extent, tenda to be self-regulating as controlled by the axial power I distribution, precluding the existence (I M ge regions of significantly reduced burnup. Among others, Nuls r* has prcvided generic analytic results of the axial burm g Mtect bast.d upon calculated and measured axial burnup distrib6tir nr e These analyses confirm the minor and generally negative reactivity effect of the axially distributed burnup. The trends observed, howcVer, suggest g the possibility of a small positive reactivity effect at high P burnup.
Calculations were made with KENO-Sa in three dimensions, based upon the typical axial burnup distribution of spent fuel (that observed at the Surrey plant was taken as representative). In these calculations, the axial height of the burned fuel was divided into a number of axial zones (6-inch intervals near the more significant
-top of the fuel), each with an enrichment equivalent to the burnup 4-13
of that zone. These calculations resulted in an incremental reactivity increase of 0.0037 6k for the reference design case.
Fuel of lower initial enrichments (and lower burnup) would have a p smaller (or negative) reactivity effect as a result of the axial variation in burnup. These estimates are conservative since smaller axial increments in the calculations have been shown to result in lower incremental reactivities *.
I I
I l
l l
l I
I h
4-14
4.5 Criticality Analyses and Tolerances u 4.5.1 Nominal Desian For the nominal storage cell design, the NITAWL-KENO-Sa calculation resulted in a bias-corrected k, of 0.9250 0.0012 (923/95%),
which, when combined with all known uncertainties and the axial burnup effect, results in a k, of 0.929 ! O.011 or a maximum k, of I 0.940 with a 95% probability at the 95% confidence level *,
I For the interim loading pattern of checkerboarded fuel and empty cells in Region 3, calculations resulted in essentially the same reactivity as the reference de.s ig n within the normal KENO-Sa statistics (maximum k, of 0.940, including all allowances and uncertainties, see Table 4.2).
4.5.2 Uncertainties Due to Manufacturina Tolerances The uncertainties due to manuf acturing tolerances are summarized in Table 4-5 and discussed below.
4.5.2.1 Boron Loadina Tolerances {
i The Boral absorber panels used in the storage cells are nominally 4
O.101 inch thick, 7.50-inch wide and 144-inch long, wit: a nominal I B-10 areal density of 0.0345 g/cm. 2 The vendors manufacturing 2
tolerance limit is, t 0.0045 g/cm in B-10 content which assures that at any point, the minimum B-10 areal density will not be less 2
than 0.030 g/cm. Differential KENO-Sa calculations for the reference design with the minimum tolerance B-10 loading results in an incremental reactivity of + 0.00614 6k uncertainty.
4-15
)
I l 4.5.2.2 Boral Width Tolerance l
The reference storage cell design uses a Boral panel with an I initial width of 7.50 ! O.06 inches. For tite maximum tolerance of 0.06 inch, the differential CASMO-3 calculated reactivity uncertainty is 0.0009 6k.
I 4.5.2.3 Tolerances in Cell Lattice Soaqina The manufacturing tolerance on the inner box dimension, which directly affects the storage cell lattice spacing between fuel assemblies, is ! O.06 inches. This corresponds to an uncertainty in reactivity of t 0.0015 Sk determined by differential C AS:!O- 2 calculations.
I 4.5.2.4 Stainless Steel Thickness Tolerances The nominal stainless steel thickness is 0.075 0.005 inch f or the inner stainless steel box and 0.035 0.003 inch for the Boral cover plate. The maximum positive reactivity effect of the expected stainless steel thickeess tolerances was calculated (CASMO-3) to be + 0.0009 6k.
I 4.5.2.5 Fuel Enrichment and Density Tolerances I The design maximum enrichment is 4.95 ! O.05 wtt U-235. Separate CASMO-3 burnup calculations were made for fuel of the maximum enrichment (5.00%) and for the maximum UO 2 density (10.50 g/cc).
Reactivities in the storage cell were then calculated using the restart capability in CASMO-3 and equivalent enrichments determined for the reference fuel burnups of 38 and 50 MWD /KgU. The incremental reactivities between these calculations and the reference CASMO-3 cases, were conservatively taken as the B
4-16 I
I
sensitivity to small enrichment and unsity variations. For the tolerance en U-235 enrichment, the uncertainty in k, is i O.0034 6k and fer fuel density is i O.0035.
4.5.3 Water-cao Scacina Bt: tween Modules 4 The water-gap between modules constitute a neutron flux-trap for the outer (peripheral) row of storage cells, calculations with KENO-Sa were made for various water-Gap spacings (Figure 4.7).
From there data, it was determinud that the incremental reactivity consequence (uncertainty) for the minimum water-gap tolerance of 1/4 inch is t 0.0045 6k. The racks ara sonstructed with the base plate extending beyond the edge of the calls. This assures that a minimum spacing of 1.75 inch between storage modules is maintained under all credible conditions.
4.5.4 Eccentric Fuel Positioninc The fuel assembly is assumed to be normally located in the conter of the storage rack cell. Infinite array calculations wero -ad-using KENO-Sa for a single (.all with the fuel assemblies centered
, and with the assembliert assumed to be in the corncr of the storagr4 rack cell (four-assembly cluster at closest approach). LMr.o
~
calculations indicated that the reactivity uncertainty could be as
- much as i O.0019 6k.
4.6 Abnormal and Aggident CQDM tions 4.6.1 Temoetature and Water Density Effects The moderator temperature coefficient of reactivity is negative; a moderator temperature of 20*C (68'F) was assumed for the reference designs, which assures that the true reactivity- will always be lower over the expected range of water temperatures. Temperature
- 4-17 4
4
~
offects on reactivity have been calculated and the results are L shown in Table 4.6. with seluble poison present, the temperature coefficients of reactivity would dif f er from those inferred from
~
the data in Table 4.6. However, the reactivities would also be substantially lower at all temperaturen with soluble boron present, and the data in Table 4.6 is pertinent to the higher-reactivity unborated case.
I 4.6.2 Dronced Fuel Aqpembly For a drop on top of the rack, the fuel assembly will come to rest horizontally on top of the rack with a minimum seperation distance from the fuel in the rack of more than 12 inches, including the potential deformation under sulsuic or accident conditions. At this separation distance, the ef f ect on reactivity is insignificant
(<o.0001 6k). Furthermore, soluble boron in the pool water would I substantially reduce the reactivity and assure that the true reactivity is always less than the limiting value for any conceivable dropped fuel accident.
4.6.3 Lateral Rack Movement Lateral motion of the rack modules under seismic conditions could potentially alter the spacing between rack modules. However, the maximum rack movement has benn determined to be less than the tolerance on the water-gap spacing. Furthermore, soluble poison would assure that a reactivity less than the design limitation is maintained under all accident or abnormal conditions.
4-18 l
s
..mn- .-
.I 4.6.4 Abnormal Location of a Puol Assembly The abnormal location of a fresh unirradiated fuel assembly of 4.95
- vtt enrichment could, in the absence of soluble poison, result in j exceeding the design reactivity limitation (k. of 0.95). This could occur if a fresh fuel assembly of the highest permissible enrichment were to be either positioned outside and adjacent to a -
utoragc rack mcdule or inadvertently loaded into either a Region 2 or Region 3 storage cell. Calculations (KENO-Sa) showed that the highest reactivity, including uncertainties, for the worst case postulated accident condition (fresh fuel assembly in Region 2) would exceed the limit on reactivity in the absence of soluble boron. Soluble boron in the spent fuel pool water, for which
) credit is permitted under these accident conditions, would assure that the reactivity is maintained substantially less than the
{ design limitation. It is estimated that a soluble poison I concentration of 550 ppm boron would be sufficient to maintain k,
! at the reference design value of 0.940 under the maximum postulated
! accident condition. Approximately 450 ppm boron would be required to limit the maximum reactivity to a k,,, of 0.95.
4.7 Existina Spent Fue_1
- I As of May 1990, there were 1596 spent fuel assemblies in storage at
- g the Donald C. Cook plant, including those now in the reactor and B their projected burnups at discharge. Figure 4.8 superimposes the enrichment-burnup combination of these fuel assemblics on the curves defining the acceptable burnup domains. As :::ay be seen in this figure, most of the spent fuel now in storage falls well into the acceptable domain for the barrier fuel (Region 2). The number of fuel assemblies meeting the enrichment-burnup criteria for storage in Region 2 is 1390 which will nea71y fill the 1447 Region 2 storage locations. Twelve fuel 3ssemblies (discharged I
4-19 I
, - , - - - ,w ,s- - ~ - - , , , , - ,w,
l I
prematurely for various reasons) will need to be kept in a Region 1 storage location, and the remaining 194 assemblies may be stored I in Region 3 locations. Future discharge batches may reasonably be expected to have a proponderance of highly burned fuel capable of being stored in Region 2 (or in Region 3 once Region 2 is filled).
An appreciable number of spent fuel assemblies have enrichment-burnup combinations well in excess of the design basis and this I provides further conservatism in the criticality safety of the spent fuel storage rack design.
I I
I I
I I
I I
I
I L
[ 4.8 Referencea I
I 1. Green, Lucious, Patrie, Ford, White, and Wright, "PSR I /NITAWL-1 (code package) NITAWL Modular Code System For Generating Coupled Multigroup Neutron-Gamma Libraries from ENDF/B", ORNL-TM-3706, Oak Ridge National Laboratory, November 1975.
- 2. R.M. Westfall et. al., " SCALE: A Modular System for Performing Standardized Computer Analysis for Licensing I Evaluation", NUREG/CR-02OO, 1979. Volume 2, Section Fil,
" KENO-Sa An Improved Monte Carlo Criticality Program with Supergrouping".
- 3. A. Ahlin, M. Edenius, and H. Haggblom, "CASMO - A Fuel Assembly Burnup Program", AE-RF-7 6-4158, Studsvik report.
A. Ahlin and M. Edenius, "CASMO - A Fast Transport Theory Depletion Code for LWR Analysis", ANS Transactions, Vol.
26, p. 604, 1977.
"CASMO-3 A Fuel Assembly Burnup Program, Users Manual",
Studsvik/NFA-87/7, Studsvik Energitechnik AB, November 1986
- 4. H. Richings, Some Notes on PWR (W) Power Distribution Probabilities for LOCA Probabilistic Analyses, NRC Memorandum to P.S. Check, dated July 5, 1977.
- 5. S. E. Turner. " Uncertainty Analysis - Burnup Distribu-tions", presented at the DOE /SANDIA Technical Meeting on Fuel Burnup Credit, Special Session, ANS/ ENS Conference, Washington, D.C., November 2, 1988
- 6. M.G. Natrella, Experimental Statistics National Bureau of Standards, Handbook 91, August 1963 4-21
--_--.--._----,----.w- - - - - - . - - - ---,---,u----,----- - - - . - - - --- ---x-~_,..,--- s---" -
~
Table 4.1
~
SUMMARY
OF CRITICALITY SAFETY ANALYSES NORMAL STORAGE CONFIGURATION I Design Basis durcips at 6.95%
0.05% initiil ensick.ent O in Region 1 50 in Region 2 38 in Region 3 Temperature for analysis 20'C (68'F)
Reference k, (KENO-5a) 0.9160 I Calculational bias, 6k O.0090 Uncertainties Bias statistics (95%/95%) i O.0021 KENO-Sa statisticu (95%/95%)
i Manufacturing Tolerances Water-gap i
i i
O.00l?
O.0064 O.0045 Fuel enrichment 2 0.0034 Fuel density i O.0035 Burnup (38 MWD /KgU) i O.0019 Burnup (50 MWD /KgU) i O.0047 Eccentricity in position 0.0019 I Statistical combingion i O.0110 of uncertainties Axial Burnup Effect 0.0037 Total O.9287 0.0110 Maximum Reactivity (k.) 0.940 See Appendix A Square root of sum of squares. .
4-22
I I Table 4.2
SUMMARY
OF CRITICALITY SAFETY ANALYSES INTERIM CHECKERBOARD LOADING Design Basis burnups at 4.95% 0 in Region 1 I O.05% initial enrichment 50 in Region 2 Region 3 -CHECKERBOARD (FRES!! FUEL AND EMPTY)
- n I Temperature for analysis 20'C (68'F)
Reference k, (KENO-Sa) 0.9168 I Calculational bias, Sk N O.0090 Uncertainties (Assumed same as the reference case)
Bias statistics (95%/95%)
I KENO-5a statistics (95%/95%)
Manufacturing Tolerances 1
i i
0.0021 O.0012 O.0064 Water-gap i O.0045 I Fuel enrichment Fuel density Burnup (38 MWD /KgU) 1 i
0.0034 O.0035 NA I Burnup (50 MWD /KgU)
Eccentricity i O.0047 0.0019 Statisticalcombingion i O.0108 l of uncertainties Axial Burnup Effect 0.0037 I
= Total O.9295 i O.0108 Maximum Reactivity (k.) 0.940 1
See Appendix A Square root of sum of squares.
I 4-23
'I
1 I
I I
I I
I Table 4.3 I REACTIVITY EFFECTS OF ABNORMAL AND ACCIDENT CONDITIONS Accident / Abnormal Conditions Reactivity Effect l Temporature increase (above 68'F) Negative (Table 4.6)
Void (boiling) Negative (Table 4.6)
Assembly dropped on top of rack Negligible (<0.0001 6k)
Lateral rack module movement (Included in Tolerances)
Misplacement of a fuel assembly Positive (0.065 Max 6k)
(controlled by soluble poison)
W 4-24 i
l I
Table 4.4 DESIGN BASIS FUEL ASSEMBLY SPECIFICATIONS I FUEL ROD DATA Outside diameter, in. O.413 Cladding thickness, in. 0.024.1 Cladding inside diameter, in. O.3734 Cladding material Zr-4 Pellet density, % T.D. 95.0 stack density, g UO 2 /cc 10.29 0.20 Pellet diameter, in. O.3659 Maximum enrichment, wt % U-235 4.95 0.05 FUEL ASSEMBLY DATA Fuel rod array 15 x 15 Number of fuel rods 204 Fuel rod pitch, in. O.563 Number of control rod guide and 21 instrument thimbles Thimble 0,D., in. (nominal) 0.533 Thimble I.D., in. (nominal) 0.499 l
I 1
4-25 1
I
I Table 4.5 Reactivity Ef f'ects of Manuf acturing Tolerances Tolerance Incremental 'eactivity, Sk Boron-10 loading ( O.004 5 g/cm*) 0.0061 Boral Width (t 1/16 inch) 0.0009 Lattice spacing ( O.04 inch) 0.0015 Stainless Thickness (t 0.005 inen) ! O.0009 Total (statistical sum) i O.0064 I
I I I I
I I
- I
- I .
I
- I '- '
- I
I l
t I
Table 4.6 EFFECT OF TD4PERATURE AND VOID ON CALCULATED REACTIVITY OF STORAGE RACK Case Incremental Reactivity Change, Sk Region 1 Region 2 20*C (68'F) Reference Reference 40'C (104*F) -0.003 -0.002 66*C (150'F) -0.009 -0.005 90*C (194'F) -0.013 -0.010 122'C (252'F) -0.024 -0.015 122'C (252*F) + 20% void -0.071 -0.061 I
0 4-27 ,
W I
I I -mm <mx r- m a
um
,i1 .
n ,i n-- - .
- i I i i n i.
.E E
- f. = .ks , ,
! i E m E m E 1 IN B E mu- E aI
+i i Em.
E -___mm
__! ! I ii N
E e E m e aa aea ema E m am am i mm sgg
.i i
i ,
. i i i i.
E m a t. m E - i 3
e
, t ; 4 j i_ ,g g . <fg E m E m is E..4,.. .:.4.m. 1
~ g' n m a i.
m E m m
1 : ,
g I
i E
i e a m i ie
-m a ie a
i I '
, a ^
d v
.) m' a a E -
i E -r ii a E ,
m -
d m 1 .. - , . 1 i i - O E mam1 m_amEXu.m~M~ MEN ~M m E_D XM N 8
= = W E-a =, =, i m ,tm, i n_m.,a rw m m-- m. M
^
E - e E a i m 1
4 ,,, i i
n e
, r ,
a E
.i
- U i e W
d $
., ,i . . 4 , -
E = a a E - E -
I e i . <.
a m E
_t g o
'l' m a_ E E _m y E m E a ii
- b. ,_ ; i . a m, 1 , ,
4 _.
E Il (gjg.. m m I- m E _i 1 l1 - U I
i.
E m _ m' mE - -
. .i E m n _- -
t m_ m ai E m m'u m a t i_
a e a mi =
_ _ i . ! L_a !
=_m g- g Emaa E am E E mI mam E ma m d
U i . i i i ,, . a a, i g_L
, K . _ m m !a1 m.
1 .
C w
- m. m E m E u m r si m < ,
I i i ii t E E m E i , , 1 o i
i i , i, i m
{e'
- 4 i
t7' -
{ -
4 , i w h
. ,i i ,i i --i i r- i i O E-E i
m E 1 m E N m i m 4 A ii i 1ee i i l F. Z Q I
i 6 E
W E I m m in E E a O --
. i E mm E i i } i i i .
i 1-, i , ii +
E E a (([Wm E m E - m E E = m m -ma @
E m mm m mm a ma maa E a m a=m E m e m E mm -J I
$ i i i ; } } i 6 6a ( .l e i{ } i. ,( 4
- : I' 4: 1 m E 1 m E m E 56 m 2
. un +c i.j . . ,4 i # . i i %
U '5
- m E u E a Lii m
. .p _ _
_ .' . i
. .i
, e i i f i ! 7 i g
. 31 41 ' i ' -.- 11 ' i u
I E -am
~
4'
[_
. mu
. 6
._."i '.
e m___E m .mam i i.
l 'i EbNYY
't' mu-
)
E -_ E
~
m__E E__E l_
I i
b a
E -m - CM eea ----m--- ---- --
m2-E E m m m.
. i > i . . - ..i.m u. ,,
E, i. . t t ,,
a
". E : 7 :s m a E
-:T] i i - + . .; ei i E1:s E a 4
- E O i E E m I E E E
. 4 . ..i. . i E
I
- E E F I E
~
E a E e ? 6 e ie e i i
E .
m a s
- m E *' E E
! ! i E
t i E :I ' E a E E E E
- I E
1
___'7 i ) i : .
3__ i !
i ae i.
tma a m_m E E aa aaa E E ama E R_m a m_m E m 4-28 I
w w - - w i e t___ -
m.m_im.m_ u m n m in mC E m ..m m ..m m ..a m R m.. m.. m m sh TmTiu.m. m. ..a m. m :n m im m a . .
+ m ml:
-'m d,-
fg m -x n a
s .i m m .
E l
-'m u a m :p s u.. a m.. :n m . ;. .._.. m m ... .. _,. . , .
} . .
^~"~ .d
-'~~'^
~
mI..a : ~~:1:~Tu mI_i
~~
n u ,__m u ;_-
m~_ m m :~I:: .:IIlm -
q
-'~~~
m m :,s m : .s u. .- -a
- i. :n m f a m_ .
,7 m ,.
l l
. f n m~ m n, a..
Zim n' a .
W:n ult._.. .; ,
m:h. m m .
sis mislu:
. . t.
m:m;m min:ulu m:m m:m m:s:E: sis:s:
m m:nin m m u u r m~~m m u _mim m m a.m_m:mim.,m m m mTu~sTm:
m~u__m.m.m,u H m.iu.m..m m..m.m.m.m.u a n -.
m;;mim.m.m.m m,,m , t 1 1 .. ..
g . . -->. >. --<.-
ol g
g g - _ . ..- . .... .-
g ..
g i - -
t~
m-~
l t l
,n- -- i l
~~~-
- n m m n
- n l
m [:: a u a _,_ _q ja
~~-
- n : s m:
E s' m.: M:s m :"-- :n m: u I m. .
m l:
its m' . ..
- m st "
-*t m mi
- '~
.... :n m :^ m. m: -M in:m mim:m:W m:m:m min' mistuku:u:sts' m:m:m::m m:m min:m:is:m:m' min:m m:m:m nm m sTETETK sTE K u lu im ..m. m.
im u. m im T u m3 m m m. m m m m~u ni m x,. mi n m i m m m ..m m .m .
- m: :n m
- -m m.- :n m: .
- m 3- .
m
~~~
- n '
~ -
w ~
- I l
i .
~ ' ^~--
u m m u.
a m. m:
e a m _..
- .: : d
!~ >
ml_ .
\ -.
\
m u m
l:::'.1
\
- m m;
- Ea a -'u m m
.t i.
si :m m !: .
_m mL -m ,' n ', m':
n .
m ,
~sinia:n w:n
. , 1 m:m;miminim:n mix m m m:m~m m:mim m:m:m m m:m m::m m:n m;m:n- m:m:m t
i m mpm m m m mTu,u,Tu:m_m_m a mim _m nim n im, mini m:in Ju:m, -ulm__.m.u .
s s': a
- n ) m;: -m-e- a m:
m: u i
J In m. m - a mlc - -
m -
a
-al:
e l
- n _".~"~
m t
i a t
d i:. - _ .
m:
'r
- ^~ 4
-' ~~r~" l - ^ ^
- s ?m::
^ <
- -~ -~~ - --
ml: :n t
"t . :n m .
m.m :m:m1m u
a:m:mMm L l
a :s > 1 '
E i m'. .
- m. _m
~.mu _. _ _m n '
- n_l m:m:n Isim:n I >
m::mim:n m:m:a
> 1 L= - at ::m l
i Fig. 4-2 INTERIM STORAGE PATTERN (CHECKERBOARD) i4is accies 2 crtts B is7s accoa s ettts Q iss turiv tocarons E sai accion i cttis
o F
l 5
55000 50000 =
~
4
- g I 45000
- +G 40000 AC :EPTABLE $y 3 : BURNUP DQMAIN y jy l :
.S +
2 3c000 ,
q7 i i / 5 g g -
V d 2 coco
/
I 2 :
/
15000 /
I l
,ggg,
/
/
/
UNACCEPTABLI:
Bl)RNUP DOMA!N
, (Requ res Storcs;e in Re; ',on 1) 50
{i,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
INITIAL ENRICHMENT, ssU-235 Fig. 4-3 ACCEPTABLE BURNUP DOMAIN IN REGIONS 2 & 3 4-30 4
_ _ __ _ _ _ . _ _ _ _ _ _ _ . _ - - - - - - - - - - - - - - - - - ~ - - - - - - - - - - - - - - - - -
)
1 l
7.50
- 0.06" BORAL IN 0.109" GAP 0.101 i 0,005" THK 0.0345 gmB-10/cm2
.:a : x :xvc yr:c av: :x. _ _. :
f o.3659 IN.
- , FUEL OD
)
Q, flNSITY 1o.29 g/cc j OOk 204 cis x is) : 0.035" ss O004 Io 'on 0:'E'f i T SHEATH OOOOA mcw o.scs is. s OOOOOh, mustro 00 o.s s iN. !
OOOOOOE io o.si2 is. ;
00000004 No. 2i 3
OOOOOOOOb !
0090000001, ~; J 21s" 0000000000A !
OOOOOOO9000G !
000000000000b l
.OOOC ^006. !
~OOO C - 3.,3 , o.o,.. ,0x , o-- 1000E !
's
^ ' N.075" SS BOX NOT TO SCAM PITCH
- 8.969" NOM!NAL FIG. 4-4 FUEL STORAGE CELL CROSS SECTION 4-31 i
U L__.J L_ J 2* WATER GAP WITH REFLECTING BOUNDARY CONDITIONS l
OR RETLECTOR (WATER OR CONCRETE) WITH ZERO FLUX BOUNDARY CONDITION I
PERIODIC BOUWDARY CONDITION l l
- - I m-- ;mo im
- m:rg:-:m::m:: m m -
m m m m __m. m m m-- __
p m m E.- -.,___ -E m __
u ,
m r - ..
g
.u:: ::p)_--_. ~m-- J 0- m '
us = :: __u- m _
st FRESHFUELjd-
--(.REQ 0N 2)--
-E: -
f-E:_( REQ 0N .3E:)
E3 REQ 3N 3
)--
EE -E: (
REGON 3 E:-
- f. u:___E:: : .1 .
- .x y a .
u4j_ _.s____ = =
=
m ____.m __ _
- m. --
,~,
'm g ___.
m m m m m m m u
~-
m m m m ::g::.u __:u :m:: JE%B;? : m : :m :
u o .
9 w g n - - '
-_ - Immuuuuuuuuuuum mumma g
2 , -
~
'm m m m m m ::;g:: u
- .g::m:: m m --'u--u E m m m m m m m m m ____
g ____,-..u_,T=__ ,T=::
m , , ,
- u - _-
f--
REQ 0N 2 --E REQCN 2 g[ ., _
"y_.i
- -- REQ 0N 3 - .
__]
E] REQON 3 E -
E: REQ 0N 3 2: - .i:
5: E '
-~u
~
m
- n m;; -~u [u::
m m m
[ m m m m m m m m m-~ m m m m~- --u-~u __ m::u-: m m m u__
PERIODIC BOUWDARY CONDtil0N FIG. 4-5 KEN 0 CALCULATIONAL MODEL
)'
l e
L
- 5.00 4
4.75 .
I 4.50 :
\
4.25
- \
I :
\
4.00 :
I $ 3.75 .
\\ v m :
3.50 -
- 3.25 -
$ i k y 3.00 -:
=
y 2.75 ;
=
\g U \
2.50 .
r : \ '
5 2.25 :
J : lh f8 2.00 1.75 } \ N 1.50 :. -
3 N 1.25 -:
N 1.00 ~_
0.75 ~ .. . ... . .... ........iii........,.o. iii . ... .
0 5 10 15 20 25 30 35 40 45 50 55 60 BURNUP, MWD /KgU Fig. 4-6 EQUlVALENT ENRICHMENT FOR SPENT FUEL AT VARIOUS BURNUPS FOR INITIAL ENRICHMENT OF 4.955 4-33 1
I L
0.940 -
.I :
I 0.933
(<
I 0.930 I .
h 0.925 2--
I <
W e
9 l 2 0.920 a :
- X f
l
(
~
g 0.915 ,
l
_2
- 0 910
I :
0.905 ,
0.900 ' . , , , ,,,, , ,,, ,,,, ,,,,
0.5 1,0 1.5 2.0 2.5 3.0 WATER GAP BETWEEN MODULES, inches Fig. 4-7 EFFECT OF WATER-GAP SPACING BETWEEN MODULES ON SYSTEM REACTMTY 4-34
u 55000 ,
Soooo .
- s I 45000 i ., #' '
?
I 3 4 coco }
- 8 URN JP DC'AAIN ACCEPTA9'_E I
[
D
/
y l
d 30000 l .
f} e 7 I
'k' !
E 25000 l
l cj 2 coco b ,
o /. .
/
g/ /-;
2 :
l 15000 14
,g
- / / UNACCEPTABL!:
I ,
/ / BURNUP DOMAN (Rey ;res Steroge in Re; ion 1) 5000 9$ ' ' 's.s ' ' ' 2.o' ' ' '2.5' ' ' 's.o' ' ' 's.5' ' ' ' 4.o' ' ' '4.s' ' ' '5.o INITIAL ENRICHMENT, zU-235 Fig. 4-8 ACCEPTABLE BURNUP DOMAIN IN REGIONS 2 & 3 SHOWING EXISTING SPENT FUEL ASSEMBLIES 4-35 ;
s
(
I APPENDIX A BDICHMARK CALCULATIC11S I
I by I
Stanley E. Turner, PhD, PE HOLTEC INTERNATIONAL l
January 1991 I
I l
I 4 1
H 1.0 INTPCDUCTION AND SUPRARY The objective of thic benchmarking study is to verify both O NITAWL-KENO-sam methodology with the 27-group SCALE c:d,t- 4t o...i p library and the CASMO-3 coddD for use in criticality st 'J tv % % '> ca.iors of high density spent fuel storage racks. B o '.h ca*qqut b.a. .nethods are based upon transport theory and have bean bencWruG aga !7st crt.t nal experiments that simula'te typict.1 spent fuel storage rack designs as realistically as possiblG.
Results of these benchmark caltt.latlecs with both methodologies ph consistent with corresponding calculations r y r.r t e in thg literature.
Results of the bench 7 ark calculations show that the 27-group (SCALE) NITAWL-KENO-Sa calculations consistently under-predict the critical eigenvalue by 0.0090 t 0.0021 6k (with a 95%
probability at a 95% confidence level) for critical experiments (9 that are as representative as possible of realistic spent fuel storage rack configurations and poison worths.
Extensive benchmarking calculations of critical experi-ments with CASMO-3 have also been reportedW , giving a mean kg of 1.0004 t 0.0011 for 37 cases. With a K-factor of 2.14(9 for 95%
probability at a 95% ccnfidence level, and conservatively neglect-ing the small overprediction, the CASMO-3 bias then becomes 0.0000
! i O.0024. CASMO-3 and NITAWL-KENO-Sa interco=parison calculations of infinite arrays of poisoned cell configurations (representative of typical spent fuel storage rack designs) show very good agreement, confirming that 0.0000 t 0.0024 is a reasonable bias and uncertainty for CASMO-3 calculations. Reference 5 also documents g good agreement of heavy nuclide concentrations f or the Yankee core I isotopics, agreeing with the measured values within experimental error.
I A-1 1
E
~ The benchmark calculations reported here confirm that l
- oither the 27-group (SCALE) NITAWL-KENO or CASMO-3 calculations are acceptable for criticality analysis of high-density spent fuel storage racks. Reference calculations for the rack designs should be performed with both code packages to provide independent Verification.
I 2.0 NITAWL-FENO Sa BENCHMARK CALCULATIONS ,
Analysis of a series of Babcock & Wilcox critical experiments W , including some with absorber panels typical of a poisoned spent fuel rack, is summarized in Table 1, as calculated with NITAWL-KENO-Sa using the 27-group SCALE cross-section library and the Nordheim resonance integral treatment in NITAWL. Dancoff factors for input to NITAWL were calculated with the Oak Ridge SUPERDAN routine (from the SCALEA system of codes). The mean for those calculations is 0.9910 ! O.0033 (1 a standard deviation of the population). With a one-sided tolerance factor correspending I to 95% probability at a 95% confidence level @, the calculational g bias is + 0.0090 with an uncertainty of ! O.0021 for the sixteen B critical experiments analyzed.
Similar calculational deviations hav? been reported by ORNLO for some 54 critical experiments (mostly clean critical without strong absorbers), obtaining a mean bias of 0.0100 t 0.0013 (95%/95%). These published results are in good agreement with the results obtained in the present analysis and lend further credence to the validity of the 27-group NITAWL-KENO-5a calculational model for use in criticality analysis of high density spent fuel storage racks. No trends in k,g with intra-assembly water gap, with absorber panel reactivity worth, with enrichment or with poison concentration were identified.
A-2
Additional benchmarking calculations were also rade for L a series of French critical experiments") at 4. 7 5 % enrichr.ent and e for several of the BNWL criticals with 4.26% enric.t ed fuel.
Analysis of the French criticals (Table 2) showed a tendency to overpredict the reactivity, a result also obtained by ORNL(IO) The .
calculated k,g values showed a trend toward higher values with decreasing core size. In the absence of a significant enrichment effect (see Section 3 below), this trend and the overprediction is
{
attributed to a small inadequacy in NITAWL-KDIO-Sa in calculating neutron leakage from very small assemblies.
l Similar overprediction was also observed for the BmiL series of critical experiments (U, which also are small assemblies (although significantly larger than the French criticals) . In this case (Table 2), the overprediction appears to be small, giving a mean k eg of 0.9990 t 0.0037 (1 a pcpulation standard deviation).
Because of the small size of the BNWL critical experiments and the absence of any significant enrichment effect, the overprediction is also attributed to the failure of NITAWL-KETO-Sa to adequately treat neutron leakage in very small assemblies.
Since the analysis of high-density spent fuel storage racks generally does not entail neutron leakage, the observed inadequacy of NITAWL-KDIO-Sa is not significant. Furthermore, omitting results of the French and BNWL critical experiment analyses from the determination of bias is conservative since any leakage that might enter into the analysis would tend to result in overprediction of the reactivity.
I I,
A-3
h r 3. CASMO-3 BE!ICm9.RM CALCUUsTIC?iS l
The CASMO-3 code is a multigroup transport theory code utilizing transmission probabilities to accomplish two-dimensional calculations of reactivity and depletion fcr BWR and PWR fuel assemblies. As such, CASMO-3 is well-suited to the criticality analysis of spent fuel storage racks, since general practice is to treat the racks as an infinite medium of storage cells, neglecting leakage effects.
g CASMO-3 is a modification of the CASMO-2E code and has been P extensively benchmarked against both mixed oxide and hot and cold critical experiments by Studsvik EnergiteknikII). Reported ana-lyses (5) of 37 critical experiments indicate a mean kg of 1.0004 0.0011 (1c). To independently confirm the validity of CASMO-3 (and to investigate any effect of enrichment), a series of calculations were made with CASMO-3 and with NITAWL-KENO-Sa en identical poisoned storage cells representative of high-density spent fuel storage racks. Results of these intercomparison calculations * (shown in Table 3) are within the normal statistical variation of KENO calculations and confirm the bias of 0.0000 :
0.0024 (95%/95%) for CASMO-3.
Since two independent methods of analysis would not be expected to have the same error function with enrichment, results of the intercomparison analyses (Table 3) indicate that there is no significant effect of fuel enrichment over the range of enrien-ments involved in pcwer reactor fuel. Furthermore, neglecting the French and BNWL critical benchmarking in the determination of bias is a conservative approach.
'Intercomparison between analytical methods is a technique endorsed by Reg. Guide 5.14, " Validation of Calculational Methods for Nuclear Criticality Safety".
. A-4
l I
I REFERENCES TO APPENDIX A I
- 1. Green, Lucious, Petrie, Ford, White, and Wright, "PSR /NITAWL-1 (code package) NITAWL Modular Code System For Generating Coupled Multigroup Neutron-Gamma Libraries from ENDF/B", ORNL-TM-3706, Oak Ridge National Laboratory, November 1975.
- 2. R.M. Westfall et. al., " SCALE: A Modular System for Perform-ing Standardi::ed Computer Analysis for Licensing Evaluation",
I NUREG/CR-02OO, 1979.
- 3. A. Ahlin, M. Edenius, and H. Haggblom, "CASMO -
A Fuel Assembly Burnup Program", AE-RF-76-4158, Studsvik report.
A. Ahlin and M. Edenius, "CASMO - A Fast Transport Theory j Depletion Code for LWR Analysis", ANS Transactions, Vol. 26, IW p. 604, 1977.
I "CASMO-3 A Fuel Assembly Burnup Program, Users Manual",
Studsvik/NFA-87/7, Studsvik Energitechnik AB, November 1986
- 4. M.N. Baldwin et al., " Critical Experiments Supporting Close 5 Proximity Water Storage of Power Reactor Fuel", BAW-1484-7, The Babcock & Wilcox Co., July 1979.
l 5. M. Edenius and A. Ahlin, "CASMO-3 : New Features, Benchmarking, and Advanced Applications", Nuclear Science and Encineerina, 100, 342-351, (1988)
- 6. M.G. Natrella, Experimental Statistics, National Bureau of Standards, Handbook 91, August 1963.
- 7. R.W. Westfall and J. H. Knight, " SCALE System Cross-section Validation with Shipping-cask Critical Experiments", AEE Transactions., Vol. 33, p. 368, November 1979
- 8. S.E. Turner and M.K. Gurley, " Evaluation of NITAWL-KENO Benchmark Calculations for High Density Spent Fuel Storage Racks", Nuclear Science and Encineerine, 80(2):230-237, February 1982.
A-5
I I
- 9. J.C. Manaranche, et. al. , " Dissolution and Storage Experiment I with 4.75% U-235 Enriched UO Rods", Nuclear Techno1cov, Vol.
50, pp 148, September 1980
- 10. S.R. Bierman, et. al., "Critigl Separation between Sub-critical Clusters of 4.29 Wt. % U Enriched UO4 Rods in Water
- with Fixed Neutron Poisons", Batelle Pacific Northwest Labora-tories, NUREG/CR/C073, May 1978 (with August 1979 errata).
- 11. A.M. Hathout, et. al., " Validation of Three Cross-section j Libraries Used with the SCALE System for Criticality Analy-l5 sis", Oak Ridge National Laboratory, NUREG/CR-1917, 1981.
i 4
- I II lI il
!I
!I
- I il d
lI
s r
Table 1 l
RESULTS OF 27-GROUP (SCALE) liITAWL-KElio-Sa CALCULATIOliS OF B&W CRITILAL EXPERIME!ITS Experiment Calculated a 11 umber k,g I O.9932 2 0.0016 II O.9915 t 0.0015 III O.9916 ! O.0013 IX O.9918 : 0.0014 X O.9923 0.0015 XI O.9919 t O.0014 XII O.9961 t 0.0015 XIII O.9960 : 0.0015 XIV O.9817 0.0015 XV O.9843 t 0.0014 XVI O.9912 t 0.0015 XVII O.9866 : 0.0013 XVIII O.9904 t 0.0014 XIX O.9861 t 0.0013 XX O.9934 0.0013 XXI O.9874 0.0014 Mean 0.9910 t 0.0014(I)
Bias 0.0090 0. CO'a3(2)
Bias (95%/95%) 0.0090 t 0.0021 (l) Calculated from individual standard deviations.
@ Calculated from k,.g values and used as reference.
A-7
I s
Table 2
[
RESULTS OF 27-GROUP (SCALE) NITAWL-KENO-Sa CALCULATIONS OF FRENCH and BNWL CRITICAL EXPERIMEN'"S
[
French Experiments
( Separation critical calculated Distance, cm Height, cm k,g O 23.8 1.0231 t 0.0036 2.5 24.48 1.0252 : 0.0043 S.0 31.47 1.0073 i O.0013 10.0 64.34 0.9944 t 0.0014 BNWL Experiments Calculated case Expt. No. k,e No Absorber 004/032 O.9964 + 0.CO34 SS Plates (1.05 B) 009 0.9988 d 0.0038
~
SS Plates (1.62 B) 011 1.0032 i O.0033 SS Plates (1.62 B) 012 0.9986 0.0036 SS Plates 013 0.9980 t 0.0038 SS Plates 014 0. 993 6 t 0.003 6 Zr Plates 030 1.0044 0.0035 Mean 0.9990 i O.0037 A-8 I
W 9
a
- i Table 3 RESULTS OF CASMO-3 AND NITAWL XENO-Sa BENCHMARK (IN*ERCOMPARISON) CALCULATIONS Enrichment (1) k Wt. % U-235 NITAWL-KENO-Sa@)jl)CASMO-3 l6kl l 2.5 0.8385 : 0.0016 0.8379 0.O'
, 3.0 O.8808 2 0.0016 O.8776 0.0032 3.5 0.9074 : 0.0016 0.9090 0.0016 4.0 0.0311 t 0.0016 0.9346 0.0035 4.5 0.9546 : 0.0018 0.9559 0.0013 I 5.0 0.9743 i O.0018 0.9741 0.0002 B
Mee.n 0.0017 i
0)
Infinite array of assemblies typical of high-density spent fuel storage racks.
@)
k from NITAWL-KENO-Sa corrected for bias of 0.0090 6k.
J
)
A-9 I _ _.- -
[ 5.0 THERMAL-HYDRAULIC CONSIDERATIONS 5.1 Introduction l
l A primary objective in the design of the high density spent fuel storage racks for the Donald C. Cook spent fuel pool is to ensure adequate cooling of the fuel assembly cladding. In the following section a brief synopsis of the design basis, the method of analysis, and the numerical results is provided.
Similar methods of thermal-hydraulic analysis have been used in previous licensing efforts on high density spent fuel racks for Fermi 2 (Docket 50-341), Quad Cities 1 and 2 (Dockets 50-254 and 50-265), Rancho Seco (Docket 50-312), Grand Gulf Unit 1 (Docket 50-416), Oyster Creek (Docket 50-219), Virgil C. Summer (Docket 50-395), Diablo Canyon 1 and 2 (Docket Nos. 50-275 and 50-323),
Byron Units 1 and 2 (Docket 50-454, 455), St. Lucie Unit One (Docket 50-335), Millstone Point I (50-245), Vogtle Unit 2 (50-425), Kuosheng Units 1 & 2 (Taiwan Power Company), Ulchin Unit 2 (Korea Electric Pcwer Ct .apany) , J.A. FitzPatrick (New York Power Authority) and TMI Unit 1 (GFJ Nuclear).
The analyses to be carried out for the thermal-hydraulic qualification of the rack array may be broken down into the following categories:
(i) Pool decay heat evaluation and pool bulk temperature variation with time.
(ii) Determination of the maximum pool local temperature at the instant when the bulk temperature reaches its maximum value.
5-1
I (iii) Evaluation of the maximum fuel cladding temperature to establish that bulk nucleate boiling at any location resulting in two phase conditions environment around the fuel does not occur.
(iv) Evaluation of the time-to-boil if all heat rejection paths through the cooling and cleanup are lost.
(v) Compute the effect of a blocked fuel cell opening on the local water and maximum cladding temperature.
The following sections present a synopsis et une n.a t hods employed to perform such analyses and a final summars 'I the r sults.
I 5.2 Spent Fuel Coolinn System Descri;;1isn I The principal functions of t: e Spent Fued C>oling System are the removal of decay heat f r- uhe spent Ic 1 st; red in the pool it serves and maintaining the c]1rity of, c 1d a low activity level g in, the water of the pool. Cleanup of pool .ter is accomplished 3 by diverting part of the flow, maintained for removal of decay heat, through filters and/or demineralizers as described in Section 9.4 of UFSAR.
l 5.2.1 System Functions The Spent Fuel Pool Cooling System is designed to remove from the l spent fuel pcol the heat generated by stored spent fuel elements.
I The system serves the spent fuel pool which is shared between the two units.
1 I
l 5-2 t
t I
I The system design incorporates two separate cooling trains. '
system piping is arranged so that failure of any pipeline does not drain the spent fuel pool below the top of the stored fuel elements.
I 5.2.2 System Description I Each of the two cooling loops in the Spent Fuel Pool Cooling System consists of a pump, heat exchanger, strainer, piping, associated valves and instrumentation. The pump draws water from i the pool, circulates it through the heat exchanger and returas it to the pool. Component cooling water cools the W t exchanger.
The clarity and purity of the spent fuel pool water is maintained by passing approximately 100 gpm of the cooling flow through a filter and demineralizer. Skimmers e ze provided to prevent dust and debris from accumulating on the surface of the water.
The refueling water purification pump and filter can be used separately oc in conjunction with the spent fuel pool demineralizer to regain refueling water clarity after a crud burst in either unit. This can prevent loss of time during refueling due to poor visibility. The system is also used to maintain water l quality in the Refueling Water Storage Tanks of both units.
The spent fuel pool filter /demineralizer is downstream of the spent fuel pool cooler. As a result, the pool purification components are subjected to water temperatures which correspond to the cooler outlets (less than 140aF). All elements of the purification system, including the resins, are qualified for 200'F design temperature.
0-3
L The spent fuel pool pump suction lines penetrate the spent fuel r
L pool wall above the fuel assemblies stored in the pool to prevent loss of water as a result of a suction line rupture. The pool is initially filled with water at the same boren concentration (2400 ppm) as in the refueling water storage tank.
The spent fuel pool is located outside the reactor containment.
I During refueling the water in the pool can be isolated from that in the re-fueling canal by a weir gate so that there is only a very small amount of interchange # water as fuel assemblies are transferred.
I 5.2.3 Performance Recuirements l The first design basis of the system is based on the normal refueling operation with a normal batch of 30 assemblies being I removed from the unit each time.
The second design basis for the system considers that it is possible to unload the reactor vessel (193 fuel assemblies) for maintenance or inspection et a time when a mz.ximum of 3518 spent fuel assemblies are assumed already stored in the spent fuel pool.
5.3 Decav Heat Load Calculations The decay heat load calculation is performed in accordance with I the provisions of USNRC Branch Technical Position ASB9-2, i
" Residual Decay Energy for Light Water Reactors for Long Term Cooling", Rev. 2, July, 1981. For purposes of this licensing l
5-4 1 - - - _ . - - - - _ - - - - - - - - - _ - - - - - - - - - - - _ - _ - - - - - - - - - - - - - - - . - - - _ - - - - - - - _ - - _ - - - - - - _ - - - _ _ -
application, it is assumed that the pool contains an inventory accumulated through scheduled discharges from 1975 to 2009 (Table 1.1.1). Further, since the decay heat load increases monotonically with reactor exposure time, an upper bound of 1260 full power operation days (approximately 3.5 years) is assumed for all stored fuel. The cumulative decay heat load is computed for the instance I of hypothetical normal discharge (21B in Table 1.1.1) in the year 2009. As shown in Table 5.4.1, the ratio of this decay heat load due to the inventory of previously stored fuel to the average I assembly operating power S is 0.3303.
This decay heat load is assumed to remain invariant for the duration of the pool temperature evaluations performed in the wake of normal and full core offloads discussed below.
5.4 Discharce Scenarios The following discharge scenarios are examined:
Case 1: Normal discharge A normal batch of 80 assemblies with 1260 days of reactor exposure time at full power is discharged in the pool at the end of a normal 18 month operating cycle. There are 43 previously discharged batches in the pool. As described I later, the normal discharge is assumed to occur at the rate of approximately 4 assemblies per hour after 168 hours0.00194 days <br />0.0467 hours <br />2.777778e-4 weeks <br />6.3924e-5 months <br /> of decay in the reactor. One fuel pool cooling train is active and running. One cooling train contains one heat exchanger and one fuel pool pump.
This case is run with the design fuel pool water flow rate I (2300 gpm) and actual available flow rate (2800 gpm).
two cases are labelled as Case la and Case Ib, respectively.
These Case 2: Normal discharge Same as Case 1 except two fuel pool cooling trains are operating.
I I
I 55
k W
, Case 3: Back-to-Back Full core offload The full core offload condition corresponds to the emergency reactor offload condition wherein the shutdown of a reactor occurs 30 days after the other reactor shutdown for a normal
^
re'neling. Two cooling trains are assumed to be operating in pc 211el af ter the shutdown. The decay time of the core in the reactor and the rate of discharge to the pool are the saae as in Case 1.
Case 4:
Same as Case 3 except only one cooling train is in operation.
I This case is listed for reference only; it is not a design basis case by the Donald C. Cook Technical Specification or the USNRC guidelines (NUREG-0800).
Detailed data on the three cases are given in Table 5.4.1 to 5.4.3.
I 5.5 Bulk Pool Temnerature i In order to perform the analysis conservatively, the heat exchangers are assumed to be fouled to their design maximum. Thus, the temperature effectiveness, p, for the heat exchanger utili:ed in the analysis is the lowest postulated value calculated from I heat exchanger thermal hydraulic codes. p is assumed constant in the calculation.
The mathematical formulation can be explained with reference to the simplified heat exchanger alignment of Figure 5.5.1.
Referring to the spent fuel pool / cooler system, the governing differential equation can be written by utilizing conservation of energy:
dT C = QL - QHX (5-1) dr QL = Pcons + Q (?) - QEV (T, ta) 5-6
I I where:
I C: Thermal capacitance of the pool water volume times water density times heat capacity), Btu /*F.
(net and Q3: Heat load to the heat exchanger, Btu /hr.
Q(t): Heat generation rate from recently I discharged fuel, which is a function of time, r, Btu /hr.
specified I Pcons = Po: Heat generation rate from "old" fuel, Btu /hr. (Po = average assembly operating power, Btu /hr.)
gHX: Heat removal rate by the heat exchanger, Btu /hr.
QEV (T,ta): Heat loss to the surroundings, which is a function of pool temperature T and ambient temperature ta, Btu /hr.
QHX is a non-linear function of time if we assume the effectiveness p constant during I temperature is the calculation. Qux can, however, be written in terms of effectiveness p as follows:
QHX " Wtce p (T - ti) (5-2) to - ti P"
T - ti I
where: s W:
t Coolant flow rate, lb./hr.
Ct: Coolant specific heat, Btu /lb.*F.
[ p: Temperature effectiveness of heat exchanger.
I I 5-7 I
I l
I g
T: Pool water temperature, F ti: Coolant inlet temperature, "F to: Coolant outlet temperature, F I p is obtained by rating the heat exchanger on a Holtec proprietary g thermal / hydraulic computer code. Q(t) is specified according to E the provisions of "USNRC Branch Technical Position ASB9-2,
" Residual Decay Energy for Light Water Reactors for Long Term i Cooling", Rev._2, July, 1981. Q(t) is a function of decay time, number of assemblies, and in-core expo.m re time. During the fuel transfer, the heat load in the pool will increase with respect to the rate of fuel transfer and equals to Q(t) after the fuel transfer.
QEV is a non-linor function of pool temperature and ambient I temperature. QEV contains the heat evaporation loss through the pool surface, natural convection from the pool surface and heat conduction through the pool walls and slab. Experiments show that the heat conduction takes only about 4% of the total heat loss (5.5.1], therefore, can be neglected. The evaporation heat and nature convection heat loss can be expressed as:
QEV = m P As + hC As 6 (5-3) where:
m: Mass evaporation rate, lb./hr. ft.2 P: Latent heat of pool water, Btu /lb.
As: Pool surface area, ft.2 he: Convection heat transfer coefficient at pool surface, Btu /ft.2 hr. F I
I I
5-8
i L
r 6 = T-ta: The temperature difference between pocl water and ambient air, 'F The mass evaporation rate m can be obtained as a non-linear function of 6. We, therefore, have m = hp (6) (Wps - Was) (5-4) where:
4 l t'ps : Humidity ratio of saturated moist air at pool water surface temperature T.
Was: Humidity ratio of saturated moist air at ambient temperature t a ho(6): Diffusion coefficient at pool water surface. ho is a non-linear function of 6, lb./hr. ft.2 27 The non-linear single order differential equation (5-1) is solved l using Holtec's Q.A. validated numerical integration code 1
"OllEPOOL".
Figures 5.5.2 through 5.5.6 provide the bulk pool temperature profiles for the normal discharge, and full core offload scenarios respectively. Table 5.5.1 gives the peak water temperature, la coincident time, and coincident heat load to the cooler and coincident heat loss to the ambient for three coses.
The next step in the analysis is to determine the temperature rise profile of the pool water if all forced indirect cooling modes are suddenly lost. Make-up water is provided with a fire hose.
Clearly, the most critical instant of loss-of-cooling is when pool water has reached its maximum value. It is assumed that cooling water is added through a fire hose at the rate of G lb./hr. The 5-9
I 4 lA k,
I cooling water is at temperature, tecol. The gevarnir.g enthalpy balance equation for this condition can be written >.
dT I [C , G(Ct)(# - Io))
dr
= Pcons + Q(I + rins) +G (Ct) (teool - T)
EV where water is ascumed to have specific heat of unity, and the time coordinate t is measured from the instant maximum pool water I temperature is reached. to is the time coordinate when the direct addition (fire hose) cooling water application is begun. rins is the time coordinate measured from the instant of reactor shutdown to when maximum pool water ter.aerature is reached. T is the dependent variable (pool water temperature). For conservatism, Ogy is assumed to remain constant after pool water temperature reaches and rises above 170*F.
I A Q.A. validated . numerical quadrature code is used to integrate the foregoing aquation. The pool water heat up rate, time-to-boil, and subsequent water evaporation-time profile are generated I and compiled for safety evaluation.
Assuming no make-up water (G = 0), the time-to-boil output results are presented in Table 5.5.2. Figures 5.5.6 through 5.5.10 show the plot of the inventory of water in the pool after loss-of-coolant-to-the-pool condition begias.
It is seen from Table 5.5.2 that sufficient time to introduce manual cooling measures exists and the available time is consistent with other PWR reactor installations.
I I
I I 5-10
I 5.6 Local Pool Water Temperature In this section, a summary of the methodology, calculations and results for local pool water temperature is presented.
5.6.1 Basis In order to determine an upper bound on the maximum fuel cladding i temperature, a series of conservative assumptions are made.
most important assumptions are listed below:
The I O The fuel pool will contain spent fuel with varying time-arter-shutdown (ts). Since the heat emission falls off I rapidly with increasing ts, it is conservative to assume that all fuel assemblies are from the latest batch discharged simultaneously in the shortest possible time and they all have had the maximum postulated years of operating time in the reactor. The heat emission rate of each fuel assembly is assumed to be equal and maximum.
O As shown in the pool layout drawings, the modules occupy an irregular floor space in the pool. For the hydrothermal analysis, a circle circumscribing the I actual rack floor space is drawn (Fig. 5.6.1).
further assumed that the cylinder with this circle as It is its base is packed with fuel assemblies at the nominal layout pitch.
O The actual downcomer space around the rack module group The nominal downcomer gap available in the pool I varies.
is assumed to be the total gap available around the idealized cylindrical rack; thus, the maximum resistance to downward flow is incorporated into the analysis (Figs. 5.6.2 and 5.6.3) (i.e. minimum gap between the pool well and rack module, including seismic kinematic effect).
O No downcomer flow is assumed to exist between the rack modules.
I I 5-11 1
1
i I
I O The ANF 17x17 fuel assembly has been used in the I analysis which, from the thermal-hydraulic standpoint, bounds all types of fuel bundles utilized in the Donald C. Cook reactor.
0 No heat transfer is assumed to occur between pool water i and the surroundings (wall, etc.)
5.6.2 Model Descrintion l
In this manner, a conservative idealized model for the rack assemblage is obtained. The water flow is axisymmetric about the vertical axis of the circular rack assemblage, and thus, the flow is two-dimensional (axisymmetric three-dimensional). Fig. 5.6.2 shows a typical " flew chimney" rendering of the thermal hydraulics model. The govez.ning equation to characterize the flow field in the pool can now be written. The resulting integral equction can I,
be solved for the lower plenum velocity field (in the radial
! direction) and axial velocity (in-cell velocity field), by using l the method of collocation. The hydrodynamic loss coefficients which enter into the formulation of the integral equation are also taken from well-recognized sources (Ref. 5.6.1) and wherever discrepancies in reported values exist, the conservative values are consistently used. Reference 5.6.2 gives the details of mathematical analysis used in this solution process.
I After the axial velocity field is evaluated, it is a straight-forward matter to compute the fuel assembly cladding temperature.
I The knowledge of the overall flow field enables pinpointing of the storage location with tne minimum axial flow (i.e, maximum water outlet temperatures). This is called the most " choked" location.
In order to find an upper bound on the temperature in a typical cell, it is assumed that it is located at the most choked location. Knowing the global plenum velocity field, the revised I
I 5-12
[ axial flow through this choked cell can be calculated by solving h the Bernoulli's equation for the flow circuit through this cell.
. Thus, an absolute upper bound on the water exit temperature and maximum fuel cladding temperature is obtained. In view of these aforementioned assumptions, the temperatures calculated in this manner overestimate the temperature rise that will actually occur in the pool. Holtec's computer code THERPOOL*, based on the theory of Ref. 5.6.2, automates this calculation. The analysis procedure embodied in THERPOOL has been accepted by the Nuclear Regulatory Commission on several dockets. The Code THERPOOL for local temperature analyses includes the calculation of void generations. The effect of void on the conservation equation, crud layer in the clad, flux trap temperature due to gamma heating, and the clad stresa calculation when a void exists, are all incorporated in THEPPOOL. The peaking factors are given in Table 5.6.1.
5.7 Claddina Temperature The maximum specific power of a fuel array qA can be given by:
qA = q Fxy (1) where:
Fw = radial peaking factor q ~ = average fuel assembly specific power l
- THERPOOL has been used in qualifying the spent fuel pools for Enrico Fermi Unit 2 (1980), Quad Cities I and II (1981),
Oyster Creek (1984), V.C. Summer (1984), Rancho Seco (1983),
Grand Gulf I (1985), Diablo Canyon I and II (1986), among others.
5-13 I
I
J The maximum temperature rise of pool uater in the most disadvantageously placed fuel assembly is computed for all loading cases. Having determined the maximum local water temperature in the pool, it is now possible to determine the maximum fuel cladding temperature. A fuel rod can produce F2 times the average
- heat emission rate over a small length, where F
- is the axial rod peaking factor. The axial heat distribution in a rod is generally a maximum in the central region, and tapers off at its two extremities.
It can be shown that the power distribution corresponding to the chopped cosine power emission rate is given by q(x) = qA sin h + 2a where:
f h: active fuel length i
a: chopped length at both extremities in the power curve x: axial coordinate with origin at the bottom of the active fuel region The value of a is given by hz a=
1 - 22 where:
__ __1/2 1 1 1 2 z= - - +
n Fz n 2 pz2 g pz y2 5-14 l
I
L where Fg is the axial peaking factor.
The cladding temperature Te is governed by a third order differential equation which has the form of d3 T d2 T dT
+ c1 -
a2 " f (X) dx3 dx2 dx
] '
where al, a2 and f(x) are functions of x, and fuel assembly gecmetric properties. The solution of this differential equation with appropriate boundary conditions provides the fuel cladding temperature and local water temperature profile.
In order to introduce some additional conservatism in the analysis, we assume that the fuel cladding has a crud deposit which results in .005 0F -sq.ft.-hr/ Btu of crud resistance, which covers the entire surface.
Table 5.6.2 provides the key input data for local temperature analysis. The results of maximum local pool water temperature and minimum local fuel cladding temperature are presented in Table 5.7.1.
The local boiling temperature of water is approximately 242*F at 26' below the free water surface and higher at lower elevations.
The location where the local water temperature reaches its maximum value is deeper than 26' below the free water surface, where the coincident boiling temperature of water is greater than 242or.
It is shown that the local pool water temperature is lower than the local boiling point and therefore, nucleate boiling will not occur.
I I
l e-1s
4 i
L Finally, it is noted that the fuel cladding temperature is considerably lower than the temperatures to which the cladding is
( subjected inside the reactor. Therefore, it is concluded that there is sufficient margin against fuel cladding failure in the
~
spent fuel pool.
I 5.8 Blocked Cell Analysis Calculations are also performed assuming that 50% of the top opening in the thermally limiting storage cell is blocked due to a horizontally placed (misplaced) fuel assembly. The corresponding maximum local pool water temperature and local fuel cladding temperature data are also presented in Table 5.7.1.
I There is also no incidence of localized nucleate boiling of the pool water or potential for fuel cladding damage.
i 5.9 References for Section 5 5.6.1 General Electric Corporation, R&D Data Books, " Heat 1 Transfer and Fluid Flow", 1974 and updates.
5.6.2 Singh, K.P. et al., " Method for Computing the Maximum Water Temperature in a Fuel Pool Containing Spent Nuclear Fuel", Heat Transfer Engineering, Vol. 7, No. 1-2, pp. 72-82 (1986).
I 5-16
L Table 5.4.1 FUEL SPECIFIC POWER AND POOL CAPACITY DATA I
Total water volume of Pool: 635645 gallons Specific Operating Power of a Fuel Assembly: 60.3E+06 Btu /hr.
Dimensionless decay power of "old" discharges: 0.3303 l
l l
l 5-17
w I
I Table 5.4.2 DATA FOR SCENARIOS 1 through 4 CAsr No, It ih 1 1 1 1 Pool thermal capacity 4.241 4.241 4.241 4.241 4.241 C x 10-0, Btu / 0F No. of Cooling Trains 1 1 2 2 1 No. of Discharges 1 1 1 2 2 Considered for the Analysis l Time between --- --- --- 720 720 B Shutdowns, hr.
Cooler Inlet Tamp., 108.4 108.4 103.9 101.4 104.7 0F ,
coolant Flow Rate / 1.49 1.49 1,49 1.49 1.49 Cooler, 10 6 lb./hr.
Fuel Pool Water 1.14 1.40 1.14 1.14 1.14 Flow Rate, 106 lb./hr.
g Temperature 0.3970 0.43 0.3975 0.3979 0.3987 g Effectiveness /
cooler, p I
I l
5-18
] _
I Table 5.4.3 I DATA FOR SCENARIOS 1 THROUGH 4 I .
Time After Shutdown when offload Expo.
I. Case Discharge No. of Transfer Begins Time Ti.me No. ID Assemblies (hrs) (hrs) (hrs I la Discharge 1 80 168 19.07 30240 '4 or lb 2 Dischargo 1 80 168 19.07 30240 Discharge 1 80 168 19.07 30240 3 {
l or 4
Discharge 2 (Full Core) 80 113 168 46.00 10080 30240 5-19
s a
l l
I I
Table 5.5.1 POOL BULK TEMPERATURE AND HEAT LOAD DATA Time Coincident Tmax to Tmax Coincident Coincident Max. Pool (after Evaporation Case CoglerDuty Bulk Temp., reactor No. 10 Btu /hr. Hegt Loss,
'F shutdown) 10 Btu /hr.
- 1a 30.241 15e.54 207 2.00 lb 30.69 156.31 206 2.578 2 32.787 131.57 198 0.689 3 50.690 143.84 222 1.395 4 45.04 176.91 225 6.887 I
I I
I I
I 5-20 u__
Y
\
l I
[ Table 5.5.2 TIME-TO-BOIL FOR VARIOUS DISCHARGE SCE!!ARIOS Time-to ' oil (hours)
Case 11 umber, G =. O GPM la 7.82 lb 8.27 2 11.52 3 5.74 4 3.02 I
I I
I I .
l
"~"
i~
L-y Table 5.6.'
PEAKING FACTOR DATA Radial Bundle Peaking Factor 1.65 Total peaking factor 2.40 I
I I
y 5-22
]
L Table 5.6.2 DATA FOR LOCAL TEMPERATURE I
Type of Fuel Assembly PWR Fuel Cladding Outer Diameter, inches 0.36 Fuel Cladding Inside Diameter, inches 0.31 Storage Cell inside Dimension, inches 8.75 Active fuel length, inches 144 No. of Fuel Rods / Assembly 264 Operating Pcwer per Fuel Assembly 60.3 P o x -6, Btu /hr Cell pitch, inches 8.97 Cell height, inches 168 _
Plenum radius, feet 29.3 Min. Bottom height, inches 4.75 Min. gap between pool wall 1.5 and outer rack periphery, inches 5-23
4 s
L-I I
Table 5.7.1 LOCAL AND CLADDIl1G TEMPERATURE OUTPUT DATA FOR THE MAXIMUM POOL WATER COllDITIOli (Case a)
Condition Water Temo., F Temn., *F No blockage 168.0 212.9 50% blockage 219.2 246.9 I
I I
l 5-24 1
L F
L r
l 1 o.
[ N
^
Y W,([ s \
W9 v
)
I /
I it t wt l
l FIGURE 5.5.1 pool Bulk Temperature Model 5-25 i
__ _ .----________1.__---_ - - - -
MT'D L_J L_J 165 -
DONALD C. COOK SFP NORMAL DISCHAh iE, ONE COOLING TRAIN, CASE la j 163 -
1 l
160 -
158 -
155 k w Q. :
i, js 153 -
t
_i o
@ 150 -
5 :
S148(
145 143 ;
I 140 -
~
138- , ,,,;i , ii,iiii , ii , ij i i i i iii irTiiii1 i i i i i i e i i i i i i i 500 i
150 200 250 300 350 400 450 0 50 100 TIME AFTER FIRST SliUIDOWN OF Tile RFACTOR. HR FIGURE 5.5.2
7-- - ; .;
l 165--
DONALD C. COOK SFP NORMAL DISCHARGE, ONE COOLING TRAIN, CASE lb 163-l :
160 2 158--
155 i tL a:
1 153I-
's" N I 3 g 150 -- ;
o -
- a. :
148 -
3 8 5 145 -
1435 f 140 1 138 -
~
135 - i i i ijiiiiji,i,,,iiiie i i i i i i ii iiiiiiiiiiiii i e i i i i i :
500 200 250 300 350 400 450 0 50 100 150 TIME AFTER FIRST StiUTDOWN OF Tile REACTOR. IIR FIGURE 5.5.3
_, _ _ . , ,_1 ,
1 l
135 -
~
I DONALD C. COOK SFP NORMAL DISCHARGE, W/O COOL!NG TRAINS, CASE 2 133 -
i l
~_
l 130 - ~
I l
tt
'~
T (128 -
% W ~
s v i 4
O 125 -
f
??
)
t 123 -
120 -
118
~
ii i g i i i i ie i e i l i i i i ie i i iiiii i i i iiiiii iiiii i i i; 500 150 200 250 300 350 400 450 0 50 100 f TIME AFTER FIRST SHUlDOWil OF THE REACTOR. IIR l
FIGURE 5.5.4
l 3fy n MMMO TO
\' <
150 -
- DONALD C. COOK SFP FULL' CORE OFFLOAD,TWO COOLING TRAINS. CASE 3 145 -
4 N 140 -
~
l _
- u. 135 -
w a h :E _
w I _
a 130 -
8 a :
125 ~
m _
120 -
~
- l 115 -
110 , i,ii iiiiie iiiii **!'I TT1i iia i *IiI3 100 '00 300 * ' ' l 400 ' ' ' ' I 500' ' ' ' [O 6 700 800 900 1000 1100 1200 l j TIME AFTER FIRST SHUIDOwtl OF THE REACTOR, HR FIGUltE 5.5.5
, _ _ c ~ , .
l 185 _
1a0 DONALD C. COOK SFP FULL CORE OFFLOAD.ONE COOLING TRAIN, CASE 4 l :
175 -
170k k 165 2 u.
a 160 2-t :E Y N g g 155 -
o -
0-x 150 -=
S m :
145 -
~
140 -
135 - J 130 1
...i....i,,,,i.i. iii.,i 125 ~ i,i,i, ..,,,,,,i ,,,,ii,ii,i.,, ..i 400 500 600 700 800 900 1000 1100 1200 0 100 200 300 TiuE AFTER FIRST StiUIDOWr1 OF THE REACTOR. IfR FIGURE 5.5.6
m nums i FT 1--i 1 5.50E+005 -
~
5.00E+005 -
COOK SFP LOSS OF COOLING SCEllARIO, CASE la 4.50E+005 0 4.00E+005 l j
cn -
C _
- @ 3.50E+005 _
O _
E
, cE 3.00E+005 0' :
- 3 o 2.50E+005 -
n o -
2.00E+005 -
v -
E :
a -
~o 1.50E+005 -
Ti :
y 1.00E+005 - \
H :
5.00E+004 \
~
0.00Et000 , ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,160 80 120 O 40 Time After Max. temp. has been reached, hr F IGURE 5.5. 7
. _ . . . . ..- ..... - - .-.- -.-~.- ....-.-. ...-... ----.-.- -.--.--..--- - ---..- --.. ..-.- ....-..-. _..
m m M M M M m m M -
l I f i
, t
- 5.50E+005 -- l 1
i 5.00E+005 - i COOK SFP LOSS OF COOLING SCENARIO CASE lb l j
4.50E+005 (
\ !
i O 4.00E+005 -~ - ,
i 6
c t
j 'c 3.50E+005 - l i '6 : l 1
E -
o -
i cr 3.00E+005 - -
i
- i i T E 1 !
- w - -
i w o 2.50E+005 - !
3: _
n -
2.00E+005 -
E :
a _
o 1.50E FOO5 - -
> ; N t
_o -
l 5 1.00E+005 -
F- _
5.00E+004 (
~
0.00E+000 i ii,,,iiii iiiiis i i i i i . . . iii i iiiiiiiii O 40 80 120 160 Time After Max. temp. has been reached, hr i
FIGURE 5.5.8 1
m M M M M M M 5.50E+005 -
~
S.00E+005 -
COOK SFP LOSS OF COOLING SCENARIO, CASE 2
{
~
4.50E+005 -
~
0 4.00E+005 -
ci _
.E -
c 3.50E+005 -
~6 :
E -
w E 3.00E+005 I -
o 2.50E+005 -
3 0 -
2.00E+005 -
v -
E :
3 -
o 1.50E+005 - l 0 :
5 1.00E+005 - 1 W -
~
5.00E+004 - \
0.00E+000 j i iiiiii,,ii,,,,,,iig . . . , , , i i i , iiii. ,,,,
40 80 120 160 0
Time Af ter Max. temp. has been reached, hr FIGURE 5.5.9 1
t l
M M M M M M M M M M M M M M M M l l
i i
i 1 5.50E+005 - -
l
\
5.00E+005 -
COOK SFP LOSS OF COOLING SCENARIO, CASE 3 t
! - 3 i
4.50E+005 - l i
i l 0 4.00E+005 (
i d, c
~E 3.50E+005 - !
'5
~
i E !
- o -
i v' e
rr 3.00E+005 - -
w u 1 A g -
o 2.50E+005 --
3 _
1 o - .
3 2.00E+005 -
o -
E : ,
, a -
t i --o 1.50E+005 - l
) !
o -
4 5 1.00E+005 - l H -
- t i
5.00E+004 - i
- L 0.00E+000 , ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,\,
! 0 20 40 60 80 t Time After Max. temp. has been reached, hr f >
l FIGU RE 5. 5.10 j
i
. - - - - . . . . ~ . . - . . - - - . - . .
m m- M M M M M M M M M m m m m m 'm m :
r i
5.50E+005 ---
5.00E+005 COOK SFP LOSS OF COOLING SCENARIO, CASE 4 4.50E+005 (
O 4.00E+005 (
si :
.E -
,g 3.50E+005 --
o _
E -
, cE 3.00E+005 (
a, w e g :
o 2.50E+005 --
3 _
O -
2.00E+005 -
v -
E :
a _
-o 1.50E+005 -
> ~
o _
y 1.00E+005 --
w 5.00E+004(
~
0.00E+000 i iiiiiiiii,. . iiiii,iii iiiiiiiiiie iii 0 20 40 60 80 Time Af ter Max. temp. has been reached, hr FIGURE 5.5.11
- - __- . _ _ _ . . _ _ _ -]
, N
~
E'S b 3
$$ 5Us
~
- 2 cEP b$ UeU E
L E3 Ed 3E W
5 d >
<; = =::
NNN'N, T '
I L
'x i s
I !
i N
' l
. I
/ ,/
=. =.,
l 3_: EU$
88 c E!!
sb t
!I!
a 5-36 ICEAL!ZATICN CF RACX ASSE!.tELY FIGURE 5.6.1
I
,r-M ,
I ;
l i 4 i
A Y A
/i /
- 7 TCUT g
[l / fi/
I I IN g ,
91
% l w i e i 1 m 2 E H
- J C b l > C
! 4 Q HEAT ACCITICN l h
1 :si !
l 9
v l 1 I
I
)
l
/D T )3y) -
l y '
' / i i;/
/i / ' '[ ' " p
~
l / / /
q /) .
w ._ V j -
. _a V I ,
I l 5-37 THERMAL CHIMNEY FLCW MCCEL FIGURE 5.6.2
>- .2 / ;
I z
- *L.
.. ..J l
.~ .
- a. .>
l .- .- ,
t- . >\
- 1. ,,
.1 . -
l 4
.: r. '.
l ,.
t I
.. .>l . .
1 ;.-
.A .A E 3 .
i.' a cx ;.
l
\, 's - I
. . i.
,.4 l .
i -
9
. d. i l p.
. Y i i ,
Y .'.
3 1.. ..\
E .
../
i,
.t
- .. ..(
g .
... yy I ( _ ~4 I.v.*
3 ccwn ccr. San I .3
.; . i .
..e
.y I
l ..
. r. + ; 7 -ll
- m .
l 3 A ; A '. -
q , . t _. -s 1 .7
, . ,r . , 5'. y -s - i s
..,-s..,
.c.s,.7-
- -4
../ .A ,
_.945
. 7. -
l ut.Nur,t 5-38 CONVECTICN CUR RENT 3
.l
'=
IN THE PCC L FIGURE 5.6.3
E l
I 6.0 STATIC AND DYNAMIC ANALYSIS OF RACK STRUCTURE 6.1 Introduction The purpose of this section is to present analyses which demonstrate the structural adequacy of the Donald C. Cook spent fuel high density rack design under normal storage and the postulated accident loading conditions as defined by and following the guidelines of the USNRC Standard Review Plan (Ref. 6.1.1). The I method of analysis presented uses a time-history integration method similar to that previously used in the licensing reports on high density spent fuel racks for Enrico Fermi Unit 2 (USNRC Docket No. 50-341), Quad Cities 1 and 2 (USNRC Docket Nos. 50-254 and 50-265), Rancho Seco (USNRC Docket No. 50-312), Grand Gulf Unit 1 (USNRC Docket No. 50-416), Oyster Creek (USNRC Docket No.
50-219), V.C. Summer (USNRC Docket No. 50-395), Dihblo Canyon I Units 1 and 2 (USNRC Docket Nos. 50-275 and 50-323), Vogtle Unit 2 (USNRC Docket No. 50-425) and Millstone Point Unit 1 (USNRC Docket No. 50-245). The analyses carried out for the Donald C. Cook racks are considerably more elaborate and exhaustive in scope and substance than those performed in the aforementioned dockets, and reflect advances in 3-D fuel rack simulation technology in the past two years. The details are presented later in this section, I after the essential elements of the dynamic model are fully explained.
The results show that the high density spent fuel racks are f structurally adequate to resist the postulated stresa combinations associated with levol A, B, C, and D conditions as defined in Refs. 6.1.1, 6.1.2, and 6.1.3.
6-1 l
l
L
[ 6,2 Analysis Outline The principal steps in performing the seismic analysis of Donald C. Cook racks are summarized below:
(
- a. Develop statistically independent synthetic time histories for three orthogonal directions which satisfy USNRC SRP3.8.4. Two time histories are considered to be statistically independent if their normalized correlation coefficient is less than 0.15.
- b. Prepare a three-dimensional dynamic model of the fuel rack which embodies all elastostatic characteristics and structural nonlinearities of the Donald C. Cook rack modules.
- c. Perform a series of 3-D dynamic analyses on a limiting module geometry type from those listed in Tables 2.1.1 I and 2.1.3 and for varying physical conditicns (such as coefficient of friction, extent of cells containing fuel assemblies, and proximity of other racks).
I d. Perform stress analysis for the critical case from the dynamic analysis runs made in the fc egoing steps.
Demonstrate compliance with ACME Code Sect. ion III, sub-I
- section NF (Ref. 6.1.2) limits.
- e. Carry out a degree-of-freedom (DOF) reduction procedure on the single rack 3-D model such that the kinematic l
responses calculated by the reduced DOF (model RDOFM)
I are in agreement with the baseline model of step (b abova. This reduced DOF model is also truly three-dimensionel.
- f. Prepare a whole pool multi-rack dynamic model by compiling the RDOFM's of 3.11 rack modules in the pool, and by including all fluid coupling interactions among them, as well as those between the racks and pool walls.
'1his 3-D multi-module simulation is referred to as a Whole Pool Multi-Rack (WPMR) model.
6-2 i
- g. Perform a 3-D Whole Pool Multi-Rack (WPMR) analysis to L demonstrate that all kinematic criteria for Donald C.
Cook rack modules are satisfied (see Section 6.8), and that pedestal compressive loads are comparable to the loads used for structural qualification per item d above. Section 6.8 gives the criteria which need to be checked.
For the Donald C. Cook racks, the principal kinematic I criteria ara (i) no rack to pool wall impact, and (ii) no rack-to-rack impact in the cellular region of the racks.
I Figure 6.2.1 shows a pictorial view of the rack module. It is noted that the baseplate extends beyond the cellular region envelope, thus ensuring that the inter-rack impact, if any, would l
first occur at the baseplate elevation. The baseplate of the rack I modules is structurally qualifiable to withstand large in-plane impact loads.
We describe each of the above analysis steps in some detail in the following sub-sections with special emphasis on the baseline 3-D dynamic model which is the building block for all subsequent analyses. We also present the results of the analysis in the l
concluding sub-section.
I 6.3 Artificial Slab Motions The UFSAR provides a single response spectrum in the horizontal direction and a single response spectrum in the vertical direction (2/3 of the horizontal) for the Design Basis Earthquake (DBE). A corresponding pair of spectra are provided for the operating Basis Earthquake (OBE).
6-3
I '
h i
Holtec's Q.A. validated time history generation ecde GE!!EQ ( 6. 3.1]
was used to generate three synthetic statistically independent time histories for the North-South, East-West and vertical directions, respectively, from the two response spectra. 5%
l damping is used for the DBE condition. Figures 6.3.1 - 6.3.3 show the DBE time history plots. Response spectra corresponding to these time histories were also generated and are shown overlaid on the design spectra in Figures 6.3.4 - 6.3.6.
The normalized correlation coefficients pij between time histories
, i and j (1 a N-S, 2 = E-W, 3 a vertical) are provided in Table 6.3.1.
The above analyses were repeated for the OBE spectra using 2%
damping. Figures 6.3.7 - 6.3.9 present the time history plots, and Figures 6.3.10 -6.3.12 show the comparison between the design spectra and the derived spectra. Table 6.3.1 also provides pij l
g for the CBE time histories. It is noted that the enveloping
, 3 requirement on the derived spectra and statistical non-coherence of artificial motions are unconditionally satisfied.
!I l
I I
!I iI lI l
l
{ 6.4 Outline of Sincie Rack 3-D Analysis The spent fuel storage racks are Seismic Class I equipment. They are required to remain functional during and after a Design Basis I Earthquake (Ref. 6.1.3). These racks are neither anchored to the pool floor nor attached to the sidewalls. The individual rack modules are not interconnected. Furthermore, a particular rack may be completely loaded with fuel assemblies (which corresponds to greatest rack inertia), or it may be completely empty. The coefficient of friction, y, between the supports and pool floor is indeterminate According (Ref.
I another 6.4.1), the results of 199 tests performed on austenitic stainless factor. to Rabinowicz steel plates submerged in water show a mean value of p to be 0.503 with a standard deviation of 0.125. The upper and lower bounds (based on twice the standard deviation) are thus 0.753 and 0.253, respectively. Analyses are therefore performed for single rack simulations using values of the coefficient of friction equal to 0.2 (lower limit) and 0.0 (upper limit), respectively. The bounding values of p = 0.2 and 0.8 have been found to bracket the upper limit of the module response in previous rerack projects.
A single rack 3-D &nalysis requires another key modelling assumption. This relates to the location and relative motion of neighboring racks. The gap between a peripheral rack and an adjacent pool wall is known, and the motion of the pool wall is prescribed without any ambiguity. However, another rack adjacent to the rack being anclyzed is also free-standing and subject to I
l motion during a seismic event.
a given rack its physical interface with neighboring modules must To conduct the seismic analysis of be specified. The standard procedure in the single rack analysis is to specify that the neighboring racks move 180* out-of-phase in 6-5 l
l
1 a
relation to the subject rack. Thus, the available gap before inter-rack impact occurs is one half of the physical gap. This
- " opposed phase motion" assumption increases the likelihood of predicting intra-rack impacts and is thus a conservative l assumption. However, it also increases the relative contribution of fluid coupling terms, which depend on fluid gaps and relative movements of bodies, making the outright conservatism a less certain assertion. In fact, 3-D Whole Pool Multi-rack analyses carrded out for Taiwan Power Company's Chin Shan Station, and for 1 GPU Nuclear's Oyster Creek Nuclear Station show that the single I
.g rack simulations predict smaller rack displacement during seisnic jg responses. Nevertheless, single rack analyses permit detailed evaluation of stress fields, and serve as a benchmark check for the much more involved, WPMR analysis results. In order to i predict the limiting ccndi'lons of rack module seismic response 4
within the framework of single rack analysis, module A4 (13x14) is analyzed. This is typical of the largest module, and is also a corner module. The corner module has larger rack-to-wall gaps
!. which will minimize the fluid coupling.
The rack is considered fully loaded or half loaded, with limiting coefficients of friction; these simulations identify the worst case response for rack movement and for rack structural integrity.
After completion of reracking, the gaps between the rack modules
$ and those between the racks and walls will be in the manner of Figure 2.1.1. We show in this report that all single rack 3-D
. simulations predict that no rack-to-rack or rack-to-wall impacts
!. will occur in the cellular region of the racks.
The seismic analyses were performed utilizing the time-history method. Pool slab acceleration data presented in the preceding
{ sub-section was used.
I 6-6
I I The objective of the seismic analy sis of single racks is to determine the structural response (otresses, deformation, rigid body motion, etc.) due to simultaneous application of the three statistically independent, orthogonal seismic excitations. Thus, 1 recourse to approximate statistical summation techniques such as the " Square-Root-of-the-Sum-of *.he-Squares" method (Ref. 6.4.2) is avoided. For nonlinear analysis, the only practical method is simultaneous application of the seismic loading to a nonlinear model of the structure.
I Thn seismic analysis of a single rack is performed in three steps, namely:
I 1. Development of a nonlinear dynamic model consisting of inertial elements.
mass elements, spring, gap, and friction
- 2. Generation of the equations of motion and inertial coupling and solution of the equations using the
" component element method" (Refs. 6.4.3 and 6.4.4) to i determine nodal forces and displacements. The Holtec computer code DYNARACK is used to solve the system of equations [6.4.5].
- 3. Computation of the detailed stress field in the rack l just above the baseplate and in the support legs is made
- using the nodal forces calculated in the previous step. These stresses are checked against the design limits given in a later sub-section.
l A brief description of the dynamic model follows.
l 6.5 Dynamic Model for The Sinale Rack Analysis Since the racks are not anchored to the pool slab or attached to the pool walls or to each other, they can execute a wide variety of motions. For example, the rack may slide on the pool floor 6-7 i
[ (so-cal?.d " sliding condition"); one or more legs may momentarily lose contact with the liner (" tipping condition"); or the rack may experience a combination of sliding and tipping conditions. The otructural model should permit simulation of these kinematic I events with inherent built-in conservatisms. Since the modules are -
designed to preclude the incidence of inter-rack impact in the cellular region, it is also necessary to include the potential for inter-rack impact phenomena in the analysis to demonstrate that such impacts do not occur. Lift-off of the support legs and subsequent liner impacts must be modelled using appropriate impact (gap) elements, and Coulomb friction between the rack and the pool I
liner must be simulated by appropriate piecewise linear springs.
The elssticity of the rack structure, relative to the base, must also be included in the model even though the rack may be nearly rigid. These special attributes of rack dynamics require a strong em7h asis on the modeling of the linear and nonlinear springs, dampers, and compression only stop elements. The term non-linear I spring is the generic term to denote the mathematical element representing the situation where the restoring force exerted by the element is not linearly proportional to the displacement. In the fuel rack simulation the Coulomb friction interface between the rack support leg and the liner is a typical example of a non-linear spring. The model outline in the remainder of this sub-section, and the model description in the following sub-section, describe the detailed modeling technique to simulate these effects, with 'nsiderable emphasis placed on the nonlinearity of the rack seismi. response.
6-8 l
a
( 6.5.1 Assumntions
- a. The fuel rack structure is a folded metal plate assemblage welded to a baseplate and supported on four legs. An odd-I shaped module may have more than structure itself is a very rigid structure. Dynamic analysis of typical multi-cell racks has shown that the motion of the four legs. The rack structure is captured almost completely by modelling the rack I as a twelve degree-of-freedom structure, where the moven.nnt of the rack cross-section at any height is described in terms of six degrees-of-freedom of the rack base and six degrees of I freedom defined at the rack top. The rattling fuel is modelled by five lumped masses
.25H, and at the rack base, where H is the rack height as located at H, .75H, .5H, measured from the base.
- b. The seismic motion of a fuel rack is characterized by random rattling of fuel assemblies in their individual I storage locations. Assuming a certain statistical coherence (i.e. assuming that all fuel elements move in-phase within a rack) in the vibration of the fuel assemblies exaggerates the computed dynamic loading on the rack structure. This I. assumption, however, greatly reduces the required degrees-of-freedom needed to model the fuel assemblies which are represented by five lumped masses located at different I levels of the rack. The centroid of each fuel assembly mass can be located, relative to the rack structure centroid at that level, so as to simulate a partially loaded rack.
- c. The local flexibility of the pedestal is modelled so as to account for floor elasticity, and local rack elasticity just above the pedestal, d The rack base support may. slide or lift-off the pool floor.
- e. The pool floor has a specified time-history of seismic I
1 accelerations along the thrae orthogonal directions.
- f. Fluid coupling between rack and fuel assemblies, and between rack and wall, is simulated by introducing appropriate inertial coupling into the system kinetic energy. Inclusion of these effects uses the methods of Refs. 6.5.1 and 6.5.2 for rack / assembly coupling and for rack / rack coupling.
6-9 l
l
_ . - - - __ . - - _ _ = _ _ -_
I
- g. Potential impacts between rack and fuel assemblies are accounted for by appropriate " compression only" gap elements between masses involved.
- h. Fluid damping due to viscous effects between rack &
assemblies, and between rack and adjacent rack, .
conservatively neglected.
- 1. The supports are modeled as " compression only" elements I the vertical direction and as " rigid links" for transferring horizontal stress. The bottom of a support leg is attached for to a frictional spring as described in sub-section 6.6. The cross-section inertial properties of the support legs are computed and used in the final computations to determine support leg stresses.
- j. The effect of sloshing is negligible at the level of the top of the rack and is hence neglected.
- k. The possible incidence of rack-to-wall or rack-to-rack impact is simulated by gap elements at the top and bottom of the rack in the two horizontal directions. The bottom elements are located at the baseplate elevation.
- 1. Rattling of fuel assemblies inside the storage locations causes the " gap" between the fuel assemblies and the cell I wall to change from a maximum of twice the nominal gap to a theoretical zero gap. Fluid coupling coef ficients are based on the nominal gap.
- m. The form drag due to motion of the fuel assembly in the storage cell, or that due to movement of a rack in the pool, has been neglected in this ar alysis for added conservatism.
- n. The fluid coupling terms are based on opposed phase motion of adjacent modules.
I Figure 6.5.1 shows a schematic of the model. Twelve degrees of freedom are used to track the motion of the rack structure.
Figures 6.5.2 and 6.5.3, 1.spectively, show the inter-rack impact f springs (to track the potential for impact between racks or f between rack and wall) and fuel assembly / storage cell impact 6-10
.I 4 1
I springs at a particular level. Si (i = 1,. 4) represent support
{
locations, pi represent absolute degrees-of-freedom, .ad gi represent degreer-of-freedom relative to the slab. a n the height of the rack above the baseplate.
As shown in Figure 6.5.1, the model for simulating fuel assembly motion incorporates five rattling lumped masses. The five rattling masses are located at the basoplate, at quarter height, at half I
height, at three quarter neight, and at the top of the rack. Two degrees-of-freedom are used to track the motion of each rattling mass in the hori:: ental plane. The vertical motion of each rattling mass is assumed to be the same as the rack base. Figures 6.5.4, 6.5.5, and 6.5.f show the modelling scheme for including rack lI elasticity and the degrees of freedom associated with rack elasticity. In each plane of bending a shear and a bending spring are used to simulate elastic effects in accordance with Ref.
6.5.1. Table 6.6.2 gives spring constants for these bending springs as well as corresponding constants for extensi'snal and torsional rack elasticity.
6.5.2 Model Descrintiqn The absolute degrees-of-freedom associated with each of the mass locations are identified in Figure 6.5.1 and in Table 6.5.1. The rattling masses (nodes 1*, 2*, 3*, 4*, 5*) are described by translational degrees-of-freedom q7-q16-1 g Ui(t) is the pool floor slab displacement seismic time-history.
B Thus, there are twenty-two degrees of freedom in t.se system. Not shown in Fig. 6.5.1 are the gap elements used to model the support legs and the impacts with adjacent racks.
.I 6-11
I 6.5.3 Fluid Coucline An effect of some significance requiring careful modeling is the
" fluid coupling of fect" (Refs. 6.5.1 and 6.5.2). If one body of mass (mi) vibrates adjacent to another body (mass m2), and both bodies are submerged in a frictionless fluid medium, then Newton's equations of motion for the two bodies have the forms (mi + M11) X1+M12 X2 = applied forces on mass mi + 0 (x12)
I M21 X1 + (m2 + H22) X2 = applied forces on mass m2 + 0 (x7 2 l .
X,1 X2 denote absolute accelerations of masses mi and m2r respectively and the notation O(x2) denotes non-linear terms which arise in the derivation.
I M11, M12, H21, and H22 are fluid coupling coefficients which I depend on the shape of the two bodies, their relative disposition, etc. Fritz (Ref. 6.5.2) gives data for Mij for various body shapes
- and arrangements. The above equations indicate that the effect of l,
the fluid is to add a certain amount of mass to the body (M11 to body 1), and an external force which is proportional to the acceleration of the adjacent body (mass m2). Thus, the acceleration of one body affects the force field on another. This force is a strong function of the interbody gap, reaching large values for very small gaps. This inertial coupling is called fluid coupling. It has an important effect in rack dynamics. The lateral motion of a fuel assembly inside the storage location will encounter this - effect. So will the motion of a rack adjacent to
! another rack if the racks are closely spaced. These effects are included in the equations of motion. For example, the fluid I 6-12
- I
coupling is between nodes 2 and 2* in Figure 6.51. Furthermore, the rack equations contain coupling terms which model the effect of fluid in the gaps between adjacent racks. The coupling terms modeling the effects of fluid flowing between adjacent racks are computed assuming that all adjacent racks are vibrating 1800 out of phase from the rack being analyzed. Therefore, only one rack is I considered surrounded by a hydrodynamic mass computed as if there were a plane of symmetry located in the middle of the gap region.
Finally, fluid virtual mass is included in the vertical direction vibration equations of the rack; virtual inertia is also added to the governing equation corresponding to the rotational degree of freedom, q6(t) and q22(t).
6.5.4 Damoinc In reality, damping (Ref. 6.5.3) of the rack motion arises from material hysteresis (material damping), relative intercomponent motion in structures (structural damping), and fluid viscous effects (fluid damping). In the analysis, a maximum of 1%
structural damping is imposed on elements of the rack structure during OBE and DBE simulations. Material and fluid damping due to fluid viscosity are conservatively neglected. The dynamic model has the provision to incorporate form drag effects; however, no form drag has been used for this analysis.
6.5.5 Imoact Any fuel assembly node (e.g., 2*) may impact the corresponding I structural mass node 2. To simulate this compression-only gap elements around each rattling fuel assembly impact, four 6-13
node are provided (see Figure 6.5.3). The compressive loads developed in these springs provide the necessary data to evaluate the integrity of the cell wall structure and stored array during the seismic event. Figure 6.5.2 shows the location of the impact I springs used to simulate any potential for inter-rack or rack-to-wall impacts. Sub-section 6.6 gives more details on these additional impact springs. Since there are five rattling masses, a total of 20 impact springs are used to model fuel assembly-cell wall impact.
I 6.6 Assembiv of the Dynamic Model The cartesian coordinate system associated with the rack has the following nomenclature:
I x = Horizontal coordinate along the short direction of rack rectangular planform y = Horizontal coordinate along the long direction of the rack rectangular planform l z = Vertical coordinate upward from the rack base Table 6.6.1 lists all spring elements used in the 3-D single rack analysis.
If the simulation model is restricted to two dimensions (one horizontal motion plus vertical motion, for example) for the
! ourcoses of model clarification oniv. then a descriptive model of the simulated structure which includes gap and friction elements is shown in Figure 6.6.1.
The impacts between fuel asremblies and rack show up in the gap elements, having local stiffness KI, in Figure 6.6.1. In Table 6.6.1, gap elements 5 through 8 are for the vibrating mass at the 6-14
I I top of the rack. The support leg spring rates Ks are modeled by elements 1 through 4 in Table 6.6.1. Note that the local compliance of the concrete floor is included in Ks. To simulate sliding potential, friction elements 2 plus 8 and 4 plus 6 (Table 6.6.1) are shown in Figure 6.6.1. The friction of the support / liner interface is modeled by a piecewisc linear spring with a suitably large stiffness Kf up to the limiting lateral load, pN, where N is the current compression load at the interface I between support and liner. At every time step during the transient analysis, the current value of N (either zero for lift-off condition, or a compressive finite value) is computed. Finally, the support rotational friction springs KR reflect any rotational restraint that may be offered by the foundation. This spring rate is calculated using a modified Bousinesq equation and is included to simulate the resistive moment of the support to counteract ro'ation of the rack leg in a vertical plane. This rotation spring is also nonlinear, with a zero spring constant value assigned after a certain limiting I loading is reached.
condition of slab moment The nonlinearity of these springs (friction elements 9, 11, 13, and 15 in Table 6.6.1) reflects the edging limitation imposed on the base of the rack support legs and the shifts in the centroid of load application as the rack rotates. If this effect is neglected, any support leg bending, induced by liner / baseplate friction forces, is resisted by the leg acting as a beam
'g cantilevered from the rack baseplate. This leads to higher 5 predicted 1 rads at the support leg - baseplate junction than if the moment resisting capacity due to floor elasticity at the floor is included in the model.
The spring rate Ks, modeling the effective compression stiffness of the structure in the vicinity of the support, is computed from the equation:
, _ . , _ . . . , , , ~ . - . . .
The spring rate Ks, modeling the effective compression stiffness of the structure in the vicinity of the support, is computed from the equation:
I 1
-~
Ks w
1
-+
K1 1
K2
+
1 K3 l where
= spring rate of the support leg treated as a K1 tension-compression member K2 = local spring rate of pool slab K3
= spring rate of folded plate cell structure above support leg As described in the preceding section, the rack, along with the base, supports, and stored fuel assemblies, is modeled for the general three-dimensional (3-D) motion simulation by a twenty-two degree of freedom model. To simulate the impact and sliding phenomena expected, up to 64 nonlinear gap elements and 16 nonlinear friction elements are used. Gap and friction elements, with their connectivity and purpose, are also presented in Table 6.6.1. Table 6.6.2 lists representative values for a module used in the single rack dynamic simulations.
I
. For the 3-D simulation of a single rack, all support elements (described in Table 6.6.1) are included in the model. Coupling between the two horizontal seismic motions is provided both by any offset of the fuel assembly group centroid which causes the rotation of the entire rack and/or by the possibility of lift-off of one or more support legs. The potential exists for the rack to
- I be supported on one or more support legs during any instant of a complex 3-D seismic event. All of these potential events may be 1g 6-16 I
I I simul ?d during a 3-D motion so that a mechanism exists in the model to simulate the real behavior.
I 6.7 Time Intecration of the Ecuatieas of Motion 6.7.1 Time-History Analysis Usino Mul i-Decree of Freedon Rack Model
- E Having assembled the structural model, the dynamic equations of motion correnponding to each degree of freedom are written by I using Lagrangc's Formulation. The systsm kinetic energy can be constructed including contributions f u the solid structures and from the trapped and surrounding fluid. A single rack is modelled in detail. The system of equations can be represented in matrix notation as (M) (q") = (Q) + (G) where
[M] - total mass matrix;
{q} - the nodal displacement vector relative to the I pool slab displacement; double prime stands for secondary derivations; a vector dependent on the given ground
{G} -
I {Q} -
acceleration; a vector dependent on (linear and non-linear) the and spring the forces coupling between masses.
The equation can be rewritten as (q"} = (M]'1 (Q) + [Ml-1 (G}
As noted earlier, in the numes.ical simulations run to verify structural integrity during a seismic event, the rattling fuel assemblies are assumed to move- in phase. This will provide maximum impact force level, and induce additional conservatism in the time-history analysis.
6-17 1
I I
- . This equation set is mass uncoupled, displace:.nent coupled at each instant in time, and is ideally r ited for numerical solution using a central difference scheme. The proprietary, USNRC accepted, computer program "DYNARACK* is utilized for this purpose.
Stresses in various portions of the structure are computed from known element forces at each instant of time and the maximum value of critical stresses over the entire simulation is reported in
__ summary form at the end of each run.
In summary, dynamic t.nalysis of typical multi-cell racks has shown that the motion of tSe structure is captured almost completely by twenty-two degree of freedom structure;
] the behavicr of a therefore, in this analysis model, the movement of the rack cross-section at any height is described in terms of the rack degrees of freedom (q1(t),...q6(t) and gl7-q22(t)). The remaining degrees of free. dom are associated with horizontal movements of the
]
] fuel assembly masses. In this dynamic model, five rattling masses are used to represent fuel assembly n_vement in the horizontal This code has been ;reviously utilized in licensing of similar racks for Enrico Fermi Unit 2 (USNRC Docket No. 50-341),
Quad Cities 1 and 2 (USNRC Docket Nos. 50-254 and 265), Rancho I Seco (USNRC Docket No. 50-312), Oyster Creek (USNRC Docket No.
] 50-219), V.C. Summer (USNRC Docket No. 50-395), and Diablo Canyon 1 and 2 (USNRC Docket Nos. 50-275 and 50-323), St. Lucie Unit I (USNRC Docket No. 50-335), Byron Units I and II (USNRC Docket Nos.
50-454, 50-455), Vogule 2 (USNRC Docket 50-425), and Aillstone
- Unit 1 (USNRC Docket 50-245), Indian Point Unit 2 (USNRC Docket No. 50-247), among others.
1 1
I
-4 e-1e
f plane. Therefore, the final dynamic model consists af twelve degrees of freedom for the rack plus ten additional mass degrees of freedom for the five rattling masses. The totality of fuel mass is included in the simulation and is distributed among the five i rattling masses.
6.7.2 Evaluation of Potential f or Tnter-Rack ILnpact since racks are usually closely spaced, the simulation includes impact springs to model the potential for inter-rack impact. To yet still retain the cimplicity of I account for this poter. . .al, simulating only a single rack, gap elements are located on the rack at the top and at the baseplate level. Fig. 6.5.2 shows tbc location of these gap elements. The baseplate location is a designated potential impact region, and the impact springs located in this region are expected to register impact loads. However, the impact is disallowed in the cellular region of the racks.
Therefore, the impact springs located at the top mast not indicate any loads at any time during the seismic event.
6.8 Structural Accentance Cri),
Th we are two sets of criteria to be satisfied by the rack modules:
l
- a. Kinematic Criterion This criterion seeks to ensure that the rack is a physically stable structure. The racks are designed to preclude . inter-rack impacts in the cellular region.
Therefore, physical stability of the rack is considered along with the criterion that inter-rack impact or rack-to-wall impacts in the cellular region do not occur.
6-19 1
I I b. Stress Limits
-] The stress limits of the ASME Code,Section III, Subsection NF, 1989 Edition are used. The following loading combinations are applicable (Ref. 6.1.2) and are consistent with the plant UFSAR commitments.
Loadine Combination Stress Limit D+L Level A service limits D+L+To D + L + To + E D + L + Ta + E Level B service limits D + L + To + Pg D + L + Ta E' Level D service limits D+L+Fd The functional capability of the fuel racks should I be demonstrated.
The abbreviations in the table are those used in Section 3.8.4 of the Standard Review Plan and the " Review and I Acceptance Applications":
of Spent Fuel Storage and Handling D = Dead weight-induced internal moments (including fuel assembly weight)
I L = Live Load (not applicable for the fuel rack, since there are no moving objects in the rack load path).
Fd = Force caused by the accidental drop of the heavicat load from the maximum possible height.
Upward force on the racks caused by postulated
=
Pf
-tuck fuel assembly E = Operating Basis Earthquake (OBE)
E' = Design Basis Earthquake (DBE)
I I 6-20 I
I - _
I
= Differential temperature induced loads (normal I To operating or shutdown condition based on the most critical transient or steady state condition).
Ta - Differential temperature induced loads (the highest temperature associated with the postulated abnormal design conditions).
I The conditions Ta and To cause local thermal stresses to be produced. For fuel rack analysis, only one scenario need be examined. The worst situation will be obtained when an isolated storage location has a fuel assembly which is generating heat at the maximum postulated rate. The surrounding storage locations are assumed to contain no fuel. The heated water makes unobstructed I- contact with the inside of the storage walls, thereby producing the maximum possible temperature difference between the adjacent cells. The secondary stresses thus produced are limited to the body of the rack; that is, the support legs do not experience the secondary (thermal) stresses. For rack qualification, To, Ta are the same.
I 6.9 Material Procerties The data en the physical properties of the rack and support materials, obtained from the ASME Boiler & Pressure Vessel Code,
-Section III, appendices, are listed in Table 6.9.1. Since the maximum pool bulk temperature is less than 200 F, this is used as the reference design temperature for evaluation of material properties.
I I
I 6-21 I
6.10 Stress Limits for Various Conditions
~
The following stress limits are derived from the guideJines of the ASME Code,Section III, Subsection NF [6.1.2), in conjunction with the material properties data of the preceding section. All parameters and terminology are in accordance with the Code.
6.10.1 Normal and Upset Conditions fLevel A or Level B)
- a. Allowable etress in tension on a net section
=Ft = 0.6 Sy (Sy = yield stress at temperature)
Ft= (0.6) (25,000) = 1L,000 psi (rack material)
Ft = is equivalent to primary membrane stresses Ft= (.6) (25,000) = 15,000 psi (upper part of support feet)
= ( 6) (106,300) = 63,780 psi (lower part of support feet)
- b. Oc the gross section, allowable stress in ahear is:
1 Fy = .4 S
(.4)y(25,000) = 10,000 psi (main rack body)
Ft= (.4) (25,000) = 10,000 psi (upper part of support feet)
= (.4) (106,300) = 42,520 psi (lower pe.rt of support 'eet) i I
I 6-22
I
- c. Allowable stress in compression, Fa I (k1)2 2 (1 - /2Ce Sy I
Fa "
kl kl 3 3 5
{( ) + [3 ( ) /8Cc] -
[( -) /8C e ]}
3 r r I where:
2 E) 12
( 2rr /
Ce = [ ]
S Y
l = unsupported length of component k = length coef ficient which gives influence of boundary conditions; e.g.
k = 1 (simple support both ends)
=
1/2 (cantilever beam)
= 2 (clamped at both ends)
E = Young's Modulus r = radius of gyration of component kl/r for the main rack body is based on the full height and cross section of the honeycomb region.
I Substituting numbers , we obtain, for both support leg and honeycomb region:
Fa = 15,000 psi (main rack body)
Fa = 15,000 psi (upper part of support feet)
= 63,780 psi (lower part of support feet)
- d. Maximum allowable bending stress at the outermost fiber due to flexure about one plane of symmetry:
I F6 = 0.60 Sy = 15,000 psi (rack body)
Fb = 15,000 psi (upper part of support feet)
= 63,780 pri (lower part of support feet)
I I 6-23 I
I
I I e. Combined flexure and comprescion:
fa Cmx fbx C myf by
+ + <1 Fa DxFb x DyFby where:
fa = Direct compressive stress in the section fxb = Maximum flexural stress along x-axis I fby = Maximum axis flexural stress along y-(mx =
Cmy = 0.85 fa Dx = 1 -
F'ex fa Dy=1-F'ey 12 rr2 E F'ex,ey =
kl I 23 (
r
)
x,y I and the subscripts x,y reflect bending plane of interest.
the particular I f. Corbined flexure and compression (or tension):
fa fx b fby
+ -- + < 1.0 0.6S y Fx b Fby I
1 6-24
The above requirement should be met for both the direct tension or compression case.
6.10.2 Level D Service Limits Section F-1370 (ASME Section III, Appendix F), states that the limits for the Level D condition are the minimum of 1.2 (Sy/Ft) or (0.7Su/Ft) times the corresponding limits for Level A condition. Su is the ultimate tensile stress at 200*F per Table 6.9.1. Since 1.2 S y is greater than 0.7 S u for the lower part of the support feet, the limit is 1.54 for the lower section under DBE conditions. The limit for the upper portion of the support foot is 2.0 under DBE conditions.
1 Instead of tabulating the results of the different stresses as dimensioned values, they are presented in a dimensionless form.
These dimensionless stress factors are defined as the ratio of the actual developed stress to its specified limiting value. With this definition, the limiting value of each stress factor is 1.0 for the OBE and 2.0 (or 1.54) for the DBE condition.
6.11 Results for the Analysis of Spent Fuel Racks Usino a Sincie Rack Model and 3-D Seismic Motion A complete synopsis of the analysis of the single rack, subject to the postulated earthquake motions, is presented in a summary Table 6.11.1 which gives the bounding values of stress factors Ri (i =
1... 7). The stress factors are defined as:
R1
= Ratio of direct tensile or compressive stress on a net section to its allowable value (nere support feet only support compression)
R2
= Ratio of gross shear on a net section in the x-direction to its allowable value I
I 6-25
!I I = Ratio of maximum bending stress due to bending R3 I R4
=
about the x-axis to its allowable value for the section Ratio of maximum bending stress due to bending about the y-axis to its allowable value i R5
= Combined flexure and compressive factor (as defined in 6.10.le above)
R6
= Combined flexure and tension (or compression) factor (as defined in 6.10.lf)
I R7
= Ratio of gross shear on a net scetion in the y-direction to its allowable value.
As stated before, the allowable value of Ri (i =1,2,3,4,5,6,7) is 1 for the OBE condition and 2 for the DBE fexcept for the lower section of th2 support where the factor is 1,54)
The dynamic analysis gives the maximax (maximum in time and in I space) values of the stress factors at critical locations in the rack module. Values are also obtained for maximum rack displacements and for critical impact loads. Table 6.11.1 presents critical results for the stress factors, and for rack-to-fuel impact load. Table 6.11.2 presents maximum results for horizontal displacements at the top and bottom of the rack in the x and y direction. For single rack simulations "x" is always the short direction of the rack. In Table 6.11.2, for each run, both the maximum value of the cum of all support foot loadings (4 supports) as well as the maximum value on any single foot is I reported. The table also gives values for the maximum vertical load and the corresponding net shaar force at the liner at essentially the same time instant, and fc;- the maximum net shear load and the corresponding vertical force at a support foot at essentially the same time instant.
I I 6-26 I l I
I 1
The results presented in Tables 6.11.1 and 6.11.2 represent the totality of single rack runs carried out. The critical case for
,. structural integrity calculationa is included. Displacements at the baseplate level are minimal.
The single rack ' lysis for run A04 gave the highest stress factors for sub quent structural integrity calculations.
Subsequent to the detailed analysis, pedestals adjacent to the pool walls were relocated from the corner cell to new locations 2 cells inboard from the edge. Since this relocation could affect the conclusions concerning rack structural integrity, the critical case of run A04 was re-considered using the new pedestal I locations. The results of that re-analysis are presented in the tables as run A94. The detailed structural integrity computations reported herein are based on the critical case for the loading I scenario investigated. Subsequent Whole Pool Multi-Rack analyses are also based on the final pedestal locations.
I The results corresponding to DBE give the highest load factors.
I The critical load factors reported for the support feet are all for the upper segment of the foot for DBE simulations and are to be compared with the limiting value of 2.0. Results for the lower portion of the support foot are not critical and are not reported in the tables.
I Analyses show that significant margins of safety exist against I local deformation of the fuel storage cell due to rattling impact of fuel assemblies.
I I
'i 6-27 1
I
s c Overturning has also been considered. This has been done by assuming a multiplier of 1.5 on the DBE hori:: ental earthquakes (more conservative than required by the USNRC Standard Review Plan) and checking predicted displacements. The horizontal displacements do not grow to such an extent as to imply any possibility for overturning.
I It is noted that the analyses of the Donald C. Cook plant fuel racks have included some asymetrically loaded racks. The results I of these studies can be used as bounding analyses for the case when a rack module is picked up and relocated when loaded asymmetrically with fuel assemblies. The results presented herein indicate that twisting or deformation that would cause loss of function or violation of safety margins will not occur during a planned rack relocation.
6.12 Imnact Analyses 6.12.1 Impact Loadina Between Fuel Assembly and Cell Wall 1
The local stress in a cell wall is conservatively estimated f rom the peak impact loads obtained from the dynamic simulations.
Plastic analysis is used to obtain the limiting impact load. The limit load is calculated as 3125 lbs. per cell which is much greater than the loads obtained from any of the simulations.
I 6.12.2 Imoacts Between Adiacent Racks All of the dynamic analyses assume, conservatively, that the racks are isolated. However, the displacements obtained from the dynamic analyses are less than 50% of the rack-to-rack spacing or rack-to-wall spacing if the pool is assumed fully populated.
i 6-28 ,
1
I I Therefo.e, we conclude that no i npacts between racks or between I racks ard walls occur uuring the 3BE event. This has been further proven by the whole Pc.ol Multi-Ra.:k Analysis discussed in Section 6.14.
I 6.13 Weld Stresses I Critical weld locations under seismic loading are at the bottom of the rack at the baseplate connection and at the welds on the support legs. Results from the dynamic analysis using the g simulation codes are surveyed and the maximum loading is used to E qualify the welds on these locations.
6.13.1 Baseolate to Rep; Welds and Cell-to-Cell Welds Ref. [6.1.2] (ASME Code Section III, Subsection NF) permits, for the DBE condition, an allowable weld stress I = .42 Su = 29,820 psi. Based on the worst case of all runs report ed , the maximum weld stress for the baseplate to rack welds is 7605 psi for DBE conditions.
The weld between baseplate and support leg is checked using limit analysis techniques. The structural weld at that location is considered eafe if- the interaction curve between net force and moment is such that a derived function of F/Fy and M/My I limiting value of 1.0.
is below a I
I I e-2e i
I
I F y, My are the limit load and moment ' der direct load only and direct moment only. F, M are the absolute values of the actual force and moments applied to the wela section. The calculated value is .637 < 1.0 based on the instantaneous peak loading. This value conservatively neglects any gussets in place to increase pedestal area and inertia.
The criticsl area that must be considered for cell-to-cell welds is the weld between the cells. This weld is discontinuous as we proceed along the cell length.
Stresses in the storage cell to storage cell welds develop along the length of each storage cell due to fuel assembly impact with l
the cell wall. This occurs if fuel assemblies in adjacent cells are moving out of phase with one another so that impact loads in I two adjacent cells are in opposite directions which would tend to separate the channel from the cell at the weld. The critical load that can be transferred in this wold region for the DBE condition is calculated as 5273 lbs. at every fuel cell connection to adjacent cells. An upper bound to the load required to be transferred is 593 lbs. Where we have used a maximum impact load of 210 lbs. (obtained from Table 6.11.1), we have assumed two impact locations are supported by each weld region, and we have increased the load by V2 to account for 3-D effects.
5.13.2 Heatino of an Isolated Cell Weld stresses due to heating of an isolated hot cell are also computed. The assumption used is that a single cell is heated, over its entire length, to a temperature aoove the value associated with all surrounding cells. No thermal gradient in the I
'l I e-20
I l
l vertical direction is assumad so that the results are l
conservative. Using the temperatures associated with this unit, analysis shows that the weld stresses along the entire cell length do not exceed the allowable value for a thermal loading condition.
I Section 7 reports the value for this thermal stress.
6.14 Whole Pool Multi-Rack (WPMR) Analysis The single rack 3-D simulations presented in the preceding sections demonstrate the structural integrity, physical stability, and kinematic compliance (no rack-to-rack impact in the cellular I region) of the rack modules. However, prescribing the motion of the racks adjacent to the' module being as noted before, analy::ed introduces an assumption of unpredictable import in the single rack r.odules. For closely spaced racks, it is possible to demonstrate, kinematic compliance only by modelling all rack modules in one comprehensive simulation which is referred to as I Whole Pool Multi-Rack (WPMR) model. In the WPMR analysis, DBE seismic load is applied (Ref. 6.1.3) and all racks are assumed fully loaded with fuel assemblies. The primary intent of the analysis is to confirm structural integrity conclusions, from 3-D single rack analysis and to ensure that hydrodynamic effects not able to be modelled in a single rack analysis do not cause unanticipated structural impacts.
The cross coupling effects due to the movement of fluid from one interstitial (inter-rack) space to the adjacent one is modelled using classical pctential flow theory and Kelvin's circulation theorem. This formulation has been reviewed and approved by the Nuclear Regulatory Commission, during the post-licensing multi-rack analysis for Diablo Canyon Unit I and II reracking project.
The coupling coefficients are based on a consistent modelling of l
6-31 l
the fluid flow. While updating of the fluid flow coefficients, based on the current gap, is permitted in the algorithm, the analyses here are conservatively carried out using the constant nominal gaps that exist at the start of the seismic event.
Such a comprehensive WPMR model was prepared for the racks shown in the module layout drawing (Fig. 6.4.1). Computer code DYNARACK was used to perform the simulations.
e In order to eliminate the last significant element of uncertainty in rack dynamic analyses, the friction coefficient was also ascribed to the support leg / pool bearing pad interface in a manner consistent with Rabinowic 's experimental data [6.4.1]. A set of friction coefficients were developed by a random number generator with Gaussian normal distribution characteristics. These random derived coefficients are imposed on each pedestal of each rack in the pool. The assigned values are then held constant during the entire simulation in order that the results are reproducible.
6.14.1 Multi-Rack Model Figure 6.14.1 shows a planform view of the Donald C. Cook spent i fuel pool. A rack and pedestal numbering scheme is set up in the figure. We set up a global x axis towards the East. Table 6.14.1 gives information on the number of cells per rack, and on the rack f and fuel weights. All racks are assumed loaded with regular fuel.
There are twenty-three racks in the pool. The cask area in the pool is modelled as a fictitious rack (Rack #24 in Figure 6.14.1).
As noted previously, the aresence of a fluid moving in the narrow gaps between racks and between racks and pool valls causes fluid coupling effects which cannot be modelled with a simulation using I 6-32 l
l l .-_ ___ _ _ _ _ _ _ _ _ _ _ _
] only a single rack. Very simply, a single rack simulation can effectively include only the hydrodynamic offects due to contiguous racks when a certain set of assumptions is used for the motion of contiguous racks. In a multi-rack analysis the far field fluid coupling effects of all racks is accounted for using an appropriate model of the pool-rack fluid mechanics. For Donald I C. Cook, the cask area was modelled assuming very large fluid gaps between racks 18 and 24 and between racks 23 and 24.
In the Whole Pool Multi-Rack analysis, used to investigate the interaction effects of all racks, we employ a reduced degree-of-freedom (RDOF) set for each rack plus its contained fuel. The purpose of the whole pool dynamic analysis, including the complete i set of racks in the pool, is to determine whether effects, not able to be considered in a single rack analysis, alter any of the conclusions that are based on the results of the 22 DOF single rack analysis. In particular, the multi-rack analysis focusses on displacement excursions of each rack and on pedestal compressive loads. The Whole Pool Multi-Rack analysis is also utilized to investigate the possibility of impacts between racks or between I racks and pool walls.
The reduced degree-of-freedom structural model for each rack is developed in a systematic way so that the important kinematic results from a dynamic analysis are in agreement with similar results from a solution obtained using the 22 DOF single rack model. The external hydrodynamic mass due to the presence of walls or adjacent racks is computed in a manner consistent with fundamental fluid mechanics principles and the use of a reduced 6-33
t l DOF fuel rack model [6.14.1]. The fluid flow model, used to obtain the whole pool hydrodynamic effect is site specific and reflects actual gaps and rack locations.
The whole pool multi-rack model includes many non-linear compression only gap elements. There are gap elements representing compression only pedestals (normally four pedestals are assumed for each rack), gap elements describing the impact potential of the fuel assembly-fuel rack interface, and gap elements tracking rack-to-rack or rack-to-wall impact potential at the top and bottom corners of the rack cell structure. In addition to the compression only gap elements, each pedestal has two friction springs associated with the compression spring. As noted previously, a random number generator is used to establish a friction coefficient for each pedestal at each instant when the pedestal is in contact with the liner.
The seismic excitation directions X and Y are shown in Figure I 6.14.1. The critical DBE event that governs the behavior of the single rack analysis is applied to the 3-D multi-rack model in the appropriate directions. Three simulations have been carried out using coefficients of friction assumed to be 0.2, to be random with a mean of 0.5 at all pedestals, and to be 0.8, respectively.
f 6 .14 '. 2 - Results of Multi-Rack Analysis Tables 6.14.2 - 6.14.4 show the maximum corner absolute displacements at both the top and bottom of each rack in x and y directions from three multi-rack runs. In Table 6.14.5, the maximum displ'acements obtained from the three multi-rack simulations are compared with a single rack analysis. In all of 6-34 i
1 I
f f these tables, the results for fuel rack 24 can be ignored since there is no real rack at that location.
I The absolute displacement values are higher than those obtained from single rack analysis. Thus, it appears essential to perform Whole Pool Multi-Rack analyses to verify that racks do not impact or hit the wall. Figures 6.14.2 - 6.14.5 show the time history of I rack-to-rack gaps for the critical racks. It is shown that the rack-to-rack dynamic gaps are greater than 1.65" during a 15 second earthquake. Detailed examination of the rack-to-rack dynamic gaps show that the racks primarily move in-phase in all three sir.nlations . That is, the entire assemblage of racks tends to move and mini 4:e changes in rack-to-rack gaps.
I Table 6.14.5 also presents peak pedestal compressive loads of all I pedestals on the twenty-three real racks. In addition to a report of maximum pedestal loads, the time history of each pedestal load
< for each rack is archived for use in the structural evaluation of the fuel pool slab and the enveloping walls of the fuel pool.
It is noted that predicted maximum pede.stal force from the multi-rack simulation giving the largest pedestal load (Run MP3 in Table 6.14.5) is lower than the result obtained from single rack analysis. The maximum instantaneous vertical foot load obtained I from single rack analysis is 183300 lbs. From the Whole Pool Multi-Rack Run MP3, we find a peak single padestal load of 190900 lbs. secause , detailed rack stress calculations are based on the single rack analysis results, ao new structure concerns are identified by the scoping Whole Pool analysis and the overall structural integrity conclusions are confirmed.
I I
y-I e-35
6.15 Bearinc Pad Analysis To protect the slab from high localized dynamic loadings, bearing pads are placed between the pedestal base and the slab. Fuel rack pedestals impact on these bearing pads during a seismic event and the vertical pedestal loading is transferred to the liner. The bearing pad dimensions are set to ensure that the average pressure impacted to the slab surface due to a static load plus a dynamic impact load does not exceed the American Concrete Institute (6.15.1) limit on bearing pressures.
The time history results from the dynamic simulations for each pedestal are used to generate appropriate static and dynamic pedestal loads which are used to develop the bearing pad si::e.
From the whole pool multi-rack analysis, the worst case loading on a pedestal (instantaneous peak load) is 183,300 lbs. (see Table 6.14.5). For a 12" x 12" pad, this gives an average instanteous pressure Pa = 1273 psi.
Section 10.15 of [6.15.1] gives the design bearing strength as fb = $ (.85 fc') E where $ = .7 and fe' = 3500 psi for Donald C. Cook. E = 1 except
{ when the sapporting surface is wider on all sides than the loaded i
area. In that case, E = (A2 /A1 ).5, but not more than 2. Al is
) the actual loaded area, and A2 is an area greater than Al which is
! defined pictorially in the ACI commentary on Section 10.15. For Donald C. Cook, 15 E5 2; if we conservatively use E = 1, then fb
= 2083 psi which is in excess of the calculated pressure Pa-Thus, significant margin is provided by the bearing pads, l
l I
6-36
{
l
L-I 6.16 References for Section 6 6.1.1 USNRC Standard Review Plan, NUREG-0800 (1981).
6.1.2 ASME Boiler & Pressure Vessel Code,Section III, Subsection NF, appendices (1989).
6.1.3 USNRC Regulatory Guide 1.29, " Seismic Design Classification," Rev. 3, 1978..
6.3.1 Holtec Proprietary Report -
Verification and User's Manual, Report HI-89364, January, 1990.
6.4.1 " Friction Coefficients of Water Lubricated Stainless Steels for a Spent Fuel Rack Facility," Prof. Ernest Rabinowicz, MIT, a report for Boston Edison Company, 1976.
6.4.2 USNRC Regulatory Guide 1.92, " Combining hodal Responses I and Spatial Components in Seismic Response Analysis,"
Rev. 1, February, 1976.
- 6.4.3 "The Component Element Method in Dynamics with Application to Earthquake and Vehicle Engineering," S.
Levy and J.P.D. Wilkinson, McGraw Hill, 1976.
6.4.4 " Dynamics of Structures," R.W. Clough and J. Penzien, McGraw Hill (1975).
6.4.5 Holtec Proprietary Reports : User's Manual, Report HI-
, 89343, Revision 0; Theory, Reports HI-87162, Revision 1, and HI-90439, Revision 0; Verification, Report HI-87161, Revision 2.
6.5.1 " Dynamic Coupling in a closely Spaced Two-Body System Vibrating in Liquid Medium: The Case of Fuel Racks,"
K.P. Singh and A.I. Soler, 3rd International Conference on Nuclear Power Safety, Keswick, England, May 1982.
6.5.2 R.J. Fritz, "The Effects of Liquids on the Dynamic Motions of Immersed Solids," Journal of Engineering for Industry, Trans. of the ASME, February 1972, pp 167-172.
6.5.3 USNRC Regulatory Guide 1.61, " Damping Values for Seismic Design of Nuclear Power Plants," 1973.
6-37 l
l
1 6.14.1 " Fluid Coupling in Fuel Racks: Correlation of Theory and Experiment", by B. Paul, Holtec Report HI-88243.
6.15.1 ACI 318-89, ACI 318R-89, Building Code Requirements for Reinforced Concrete, American Concrete Institute,
,i Detreit, Michigan, 1989.
I I
I l
l l
l l
l .
I 6-38
]
u Table 6.3.1 CORRELATION COEFFICIENT I Value 9.f._.gij Time Historv Group D31 931 N-S and E-W (1,2) 0.0146 0.1056 N-S to Vertical (1,3) 0.1269 0.0956 E-W to Vertical (2,3) 0.01016 0.1060 I
I I
I l
6-39
F Table 6.5.1 W DEGREES OF FREEDOM Displacement Rotation Location Ux Uy U2 6x Oy 6:
(Node) 1 P1 P2 P3 94 95 96 2 P17 P18 P19 q20 921 922 Point 2 is assumed attached to rigid rack at the top most point.
2* P7 P8 3* P9 P10
- >
- Pil P12 5* P13 914 1* P15 P16 -
where:
pi = qi(t) + U l(t) i = 1,7,9,11,13,15,17
= qi(t) + U 2(t) i = 2,8,10,12,14,16,18
= qi(t) + U 3(t) i = 3,19 Ui{t) are the 3 known earthquake displacements.
I
!' 6-40 1
a Table 6.6.1 NUMBERING SYSTEM FOR ..P ELEMENTS AND FRICTION ELEMENTS I. H_onlinear Sorines (Gao Elements) (64 Total)
Number Node Location Descriotion 1 Support S1 Z compression only element 2 Support S2 Z compression only element 3 Support S3 Z compression only element I 4 5
Support S4 2,2*
Z X
compression rack / fuel element only assembly element impact I 6 2,2* X rack / fuel element assembly impact I 7 2,2* Y rack / fuel element assembly impact 8 2,2* Y rack / fuel assembly impact element B
9-24 Other rattling masses for nodes 1*, 3*, 4* and 5*
25 Bottom cross- Inter-rack impact elements section of rack (around edge)
I .
Inter-rack Inter-rack impact impact elements elements
. Inter-rack impact elements I .
Inter-rack impact Inter-rack _ impact elements Inter-rack impact elements elements 44 Inter-rack impact elements
- 45 Top cross-section Inter-rack impact elements
. of rack Inter-rack impact elements
. (around edge) Inter-rack impact elements
. Inter-rack impact elements
. Inter-rack impact elementu
. Inter-rack impact elements
. Inter-rack impact elements 64 Inter-rack impact elements 6-41 i
]
i a
Table 6.6.1 (continued)
NUMBERING SYSTEM FOR GAP ELEMENTS AND FRICTION ELEMENTS I II. Eriction Elements (16 total) i Number Node Location Description 1 Support S1 X direction friction I 2 3
Support S1 Support S2 Support S2 Y
X Y
direction direction direction friction friction friction 4
Support S3 i 5 6
7 Support S3 Support S4 X
Y X
direction direction direction friction friction friction 8 Support S4 Y direction friction I 9 10 S1 S1 X Slab moment Y Slab moment 11 S2 X Slab moment i 12 13 14 S2 S3 S3 Y Slab moment X Slab moment Y Slab moment 15 S4 X Slab moment i 16 S4 Y Slab moment i
I l
6-42
I
!I Table 6.6.2 TYPICAL INPUT DATA FOR RACK ANALYSES (lb-inch units)
I Support Foot Spring 4.91 x 106 Constant Ks (#/in.)
Frictional Spring 1.837 x 109
=
Constant Kg (#/in.)
Rack to Fuel Assembly 1,38 x 105 (x-direction)
Impact Spring Constant (#/in.) 1.61. x 10 (y-direction)
Elastic Shear Spring for 5.986 x 10 6 x-direction)
Rack (#/in.) 4.866 x 10 6 ((y-direction)
Elastic Bending Spring 5.458 x 1 10 (x-: plane) for Rack (#-in/in.) 4.71 x 10 0 (y-z plane)
I Elastic Extensional Spring
(#/in.)
4.074 x 107 Elastic Torsional Spring 1.322 x 109
(#-in./in.)
Gaps (in.) (for hydrodynamic I calculations)
I I
I I
6-43 I
I
I i
Table 6. 9.1 RACK MATERIAL DATA (200'F)
~
Young's Yield Ultimate Modulus Strength Strength Material E (psi) Sy (psi) Su (Psi) 304 S.S. 27.9 x 10 6 25000 71000 I -.
Section III Table Table Table Reference I-6.0 I-2.2 I-3.2 5
SUPPORT MATERIAL DATA (200 F)
Material 1 SA-240, Type 304 27.9 x 106 25,000 71000 (upper part of support psi psi psi feet) 2 SA-564-630 27,9 x 106 106,300 140,000 l (age hardened at psi poi psi 1100*F) 6-44 l
l
-- )
y l
Table 6.11.1 STRESS FACTORS AND RACK-TO-FUEL IMPACT LOAD l
l Rack / Fuel Impact Load Per Cell at Worst Location Along Height Critical Location B, B2 B3 B4 Bs B6 Br Bun Remarks (1bs)
.159 .166 .198 .231 .027 a03 DBE 180.2 .018 .023
- = .2 182 cells .417 .442 .078
.274 .074 .167 .161 loaded with reg. fuel
.172 .178 .204 .239 .033 a04 DDE 179.6 .018 .025 p= .8
.431 .460 .095
.284 .079 .214 .172 182 cells loaded with reg. fuel
.090 .073 .109 .127 .013 a30 y = 0.2 190 .012 .012 91 cells .281 .299 .052
.181 .046 .118 .106 loaded with reg. fuel
.094 .090 .109 .127 .015 a32 y = 0.2 209.8 .012 .011 91 cells .110 .271 .289 .049' loaded tiith .176 .045 .112 reg. fuel
.018 .168 .121 .187 .219 .032 a94 Same as a04 174.8 .018 with reloca- .483 .522 .113 ted pedestals .325 .056 .250 .118 Upper values are for rack cell cross-section just above baseplate.
Lower values are for support foot female cross-section just below attachment to baseplate.
m m- M M . M M M M M m M m M M M M M Table 6.11.7 Rack Displacements and Support Loads (all loads are in Ibs.)
FIDOR LOAD MAXIMUM MAXIMUrl SilEAR (sum of all VERTICAL LOAD AND support feet) LOAD COINCIDENT ,,
in a rack (1. foot) VERTI" DX DY RUN (lbs.) (Ibs.) LOAD fin.) fin.)
5 1.549x10' 30212 (1.511x105) a03 Full load 3.510x10 .0609 .0562 y = 0.2 .0084 .0105 DBC, Reg. Fuel a04 Full load 3.510x10' 1.60Ex10' 35832 (9. 791x10') .0679 .0583 s y = 0.8 .0015 .9012 i
b DBE, Reg. Fuel a30 Italf load in 1.883x10 5 1.021x10 5 20108 (1.00Sx10 )
3
.0520 .0450 1 Pos. x .0010 .0008
) y = 0.2 DBE, Reg. Fuel a32 Italf load in 1.883x10' 9.973x10' 19389 (9. 71x10') .0482 .0515 Pos. y .0055 .0080 y = 0.2 DBE, Reg. Fuel a94 SLme as a04 3.508x10 3 1.833x10 5 44406 (1. 4 8 29x10') .0678 .U778 l with reloca- .0014 .0018 ;
pedestals The value in parenthesic is the vertical load at the instant when the shear load is maximum. The maximum vertical and shear loads generally do not occur at the same instant.
! Upper vamues are top movements; lower values are baseplate movements (not l necessarily at the same time). >
l .
L 4
Table 6.14.1 RACK 11UMBERIllG NID WEIGilT IliFORMATIO!!
Rack llo . of Weight of Weight of A fd:lla Rack, lb. Fuel Assembiv, lb 2
( 1 2
182 168 25700 23700 1550 1550
. 3 168 23700 1550 ,
4 182 25700 1550
{ 5 182 25700 1550 6 182 25700 1550 7 156 22500 1550 8 144 20900 1550 9 144 20900 1550 10 156 22500 1550 11 156 22500 1550 12 156 22500 1550 13 143 20800 1550 14 132 19300 1550 15 132 19300 1550 16 143 20800 1550 17 143 20000 1550 18 143 20800 1550
( 19 182 25700 1550 20 168 23700 1550 t 21 168 23700 1550
{ 22 166 23900 1550 23 120 17700 1550 24* 0 0 0
(
l fictitious 6-47 i
i
. ...s
~
/
Table 6.14.2
[ MAXIMUM DISPLACEMEdTS FROM WPMR RUli MP1 (Friction Coefficient = 0.2)
] rack uxt uyt uxb uyb 1 .7004E-01 .7756E-01 .6235E-01 .7303E-01 2 .7506E-01 .5227E-01 .6494E-01 .3936E-01 1 3 .8464E-01 .7521E-01 .6897E-01 .6619E-01 4 .5943E-01 .5218E-01 .4960E-01 .3597E-01 5 .5131E-01 .5306E-01 .4290E-01 .4496E-01 1 6 .6793E-01 .9512E-01 .5135E-01 .9095E-01 7 .4733E-01 .8928E-01 .3978E-01 .7830E-01 8 .4856E-01 ,7065E-01 .3607E-01 .5917?. 01 9 .4533E-01 .6377E-01 .3196E-01 .5192E-01 1 10 .3830E-01 .5754E-01 .2848E-J1 . 4 3 5 4 E -01 11 .4224E-01 .5336E-01 .3659E-01 .43290-01 12 .6411E-01 .9620E-01 .4885E-01 .842)E-01 13 .7253E-01 .1079E+00 .6568E-01 .9505E-01 14 .4602?-01 .1114E+00 .3650E-01 .9847E-01 15 .3557E-01 .1079E+00 .2634E-01 .9325E-01 16 .3467E-01 .9211E-01 .2817E-01 .8608E-01
.1 17 .5755E-01 .4429E 01 .532GE-01 .3140E-01 18 .1011E+00 .1301E+00 .8596E-01 .9693E-01 19 .6980E-01 .1125E+00 .6341E-01 .8575E-01 20 .6878E-01 1 21
.8202E-01
.8404E-01
.8680E-01
.1455E+00 .6800E-01
.6048E-01
.1229E+00 22 .8173E-01 .1057E+00 .7111E-01 .9050E-01 23 .5647E-01 .6598E-01 .4812E-01 .6156E-01 1
uxt= absolute value of maximum rack corner displacement in x-direction at rack top; uyt= absolute value of maximum rack corner displacement in y-direction at rack top; uxb-absolute value of maximum rack corner displacement in I x-direction at rack baseplate; uyb= absolute value of maximum rack corner displacement in y-direction at rack baseplate.
i 6- 48 l
m
+
Table 6.14.3 MAXIMUM DL _ LACEMEllTS FROM WPMR RUll MP2 (Random Friction Coetficient) rack uxt uyt uxb uyb
.6524E-01 .4772E-01 .3373E-01 .2303E-01 I 1 2
3
.1423E+00
.1247E+00
.5829E-01
.4122E-01
.1442E+00
.1161E+00
.4598E-01
.2566E-01 4 .1860E+00 .6628E-01 .1859E+00 .3161E-J1 5 .1106E+00 .6379E-01 .1091E+00 .2673E-01 1 6 .9642E-01 .7250E-01 .8330E-01 .6348E-01 7 .4742E-01 .6267E-01 .3334E-01 .5443E-01
.1801E+00 I 8 9
10
.1275E+00
.2336E+00
.5755E-01
.3974E-01
.7640E-01
.1819E+00
.1207E+00
.2336E+00
.4534E-01
.2115E-01
.5527E-01 11 .1710E+00 .8644E-01 .1712E+00 .6245E-01 12 .4015E-01 .4740E-01 .2869E-01 .2678E-01 1 13 .1088E+00 .1034E+00 .1030E+00 .1040E+00 14 .1439E+00 .4029E-01 .1282E+00 .1865E-01 15 .6218E-01 .5620E-01 .6029E-01 .3386E-01 1 16 .3322E+00 .5413E-01 .3374E+00 .3677E-01 l' .1727E+00 .5385E-01 .1727E+00 .4896E-01 18 .1269E+00 .1958E+00 .1223E+00 .1913E+00 19 .8411E-01 .8106E-01 .6365E-01 .750SE-01 1 20 .8402E-01 .6480E-01 .5976E-01 .4419E-01 21 .1280E+00 .4742E-01 .1281E+00 .3530E-01 22 .8427E-01 .4951E-01 .7430E-01 .2335E-01 1 23 .2389E+00 .6473E-01 .2388E+00 .5758E-01 I uxt= absolute value of maximum rack corner displacement in x-direction at rack top; uyt= absolute value of maximum rack corner displacemeist in y-direction at rack tops 1 uxb-absolute value of maximum rack colar - displacement in x-direction at rack baseplate; uyb= absolute value of maximum rack corner displacement in y-direction at rack baseplate.
6-49
I I
Table 6.14.4 MAXIMUM DISPLACEMENTS FROM WPMR RUN MP3 (Friction Coefficient =0.8) rack uxt uyt uxb uyb 1 .2035E+00 .1702E+00 . 1987E+00 .1774E+00 2 .2751E+00 .5173E-01 . 2732E+00 .1658E-01 3 .2637E+00 .5740E-01 . 2638E+00 .4010E-01 4 .1363E+00 .5449E-01 . 1321E+00 .2788E-01 5 .1333E+00 .8237E-01 . 1273E+00 6876E-01 6 .1720E+00 .1514E+00 . 1609E+00 .1617E+00 7 .2425E+00 .8747E-01 . 2461E+00 .8782E-01 8 .1785E+00 .6039E-01 . 1784E+00 .4260E-01 9 .1519E+00 .4434E-01 . 1506E+00 .3129E-01 10 .8112E-01 .5007E-01 . 7887E-01 .2883E-01 gg 11 .1146E+00 .7975E-01 . 1117E+00 .5071E-01 12 .1005E+00 .1602E+00 . 9143E-01 .1601E+00 13 .1604E+00 .1310E+00 . 1633E+00 .1073E+00 14 .7786E-01 .7618E-01 . 7823E-01 .5953E-01 15 .8616E-01 .5521E-01 . 8214E-01 .3148E-01 16 .9843E-01 .4780E-01 . 1024E+00 .2903E-01 17 .8975E-01 .7115E-01 . 3063E-01 .7056E-01 18 .1418E+00 .4416E+00 . 1089E+00 .4526E+00 1 19 .1959E+00 .1720E+00 . 1922E+00 .1806E+00 20 .2741E+00 .5563E-01 . 2727E+00 .3118E-01 21 .2117E>00 .5159E-01 . 2120E+00 .2287E-01 22 .2361E+00 .6081E-01 . 2242E+00 .3986E-01 23 .1016E+00 .7703E-01 . 1033E+00 .7364E-01 uxt= absolute value of maximum rack corner displacement in x-direction at rack top; uyt= absolute value of maximum rack corner displacement in I y-direction at rack tops uxb= absolute value of maximum rack corner displacement in x-direction at rack baseplate; uyb= absolute value of maximum rack corner displacement in y-direction at rack baseplate.
I
-I 6-50
.I
I~
L,
)
i Table 6.14.5 MAXIMUM RACK DISPLACEMEllT A!!D FOOT LOAD Maximum Maximum i Run Remarka Rack Corner Displacement inch Foot Pedestal Force, lbs.
a94 Single Rack 0.0778 183,300 Analysis WPMR, y = 0.2 0.1455 (Rack #21 in y)
I MP1 157,400 (Rack #19, Foot 4)
MP2 WPMR, Random y 0.3322 (Rack #16 in x) 170,900 I (Rack #19, Foot 4)
MP3 WPMR, y = 0.8 0.4416 (Rack #18 in y) 180,900 (Rack #5, Foot 2)
~
I l
I
- u = m euen coe m der.t 6-51
____.__.m.__m_.___-___--________mm__ - _ _ _ _ - _ - - - - - - _ _ _ . _ _ . _ _ . _ _ _
I 5
I 1 l
t i
TYPICAL CELL WALLS
/VY1 A A A A A / /
/) AAA AA /4 / /
i g AAAAAAA/
E /1^A ^A '!A ^/0,^/3 '
/1 /i /\ / /\ A A /\ /
A A A A A A /iA /
- I L I
BASEPLATE I H
~
i u ,x / 7 BASEPLATE
]
I BEARING PAD Figure 6.2.1 Pictorial View of Rack Structure
' 6-52
t o
_o i ? ~
D 4 r-l y -
m f
~
~> o E -
I
% M ~
g >
-o 8 o s I
m
- =
l -
~o 8
_o e e
_ a
=
~
W d
I O C V) U o-
_a c z I -
_-- oe a
_ 1h g e rg E O 1 g e x C
-o -
-- D '
e
= l.4
- __ =
m
_o e
__ - _ . - -o ,~
I . .
_ ---.-=.-
_n c- - - -
l
~-
9 m -o I
': ' r-c iiiiiii ,
iiiiiiiiiiii iiiiiiiiiiiiiiiiiii -
l o m
o, o
o o
e 00 ao d d d d oi l i I s,6 'N011VB31300V l ___--___
e-s3
)
o
. _o N
? D L - _ -
N
.o
- _o l w _O l -
~
-o l _
~
-?
m
~ -
g
~_
o w l
i
'7 me- .
_ ']: s =
s 1
- 2 og 3 m --
-o e l --
_b y
8
5 E
. .n .
~ '_ao =
l - _o n
l -
o i
-o s g _.
_n g c -
s
==-
- o
% y
-o
_ r-l1 I I I I i i i I l l I I
- I I I I I iiI I I I I I I l1 ( l i I I I I I _
o o o O 00 N ,-- o e- (10 d d d d dI I I I s,6 'NOllVN37300V
-I o
I ,
i
_a o
g ,-
I C, s
5 _
0, p- -o 3 %
< > 'no e ,
~
I -
8
_o e m
_o I. ,- -
- m 5
I
-- =. =
e- o
_o --
I -
_o g
_ c) g g
g a
I -e-r -
a O
tj u
o- g I
~
_S $ $
e =
_ =
f o s I
-o =
_o
-- _ _ -m i
-~ -_ e
~
~
o a
!E -O g
- R O _n a E
i .c -
_o
%, -O 3 .-
g da
? -
g iiiiiiiiiinii,,,,,,f.,,,,,,,,,,,,,,,,,,
4 8 o o o o oo m ,- o ,- cao 6 6 lI 1 1 1T s,6 'N011VN n300V
- E
- E 6 33
N J -O
! v-L -
I i
w C
1
= -
,?
[ ,
Es ~
r I n r
-O = =
5@
a u 1 M a :
m 2
C >
- - =
CO C E s5 -
l N m 5:
6 o N E-0 -
l
~~
s z
8?
+ 9
, e m
=
8 g % _
1 1
liiiiiiiiijiiiiii6i il6iiiiiieiliiiiiiiiijiiiiii3 4 i O O O O O Qi O. 00 0 . t. N. (D O c o o o o or s,6 'N0l1VB31300V 6-56
)
1 -O
~
- r-
~
! i i
3 -
$E c1
=e
<m
- t
) E, z .
I U~
O E5
- r- mz
- 5E
-=
N
= =>
I _
z
. s' o
NO e- m
[p _
=
s x-8*
c - a:
a: e r cc O -
w = t-4 l e
~
e E I -
a g _ o
[
l :
I _
O O O O O 07 I o. o. o. t. N. O.. O I e- 0 O o o O e- i
' O, f) 'N011YB31300Y l s- s7
N J
-O
- v-p
+ -
x p
g W
- 2.
f m
} -
8 l i S
J >
5
- c I lr o
z O
m I -O
- e-
~
3 5
S h a
u as oc
= J . 0. ,
W i
=
o oae W ~t
~
~* $ E I '
s m
d 8
5 l ~
E iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii O O O OT O . T. N.
o Oo O O o-s,6 'N011VB31300V 6-58
O
, _O 1 <
> D
/
~
- -O
_O
-.c_
n
>=
- =
0
-O E-m
- O -
% e- =
,- w E
r g ;
r
-o
- -Q d, $
m x U
u
- o ,
2 oe m
-O . a
_bw ,
t
- - w 22= E- =
. v o
-O m
.- D d
. m
=
D 3
_ o w s -O s n
< k _
m_ _ .O f -O d _c 4
5 -
5 1 4 : 4 4 1 l1 1 1 1 6 l i jil ial il l jl l i l i t l et (I I I I I I I I e _
g) n N b cqO O O O O t_O I I I s,6 'NOl1V8313DOV 6 -59
)
o
_o O
c ) ~. _ '~
N _
r -o
_o
~_
n
- _o =
_o ,_.
2 m
~ ,- :
= -
n -
=
._ z.
o z
w
-o d E aam :-
.s
- _ - 2 e ;
_ . z e U U
_o- <
, _ __ -o
_b-y ~
u
- - _ g
_ i
~
e_
o
-o .
_c c m
c U
m-o R
_o m _n
_o
-o s> _-
IIiiiI4ill 4 i ii i IIi jiiiii ii1l1ii1111Iijiiiiiiiii _
o n a o ca o o o 0 0 'd d 6 o o oI 'i l l )
s,6 'NOllV831300V 6 -60
m
! o
, _o b
D c
C l
/ -
o
%::=._ o >.
C n z o
I . -- - ~
-v - p tn 1 : E i -
-o m
5
_o
_- s-r-
I
~ ~
_._ - p 6
[
__.m 3
- o
-o g
u
_ 0) dy I - - -
C
- In <
-.cc u
e s l _
. =
s
-o o '."
- _b .',
m.
I e - -
8 e
l ~o A
~_ _a
- D a
_=--- = ,
s
--- s
[ -
_g
_o w _n e -
2 o
m'
-o 7 - c-ll l 4 4 4 l l l l l1 4 I I I I iI I l l l I I I I I 4 ll l 4 I I I I I I ,,,,
o in 0 0 00 0 0 .O
.. o.
O O o o l l Ol I 6-6
's# 'N011V831300V
2
" l
+ 1 e
0 i
M i i
~ )
g e
n M i p
i m
E da M
M I T
2 D(
N AM M "
MR U
SS r NN M '
G' I D SE
' EV DI R
M \\ '
Iz I
LE AD T
NY kr ' Y C
OR ZO M Il
+
l' -
N E
I T RS U OI y ' Q l i
1 1
J[ : E R
M F 0
1 m f/ 3 i
1 6
e
' r u
M '
i g
F -
a >
m a ;
_ - ~ :
___ _~_ _ : ,;_.. . _ _ - - - _ ~
0 0 0 0 '-
6 .
4 .
2 9 0 .
O O O 01 g C L W_O zOi<xJJ_JOO t L _L g
aa" _
lI i: f l
L O c-i 1
e I -
m o
f ?.x 5 l > -
ec
=z r
< .-I,.)
rx 5 ES R
h --
O D=
xx G3
~
B -
r" -
=$
w=
l n mm
- ~
- h
?
K O
o=
N.,
EE I
Y
_ $ SE
~w ~
l -
m
, e 1 ~
v b
E, C
iiiiiiiiiiiiniiiiiiiiiiiiiiiiiiiii O O O OT O 4 N. (1
. .C O O O O c-S,0 'N0llVB31]OOV 6 -63
.~x.- -
- - - . _ . . . . . - .. . - -. - - . - . . . - - - . . . . ~ - - - _ . - - . . _ - .- - ~ . . . - . - - .
....-.---ns..--a-n -. ..
- I 1
i I '
N
! - O a - e-1g
?
il i
l' i
I
~
\
i l, %
' )
l C i E=
) N _I
- n ==
i5 - z' E
l E' n ao i 2
<~
- i. - 0 =<
tI
'E 35 mb a - W'e 2
"E a
- E
- <Q OW Q
., e- ~
l -
O %%
$ J S .
I E ~
i Di a
.m V
t
+ -
e 4
1 -
a
- I. .
l4il l 44 l l 1 ll l I IIIIIIl3 I I I II III lIIIIIII IIlIIIIIII I I 4
o o o o o 07
- g$
to. st.
to N . .
C.L o o
- o o o o or jg S,0 'N0l1VB31300V c-e4
., - .,e-,--.-.m. _.,_r,_%..._..,,,__y m_ ., . - ,_ ,,r_,,, ,,3r_,-,y.-.--y...,- -
p , - ~ + - . , , .
3 39 G, NC I n - u,
/
~ ~ ~ ~ ~
= C-e, /
I e, an I
I a- _
Rack /
Geometric /
Centerline ,
l H
\
yg =
I h 11
/
l F" %;
fk,~, Pn
! Ps d
R/ l l a; '5S, u 9" W Es a l *
! [4 A ,- v" ~~h G /
p,g/ Pa lP Long Direction Support
[
I k s, 9 n5s, N Typical Friction Element P,
Figure 6.5.1 SCHEMATIC MODEL FOR DYNARACK 6-65
m Typical Too Impact Element i
/ s A p, W ,4 fG465-h s brmv una4 l h o Amw4 W /
l ,
1 4M
/7)7 /
- b / m <~
l Rack Structure Typical Bottom Impact Element
)
6 w w na I
6 .
w I W/ - --
lmn4
/777 17??
I Figure 6.5.2 RACK-TC-RACK IMPACT EFRINGS 6-66
s L
Y 1 4 1 CELLWALL 7
I P _
MAI51 I - - - - -
X5 ,, /
p ,. t--. - - ,
l' , , ,, /
/ / 4 FUEL ASSEY.5LY!CE' L
$ IMP ACT 5? F.lt'G l /
. ~_ j
/
l Yg
/
/
I /
/
/
/
l /
/
>- X I
FIGURE 6.5.3 IMPACT SPRING ARRANGEMENT AT NODE i 6-67
L l
l
~
1 p :c %.
~ :-
J l
/ M cc
= ,
3 l 2 5 $
I "
h 6 E I J' !c !
=
I :
=
w g O r m
- 7 C 1 /
l : :
i -
, /
7 I
i 6-68 l
s L
I L
(
( 1,7 4
3 h L -
2
) 9 2)
G. as 2
- s. 4 3
FIGURE 6.5.5 RACK DECREES OF FREEDOM FOR X-Z PLANE BENDING 1
6-69
. . __ . - . _ - .- - _ _ . - .. _. _ - -._ . __. .-_._. _ - .. - ... - - ~ _ _ -.-. _ - -
1 i i f
f i
i
!I i
lI 18
' A L
I i lg q lg 20 e
9' L i 3 (O g
/
(\
l t
FIGURE 6.5.6 RACK DEGREES OF FREEDOM FOR
' Y-Z PLANE BENDING i '
ll a
i lI
- 5 l
6-70
- t lI
,!I k~
FUEL ASSY/ CELL
) IMFACT'SPRINC , 3:
) 1
' NiC h A 4 1
- 0. 5H F/2 s
-N1B NY- T
<- 1 R.t.CI N C.G. q I ds -%1 QV- _ T 1 1 l
l 0 . 2 511
}!/2 s l
-N1 [Y- T 1
I TTPICAL RATTI.I5G P. ASS .
5 .
FRICTIC:: -y y@ T f?RI.w, @.(yC 7g3 g SPRING, I l
- + , w E a t
. . .. -n FOUNDATION ROTATIONAL CQ9LIANCE SPRING, K g
FIGURE 6.6.1 2-D VIEW OF RACK MODULE 6-71
I I
I -N I ' '
RACK 24 j RACK 6 RACK 12 RACK 18 (NOT REAL) I X 4 3 4 3 4 2 1 2 1 2 1 2 1 RACK 5 RACK 11 RACK 17 RACK 23 3 4 3 4 3 4 3 4 2 1 2 1 2 1 2 1 l 3 RACK 4 4 3 RACK 10 4 3 RACK 16 4 3 RACK 22 4
2 1 2 1 2 1 2 1 RACK 3 RACK 9 RACK 15 RACK 21 3 4 3 4 3 4 3 4 2 1 2 1 2 1 2 1 RACK 2 RACK 8 RACK 14 RACK 20 l
3 4 3 4 3 4 3 4 2 1 2 1 2 1 2 1 X
g RACK 1 RACK 7 RACK 13 RACK 19 3 4 3 4 3 4 3 4 E c FIGURE 6.14.1 RACK AND FOOT PEDESDAL NUMBERING FOR D. C. COOK MULTl-RACK MODEL I .
I 6-72 I
. ~~~
w 2.03 _
COOK POOL MULTI-RACK SEISMIC At1ALYSIS, RUN MP2 (Random Cof.)
' 2.00 -- RACK '16 TO RACK 17 NORTH CORNER DYNAMIC GAP AT RACK TOP m
h 1.98 _
4
.O -
K 1.95 - -
4 o : .
l w 0 1.93 -
s -
Z' I ,
g -
1 y 1.90 -
V- _
e -
i I - ,
S .1.88 -
x 4
) Y, 1 s 1.85 -: ~
[y K4 ll l
3 m
1.83 _
O 1.80 ....g ..g iiiisisi sisig. g.. .g.. isis issi i . gsi TJiis , 1 gsiii ..3 0 1 2 .3 4 5 6 7 8 0 10 11 12 13 14 15 16 TIME. SEC.
FIGURE 6.14.2 l
2.03 -
I COOK POOL MULT!-RACI SEISMIC At1ALYSIS, rut 1 MP2 (Randora Cof.)
2.00 - RACK 16 TO RACK 17 SOUTH CORtJER DYtlAMIC GAP AT RACK TOP 198 --
E E 1.95 - -
t3.
0 I
m 9 1.93 -
o -
1.90 -
c ?
[D 2b 1.85 _
1.83 _
1.80 -
i i i i i e i i rTrrn i i i i i ; i e i i i i i i i i i i i i i rrn i i i i i irrrrTTvrrr rri i i i e i i ; s i n i i i i i i , , i , i 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 IlME, SEC.
FIGURE 6.14.3
g
'- g W 1 n W~ ~ l I U 1
2.06 COOK POOL MULTI-RACK SEISMIC ANALYSIS, RUN MP3 (Cof.=0 t) 2.00 -
-' RACK 12 TO RACK 18 WEST CORNER DYtMMIC GAP AT RACK 1:'P l
m 1.95 _
I. -
O _
E 1.90 - -
m o : -
', O 1.85 - a m
a
?
g[
t f 6 -
y 1.80 - l o _
$, i .75 _ ! k 4 f Y
o -
y h
Y -
i 1.70 -
I i g 1.65 _
W m
1.60 --~rri i j i i i i g i i i j i i i i j i i i i g i e i i j i i i i [ m ' ' l ' ' ' ' l ' ' ' 'Trn ri , i i s i s i i i i i i i i i i i i i i s i i .
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 TIME, SEC.
FIGURE 6.14.4
.. = = =
Q.*' M " ' WM M M 2.05 _
9 COOK POOL MULTI-RACK SEISMIC ANALYSIS, RUN MP3 (Cof.=0.8) 2.00 - RACK 12 TO RACK 18 EAST CORNER DYNAMIC GAP AT RACK TOP 1.95 _
E 1.90 -
7 -
0f f
$ 9 185 -
a z
,3 b -
(
p f
i 1.80 -
7 o
P 1.75 -
b :
1 j
1.70 - {
t 1.65 _
1.60 i i i i gii igiisigii iiii iiigiiiigiiii i,ii , ,i iiiigiiiigiiiigirr,ieieigiiiig 5 6 7 8 9 10 11 12 13 14 15 16 0 1 2 3 4 TIME. SEC.
FIGURE 6.14.5
~
L l 7.0 ACCIDENT ANALYSIS AND MISCELLANEOUS STRUCTURAL EVALUATIONS 7.1 Introduction This section provides results of accident analyses perfornied to demonstrate regulatory compliance of the new fuel racks.
There are several types of accidents which could potentially affect the spent fuel storage pool. Installation of the proposed high density racks will enable the storage of increased amounts of spent fuel in the Donald C. Cook spent fuel pool. Accordingly, I accidents involving the spent fuel pool have been evaluated to ensure that the proposed spent fuel pool modification does not change the present degree of assurance to public health and safety. The following accidents and miscellaneous structural evaluations have been considered:
Refueling accident - Dropped Fuel I
- Local Cell Wall Buckling Analysis of Welded Joints due to Isolated Hot Cell Crane Uplift Load 7.2 Refuelina Accidents This section considers three (3) accidents associated with the handling of fuel assemblies.
7.2.1 Droceed Fuel Assembly The consequences of dropping a new or spent fuel assembly as it is being moved over stored fuel is discussed below.
- a. Droceed Fuel Assembly Accident I A fuel assembly is dropped from 36" above the top of a storage location and impacts the base of the module.
Local failure of the baseplate is acceptable; however, the rack design should ensure that gross structural failure does not occur and the suberiticality of the adjacent fuel assemblies is not violated. Calculated 7-1 i MiA _ m
results show that there will be no change in the spacing l between cells. Local deformation of the baseplate in the
! neighborhood of the impact will occur, but the dropped assembly will be contained and not impact the liner. We
- show that the maximum movement of the baseplate toward I the liner after the impact is less than 1.52". The load
'E transmitted to the liner through the suoport by such an 3 accident is well below that caused by seismic loads,
- b. Droceed Fuel Assembly Accident II One fuel assembly is (assumed dry weight = 1550 lbs.)
dropped from 36" above the top of the rack and impacts the top of the rack. This is a more severe condition
- I
- than the currently postulated ap of 1616 lbs. from a height of 15" above the top f the rack. Permanent deformation of the rack is acceptable, but is required lI
)
to be limited to the top region such that the rack cross-sectional geometry at the level of the top of the active fuel (and below) is not altered. Analysis shows
'I that although local def ormation occurs, it is confined to a region above the active fuel area. The region of permanent deformation is to a depth 5.34" below the top of the rack.
- c. Droceed Fuel Assembly Accident III This postulated accident is identical to (b) above except that the fuel assembly is assumed to drop in an
. inclined manner on top of the rack. Analyses show that l
the straight drop case (case b above) bounds the results.
i t 7.3 Local Buckline of Fuel Cell Walls This subsection and the next one presents details on the secondary j stresses produced by buckling and by temperature effects.
I The allowable local buckling stresses in the fuel cell walls are obtained by using classical plate buckling analysis. The following formula for the critical stress has been used based on a width of cell "b": (See Figure 7.3.1.)
- n 2 Et2 a=
c^
12 b2 (1.p) 2 l 7-2 i L 1
i 1
I where E = 27.9 x 106 psi, p = 0.3, (Poison's ratio), t= .075",
I b = 8.75". The factor a long panel.
is suggested in (Ref. 7.3.1) to be 4.0 for For the given data acr = 7411 Psi It should be noted that this stability calculation is based on the I applied stress being uniform along the entire length of the cell wall. In the actual fuel rack, the compressive stress comes from consideration of overall bending of the rack structures during a seismic event and as such is negligible at the t.ac k top and maximum at the rack bottom. It is conservative to apply the above equation to the rack cell wall. if we compare acr with the maximum compressive stress anywhere in the cell wall. As shown in Section 6, the local buckling stress limit of 7411 psi is not violated anywhere in the body of the rack modules, since the maximum compressive stress in the outermost cel? is a = 3585 psi. (From Table 6.11.1 for R6 = .239, the stress at the base of the rack under combined direct plus bending loads is a = R6 x allowable stress).
l 7.4 Analysis of Welded Joints in Rack due to Isolated Hot Cell In this subsection, in-rack welded joints are examined under the loading conditions arising from thermal effects due to an isolated hot cell.
A thermal gradient between cells will develop when an isolated I storage location contains a fuel assembly emitting uaximum We postulated heat, while the surrounding locations are empty.
can obtain a conservativo estimate of weld stresses along the l length of an isolated hot cell by considering a beam strip (a cell l wall) uniformly heated and restrained from growth along one long l
N 7-3 1
edge. The strip is subject to a uniform temperature rise AT =
l 59.66 F. The temperature rise has been calculated from the differcnce of the maximu: local water temperature and bulk water temperature in the spent fuel pool. (see Tables 5.5.1 and 5.7.1).
Then, using a shear beam theory, we can calculate an estimate of the maximum value of the average shear stress in the atrip (see Figure 7.4.1).
The final result for wall maximum shear stress, under conservative restraint assumptions is given as eat l Imax "
.931 I where a = 9.5 x 10 6 in/in F 1 Therefore, we obtain an estimate of maximum weld shear stress in an isolated hot cell as Imax = 16984.8 l
Since this is a secondary thermal stress, it is appropriate to compare this to the allowable weld shear stress for a faulted event I < .42S u = 29820 psi. In the fuel rack, this ma::imum stress occurs near the top of the rack and does not interact with any other critical stress.
I 7.5 Crane Unlift Load of 3000 lb.
A local uplift load of 3000 lb. (UFSAR limit is 2950 lb.) will not induce any uplift stresses in the rack which are more severe than the limiting conditions discussed in the foregoing. This choice of load should be an upper bound load on the maximum load that can be applied to a struck fuel assembly during removal.
7-4
i 7.6 References for Section 7 7.3.1 " Strength of Materials", S.P. Timoshenko, 3rd Edition, Part II, pp 194-197 (1956).
I I
I I
I
<g I
I I
I I
I I
I 7-5 I - - -_
I I
I .
,
1 e
" b e I i i ?
I F IGURs 7. 3.1 I LD ADING ON R ACK W ALL I
$ + +-t A
Heated C ell Wall I =x e I ,-------------- "
< Aw "n
eid Lin e v
Y ric uas 7.4.1 g I W E L D E D J O l N T I N R' A C K I 7-6 I _
r I
I 8.0 STATIC AND DYNAMIC &llALYSIS OF FUEL POOL STRUCTURE 8.1 Introduction The Donald C. Cook spent fuel pool is a safety related, seismic category I, reinforced concrete structure. In this section an abstract of the analysis to demonstrate the structural adequacy of I the pool structure is presented. The object of the analysis is to demonstrate the compliance of the pool slab and confining walls to the applicable design codes and to NRC regulations for the condition of increased loadings due to high density fuel storage.
The loading on the pool structure la produced by the following discrete components:
a) Static Loadina 4
= Dead weight of pool structure plus pool water 1)
(including hydraulic pressure on the pool walls).
- 2) Dead weight of the rack modules and fuel assemblies stored therein.
b) Dynamic Loadina
- 2) Inertia loads due to the slab, pool walls and contained water mass which arise during a DBE or I OBE event.
Thermal Loadina c)
- 1) Mean temperature rise and temperature gradient across the pool slab and the pool walls due to temperature differential between the pool water and the atmosphere external to the slab and walls.
I I
8-1 I
I
l l
I t
I The spent fuel pool is analyzed using the finite element method.
The results for the above load components are combined using factored load combinations mandated by NUREG-0800, the Standard Review Plan (SRP), Section 3.8.4 (Ref. 8.1.1). It is demonstrated I that for the critical factored loan combinations, structural integrity is maintained when the fuel pool is assumed to be fully loaded with high density fuel racks with all storage locations occupied by fuel as s erblies . The general purpose finite element code ANSYS (Ref. 8.1.2) is utilized to perform the analysis.
The cr3 tical regions examined are the fuel pool slab and the most I critical wall sections adjoining the pool slab. Both moment and shear capacities of the critical regions ar, checked for structural integrity. Also evaluated is local punching integrity l in the vicinity of a fuel rack bearing pad. Structural capacit; evaluations are carried out in accordance with the requirements of g the American Concrete Institute (ACI) (Refs. 8.1.3 and 8.1.4). In this analysis, the load factors of SRP Section 3.8.4 have been I used together with the allowable concrete and reinforcement loads as called for by the American Concrete Institute. This constitutes the most conservative approach to the structurel qualification of the pool structure based on a static load qualification method.
8.2 General Features of the Model i The fuel pool model is constructed using information from design basis Donald C. Cook auxiliary building structural drawings . A description of the portion of the pool modelled for analysis is given in the following.
8-2 1
The fuel pool slab is a 5'-2 1/2" thick reinforced concrete slab with inside dimensions 39'-1 9/16" wide and 58'-3 1/8" long. The I slab is located at elevation 600'-605'-2 1/2" and its long direction is aligned along the plant East-West direction. The East edge of the slab has a 5'-2" thick vertical reinforced wall which extends above the slab and is mcdeled to level 650'. The West edge of the slab has a 6' thick wall from level 605'-2 1/2" to level 650'. Yhe West wall separates the fuel pool from the fuel transfer canal which is not modelled; however, the discontinuity I in the wall structure in the center of the West wall is included.
All wall modeling is done to level 650', and we assume free edges at this level. The North wall is a 6' thick wall extending from the slab to level 650'. The South edge of the slab has a 5' thick wall extending up to level 650'. It is clear from the above description that the South wall has the largest length to thickness ratio, and therefore, may represent a limiting condition I of structural strength. The foundation mat is at elevation 584' and the pool slab and upper walls are supported on the foundation mat by walls and columns around the periphery. The North edge of the slab is supported by a continuous 3'-0" thick wall, while the East edge is partially supported along its length by a 2'-6" thick wall. There are three vertical columns 'acated at the Southeast and Southwert corner of the slab, and intermediate along the South edge. There is also a portion of a wall below the South edge at one location. The floor slab has interior vertical support provided by a 2'-0" thick vertical wall providing vertical restraint in both the North-South and East-West direction over a substantial length of slab. In addition, there is a 25' span standard W14 x 158 wide flange beam from the slab North edge supporting wall to give additional pool slab support. This propped beam is skewed toward the East 16' from the North edge.
8-3
J The entire beam (the straight part plus the skewed part) is supported vertically by four TS10" x 10" square tubes. Each tubular column has also been stiffened by four 8" x 3/8" plates.
Figure 8.2.1 shows a schematic of the above geometry.
The pool slab is assumed to be loaded with 23 high -iensity fuel racks having a total of 3616 cells. For analysis purposes, each cell is assumed to contain a 1550 lb. weight fuel assembly. As noted previously, all fuel pool walls above the pool slab are I assumed to have a free edge at level 650'. Lateral restraint is provided to the vertical walls at certain locations above the 605' level. This restraint simulates the effect of adjacent structure which is not included in the modelled envelope. Figures 8.2.2 and 3.2,7 show layouts of the entire 3-D finite element model. The gridwork in different regions shows the totality of elements used.
Shell elements are used to model the slab and walls, while beam I elements are used to nodel the columns.
The finite element model is constructed using the ANSYS classical shell element STIF63 and the beam element STIF44 of the ANSYS finite element code. The shell element thickness in the various regions of the structure is the actual thickness of the structure at the location. The finite element model is prepared for the analysis of both mechanical load and thermal load. The effects of the reinforced concrete (cracked or uncracked) are accoun.ed for in the finite element model by establishing an appropriate
! effective modulus for each shell element and effective inertias for the column elements. Effective moduli are defined for each local in-plane axis fer the shell elements. The different moduli reflect the fact that different reinforcement geometries may be used in perpendicular directions of the plate-like sections when 8-4 1
9 the different concrete section assumptions (cracked or uncracked) are applied to the slab and walls. Only major reinforcement which affects the plate and shell-like behavior of the structure is incorporated into the definition of the effective moduli; additional local reinforcement in varici sreas of the pool structure are neglected in the defining of the effective moduli.
However, such local reinforcement is accounted for in the strength evaluation after results are obtained. The non-homogeneous nature of the reinforcement is taken into account by defining different material types as necessary to reflect the varying values of effective moduli in different regions. The concrete section I assumptions (cracked or uncracked) are fully in accordance with the requirements of American Concrete Institute (Refs. 8.1.3 and 8.1.4). In accordance with Ref. 8.1.4, we assume uncracked section properties for the mechanical load analyses (including load factors). For the thermal analyses, it is shown that the thermal gradients will always yield a cracked section if the uncracked stiffness is used; therefore, an iterative solution is used to I show that cracked section properties should be used for the thermal analyses.
The effective properties for the elements used in the finite element model are calculated using standard procedures for reinforced concrete sections to define equivalent effective homogeneous materials having the appropriate stiffness and I strength.
I 8-5
~
L 8.3 Loadina Conditions I~
L_
In order to evaluate the response due to the different load mechanisms outlined in Section 8.1, the following finite element analyses are carried out. Six loading cases are defined below which enable us to obtain the moments and shears for factored loadings by linear combination.
- 1. Dead loading from concrete, reinforcement and 40' of hydrostatic head. The loading is applied as a 1.0g I vertical gravitational load for the structure and a l surface pressure on the slab and walls for the hydrostatic head.
- 2. Dead loading due to weights of rack plus full fuel lead.
These loads are applied as a uniform static pressure applied to the slab.
} 3. Seismic vertical loading due to racks plus fuel load applied as an effective sustained pressure on the floor slab pedestals. The loading applied is obtained from the -
1 3-D whole pool multi-rack analysis deocribed in Section 6 of this report. From the results of that analysis, we take the stored time history of each pedestal load and I define an effective sustained pool pressure load which yields t5.e same total impulse over the time duration of the seismic event. The details of developing this effective sustained pressure load are presented later.
I We develop effective sustained vertical pressure loads for both OBE and DBE events and then perform appropriate finite element analyses.
- 4. Seismic horizontal loading due to structure weight (including reinforcement). The loading is applied as a I lg horizontal and vertical acceleration applied to the structure plus a hydrodynamic pressure equivalent to an acceleration of all of the water mass against the weakest wall. The acceleration level is obtained from the applicable response spectra and is taken as the peak j g level on the spectra at frequencies above the lowest natural frequency for the structure. A separate AliSYS frequency analysis simulation is carried out to establish the dynamic characteristics of the structure.
8-6
I
- 5. Seismic horizontal load due to shear loads from each of the pedestals. This loading is obtained by using the static + effective dynamic loads developed for case 3 above and assuming a coefficient of friction = .8. The I direction of these loads is set so as to develop stresses that maximize the load combinations necessary to satisfy structural integr# ty requirements discussed below. In this load case we also impose a lateral
.I pressure on the weakest pool wall to simulate hydrodpamic effects from fluid coupling due to rack motion relative to the wall.
- 6. A mean temperature rise plus a thermal gradient is applied acrors the walls and floor slab to simulate the I heating effect of the water in the pool. This gradient is calculated based on the maximum wall temperature deduced from the pool bulk temperature calculations for the licensing basis scenarios presented in Section 5 of I this report.
For subsequent discussion of structural integrity checks using various mandated load combinations, we refer to the above individual finite element load cases as " case 1-6", respectively.
As noted above, in addition to the static analyses using the I developed finite element model, we also perform a frequency analysis of the pool structure assuming that all contained fluid is attached to the pool slab. Uncracked section properties are
.I used here. This frequency analysis is used to determine the lowest pool structural frequencies so as to establish appropriate seismic amplifiers to apply to load cases 1 and 4. These seismic I
I I
8-7 I
I
I I- amplifiers are obtained from the response spectra of the seismic I event and multiply the results of load cases 1 and 4 when forming the mandated load combinations.
As noted above, the case 3 loading involves the determination of an effective pressure load to represent the seismic load on the slab due to the racks plus fuel. The method of determination of this effective pressure is described below.
I As noted previously, the Holtec 3-D dynamic simulation code DYNARACK is used to simulate the seismic response of the entire fuel pool containing multiple racks. The vertical load time history from each pedestal on each rack is saved in an archival file. For the pool slab structural analysis, which is based on static analyses, we compute an effective static load increment based on averaging of the time history. Figure 8.3.1 is used to I illustrate the concept where the total pedestal load is considered as the static load (Fs in Figure 8.3.1 plus a time varying component). Note that in Figure 8.3.1 a zero load during a portion of the time means that the pedestal has lif t.ed of f. We define an effective static load for the purpost-e of pool static analysis and structural qualification as follows:
- a. From the archival pedestal load time history we may, at I each point in time, determine the total pool load FT by stunming the total loads for each pedestal.
I b. At each point in time i, we can define the dynamic load increment for the pool as FT -
Fs = DFi where F3 now represents the total static load on the slab. We keep track of the number of time points i where DFi > 0.
.I c. An equivalent static pool load (seismic adder to the static pool load) is defined as SEISMIC ADDER = SUMDFI /SUMNIi I
'~'
- I .
I I
J where SUMDFI is the sum of all of the non zero DFI and SUMNi is the total number of points in the time history
]
1 where the dynamic pool load increment is greater than Zero.
g d. In forming the appropriate load combinations mandated l for structural integrity checks, the calculated " seismic adder" divided by the pool area, is used as the offective seismi pressure on the slab.
Of all loading conditions mandated in Ref. 8.1.1, the factored loads which apply to this structure and are deemed critical are:
A. 1.4D + 1.9E B. .75 (1.4D + 1.9E + 1.7To)
C. D + E' + To where:
D = Dead load 1 E' = Design Basis Earthquake (DBE)
E = Operating Basis Earthquake (OBE)
To = Steady State Thermal Load The appropriate load cases are formed from the individual finite element analyses as follows:
D = case 1 + case 2 E' = DBE amplifier x case 1 + DBE amplifier x case 4 + case 3 (for DBE) + case 5 (for DBE)
E= OBE amplifier x case 1 + OBE amplifier + case 4 + case 3 (for OBE) + case 5 (for OBE)
To = case 6 Load combinations are formed using absolute values where necessary so as to maximize critical stress resultants.
l 8-9 l
- I il
~
As noted above, for analysis of fuel pool structural integrity, i the seismic amplifiers are based on the peak g level responses at the lowest resonant frequency that are obtained from the plant acceleration response spectrum. We show that this is conservative.
8.4 Results of Analyses I The ANSYs postprocessing capability is used to form the
- appropriate load combinations identified above and to establish the critical bending moments in various sections of the pool structure. The ultimate moments for each section are computed
- using allcwable limit strength levels as described in Ref. 8.1.3.
. For Donald C. Cook, the following limit strengths for concrete and for reinforcement are used in the computation of limit (ultimate) moments.
concrete oc = 3500 psi (compression) reinforcement = oy = 40000 psi (tension / compression)
I In each section, we define the safety margin for bending as the I ultimate bending moment divided by the calculated bending moment (from the ANSYS postprocessing of the required load cases). Table 8.4.1 summarizes the results obtained from the finite element analyses and shows minimum safety margins on each section of the structure. Note th2t these are safety margins based on the factord load conditions as mandated in Ref. 8.1.1 and need only satisfy a limit 2: 1.0.
I I
g 8-10 I
I I
L
.g I The floor slab perimeter is also checked against gross shear i failure under factored load conditions. Local bearing strength and punching shear calculations are performed in accordance with (Ref. 8.1.3).
8.5 Pool Liner I The pool liner is subject to in-plane strains due to movement of the rack support feet during the seismic event. Calculations are made to establish that the liner will not fail due to cyclic straining caused by the rack foot loading. An ANSYS analysis of a I_ liner plate section subjected to vertical and hori:cntal static pedestal loading is carried out. The time history result for the pedestal loading is then used to evaluate the number of stress cycles to be expected in the liner for each event. The cumulative damage factor (CDF) is computed and shown to bc less than 1.0 in critical regions of the liner and attachment locations. The number of stress cycles used in the CDF evaluation is based on 1 I DBE and 20 OBE events.
8.6 conclusions Critical regions affected by loading the fuel pool completely with high density racks are examined for structural -integrity under bending and shearing action. It is determined that adequate I safety f actors exist assuming that all racks are fully loaded with normal (unconsolidated) fuel and that the factored load combinations are checked against the appropriate structural design strengths. It is also shown that local frictional loading on the liner results in in-plane stresses that are low enough so that liner fatigue is not a concern.
8-11 s
I
8.7 References for Section 8
[ 8.1.1 NUREG-0800, SRP for Review of Safety Analysis Reports for Nuclear Power Plants, Section 3.8.4, July 1981.
1 8.1.2 ANSYS User's Manual, Swanson Analysis Rev. 4.3, 1987.
I 8.1.3 ACI 318-89, ACI 318R-89, Building Code Requirements for Reinforced Concrete, Anterican Concrete Institute, Detroit, Michigan.
8.1.4 ACI349.lR-80, Reinforced Concrete Design for Thermal Effects on Nuclear Power Plant Structures, 1981.
I I
I I
I
~
I l
I l
8-12 E
[
Table 8.4.1 SAFETY FACTORS FOR BENDING OF POOL STRUCTURE REGIONS REGION FACTOR OF SAFETY
- Slab 1.23 North Wall 1.00 East Wall 1.08 South Wall 1.26 West Wall 1.05 The factors of safety have been obtained using conservative assumptions on mechanical and thermal load distribution. They represent factors of safety over the values required by NUREG-0800.
8-13
5 I 4 i
4 4
E I
i t
EAST WALL i ,-/i
- f. ./
l NOR
/ l / '
ALL j SouT ALL l /.
f.
i
. 7' .
WEST WALL
,. , e -/
h,' /; .
l
- - . . l l l l . l l
' . . . l
- ; ,
- , . : . ;; ,, ;;(.
.p... . . . .. .. . ...,,,'
l . . ... ..-,' l ! .
I l IV ... ll .. ,
t . ..,.-
.'.- l
~.
. b.'m'.*r.'.'.,yI,.. l . b.,5 . . .?\ \
~
..' i:,-
..'.... ............j:4 .- .,';- ;
- I 3' ll l
L,
[ .
,.- P' 2
- . . .k '
s ,sassa,a .
k *. / ll ,.
/ -
'l ' *n%%d, W I
I -
siouse e.2., iSoseTaic view oe coox seen1 eue eoot 8-14 I
o ..
l I \
\ \ \ \ \ \ \ \ ,-
\
\ \ \ \ \ \ \ $,_Zhi '\ / ,
.i \ \ \ \ \ \_ \ 's l
f.
/ h\ \ t i\\ % \ \ [t ss \ '
\ \+\ ;-n, s\ \!\/\l%
\ x / \f h '/ )
g ',
! +
i i . i tt \/ lLfl\ il l ! / v' / ' AX r>'s/ / ~ ,'(
l
\ \ \' i F i \
\ ; \ir71.\
\\
\\\\ ==;\
\/\/t//'
\(/\/
\. i \ \T- ;s \ \ /
\Y_/ \' '
I \
\
\ \ \ \ r_+ \
\ \ \bi
\\
\\
\\
__- -\
\
\\
\
\ \\ \ } __
\\ -- 4 \
I e
I -
y I
I FIGURE 6.2.2 OVERALL FINITE MODEL OF COOK POOL TOP VIEW t
8-15
1 - ."; 13'25 4. 4 '
I O 1 1 51 m c. . M : 20 L g F? , l . '
.. s
, , ; d 7 lLL .. ' a .w
~'
L --.3
- ,, n-2
~' T *.:
7.2 37 ,
f .,' =,?? .
l \ i \ i \\
i l'/ '! \ L l{
,. ' \ \ i O\ \ \ \ \I
/
_fllr I F l l li ? I i! ! I l
l_ s ' ' l l l ll l l ?' l I l II l I l
'g / l ._ .l j - j fjf, ,
// li i 11 : i i f f I' i Ii l l'% a I it t i a
i I
I FIGURE 8.2.3 OVER ALL FINITE MODEL OF CCCK PCOL BOTTOM VIEU 8-16
I I
'I I e t'
[pf'
.0 sN , .
A4 tw,
, .k N
i N RW x,N tk, i
.g l
L:t ,c,c W,y,q.:e.: 4 k,7.?)q d R. is.fg $
'k'w.f'fi l
i -
iv'.,:@s.k:)ss KN t ..S. , l.
'N M8x.xWSRs N,3 qsM:Rq
,4 , 0 3
- 1S .
s i i
,e r ,' s, i 5 ( il I L i /
I i
/ i i-k l.
?
r J.
It IlP )( l i
ll Uo 1l5 / !.
i"i
'I l
,i i!
V i
. l
' 0 TIME 'r.
I FIGURE 8.3.1 PEDESTAL LOAD VS. TIME (Positive Load Means Pedestal in Contact with Liner) 1 8- 17
, _ _. - - _ _ . _ _ _ _ _ _ _ . . _ _ _ - - - _- - _ - - - - - - - - - - - - - - - - - - - - - - - - - - ~ - ~ ~--
b 9.0 RADIOLOGICAL EVALUATION
!.1 Fuel Handlina Accident 9.1.1 Assumptions and Source Tern Calculations I An evaluation of the consequences of a fuel handling accident has been made for fuel of 5.0 wt% initial enrichment burned to 60,000 MWD /MTU, with the reactor conservatively assumed to have been operating at 3411 MW thermal power (38.8 MWD /KgU specific power) prior to reactor shutdown. Except for the fuel enrichment and discharge burnup, the assumptions used in the evaluation are the same as those previously reviewed and accepted by the USNRC. As in the previous evaluaticn, the fuel handling accident was I conservatively assumed to result in the release of the gaseous fission products contained in the fuel-rod gaps of all the rods in the peak-power fuel assembly at the time of the accident. Gap I inventories of fission products available for release were estimated using both the assumptions identified in Regulatory Guide 1.250) and those in NUREG/CR-SCO9@). NUREG/CR-5009 has confirmed that the Reg Guide 1.25 assumptions remain conservative for extended burnup except for I-131, for which the release, fraction was reported to be 20% higher.
Most of the gaseous fission products having a significant impact on the off-site doses are the short-lived nuclides of Iodine and Xenon which reach saturation inventories during in-core operation.
These inventories depend primarily on the fuel specific power over the few months immediately preceding reactor shutdown. In the highest power assembly, the specific power and hence the inventory of Iodine and Xenon will be directly related to the peaking factor (assumed to be 1.65 per Reg. Guide 1.25).
(
9-1
~
The inventory of long-lived Kr-85 (10.73 year half-life), however, l- is nearly proportional to the accumulated fuel discharge burnup and hence is independent of the peaking factor. Because Kr-85 is a weak beta emitter, it has only a minor impact on off-site doses, primarily affecting the whole-body beta dose. The off-site radiological consequences are dominated by the short-lived radionuclides (which are at saturation concentration independent of fuel burnup). In the present analysis, the calculated doses are higher and more coservative than those of the previous evaluation because (1) the analyses reported here use higher gap inventories based on Reg Guide 1.25 assumptions and (2) the use of the up-dated ORIGEN-2 codeW for calculating the fission product inventories.
t Results of the evaluation confirm that the off-site doses remain t
within the regulatory limits.
The present evaluation uses values for the 2-hour atmospheric y dispersion factor (X/Q) and filter efficiencies that have previously been reviewed and accepted. Core inventories of fission products were estimated with the ORIGEN-2 code based upon a reactor power of 3411 MWt and fuel with an initial enrichment of 5.0% U-235 burned to 60,000 MWD /MTU. Calculations were made for ,
100 hours0.00116 days <br />0.0278 hours <br />1.653439e-4 weeks <br />3.805e-5 months <br /> cooling time as the source term for the fuel handling
[
accident. The release fraction of the core inventories assumed -
P to be in the gap by both the Reg Guide 1.25 and NUREG/CR5009 assumptions are listed in Table 9.1.
The following equation, from Reg Guide 1.25, was used to calculate the thyroid dose (D) from the inhalation of radiciodine, Fg Ii FPBR (x/Q)i m DF p DF g summed over all Iodine radionuclides.
l 9-2 C
l I
F, = fraction of fuel rod B= Breathing" rate =
Iodine inventory in gap 3.47 x lo cubic p space meters per second Ij = core Iodine radio-nu-
\' clide inventory at time R, = Dose conversion of the accident (cu- factor (rads / curie) ries) from Reg. Guide 1.25 F= f r,1ction of uore dam-aged sc . to release (X/Q) = atmospheric Iod! . ~2 in the rod gap diffusion factor 3
(1,'193) ( 3.15 x 10" s e c / m )
P= Core peaking factor (1.65) DF p = effective Iodine decontamination DFt= effective Iodine factor for pool decontamination factor watar (= 150) for filters (= 10)
The gap inventories listed in T;ble 9-1 are the product of I i (core inventory) and F, (the fraction existing in the gap) .
The function used to calculate the external whole body dose from beta (D ) or gamma (Dr) radiation in the cloud uses many of the s
I terms defined above and is given by:
Da = I O . 2 3 (x/Q) FPGi E 3, and Dr = y 0. 2 5 (x/Q) FPG 3 Er i I where G3 is the gap inventory of the gaseous radionuclides of Xe g and K- and the functions above are summed over all the noble g3ses.
W Ey and Er are the average energies of decay ' beta and ganna radiation rsspectively) for the various radionuclide?. These functions assume the noble gas decontamination factors in water and the charcoal filters are 1.C. The gap inventories of radiciodine i
I 9-3 I
B make a negligi' contribution to the whole body deses, D3 or i because of the large decontamination factors appropriate to the iodines.
9.1.2 Results A summary of the assumptions used to evaluate the fuel handling I accident is given in Table 9-2. The minimum time a'lter chutdown when fuel assemblies would be moved was conservatively assumed to be 100 hours0.00116 days <br />0.0278 hours <br />1.653439e-4 weeks <br />3.805e-5 months <br /> as identified in the Technical Specifications. At 100 hours0.00116 days <br />0.0278 hours <br />1.653439e-4 weeks <br />3.805e-5 months <br /> after shutdown, the two-hour dose at the site boundary, for a fuel handling accident releasing all of the gaseous fission product radioactivity in the gaps of all rods in the highest power assembly, are as follows:
Two-Hour Site Boundary Dose NUREG/CR-5009 Reg. Guide Previous I, Method 1.25 Arialysis Inhalation thyroid dose = 7.07 Rads 5.97 Rads 2.15 I Whole body beta dose, D p = 0.36 Rads 0.70 Rads -
Whole body gamma dose, Dr = 0.31 Rads 0.58 Rads 0.51 1 These doses are well within the limits of 10 CFR Part 100 in I conformance with the acceptance criteria of SPP 15.7.4.
July 1981)W.
(Rev.1, I
i -
I I 9-4 I
P L
9.2 Solid Padwaste The neceseity for resin replacement is determined primarily by the requirement for water clarity and the resin is normally changed
{ about once a year. !!o significant increase in the volume of solid radioactive s 'ates is expected with the expanded storage capacity.
g During reracking operations, a certain amount of additional resins B may be generated by the pool cleanup system on a one-time basis (perhaps 10 to 30 cubic feet).
9.3 Gacecus Pe] eases Gascous releases from the fuel storage area of the auxiliary building are combined with other plant exhausts. liormally, the contribution from the fuel storage area of the auxiliary building g is negligible compared to the other releases and no significant B increases are expected as a result of the expanded storage capacity.
9.4 P fq n g n n e l Exposuren 1
During normal operations, personnel working in the fuel storage area may be exposed to radiation from the spent fuel pool.
Operating eXDerience has shown that the area radiation dose rates, g which orig siate primarily from radionuclides in the pool water, are B generally less than 1 mrem /hr but may temporarily increase to 2.5 -
3 mrem /hr during refueling operations. Ito evidence has been observed of any crud deposition around the edges of the pool that might cause local areas of high radiation.
I 1
9-5
/
Radiation levels in zones surrounding the pool are not expected to
( be significantly affected. Existing shielding around the pool (s ater depth and concrete walls) provide more than adequate protec-l tion, despite the slightly closer approach to the walls of the pool.
Typical concentrations of radionuclides in the poal water are shown I in Table 9.3. During fuel reload operations, the concentrations will increase due .to crud deposits spalling from spent fuel assemblies and to activities carried into the pool from the primary system. While these effects may increase the concentrations (as much as a factor of 10), the pool cleanup system soon reduces the concentrations to the normal operating range. No evidence has been 1
seen of any significantly higher radiation deses near the edge of the pool that might suggest the accumulation of crud deposits.
1 i operating experience has shown that there have been negligible concentrations of eirborne radioactivity and no increases are expected as a result of the expanded storage capacity. Area '
monitors for airborne activities are available in the immediate vicinit.y of the spent fuel pool.
No increase in radiation exposure to operating personnel is -
expected and therefore neither the current health physics program nor the area monitoring systems need to be modified.
9.5 Anticipated Exposure Durinq Rerackinq Total occupational exposure for the reracking operation is ,
estimated to be between 6 and 11 person-rer., as indicated in Table 9.4. While individual task efforts and exposures may differ from those in Table 9.4, the total is believed to be a reasonable estimate tor planning purposes. Divers will be necessary to remove 9-6
m certain underwater appurtenances. These appurtenances are well L removed f or the stored fuel which minimizes the radiation dose rate to the divers. Carefu nonitoring and adherence to pre-preparea procedures will assure ' hat the radiation dose to the divers will
{
be maintained ALARA. All of the reracking operation will utilize
~
detailed procedures prepared with full consideration of ALARA principles. Similar operations have been performed in a number of 1 facilities in the past and there is every reason to believe that reracking can be safely and efficiently accomplished at the Donald C. Cock 11uc. le ar Plant, with minimum radiation exposure to personnel.
The existing radi, tion protection program at the Cook Nuclear Plant is adequate fur the reracking operations. Where there is a potential for significant airborne activity, continuous air samplers will be in operation. Personnel wear protective clothing I and, if necessary, respiratory protective equipment. Activities are governed by a Radiation Work Permit and personnel monitoring equipment will be assigned to each individual. As a minimum, this includes thermoluminescent dosimeters and pocket dosimeters.
. Additional personnel monitoring equipment (i.e., extremity badges or alarming dosimeters may be utilized as required. Work, personnel traffic. and the movement of equipment will be monitored and controlled to minimize contamination and to assure that exposures are maintained ALARA.
In reracking, the existing storage racks will be removed, decon-taminated as much as possible by washing and wipe-downs, packaged and shipped to a licensed processing / disposal facility. Shipping containers and procedures will conform to Federal DOT regulations and the requirements of any State DOT office through which the shipment may pass.
9-7
9.6 Peferences for Section 9 o
l L
- 1. Reg. Guide 1.25 (AEC Safety Guide 25), " Assumptions used for evaluating the potential radiological consequences of a fuel
[ handling accident in the fuel handling and storage facility for boiling and pressuri::ed water reactors".
- 2. C. E. Beyer, et al. , " Assessment of the Use of Extended Burnup Fuel in Light Water Power Reactors", NUREG/CR-5009, Pacific Northwest Laboratory (PNL-6258). -
- 3. A.G. Croff, "A User's Manual for the ORIGEN2 Computer Code",
ORNL/TM-7175, July 1980 (ORIGEN =.02NL Icotope Generation and 1 Depletion)
- 4. Section 15.7.4, " Radiological Consequences of Fuel Handling Accidents" NUREG-0800, Section 15.7.4, Rev. 1 July 1981 1
1 I
l l
l I
I I
9-8
6m m EMI m Table 9-1 INVENTORIES AND CONSTANTS OF SIGNiFICANT FISSION PRODUCT RAD!0NUCLIDES L
TOTAL GAP iflVENIORY, CURIES __
.6 SHUiOOwn DECAY NUREG/CR-5009 Reg. Guide 1.25 DOSE CCNVERSION E (uEV) E (uEV)
NUCLIDE CONST.
INVENTORY 100 hrs 100 hrs Ri CURIES A.1/Nas 7.5 E + 6 6.3 E+ 6 f.48E+E 0.186 0.389 1-131 9.0 E+7 3.591 E-3 Negligible
- Negligible 5.35E + 4 I-132 1.3 E+ 8 3.013E-1 6.3 E+ 5
- 6.3 E+5 A0E+5 0.419 0.597 I-133 1.8 E+ 8 3.332E-2 Negligible Negligible 2.5 E + 4 -
l-134 1.9 E+ 8 7.905E-1 Y Negligible
+
Negligible 1.24E+5 0.394 1.456 e 1-135 1.7 E+8 1.048E-1 Kr-85M 1.547E-1 Negligible
- Negligib e 1.9 E+7 2.0 E+ 5 4.2 E + 5 0.251 0.002 Kr-85 1.4 E+6 7.376E-6 _
Kr-87 5.451 E-1 Negligible Neg!igible 3.6 E+7 5.0 E+7 2.4 42E-1 Negligible lleg!igible Kr-88 l
7.9 E + 4 l - 0.163 Xe-131M 1.0 E + 6 2.427E-3 7.9 E+ 4
- 0.233 5.6 E+ 6 1.319E-2 1.5 E+5 I .5 E + 5 xe-1.stu S.50GE-3 5.1 E+ 6 1.0 E f 7 0.102 0.081 X*-133 1.8 E+ 8 Negligible 0.309 0.262 Xe-135 3.9 E+ 7 7.626E-2 Negligible NO RELEASE ERAC110N GIVEN - ASSUMED SAME AS REG. GutDE 1.25
7 Table 9.2
+ DATA AND ASSUMPTIO!!S FOR THE EVALUATIO!!
OF THE FUEL HANDLING ACCIDE!1T
_s 1. Source Term Assuretions VALUES Core power level, MWT 3411 Fuel burnup, MWD /MTU 60,000 Analytical method ORIGEN
- 2. Release Assumntions Number of failed fuel all rods in 1 rods of 193 assemblies Fraction of core Pea. Guide 1.25 inventory released to gap (NUREG/CR-5009 % % of the Iodine - 10 i release of Iodine-131 is reported to be 20%
higher)
% of the Xenon
% of Kr-85 10 30 Assumed power peaking 1.65 factor i Inventory in gip available for release Table 9.1 i Pool decontamination factors For Iodines 150 For noble gases 1 Filter decontamination factors For Iodines 10 For noble gases 1 Atmospheric Dispersion, 3.15 x 10" sec/m 3 (x/Q)
Breathing rate 3 3 . 4 7 x 10" m / s ec 9-10 l
s I
1 i
l Table 9.3 Typical Concentrations of Radionuclides in the Spent Fuel Pool Water Concentration Fuclide ag;/ini Ag-110M 4.6 x 10'5 Co-58 1.5 x 10'3 4.4 x 105 i Co-60 Cs-134 3.2 x 10" Cs-137 6.4 x 10" l
I I
l 9-11 M
i 4
1 1
I i
L Table 9.4 PRELIMINARY ESTIMATE OF PERSON-REM EXPOSURES DURING RERACKING Number of Estimated Etan Egrsonnel Hours Exposure W
{
Remove empty racks 5 40 0.5 to 1.0 Wash and Decon racks 3 10 0.00 to 0.2 Clean and Vacuum Pool 3 25 0.3 to 0.6 Remove underwater 4 5 0.4 to 0.8 appurtences Partial installation 5 20 0.25 to 0.5 of new rack modules
)
Move fuel to new racks 2 150 0.8 to 1.5 Remove remaining racks 5 120 1.5 to 3.0 Wash and Decen racks 3 30 0.2 to 0.4 Install remaining new 5 35 0.4 to 0.8 rack modules Prepare old racks for 4 80 1.0 to 2.OG) shipment Total Exposure, person-rem 6 to 12
- 0) Assumes minimur dose rate of 2 1/2 mR/hr (expected) to a maximum of 5 mR/nr, except for pool vacuuming operations which assumes 4 to 8 mR/hr and diving operations which assume 20 to 40 mR/hr.
G) Maximum expected exposure, although details of preparation and packaging of old racks for shipment have not yet teen deter-mined.
9-12
I '
g 10.0 11L-SERVICE SURVEILLANCE PRO _QBAM I 10.1 Purnose I This section describes the programmatic commita nts made by I
Indiana Michigan Power Company (I&M) for in-service surveillance of the Boral neutron absorption material to comply with the previsions of Section IV (8) of the OT Position Paper (Ref.
10.1.1).
All material used within a storage system for spent nuclear fuel are qualified to a level of performance predicated upon calculated
,cors t case environmental conditions and are based on accelersted I testing of the materials to levels of service life corresponding to that environment. Because such environmental compatibility testing in the laboratory conditions is accelerated, it is prudent that each of the system components be monitored to some extent throughout the service life to assure that the actual in-service performance remains within acceptable parameters as de2ined by the I accelerated testing.
throughout the service life is relatively easy, however, the For many of the materials, monitoring neutron absorbing material is encased in a stainless steel jacket precluding a direct visual or physical examination during the in-service condition.
I The coupon surveillance program presented herein is intended to I provide a definitive assessment of the present physical integrity of the neutron absorber, as well as inferential informaticn to detect future degradation.
I 10-1 I ,
I
I The coupon surveillance procedure consirts of preparing twelve neutron absorber coupons carefully encased in a stainless steel metal jacket, and suspending them from a " coupon tree".
The coupon tree is placed in the conter of a group of freshly discharged fuel assemblies each time a new batch is discharged to I the pool. The group of assemblies surrcunding the coupon tree shall be thot. e which have the above-average values of radial peaking factor. The object, cf course, is to subject this " tree" to the maximum radiation exposare in the fuel pool in the minimum amount of time.
I Further details are provided in the following.
I 10.2 Counon Surveillance 10.2.1 Descriotion of Test Counons The neutron absorber used in the surveillance program shall be representative of the material used within the storage system. It shall be of the same composition, produced by the same method, and certified to the same criteria as the production lot neutron abscrber. The sample coupon shall be the same thickness as the neutron absorber used within the storage system and shall reet the referenced Holtec drawing dimensional requirements. Each neutron I absorber specimen shall be encased in a stainless steel jacket of an alloy identical to that used in the storage system, f m ed so as to encase the neutron absorbing meterial and fix it in a position and with tolerances similar to that for the storage racks. The jacket would be similar to that for the storage racks.
I 10-2 i
I I
I I The jacket would be closed by quick disconnect clamps or screws I with lock nuts in such a manner as to retain its form throughout the use period and also allow rapid and easy opening without contributing mechanical damage to the neutron absorber specimen contained therein.
Consistent with the USNRC OT Position Paper (reference 10.1.1),
requirements of a statistically acceptsble sample si::c, a total of twelvo jacketed neutron absorber specimens, shall be used.
10.2.2 Benchmark Data The following benchmark tests shall be perf or.acd on test coupons derived from the same production run as the actual nectron absorber panels.
(i) Length, width, thickness and weight measurements (11) Wet chemistry Neutron attenuation measurement (optional)
I (iii) 10.2.3 Couoon Reference Data I
Prior to encasing the coupons, each coupon shall be carefully calibrated. Their width, thickness, length end weight shall be carefully measured ... recorded. The wet chemistry will be performed on a strip taken from the same Boral plates from which the coupons are made to provide a benchmark B-10 loading data.
p
.s Three points on each coupon will be designated for neutron attenuation measurement. Neutron attenuation measurements at those three points will be made and recorded.
I 10-3 I
l i
10.2.4 ltecelerated Survei11ang_g At the time of the,first off-load of spent fuel, the coupon tree is surrounded by storage cells containing fuel assemblies from the
(
peak power region of the reactor core. At the time of the second off-load of the fuel assemblies, the tree is withdrawn from the fuel pool and one coupon is taken for evaluation. The specimen strip is replaced in the fuel pool in a new location, where it is again surrounded by peak power region fuel assemblies. The storage cell that was vacated may now be used to store a fuel assembly. This arrangement is repeated at the first two off-leads of fuel and after that, every third outage. By evaluation of he specimens, an accelerated monitor of environmental effect, on the neutron absorber will be obtained.
10.2.5 Post-Irradiation Tests
- Coupons removed from the pool will be tested for dimensional, neutren attenuacion, and wet chemistry changes using the same procedures which were used in initial benchmarking to minimize the pctential for instrument errors.
10.2.6 Accepta_nce Criteria A plant procedure will be developed to execute the commitments made in this licensing submittal. Equipment requirements, step-by-step instructions for executing inspections and acceptance criteria will be described in that procedure for use by plant personnel.
10-4 I d
L 10.3 Refereng,es for SecticA.M.
( 10.1.1 OT Position for lieview and Acceptance or Spent Fuel 5torage and !!andling Applications", by Brian K. Grimes, USNRC, April 14, 1978, and Revisien dated January 18, 1979.
{
O C
- 9
[
I i
1 1
I i
10-5
I l
11.0 ENVIRONMENTAL COST / BENEFIT _bSSESSMENT 11.1 Introductinn The specific need to increase the existing storage capacity of the spent fuel pool at the Donald C. Cook Nuclear Plant is based on the continually-increasing inventory I
in the pool, the prudent requirement to maintain full-core offload capability, and a lack of viable economic alternatives.
I f The inventory increase can be inferred from tne fuel assembly discharge schedule contained in Table 11.1.
The proposed project contemplates the reracking of spent fuel pool I with free-standing, high density, poisoned spent fuel racks. The engineering design and licensing will be completed for a full reracking of the pool, which is currently only partially racked.
Engineering and design will also be enmpleted to accommodate c)nsolidated ruel. The licensing effort for consolidated fuel will, however, be pursued at a later date if consolidation is chosen to accommodate future storage needs.
II 11.2 Proiect Cost Asset:sagnt The total capital cost for the rerack project is estimated to be approximately $14.1 million.
r Many alternatives were considered prior to proceeding with reracking, which is not the only technical option available to increase on-site storage capacity. Rcrac.cing does, however, enjoy a cost advantage over other technologies, as shown:
( 11-1
1 F
Capital Costs L Tvoe of Storace S/Kc0 Rerack $ 2 0(I)
Fuel consolidation $20 - 34(2)
~
Dry cask storage $4 5 - 110(2)
Storage vault $4 0 - 9 0(2)
New pool $115(3)
I There are no acceptable alternatives to develop off-site spent fuel storage capacity for the Cook Nuclear Plant. First, there are no commercial independent spent fuel storage facilities operating in the U.S. Second, the adoption of the Nuclear Waste Policy Act (NWPA) created a de f acto throw-away nuclear fuel cycle. Since the cost of spent fuel reprocessing is not offset by the salvage value I of the residual uranium, reprocessing represents an added cost for the nuclear fuel cycle which already includes the NWPA Nuclear Waste Fund fees. In any event, there are no domestic reprocessing facilities. Third, I&M has no other operating power plant; therefore, shipment of spent fue,1 from the Cook Nuclear Plant to other system nuclear power plants is not possible. Fourth, at
$600,000 per day replacement power cost, shutting down the Cook Nuclear Plant is many times more expensive than simply reracking the existing spent fuel pools.
I
February 1989 (3)
Actual estimated cost per KgU of storage space gained for this project i
11-2 f - _
11.3 Resource Crnmitment L
The expansion of the spent fuel pool capacity is expected to require the following primary resources:
Stainless steel 360 tons.
Boral lieutron Absorbar 30 tons, of which 30 tons are Boron Carbide Powder and 20 tons are aluminum.
I The requirements for stainless steel and aluminum represent a small fraction of total world output of tnese metals (less than .0001%).
Although the fraction of world production of Boron Carbide required for the f abrication is somewhat higher than that of stainless steel or aluminum, it is unlikely that the commitment of Boron Carbide I to this project will affect other alternatives. Experience has shown that the prtduction of Boron Carbide is highly variable and depends upon need, and can easily be expanded to accommodate worldwide needs.
11.4 Environment Assessment Due to the additional heat-load arising from increased spent fuel pool inventory, the anticipated maximum bulk pool temperature increases from a previously-licensed 140*F to app:.oximately 160 F, as detailed in the calculations described in Section 5.0 of this report. The resultant total heat-load (worst case) is 35.5 million BTU /HR, which is less than 0.5% of the total plant heat .'.oss to the environment.
The not result of the increased heat loss and water vapor emission (due to increased evaporation) to the environment is negligible.
11-3
w Table 11.1 ,
DONALD C. COOK NUCLEAR PLANT WORST CASE SPENT FUEL IlWENTORY i
)
ASSEMBLIES l MR IN STORAGE 1991 1362
-l 1992 1518 1993 1678 1 1904 1838 1995 1918 Lose full core discharge capability with current capacity 1996 1998 l 1997 2158 Lose normal dircharge capability with current capacity 1998 2318 1 1999 2318 2000 2478 2001 2638
, 2002 2798 2003 2798 l 2004 2958 2005 3118 2006 3198 I 2007 3278 2008 3438 Lose full core discharge capability with proposed rerack 2009 3598 2010 3678 Lose normal discharge capability with proposed rerack 2011 3758 l 2012 3918 2013 4078 l 2014 4158 2015 4351 2016 4431 2017 4,624 l
l 11-4
. ..