ML20238C370

Forwards Proposed Revs to FSAR Re Spent Fuel Racks to Be Added to West Pool & Summary Rept on Rack Design & Layout & Seismic & Criticality Analyses for NRC Review.Installation Will Begin in Apr 1988

ML20238C370
Person / Time
Site:	Vogtle
Issue date:	12/23/1987
From:	Bailey J GEORGIA POWER CO.
To:	NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM)
References
GN-1422, NUDOCS 8712300231
Download: ML20238C370 (145)

Text

{{#Wiki_filter:_ .f Georgia Fower Company Fost Office Box 282 Waynesboro, Georgia 30830 j Telephone 404 554-9961 404 724-8114 Southern Company Services. Inc. Post Office Box 2625 Birmingham, Alabama 35202 l Telephone 205 870-6011 VOgtie Project I i December 23, 1987 1 U. S. Nuclear Regulatory Commission File: X7BC35 ATTN: Document Control Desk Log: GN-1422 Washington, D. C. 20555 NRC DOCKET NUMBER 50-425 CONSTRUCTION PERMIT NUMBER CPPR-109-V0GTLE ELECTRIC GENERATING PLANT UNIT 2 V0GTLE SPENT FUEL RACKS Gentlemen: The Vogtle Electric Generating Plant FSAR describes two ' spent fuel storage racks with storage capacity for 288 spent fuel assemblies located in the east (Unit I side) spent fuel pool. Georgia Power Company is now providing a description of the spent fuel storage racks to be provided in the west (Unit 2 side) spent fuel pool. When completely installed, these racks.will provide additional storage capacity for approximately 2098 spent fuel assemblies and will serve both units. No changes are being made with respect to Unit 1. The two spent fuel racks that were previously licensed with Unit I will remain, and be used for temporary storage of spent ' fuel discharged from Unit 1 prior to its transfer to the additional storage racks being provided in the west pool. The new racks will be free standing, high density, honeycomb design using boraflex as a neutron absorber. The method of analysis uses a time-history integration method similar to that previously used in licensing reports for high density spent. fuel racks used at other plants. The seismic analysis has been performed for various amounts of fuel, symmetrically loaded into =j the racks with the pool dry and with the pool flooded. Additional seismic analysis for non-symmetrically loaded racks. in the dry condition are also included. The racks may be used for storage of new fuel in the dry or flooded condition and. spent fuel in the flooded condition. The sepa' ration of. the racks will be such that racks containing nuclear' fuel 'will not impact other racks or the sides. of the pool as a result of an earthquake. - For the purpose of dry storage of the new fuel for Unit 2,. racks may be. placed in a temporary location with an increased separation distance such j that they would not impact each other, or the pool wall ~, in the-event of an earthquake. After initial fuel loading, they would be moved to their final location as shown in the FSAR. \\ 8712300231 871223 g\\ PDR ADOCK 05000425 A PDR E

File: X7BC35 Log: GN-1422 December 23, 1987 Attachment A to this letter provides the ' proposed : revisions-to the FSAR to describe the spent - fuel racks. to be,added into the west pool. The revision is in the. form of a mark-up of the current FSAR pages in _ order to clearly indicate the changes that will be made in order to describe the racks to be added for the west pool. Installation will begin in April 1988 and a sufficient number 'of racks will be available by October, 1988 to store the new fuel for Unit 2. The remainder of the racks are planned to be installed after the fuel has been loaded. into the reactor but before spent - fuel is placed into the west pool. Attachment B provides a summary report on the rack design and layout, seismic analyses, and criticality analyses. These reports contain more detailed information to aid. performance of the NRC's review. Georgia Power requests that the NRC schedule the review of these spent fuel racks as soon as ;possible. The. current schedule shows that new fuel will arrive on site in October 1988. If the. NRC determines.that additional information is needed for completion. of the review, Georgia ' Power will be pleased to organize 'a. meeting at which additional technical information can be provided. Sincerely, .h. J. A. Bailey Project Licensing Manager JAB /wkl Attachments xc: NRC Regional Administrator l NRC Resident Inspector J. P. O'Reilly P. D. Rice L. T. Gucwa R. A. Thomas B. W. Churchill, Esquire J. E. Joiner, Esquire J. Hopkins (2) G. Bockhold R. Goddard, Esquire R. W. McManus Vogtle Project File

e --, 9 l l l l l ATTACHMENT A PROPOSED FSAR CHANGED PAGES 1 I I \\

6 5 1 2 s b t a n )e e km t m o ( o N C r - o nl t oaa I l l )irtn I i y ng V V l Jvei (nms E eD d ye tt )ea Y N N N N N N Y Y N N N N N Y Y iflae (SR ts )i Y Y N N N N N Y Y N N N N N Y Y ) hL 7 ( - Q 9 l - e .acd P O ipuo F I rC N M )ct gnsn g g g g g g g l l g g g g g g g 1 (i no f f f f f f f i l f r f f r f r roi o a m m m m a t I m a a m o a a 9 PCg r T dso E ndt H )ar a E an fsdg 5 6 6 6 6 6 6

• )

5 6 6 6 6 6 5 5 eni S (das ( ot e CSQ 1 y cr ig 2 m2 ) es1 1 2 2 2 2 2 2 1 1 2 2 2 2 2 1 1 (il 2 ea SC 3 y Pts )Ges b E dtf a 0 6 6 6 6 6 6 0 0 6 6 6 6 6 0 0 L (Val SC B A yt t A A A A A A A A A A A A A A A A T )i u N N N N N N N N N N N N N N N N cl o (ar uG Q fL o B e1 W W W W W W W W W W W W W W W W ) bcgrg (uu o H_ S 2 t ni 8 B B ion F C C C C C C C C C C C C C F F U )t aa1 (cot li n B B B U f C C C C C C C C C C C C C F F M E g n s I m Y ne i d rd ly r S t e g es S eg l gk g o e on et b n .n l o e e tt pa u g luc n r l ta gr m o il l t gi o ade de di d di datdao meei sn G sr f a prel el et etes ersekp rsrs srd - ug e ye N o e r tbytb tf tfts t e tcp asan lagaefn n Sn I dt reoneeaal aa ai aiai aptaaus al t onnhn i e r o L ns ntol grcbrc rl rl rmdrorrpsrl p D a sAisibag mg g g g l g og aata tt ri itl lam N lkMh t mredeedyed ededeedpeddlisidscetlhnd po A wecCcCaiot ast aat ag t at ait apt aul diduiatfecen iC H euaI aCthtnesnernei nenehneunetoaoatoeniuapa c NfrSmRsT sI has htIhr IhIhsIhsIhscRhRshRilT mSh nd L U 0 1 s 5 6 in E 2 3 li 1 1 r a_ ( 1 2 3 84 5 6 7 8. 9 1 1 1 1 P .3G.

nDG, i

VEGP-FSAR-4 4.3.2.6 Criticality of the Reactor During Refueling The basis for maintaining the reactor subcritical during refueling is presented in paragraph 4.3.1.5, and a discussion of how control requirements are met is given in paragraphs 4.3.2.C and 4.3.2.5. 1 Criticality of fuel assemblies outside the reactor is precluded by adequate design of fuel transfer, shipping, and storage I facilities and by administrative control procedures. The two principal methods of preventing criticality are limiting the fuel assembly array size and limiting assembly interaction by fixing the minimum separation between assemblies and/or inserting neutron poisons between assemblies. The design basis for preventing criticality outside the reactor is that, considering possible variations, there is a 95-percent probability at a-95-percent confidence level that the effective multiplication factor (kef f) of the fuel assembly array will be less than 0.95 as recommended in ANSI N210-1976. The following conditions are assumed in meeting this design bases: --W wEO ) fe+ + ke. cast Sped +w \\ poet eock4 ThefuelassemblycontainstnehighestenrichmentoD[",+ g\\ A. 4.3 weight percent 3U-235 without any control rods or go any noncontained BP and is at its most reactive point spen + el. in life. Po*l eacks and 3. r B. For flooded conditions, the moderator is pure water at "*'akt the temperature within the design limits which yields P'"*'d EF the largest reactivity. +kt ht") Mud rack s C. The array is either infinite in lateral extent or is surrounded by a conservatively chosen reflector, whichever is appropriate for the design. D. Mechanical uncertainties are treated either by using worst-case conditions or by performing sensitivity studies and obtaining appropriate uncertainties. E. Credit is taken for the neutron absorption in structural materials and in solid materials added specifically for neutron absorption. F. Where borated water is present, credit for the dissolved boron is not taken except under postulated accident conditions, where the double-contingency principle of ANSI N16.1-1975 is applied. This principle states that it shall require at least two unlikely, independent, and concurrent events to produce a criticality accident. 4.3-38 Amend. 25 9/86

VEGP-FSAR-4 For fuel storage application, water is usually present. However, the design methodology also prevents accidental criticality when fuel assemblies are stored in the dry condition. For this case possible sources of moderation such as those that could arise during firefighting operations are included in the analysis. The design basis k is 0.98 as eff 9.3. 2.(o,l CN+ical% besi s Nkog oph Ac beh Eco+ b tkeMe# b^+ FLd bd recommended in ANSI N216-1976. The design method which ensures the criticality safety of fuel assemblies outside the reactor uses the AMPX system of codes'2228' for cross-section generation and KENO _IV(14) for reactivity determination. gg4 g gy The 218 energy group cross-section library'22' that is the common starting point for all cross-sections'has been generated I from ENDF-B-IV data. The NITAWL program <ta) includes in this l library the self-shielded resonance cross-sections appropriate for a particular geometry. The Nordheim Integral Treatment is used. Energy and spatial weighting of cross-sections is performed.by the XSDRNPM program,'13) which is a one-dimensional S transport theory code. These multigroup N cross-section sets are then used as input to KENO IV,524) which is a three-dimensional Monte Carlo theory program designed for reactivity calculations. l A set of 27 critical experiments has been analyzed using the above method to demonstrate its applicability to criticality analysis and to establish the method bias and variability. The experiments range from water-moderated oxide fuel arrays separated by various materials that simulate light-water reactor (LWR) fuel shipping and storage conditions (15)(25) to dry, harder spectrum uranium metal cylinder arrays with various interspersed materials'17' that demonstrate the wide range of applicability of the method. Some descriptive facts about each of the 27 benchmark critical experiments are given in table 4.3-4. The average k,fg of the benchmarks is 0.9998, which demonstrates that there is virtually no bias associated with the method. The standard deviation of the k values is 0.0014 ok. The 95/95 525 one-sidedtolerancekimitfactorfor27valuesis2.26. There ef is thus a 95-percent probability with a 95-percent confidence level that the uncertainty in reactivity due to the method is not greater than 0.0032 ak. 125 The total uncertainty-(TU) to be added to a criticality calculation is: S ~7.-. TU = (ks)2 (ks)2 +II (ks)2 method KENO mech. 25 4.3-39 Amend. 25 9/86

VEGP-FSAR-4 where: (ks) 0.0032 as discussed above. l25 d (ks) = the statistical uncertainty associated with KENO the particular KENO calculation being used. (ks) = a series of statistical uncertainties mech associated with mechanical tolerances, such as thicknesses and spacings. If worst-case assumptions are used for tolerances, this term 2 will be zero. The criticality design criteria are met when the calculated effective multiplication factor plus the TU is less than 0.95 or, in the special case defined above, 0.98. These methods conform with ANSI N18.2-1973, Nuclear Safety Criteria for the Design of Stationary Pressurized Water Reactor Plants,.Section 5.7, Fuel Handling System; ANSI N210-1976, Design objectives for LWR Spent Fuel Storage Facilities at Nuclear Power Stations, Section 5.1.12; ANSI N16.9-1975, Validation of Calculational Methods for Nuclear Criticality safety; NRC Standard Review Plan, Section 9.1.2, Spent Fuel Storage; and the NRC guidance, Review and Acceptance of Spent Fuel Storage and Handling Applications. 2N16lt.T 4. 3. 2, 6,,1 4.3.2.7 Stability 4.3.2.7.1 Introduction The stability of the PWR cores against xenon-induced spatial oscillations and the control of such transients are discussed extensively in references 6, 18, 19, and 20. A summary of these reports is given in the following discussion, and the design bases are given in paragraph 4.3.1.6. In a large reactor core, xenon-induced oscillations can take place with no corresponding change in the total power of the core. The oscillation may be caused by a power shift in the core which occurs rapidly by comparision with the xenon-iodine time constants. Such a power shift occurs in the axial direction when a plant load change is made by control rod motion and results in a change in the moderator density and fuel temperature distributions. Such a power shift could occur j in the diametral plane of the core as a result of abnormal control action. l l 4.3-40 Amend. 25 9/86 I I

4.3.2.6.2 - Criticality Design Method Outside Reactor for the West Spent Fuel Pool Racks The principal calculational method which provides assurance of the criticality safety of fuel assemblies in the west spent fuel pool racks is the AMPX-KEN 0 code package, using the 27-group SCALE cross-section library (Standardized Computer Analysis f9rlicensing Evaluation). The AMPX calculation used l the NITAWL program L 57 to calcuTate U-238 self-shielding with the Nordheim) integral treatment and to compile cross-sections for input into KENO-IV (14 which is a three-dimensional Monte Carlo program for reactivity determination. Calculations with the transport theory program, CASM0-2E are used for independent verification of the AMPX-KEN 0 calculation and for the sensitivity studies of mechanical tolerances. Benchmark calculations on critical experiments with configurations similar to that of the racks were used to establish the calculational bias and uncertainty of +0.0106 +0.0048 6K (95% probabilty at a 95% confidence level) which is in good agreement with the values reported by the Oak Ridge National Laboratory. The total uncertainty to be added to the criticality calculation is calculated using_the same equation as given in 4.3.2.6.1, with the exception j that (KS) method is 0.0048. The design criteria and methods of criticality safety analysis are in conformance with the General Design Criterion 62 - Prevention of Criticality in Fuel Storage and Handling; NRC Letter of April 14,1978 - 0T Position l for Review and Acceptance of Spent Fuel Storage and Handling Applications; USNRC Standard Review Plan,'NUREG-0800, Sections 9.1.1 (New Fuel Storage) and 9.1.2 (Spent Fuel Storage); ANSI ANS-8.17-1984, Criticality Safety Criteria for the Handling, Storage and Transportation of LWR Fuel Outside Reactors; Regulatory Guide 3.41, Validation of Calculational Methods for Nuclear Criticality Safety; Regulatory Guide 1.13, Spent Fuel Storage Design Basis (Proposed Rev. 2), December 1981, ANSI N210-1976, Design Objectives for Light Water Reactor Spent Fuel Storage Facilities at Nuclear Power Plants.

VECP-FSAR-9 9.1.2 SPENT FUEL STORAGE 9.1.2.1 Design Bases ((( h tkt. east JAed beI Pool g spent fuel is stored in hi h density racks. Each rack consists l2 A of several cells fastened together through top and bottom supporting grid structures Aufhese modules are free-standing, g neither anchored to 'k % $**a center-to-center spacing of h he.ta cf the floor nor braced to the 22 vall. The rack area *y A 10.6-in. as shown in figure 9.1.2-28. The spent fuel storage racks include storage locations for 288 fuel assemblies in the ensi "u2 u I pool;r-Thcre arc no sterage reck; in the Unit-2 pcci at '~this time. Spent fuel pool cooling is discussed in subsection 2 Jees4 9.2.2, fuel handling building ventilation in subsection 9.4.2, and fuel handling building fire protection in appendix 9A, fire area 1-AB-LD-B. 9.1.2.2 Facilities Description The spent fuel storage facility is designed to the guidelines of ANS 57.2. The spent fuel storage facility is located within the Seismic Category 1 fuel handling building. The facility is protected from the effects of natural phenomena such as earthquakes, winds, tornadoes, floods, and external missiles. l3 The facility is designed to maintain its structural integrity following a safe shutdown earthquake and to perform its intended function following a postulated hazard such as fire, internal missiles, or pipe break. Each unit is provided with its own spent fuel pool. The units share a common cask loading and-washdown area.

4 The spent fuel pool provides storage space for rradiated. spent fuel.

Each nuclear unit has a separate pool. The pool is approximately 41 ft constructed of rei forced concrete, and lined with 1/4- ./ ick stainless steek. The normal water l3 volume of the pool out 447,030 gal of borated water with a nominal boron concentration of 2000 ppm. Figures 1.2.2-18 through 1.2.2-20 show the spent fuel pools and the cask loading area. The spent fuel racks are vertical modules designed to hold Westinghouse 17-by-17 fuel assemblies in(arrays. 12 b-, 12. I 22 wrious tw-+ C3 l3 Contiguous to each spent fuel pool is a short canal leading to j the fuel transfer canal. The fuel transfer canal is connected to the refueling canal inside the containment by the fuel transfer tube. All portions of the spent fuel transfer operation are completed underwater, and the waterways are of Amend. 3 1/84 Amend. 22 2/86 9.1.2-1 Amend. 25 9/86

Insert A while each rack in the west spent fuel pool consists of an assemblage of cells interconnected to each other along their contiguous corners to produce a honeycomb cellular structure. Insert B In the west pool, rack arrays have a center-to-center spacing of 10.40 inches in the east-west direction and 10.58 inches in the north-south direction (Fig. 9.1.2-2B). There are a total of 2098 storage locations I in the west pool. Insert C { A total of 2386 storage locations will be provided. The west pool will initially contain at least 2 racks with storage space for 198 fuel assemblies. The design allows the addition, during or before plant operation of any number of racks up to a total of 20 racks in the west pool as shown in Fig. 9.1.2-6. l s L__________

~ [s,p" c3 e %/ VEGP-FSAR-9 /g kgSes sufficient depth to maintain a minimum of 10 ft of shielding water above the spent fuel assemblies. A metal gate with gasket assembly separates the short spent fuel pool canal from the fuel transfer canal. This allows the transfer canal to be drained without interfering with the water level in the fuel pool. Subsection 9.1.3 further addresses the minimum water level in the spent fuel pool. Common to the spent fuel pools and accessible by small canals is an approximately 47-ft-deep cask loading pit. The canals are separated from the spent fuel pools by metal gates with gasket assemblies. The cask washdown enclosure is an epoxy-coated internal structure of the fuel handling building provided for decontamination of the shipping cask before it leaves the VEGP. The spent fuel bridge crane traverses the spent fuel pools and the new fuel storage facility. It is used in the movement of both new and spent fuel assemblies. This crane also has access to the adjoining canals. The cask handling bridge crane traverses the auxiliary building and a portion of the fuel handling building. The cask handling crane's path is perpendicular to the path of the spent fuel bridge crane and is designed such that the cask crane cannot pass over the spent fuel pools. This precludes the movement of heavy loads (other than those associated with the spent fuel bridge crane) over the spent fuel pools, in accordance with Regulatory Guide 1.13. The cask handling crane is used for operations involving the spent fuel shipping cask. During fuel handling operations, a controlled and monitored ventilation system removes gaseous radioactivity from the atmosphere above the spent fuel pools and processes it before discharge through the plant vent. Refer to subsection 9.4.2 for a detailed discussion of the fuel handling building heating, ventilation, and air-conditioning system and section 11.5 for process radiation monitoring. The spent fuel pool is provided with a Seismic Category 1 backup water supply. Water can be either pumped or gravity-fed to the pool from the reactor makeup water storage tank. The reactor makeup water pumps are nonsafety-related Seismic Category 1 pumps which can be aligned to the emergency non-1E buses. In the event that the pumps fail to function, water can flow through the nonfunctioning pumps to provide makeup to the pool. All intervening piping is designed to Seismic Category 1 requirements. 9.1.2-2

l ~23, Y,o i VEGP-FSAR-9 k O g h /a$ 9.1.2.2.1 Spent Fuel Rack Design r. E A. Applicable Codes, Standards, and Specifications The racks are designed and fabricated to applicable portions of the following Nuclear Regulatory Commission (NRC) Regulatory Guides, Standard Review Plan Sections, and published standards. 1. April 14, 1978, NRC Position for Review and Acceptance of Spent Fuel Storage and Handling Applications, as amended by the NRC letter dated January 18, 1979. 2. NRC Regulatory Guides 1.13, Rev. 2 Spent Fuel Storage Facility Design Basis Dec. 1981 (Draft) 1.25, Assumptions Used'for Evaluating March 1972 the Potential Radiological Consequences of a Fuel Handling Accident in the Fuel Handling 2_- and Storage Facility for Boiling and Pressurized Water Reactors 1.26, Rev. 3, Quality Group Classifications Feb. 1976 and Standards for Water Steam and Radioactive Waste Containing Components of Nuclear Power Plants 1.29, Rev. 3, Seismic Design Classification Sept. 1978 1.92, Rev. 1, Combining Modal Responses and Feb. 1976 Spatial Components in Seismic Response Analysis 1.124, Rev. 1, Service Limits and Load Jan. 1978 Combinations for Class 1 Linear-Type Cornponent Supports 3. Standard Review Plan - NUREG-0800 Rev. 1, July 1981 Section 3.7, Seismic Design Rev. 1, July 1981 Section 3.8.4, other Seismic Category I Structures 9.1.2-2a Amend. 25 9/86 I, ___________J

VEGP-FSAR-9 Rev. 3, July 1981 Section 9.1.2, Spent Fuel Storage Rev. 1, Section 9.1.3, Spent Fuel Pool July 1981 Cooling System NRC Branch ASB 9-2, Residual Decay Energy Technical Position for Light Water Rectors for Rev. 2, July 1981 Long Te'rm Cooling 4. Industry Codes and Standards ANSI N16.1-75 Nuclear Criticality Safety in Operations with Fissionable Materials Outside Reactors ANSI N16.9-75 Validation of Calculational Methods for Nuclear Criticality Safety ANSI N210-76 Design Objectives for Light Water Reactor Spent Fuel Storage Facilities at Nuclear Power Stations ASME Section III-80 Nuclear Power Plant Components (through Summer 1991 Ad dendum foo 4bt east pool eacks l MO2 Addendum) o,WA f\\ S BNG Sec+ ion III - 83 theoug k km meo 19 81+ /\\ ad e.nd% for +he. west pool eacks ) ACI 318-63 Building Code Requirements for Reinforced Concrete B. Seismic and Impact Loads The spent fuel racks are designed using the seismic loading described in this section. Seismic analysis of the fuel storage racks is performed by the time-history method. The time histories and i response spectrum utilized in these analyses represent i the responses of the pool structure to the specified ground motion. The seismic analysis of the racks is j i performed with a damping value of 4 percent for safe shutdown earthquake (SSE) and 2 percent for operating basic earthquake (OBE). l j l Maximum dynamic forces and stresses are calculated for l the worst condition as determined by combination with forces and stresses computed in accordance with paragraph C. 1 9.1.2-2b Amend. 25 9/86 1 l l b

VEGP-FSAR-9 Deflections or movements of racks under earthquake loading are limited by design such that the racks do not touch each other or the spent fuel pool walls, the racks are not damaged to the extent that nuclear parameters are given in paragraph 9.1.2.3 are exceeded, and the fuel assemblies are not damaged. The interaction between the fuel elements and the rack is considered, particularly gap effects. The resulting impact loads are of such magnitudes that there is not structural damage to the fuel assemblies. C. Loads and Load Combinations Aoxa B ._pfhe. tables 9.1.2-1gloads n load combinations that are considered in the analy i% of the spent fuel racks include those given in t e NRC, OT Position for Review and Acceptance of Spent Fuel Storage and Handling i Applications, dated April 14, 1978, as amended by the l NRC letter dated January 18, 1979. 25 It is noted from the seismic analysis that the l magnitude of stresses varies considerably from one geometrical location to the other in the model. Consequently, the maximum loaded major rack components are analyzed. Such an analysis envelops the other areas of the rack assembly. the_ ca.s+ pwlbods ** The margins of safety forAthe multidirection seismic event are produced by combining x-direction, y-direction, and z-direction loads by the square-root-of-the-sum-of-the-sqv, ares (SRSS) method. & +he. eu+ p wl e+c.ks The loads used 4n the structural analysis are loads from the seismic r.odel which have been adjusted by peaking factors from the structural model to account for the stress gradients through the rack module. D. Design and Analysis Procedures for Spent Fuel Storage Racks b s et- + b4 G The seismic and stress anal sgs of the spent fuel rack modules considers the various conditions of full, partially filled, and empty fuel assembly loadings for the wet pool case, and conditions of partially filled and empty fuel assembly loadings for the dry case. The l racks are evaluated for both operating basis earthq ake (OBE) and safe shutdown earthquake (SSE) conditions and meet Seismic Category I requirements. A detailed L a cidi M on,, o. A cm\\g si s eV a w e.s + p oc> l d o.c k. "i n + h e cbg concW ien oath Al\\g lem ded d+b. new hel *S peb 9.1.2-2c Amend. 25 9/86 E__ _

l Insert D For the west pool racks, the margins of safety for the multidirectional seismic event were evaluated by applying statistically independent acceleration time histories in three orthogonal directions concurrently. Simultaneous application of the seismic slab motion dispenses with the need to perform statistical summations (square-root-of-the-sum-of-squares). j i 1 1 l l I 1

7 S i I } \\ \\ ' :! - ' VECD+FSAR-9 t; V stress analysis is performed to verify the acceptability of the critical load componente and pa+hs [,l under normal and faulted conditions. The racks rer $

1. _

j freely on the pool floor and are evaluated p decermine,9 5 that under all loading conditions they &r, ph)1mpact each other nor do they impact the pool %laJ l 1 The dynamic response of the fuel rack assembly during e,/ seismic event is the condition which produces the v.1 governing loads and stresses on the stria.ture. y' The' '\\ seismic analysis of a free-standing fuel rack is a, time-history analysis performed on a nonlinear mdsel: y' ?._ l, Easi 1 ~' Qen+ Fue.1 bl hck A The time-history analysis is pe%ns rformed on a single cel,1 nonlinear model with the effective properties of an, average cell within the rack module. The nonlinear y model is shown in figure 9.1.2-4 A. s ,A The effective single call properties are obiined from >l a structural model cf the rack modu)cs, as shown in figure 9.1. 2-5. r N I\\ g The details of che structural moddl ~ model are discussed in the,followjng paragraphs. 2p A.id the seismic U \\ 7; l The structural model, shown in fiqmre 9.1 I.-5 is a Ai finiteelementrepresentationo(ftken'pMasse,mbly consisting of beam elements intercm nected at a finite [{ number of nodal poiras and general $ Mss matrix elements. The. beam elements model th'e bea[a action of ( the cells, the atiffening effect of the %.ds, and the L: 3~ / supporting effect of the support pads. The general s' mass matrix elements represent the hydrodynamic mass of the rack module. The beams which represent the cells are loaded with equivalent seismic loads, and bhu modes produces the structural displaceuxnts and M ernal load a distributions necessary to calculate the effective 7 structural properties of an average cell within the rack module. the internal load and stress distributions of thisIn addition to Kne model are used to calculate stress peaking factors to account for the load gradients.w.% thin the rack? nodule. \\ ! i 9' x The nonlinear seismic model, is composed of the effective prof weies from theshownyn figure 9.1.2-4 structural model with additional elements to account for hydrodynamic mass of the fuel (set to vi ue of W ro in the case of the dry pool), the gap betwnfn the fugl and cell, and the support pad boundary contitions of a free-standing rack. Ci The elements of the no'nlinpar model follow. p A. l f ( a*, 9.1.2-2d Amendh25 9/86 L l

VEGP-FSAR-9 The fuel assembly is modeled by beam elements and rotational spring elements which represent the 4 structural and dynamic properties of the fuel rod i bundle and grid support assemblies. The cell' assembly is represented by beam elements and

l rotational springs which have structural properties of an average cell within the rack structure.

The water within the cell and the hydrodynamic mass of the fuel assembly are modeled by general mass matrix elements connected between the fuel and cell (values of these elements are set to zero in the case of the dry pool). "he gaps between the fuel and cellgare modeled by dynamic gap elements which are composed of a spring and damper in parallel, coupled in series to a concentric gap. The properties of the spring are the impact ) stiffness of the fuel assembly grid or nozzle and cell wall. The properties of the damper are the impact 25 damping of the grid or nozzle. The properties of the concentric gap are the clearance per side between the fuel and cell. The hydrodynamic mass of a submerged fuel rack assembly is modeled by general mass matrix elements connected j between the cell and pool wall (value is set to zero in i the case of the dry pool). i The support pads are modeled by a combination of J ; dynamic frictden elements connected by a " rigid" base 'I beam arrangeme:Jc e'Jeh produces the spacing of corner. i support pads. The cell and fuel assemblies are located, in the center of the base beam assembly and form a ' y h, model which represents the rocking and sliding s g-1 characteristics of a rack module. i The nonlinear model is run with simultaneous inputs of; the vertical and the most limiting horizontal acceleration time-history values. The damping values used in the seismic analysis are 4 percent damping for. l SSE and 2 percent for OBE. In addition, the model is run for a range of friction coefficients (0.2 and 0.8) to obtain the maximum values. The results from these runs are fuel to cell impact loads, support pad loads, i support pad liftoff, rack sliding, and fuel rack structure internal loads and moments. Theativalues are searched through the full time in order to obtain the 9 maximum values. The internal loads and stresses from t' 9.1.2-2e Amend. 25 9/86' o a i

') l l l 1 VECP-FSAR-9 'l L' the seismic model are adjusted by peaking factors from the structural model to account for the stress gradients through the rack module. Consequently, the maximum loaded rack components of each type are analyzed. Such an analysis envelops the other areas of the rack assembly. The maximum stresses from each of the three seismic events are combined by the SRSS method. In addition the results are used to determine the rack response for full, partially filled, and empty rack module loading conditions for the wet pool case, and for partially filled and empty rack module loading conditions for the dry $a& Ana.\\gsiS ocl case.

2. Wex+ Sp eM heA %\\

.n. ,((I[*[ Structural Acceptance Criteria for Spent Fuel Storage Racks The fuel racks are analyzed for the normal and. faulted lead combt: nations of paragraph C in accordance with the IIRC., OT Positions for Review and Acceptance of Spent Fue Storaga and Handling Applications. The major normal and upset condition loads are produced by the o}W -ati ng oasis earthquakes. The thermal strenses due to rack relative expansion are calculated and ccubined with the appropriate seismic loads in accordance with the NRC, OT Position for Review and 25 i Acceptance of Spent Fuel Storage and Handling Applications'2' (with clarifications as noted in Table 9.1.2-k foe. ake, east r-acks sA To.b\\c. 9.1.1.-IB b +ke. we,s+ t acks. The f aulted conditi)on loads are produced by the safe chutdown earthquakes and a postulated fuel assembly drop accident. The computed stresses are within the acceptance limits identified in the NRC, OT Position for Review and Acceptance of Spent Fuel Storage and Handling Application'** (with clarifications as noted in T s.ble 9.1. 2 ~1 ?. A 4 de. c as+ sack s a.d ~To. Me 9.1, Z.- IB Mo e-i h e. wt si the resu)lts of the seismic and structural r acks. In summary, analysis show that the VEGP spent fuel storage racks meet all the structural acceptance criteria adequately. E. Paal Hrtndling Crane Uplift Analysis An analysis is performed to demonstrate that the rack can withstand a maximum uplift load of 5000 lb. This load can be applied to a postulated stuck fuel assembly without violating the criticality acceptance 9.1.2-2f Amend. 25 9/86 u 1 -

I Insert E ') The details of the structural model and the structural analysis are l described in the following paragraphs: The seismic analysis is performed in three steps: a. Development of a non-linear dynamic model consisting of inertial mass elements and gap and friction elements. l 1 b. Generation of the equations of motion and inertial coupling and solution of the equations using the " component element time integration scheme" to determine nodal forces and displacements. c. Computation of the detailed stress field in the rack (at the critical location) and in the support legs using the nodal forces l calculated in the previous step. These stresses are checked j against the design limits given in Table 9.1.2-1B. j Since the racks are not anchored to the pool floor or attached to the l pool walls or to each other, they can execute a wide variety of rigid body motions. For example, the rack may slide on the pool floor (so-called " 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 structural model permits simulation l of these kinematic events with inherent built-in conservatism. Since l these rack modules are designed to preclude the incidence of inter-rack impact, it is also necessary to include the potential 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 modeled using appropriate impact elements, and Coulomb friction between the rack and the pool liner must be simulated by appropriate piecewise linear springs. These special attributes of the rack dynamics require a strong emphasis on the modeling of the linear and non-linear springs, dampers, and stop elements. These considerations lead to the following attributes of the analysis model: a. The fuel rack structure is a folded metal plate assemblage welded to a baseplate and supported on a minimum of four legs. The rack structure itself is a very rigid structure. Dynamic analysis of typical multicell racks has shown that the motion of the structure is captured almost completely by the behavior of a six degrees-of-freedom structure. Therefore, the movement of the rack cross-section at any height is described in terms of the six degrees-of-freedom of the rack base. The rattling fuel is modeled by five lumped masses located at H,.75H,.5H,.25H, and at the rack base. (Fig. 9.1.2-4B and Fig. 9.1.2-40) l l l L__________.

Insert E (Continued) b. The seismic motion of a fuel rack is characterized by random rattling of fuel assemblies in their. individual storage locations. Substituting the assemblage of rattling masses by an' effective. dynamic mass group _ greatly. reduces the required degrees-of-freedom needed to model the fuel assemblies which are represented by five lumped masses located at different 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 (Fig. 9.1.2-4C). c. The local flexibility of. the rack-support interface is modeled conservatively in the analysis (spring KR in Fig. 9.1.2-48). d. The rack base support may slide or lift off the' pool floor. e. Fluid coupling between rack and assc.:blies, and between rack and adjacent racks, is simulated by. introducing appropriate-inertial coupling into the system kinetic energy. f. Potential impacts between rack and assemblies are accounted for by appropriate " compression only" gap elements between masses

involved, g.

Fluid damping between rack and assemblies, and between rack and adjacent rack, is conservatively neglected. h. The supports are modeled as " compression only" elements for the vertical direction and as " rigid links" for dynamic analysis. The bottom of a support leg is attached to a' frictional spring as shown in Fig. 9.1.2-4B. The cross-section inertial properties of the support legs are computed and used in the final _ computations to determine support leg stresses.

i. The possible incidence of inter-rack impact is simulated by l

a series of gap elements at the top and bottom of the~ rack in the two horizontal directions. The most conservative case of adjacent rack movement is assumed; each adjacent rack is assumed to move completely out of phase with the rack being analyzed.

j. The form drag opposing the motion of the fuel assemblies in the storage locations is const vatively neglected in the results reported herein.

k. The form drag opposing the motion of the fuel rack in' the water is also conservatively neglected in the results reported herein.

Insert E (Continued) 1. The rattling 'of the-fuel assembliesLinside'the storage' locations causes-the'" gap" between the fuel assembliesLand the cell wall to change from a maximum of twice the. nominal gap to a theoreti_ cal i zero gap. Therefore, the fluid coupling ' coefficients utilized are based on non-linear. vibration theory. Studies in the literature show that inclusion of-the non-linear effect (viz. vibration l amplitude of the same order of. magnitude as-the gap) provides a more accurate characterization of-the equipment. response, m. The cross coupling effects due..to the' movement of fluid from. one interstitial (inter-rack) space to the adjacent one is'modeled using potential flow and Kelvin's circulation theorem. L Figure 9.1.2-4D shows a schematic of the model. Six degrees-of-freedom ~ are used to track the motion of the: rack structure.. Figure 9.'1.2 4C ~ shows the fuel assembly /stroage cel1-impact springs at-a particular level. An important. feature of 'the rack analysis is' incorporation 'of_ the fluid coupling effects. The fluid coupling. forces are a strong. function'of the interbody gap, reaching large values' for very small gaps. The lateral motion of a fuel assembly inside the storage' location encounters the fluid coupling effect. So does the motion _of a rack adjacent to another rack. These effects are included in the equations;of motionc 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' 180. out of phase from 'the rack 'being analyzed. Therefore, only one rack is considered surrounded by. a hydrodynamic mass computed as if there were a plane of symmetry located in the middle of the gap region. The fluid virtual mass is included in the vertical. direction vibration equations of the rack; virtual inertia is also added to the goye(ning 6 equation corresponding to the rotational degree-of-freedom, q tti. Damping of the rack motion arises from material hysteresis (material damping, relative intercomponent motion in structures structural damping), and fluid drag effects (fluid damping). In the analysis,.a maximum of 4% structural damping is' imposed on elements of the rack structure L during SSE seismic simulations, and 2% for OBE simulations. Material and fluid damping are conservatively-neglected...The dynamic model constructed' in this manner is employed to evaluate the rack module. structural-response for limiting values of the interface coefficient of friction h = 0.2 ' and 0.8), and a number of conditions of rack. loading.

VEGP-FSAR-9 criterion. Resulting stresses are within acceptable stress limits, and there is no change in rack geometry of a magnitude which causes the criticality acceptance criterion to be violated. G. Fuel Assembly Drop Accident Analysis In the unlikely event of dropping a fuel assembly, accidental deformation of the rack does not cause the criticality acceptance criterion to be violated. Fuel assembly drop accidents involving nonirradiated new fuel, which may be stored in the spent fuel racks in a dry pool do not effect criticality criteria; therefore, analysis considers only the case of a dropped spent, irradiated fuel assembly in a flooded pool and takes credit for dissolved boron in the water. l For the analysis of a dropped fuel assembly, thr:0 4ulo accident conditions are postulated. The first accident condition conservatively assumes that the weight of a fuel assembly, control rod assembly, and handling tool (2300 lb total) impacts the top of the fuel rack from a drop height of 3 ft. Calculations show that the impact energy is absorbed by the dropped fuel assembly, the 25 cells, and rack base plate assembly. Under these faulted conditions, credit is taken for dissolved boron in the water, and the criticality acceptance criterion l is not violated. J_ ~. AA A.Mody -The re-ond accident-conditicn ic an inclined drop o top of the rack Results are the__same as for_the first condition. w a.s conribe+o.3 h the. cas+ pa\\ Fack] .r e cc,+ A The third accident condition assumes that the dropped assembly and tool (2300 lb) falls straight through an empty cell and impacts the rack base plate from a drop height of 3 ft above the top of the rack. The results of this analysis show that the impact energy is absorbed by the fuel assembly and the rack base plate. I Criticality calculations show that keff s 0.95 and the acceptance criterion is not violated. H. Fuel Rack Sliding and overturning Analysis Consistent with the criteria of the NRC, OT Position for Review and Acceptance of Spent Fuel Storage and Handling Applications'5) the racks are evaluated for l 9.1.2-2g Amend. 25 9/86

l VEGP-FSAR-9 I I overturning and sliding displacement due to earthquake conditions under the various conditions of full, j partially filled, and em for the wet pool case,'an3'p,pty fuel assembly loadings artially filled and empty fuel loadings for th,e dry poo case. / gnt The nonlinear model>dese,ibed in paragrapr D is us d in i this evaluation \\to accjunt for fuel-to-rac k impact s loading, hydrody'namic forces (in the wet p' col e only), and the nonlinearity of sliding friction interfaces, ) The horizontal resistive force at the interface between j the rack module and pool floor s produced by 2'i friction. A range of friction co ficients (p = 0.2 and 0.8) is used in the angly @. low coefficient of friction (p = 0.2 ) producep ma kimum ack base horizontal displacement o# sliding, hile a high value I (p = 0.8) produces maximun. rac hori ontal, +urning force. ( es, f^ \\ e.s The fuel rack nonlinear tims-his y an lysiq she s i that the fuel rack slides a minimal dis ance. J his i distance, combined with the rack structu Eflection i and thermal growth, is less than the rack-to-rack or rack-to-wa'l clearances. Thus, impact'between adjacent rack modules or between a rack module and the pool wall l is prevented. The factor of safety against overturning 1 is well within the values permitted by Section 3.8.5.II.5 of the Standard Review Plan. l \\ i l 1 l 1 9.1.2-2h Amend. 25 9/86 j I l f

i VEGP-FSAR-9 c.1.2.3 Safety Evaluation The design and safety evaluation of the spent fuel racks is in accordance with the Nuclear Regulatory Commission position paper, Review and Acceptance of Spent Fuel Storage and Handling Applications. The racks, being Nuclear Safety Class 3 and Seismic Category 1 structures, are designed to withstand normal and postulated dead loads, live loads, loads resulting from thermal effects, and loads caused by the operating basis earthquakes and safe shutdown earthquake events. The design of the racks is such that kef f remains less than or equal to 0.95 under all conditions, including fuel handling accidents. Because of the close spacing of the cella, it is impossible to insert a fuel assembly in other than design locations. Inadvertent insertion of a fuel assembly between the rack periphery and the pool wall or placement of a fuel assembly across the top of a fuel rack is considered a postulated accident, and as such, realistic initial conditions such as boron in the water ca taken into account. This condition has an acceptable eff f less than 0.95. Should the 22 spent fuel storage be used fo w fuel storage in the dry condition, keff will be iess than or equal to 0.98. The racks are also designed with adequate energy absorption capabilities to withstand the impact of a dropped fuel assembly from the maximum lift height of the fuel handling machine. Handling equipment (fuel building crane) capable of carrying i loads heavier than a fuel assembly is prevented by interlocks cr d or administrative controls, or both, from traveling over the d' oF fuel storage area. The fuel storage racks can withstand an de uplift force}(4000 lb)] equal to the uplift capability of the j Ar fuel handling machine @ l i All materials used in construction are compatible with the storage pool environment, and all surfaces that come into contact with the fuel assemblies are made of annealed austentic stainless steel. All the materials are corrosion resistant and will not contaminate the fuel assemblies or pool environment. Venting of the boraflex can be accomplished l through the holes in the corners of the wrapper. I Design of the facility in accordance with Regulatory Guide 1.13 ensures adequate safety under normal and postulated accident conditio s. i Ad cussion of the methodology used in the criticality yshg is rovided in paragraph 4.3.2.6. j i ana CJ l 1 l Amend. 22 2/86 9.1.2-3 Amend. 30 12/86

OVC VEGP-ESAR-9 4 O o / r f 4,,3 c REFERENCE 9 l-Nuclear Regulatory Commission, letter to All Power Reactor Licensees, from B. K. Grimes, "OT Position for Review and 25 Acceptance of Spent Fuel Storage and Handling Applications", April 14, 1978. l l 1 l l I l 9.1.2-4 Amend. 25 9/86

i i VEGP-FSAR-9 I l TABLE 9.1.2-1A(SHEET 1 OF 2) LOADS AND LOAD COMBINATIONS-Eas+ Pool j 1 l Load Combination Acceptance Limit D+L Normal limits of NF 3231.la D + L'+ P Normal limits of NF 3231.la g D+L+E Normal limits of NF 3231.la D+L+T Lesser of 2S y or Su stress o range (see Note 3) D+ L+T +E Lesser of 2Sy or Su stress ) range (see Note 3) l \\ l D + L + T, + E Lesser of 2Sy or S stress I range (see Note 3)u l 25 D+L+T +P Lesser of 2Sy or S stress g g range (see Note 3)u D + L + T* + E' Faulted condition limits of NF 3231.1C (see Note 4) D+L+F d The functional capability of the fuel racks shall be demonstrated Notes: 1. The abbreviations in the table above are those used in i Standard Review Plan (SRP) Section 3.8.4 where each term is defined except for Ta, which is defined here as the highest temperature associated with the postulated abnormal design conditions. Ed is the force caused by the accidental drop of the heaviest load from the maximum possible height, and i l Pf is the upward force on the racks caused by a postulated I stuck fuel assembly. { 2. The provisions of NF-3231.1 of ASME Section III, Division I, shall be amended by the requirements of Paragraph c.2.3 and 4 of Regulatory Guide 1.124, entitled " Design Limits and Load Combinations for Class A Linear-Type Component Supports." L_--------

VEGP-FSAR-9 TABLE 9.1.2-14 (SHEET 2 OF 2) 3. The application of.this acceptance limit for the combination of primary and thermal stresses will typically limit the stresses to'Sy. However, when proper justification is provided to show that the thermal stresses are 2! self-limiting, the combined stresses may exceed Sy provided-the lesser of 2 Sy or Su stress range limit is met. 4. For the. faulted load combination, thermal loads will be neglected when they are secondary.and self-limiting in nature'and the material'is ductile. l 3 l 'l i l 1 l 1 i l l l l

IbRL E 3.I,1.-IB

LoAbs, AM6 LoAb CombMATlom-Wesr Pool Loadina 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 W ) D + L + Ta + E' Level D service limits D+L+Fd The functional capability'_ of the fuel racks should 1 be demonstrated. j l 1 where: 1 D Dead weight-induced stresses (including fuel = assembly weight) i Live-Load (0 for the structure, since there L = are no moving objects in the rack load path). 'i Fd Force caused by,the accidental drop of the = heaviest load from the maximum possible height. Pg Upward force on the racks caused by = postulated stuck fuel assembly-E = Operating Basis Earthquake E' = Safe Shutdown. Earthquake To Differential temperature induced loads = (normal or upset condition) Ta Differential temperature induced loads = (abnormal design conditions)

i 3 ?.. o u 3 3 y i i 1 I l *'*

• '*i.

_.1

• q

_;Ll I l i l I i l I t q i I l l J i i I l i l 173.25 IN. j l l l I i l l t 1 I -i .J Jp J i ,I l 1 t i a i 1 i i I i I l j _l } l l l l l i i < j i I .I _l -I l l i i i i I l _i _l l i I l l l F i i l t a l i 1 i q'. 7 i i i i e 1 i i T i e 4' l I,. 4 ' L4 j IWIW{ 1, [_f i I Amend. 3 1/84 Amend. 25 9/86 voorte 6 SPENT FUEL STORAGE RACK ARRAY, GeorgiaPower E t.ECTRIC GENERATING PLANT CAST E*03 uu,T i Ano unit 2 SIDE VIEW FIGURE 9.1.2-1A 133-9

I 'hW6 ' ~ Wh q --/WWWWWW / /WWn WWWWW,,/ /,, 'W W /.-/ W W W,/,~, /wwww /.- wy /WW /: 'W. 'WWWW, y,/ /; /WW)-/: C ', f4,/ / W W W M / W -,' W W,/ f /JJ s r- / s- /_ /./ s e' 8 s' ,s. i. I ssu i. i s. i= i n. d' p' p i se i. i-i.

1. l s'

M' e' s8 e d' =- i= i-s. ii.e i. i i., ,f ed ud d' d' 1 .e p' i-- i. i nn i. i-i. I I i n. i-i i i i. M f f I W s' i l d ,,,1 ,,/ ,,/ ,,, i / p ,,/ , / Base / Plate / ,.,-,_,.,.__,,/ g ADJUSTABLE SUPPORT LEG (TYP) MEL bMkAM AG " byLE BIT cot @T b DY, Iso meTAtc %EW Fi suu

9. l.1 - I B

1 i' ~ l ~ i 7.5" 50RAFLEX ? TYPICAL FOR i SIDES EXCEPT AS NOTEDIII i 1 I 1 A3" h c r NOMINAL a ' h8.80" SQ P I y DETAIL "Ao CELL CENTER TO CENTER (10.6") d u) Er 0.90" GAP 0.075 INNER CELL PALL 1r 0.078" BORAFLEX [ bi r 10A20 GM410/cm ) 2 y - #5ti 1 t ~ (1) OUTSIDE CELL WALLS OF CERTAIN PERIPHERAL CELL ROWS DO NOT INCLUDE BORAFLEX AS SHOWN IN FIGURE 9.1.24g DETAll "A" Amend. 25 9/86 j W+ A '?R9.Lvv'Lt SPENT FUEL STORAGE VOGTLE i l Georgia Power - ELECTRIC GENERATING PLANT CELL NOMINAL DIMENSIONS unit i ANO unit FIGURE 9.1.2-2A 433 9

y 7.75-1 ' P O l'S O N I 10.40 = PITCH .0 7 5 THK. BOX l l l l s l j 1 i l a. l r a t l 8 1 l i '$ 3 - to I i i i a z i ) a - -.._ _1-1 e u a g g .= o n. j w I l 8.7f 4.03 g '+ l SQ j . 2 ' G"- y COVER-j z ) r t l t j i ? i l ( ) i -... - ~ - _.. - - N l .0 96 TH K. .0 90 T H K. STRIP I'2 7.75" WIDF Il3GA ) NOM. ANGLE ~ J GAP i 1 i i i 1 i ! FIGURE 5 ) W EST Poo t-JPeNT Fuel STeeiGE - I C ELL N o rn i w AL 6 m e a.ste td.s i F, caw 3. l.2 -1B i -l

2 k N M dn T e S. m L A L A EA N TW O P I n. t S s f l O o I 8 , 5 d 0 P e r 6 4 O p N e o. 7 _ _ [ 2 + _. [ N O I _x S N O S O G P P O O N 2 _.u i _. N N N O O S I \\' I 1 S O O P \\ P N O S I O P ON S K t C t A A H W L H E I U M F O T N N i E P Pl S

c., I & L L"d r, -igh J Y L 1,"y r j J l 1. 'u,l.. \\-t,6 / u,.- La f ,\\.q., -(5))- , ",f r / 62

• "/

un e e \\Y(g/l-u, L9 r ca

e.,

-/W T-a x . 'L &g u n .E 4 .}; N,W7-u, c-s ""b l e 1 u, I \\ -{I G 't, / l n e.1 / um o ( r.,L ,\\g {,r' / Lj _I c., / rw l r_9 af i n g /-Me -{I g ;. r, p L _1A y.9 4Ld b \\ ~ ~ ~,~ ,r..,,-,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, i Amend. 25 9/86 CAST Spent Fuct Al i< x k'.s "" E NONLINEAR SEISMIC MODEL Georpia Pbwer ' ' ' * " " * ' " ' " ^ " " ' ^ " ' 'O UNIT 1 AND UNIT 2 FIGURE 9.1.2-4A 433 9

FUEL ASSY/ CELL IMPACT ~ SPRING, K - a h JL 0.25H yhp y H 0.25H RACK N C.G. %ll% 9 U s h H 0.25H H/2 M Ek a TYPICAL RATILING MASS 0.25H i 5 FRICTION Eb y SUPPORT LEG 9 SPRING, K INTERFACE S SPRING, K f l l M s a 2 ,,L. 3,,, s FOUNDATION ROTATIONAL COMPLIANCE SPRING, K ~ R WEST JPENT FtAEL Pcot it AC-k - FICL'".E 49 TWO DIMENSIONAL VIEW OF THE SPRING-MASS SIMULATION Pisute 9. i.1 - 4 B h

l l Y CELL WALL MASS i XB i y / / / FUEL ASSEMBLY / CELL IMPACT SPRING / / / / YB / / / / / > X l W E,5T .5 9u WT P ue L Poot id AC.K IMPACT SPRING ARRANGEMENT AT NODE i-FIGURE 9; 1, L-4 C,

L G3 7 l /2* t y p8 a P7 H/4 g6 8 3 o P10 n RACK GEOMETRIC '9 CENTER LINE N H/4 H u4 > P12 a H/4 P11 5* o 4 - pj4 h / H/4 b + S4 p13 45 & y -->.g2 p X A Y ? 6 B ge LONG DIRECTION / SUPP'T / / 44 TYP. FRICTION ELEMENT 4 / 4 W EST.JPetJT Fuck Pool s( A m SCHEMATIC MODEL FOR DYNRACK FIGURE 9,l', Z_ I D t l

-~ % w _ 1 1 s ~ s . y = m ~ _ i m N s ,s m c-i ~ m m x ~ m, ~ V5J 5 45C5CTQw 1 Q5 ?$2Cb% % Amend. 25 9/86 East Speut f%eg poog g gg., voorts STRUCTURAL MODEL j Geogia Pow'er d NIT 1 AND UNIT 2 ' '"'c ' ***aar'ao r'a=r i U i FIGURE 9.1.2-5 f 4339 l -t 'l 3 ^^ ^ - - - - - _ _ _ _ _ _

---

'-----._.m._____

l l ,? 'h o 2 L 'a + m n_ = P 'f f 'f l ~ m 0 4 4 4 e 21 o 29 9 9 o uT F"G a x i x x G O i x D 2 x 02 2 9 x l 4~ T O ~i, k X 1 H x l 1 m 1 J Y m 4 2 ~ F. A l o 42 2 4 4 f =t L 1 2 2 P E I 1 1 i 1 S E E { [ [ [ tad G 4 sR 0 4 0 4 4 31 o =t O G m 0 9 9 39 1 m l 1 9 1 x x aT1 x x x sSF i u x D2 02 C2 x D2 x ~k ( g [ 1 1 1 1 r 4 4 4 4 ~f u 2 c 2 1 2 2 I u 1 1 L O t 4~ [ [ "f c l o F x xf 0 4 r 54 3 9 4 4 41 9 69 i 9 o l - 0 x x x i x x 1 AI A ll Bi x u l x l [ [ ~h .) 1 ~ t 3 3 3 ~f o 3 n u i l i 1 i G + i i i_ i l '0 r a 5 f ~ [ j j 4 4 0 4 0 2 9 4 59 9 31 o 9 O 1 x x x i x 4x Ai x Ai Bi x l E x x {' [ f { l l l 4 3 i m 3 3 3 f l l i i i i i l i i ~f [ { ~ 4 N 0 o 4 4 4 0 9 9 49 9 21 o 1 l s a f l s x x x x x x x x +- l ~ BI [ Al [ Al { Bi [ ie I I l 3 3 3 3 f i i i 2 i i i i i + Qt m f' e =_ J 4

• y

'u i m. Il ll

l i i l 1 I l 4 J 1 I I 1 ATTACHMENT B i l

SUMMARY

REPORTS ON SPENT FUEL RACK I DESIGN, SEISMIC ANALYSIS, CRITICALITY ANALYSIS 1 1 l l 'l i i 1 I

i ] i I a l SYNOPSIS OF MODULE DESIGN'AND LAYOUT I l 4

? l 1.0 GENERAL i i The spent fuel storage racks scheduled to be installed in the West pool consist of 20 free standing modules arranged as shown in Figure 1. Table 1 gives the cell count and module I.D. data. i The storage cells consist of 8.75 inch nominal prismatic openings formed by seam welding precision formed channels. These cells are interconnected using longitudinal angle connectors to form a honeycomb construction structure. Each fuel storage location incorporates the Boraflex neutron absorbing material (boron carbide powder uniformly dispersed in a polymeric matrix) which is held in place by a stainless steel sheathing. The Boraflex encasement method provides for unconstrained in-plane contraction (or expansion) of the poison material and lends complete lateral support to it to protect it from slumping. The material is not sealed since it is compatible with the pool environment. The fuel storage cell walls, as well as all other structural fabricated from SA240 Type 304L stainless steel. components, are The only exception is the bottom portion of the support spindle which is made out of precipitation hardened stainless steel (ASME SA564-630). The cell pitch is 10.58 inch in the north-south direction, and 10.40 inch in the east-west direction. Table 2 gives the essentials of rack construction data. These racks are free standing and are not inter-connected to each other or the pool walls. Each rack is equipped with a minimum of four adjustable support feet (Figure 2). l 1 l l J

Table 1 i l CELL AND MODULE DATA j i NUMBER OF CELLS Total Number of Number Module No. of East-West North-South Cells Per 'of I.D. Modules Direction Direction Module Cells i A 6 11 9 99 594 B 4 11 10 110 440 C 3 12 9 108 324 ) D 3 12 10 120 360 E 1 11 10 94* 94 F 1 11 10 90** 90 G 1 12 9 104*** 104 H 1 12 10 92**** 92 Total: 2098 16 cells removed from the north-east corner 20 cells removed from the north-west corner 4 cells removed from the south-east corner 28 cells removed from the north-east corner l l l l 1 l l l l l l

I

2. y -. gr

-; x hi. e Ti

s u

lf y \\ ' d,. '.'y/ g Y L i(<

p i

.g -',.. y 4 b% Table 2 j,. j \\ 'r l i f l-SPENT FUEL RACK DIMENUIONS*' ( q., - cg Fuel Type: Westinghouse 17.h-17 !( 6 ), ' Rack Type: Honeycomb constriction; [ density t 'poisfnsd high .l Center-to-Center Spacing: N.- North-South: 10.58 East-West: 10.40 s 1 Cell inside. dimension: 8.75 [ $* g, g *,./' q 7 cell wall thickness: 0.075 g 1, Poison length: 139 , Poison width: 7.75 g tJ- . j '3' Poison thickness: 0.075 i i i h i c B-10 Leading: 0.020-(gm/sq.cm)- Poison sheathing thickness: 0.020 1 Total Height of 175-3/8" nom. Rack from Pool Floor Bearing Plate Bearing Plate 1-1/4" thick c. i 1 A s 't, ( i

4. P :,: ]

(

• All dimensions in inches unless otherwise indicated.

M a -1 I 1 l 'N i i q. p-h 3 ~. l l i l + L___--------------------------------- ~ - ~ ~' ~ ~

/ i i- ) 2.0 FABRICATION OF THE RACK MODULE In this section, we introduce the constituent elements of the Vogtle PWR rack modules by. describing them in the sequence in, which they are brought into the fabrication proce'ss. The rack module manufacturing begins with fabrication of the so-called box. The " boxes" are fabricated from two precision formed. channels by seam welding them.togethertin a seam welding machine equipped with copper chill bars, and pnetunatic clamps to minimize distortion due to welding heat input. Figure 3 shows the box. L The minimum weld penetration shall be'80% of the box metal vage which is 0.075" (~ gage). The boxes are manufactured to 8.75" I.D. (inside dimension) .03". 41 Each box constitutes a storage location. The next step in the manufacturing process entails affixing dte. poison sheet (7-3/4" wide x 139" long) on each side of.the'tox. To locate the poison in its correct elevation, 13 g a g e y' x 2" (min.) x 7-1/2" wide stainless steel flats are positionednat the j bottom of the box, and spot welded to it. Similarly, 13 gage x 2" high x 5-1/2" wide strips are located 1/2" (nominally) above the top of the " poison" prevents the poison frem sliding upwards during shipment and handling of the racks. The picture frame is completed by attaching angles (or flats, as needed by the grid pattern) to the sides of the box as shown in the plan view. Some z corners require angles; others require flats as will become clear from the following. These angles are' fusion welded to the box, { and the flats are spot. welded to it. ,[ ] 4 4 0 4 m____.___ _ __ _ _ _ _ _ _ _ _ _ _ - - - -. _ - _ - - - - -. ~. - - - - - - - - - - - -. - ' - - - - ^ -

4 The poisen sheet can now be installed in the picture frame space, j and covered with .020" thick (nominal) sheathing. The " sheathing" overlaps the picture frame stripa at the bottom, side and top, and is spot wolded to them. Finally, 1-1/2" diameter flow holes are punched near the bottom on all four sides of this

• composite kox assembly".

The top of the box is equipped with a lead-in as discussed later (Fig. 4). Haring fabricated the required number of the composite box assenblies, they are joined together in a fixture in the manner shown in Fig. 5. The pitch between the box centerlines is 10.40" in one principal direction and 10.58" in the other principal direction. The fabrication procedure in either direction is ) identical, since the protruding angles from adjacent boxes overlap and are fillet welded to each other. Figure 5 shows an array of boxes attached to each other. Fig. 6 also illustrates that the joining pattern results in a well designed shear flow path; and essentially makes the box assemblage into a multi-flanged beam type of structure. In the next step of n.:inufacture, the "!:ane plate" is attached to the bottom edge of the boxes. The base plate is a 5/8" thick austenitic stainless steel plate stock which has 6" hole (Ref. Fig., 4) curned out in a pitch identical to the box pitch. The base ple".e is attached to the cell insemblage by fillet welding the box edge to the plate by reaching in through the bottom hole using a " goose neck" welding head (2 sides only; other 2 sides welded from outside) (Ref. Fig. 4). In the tital s?.ep, adjustable leg supports (shown in Fig. 6) are i welded to the underside of the base plate. The adjustable legs } 5 t o .-a

t i 1 provide a i 1/2" vertical height adjustment at each leg. location. The lifting bosses are also attached to the underside of the base . plate using suitable filled welds. Finally, a 300 lea'd-in." cap" is inserted and fillet welded to the walls for forming a smooth lead-in. The manufacturing of the rack module culminates.with l visual examination of welds in accordance with the design drawings. l l l 1 l l .i 1 l l 3 i 6 l l l

~ ,? 0 [ a r + 2 = 'y + ~f "f 0 4 4 4 o 2 9 9 9 O } 2 1 x 4 O I x i x x G D 2 x C2 2 9 x l k~ 't, 1 "k x 1 H x 1 4 4 2 ~ 2 2 4 4 { 1 2 2 1 1 1 + i I ~ ~ j k [ - 0 4f 4 0 4 4 31 O o 9 9 39 l 1 s l 9 x x x x x i x D2 C2 C2 x Da x ~ 'k k ~k k 1 l 1 1 4 4 4 4 ~f z 2 i l 2 2 I 1 + 1 + O 4 "h ~ xx[ "f y O l F 0 P i54 3 9 4 4 41 4 9 69 9 O l - o x x x i x x i Al Ai Bi x i x l h h ,h l I ~ ~ { E ~ O 3 3 3 3 fR U i l l l i G i i I l I '0 F + 5 ~ ~ f k h k 4 4 0 4 o 2 s 4 59 9 31 0 'x 9 O x x 1 s x 4 x Ai x Ai Bl x l E xx{ f f l l I h 4 3 ~ i 3 3 3 { l - i l l l 1 I + I ~f 'g "f + { 4 N 0 o 4 4 4 0 L i 9 9 49 9 2 0 1 l l 1 1 x x x x x x x x Bl AI f Ai { Bi i ~ l ~ ~ { l h l 3 3 s 3 [ i l i i u 2 l l + Ttp { r

• {4 -

= 'u

1 1 \\ l o /WWWWWWWJy, - /Wi/WWWWWi 1 'g / /w /w-x 'wwww i 1 /WW 'WWWW&g'/ /: 'WWWW 'wwy,/ /-' 'WWW: 'WWWW,/ /WW- 'W, /W/-- /, W,/ / /J / s- / ,s /,/,- l ,a, ps ' s' ) ( ) , s. A p0 d' i.eu i. i i.e ,i. sell i s' 8' j d' s' l w.a s. n is i p l l [ I l j ,d d' d' d' 1 /' j s' Isas I ,gs i s' d' s' j, j ~ ,...L ,, / ./ ~' ,,,,,/ ,,/ ,,,. / ,,/ 30Se / Plate / j g ADJUSTABLE SUPPORT LEG (TYP) fig. 2 A TYPICAL MODULE 8

1 s \\ \\ \\ ) / g,75' i \\ \\ / - / A,>, VELD f FIGURE. gggg VELDlHG pggcis10H YORMED CUANNEL 9

, i l FT e ' ins-i. wEL: 7 " L:. K .,._w s a idl S! des (2 SIDES p-L t 7 .gj g f f 4 [ g [ INSIDE AND 2' SIDES OUTSIDE) e [ 4 l l a i ~--- BASE Pt ATE m e e A i ^ - s:. _ 8.75"_ l "*3 i VAFlABLE PITCH * ""l""-l" FUEL ASSEMBLY i l 2" REF. l O - T. dd A i n ,n., n,,\\ s y V SMCCTH j a . r - ' --,j i l L_ __1 I

t- - {- "i J

= e n __I_q ~ h l l PO! SON I l j 'O '. b _ I~ !_j o 1 L r_a _ _, I _i L_ _ _M [ l--{'- l 1.~ I 2 DI A. l g Flow McLa i L___J i r- - t- - y Y l E,lI 5 ,W ,8 i,8 sAs. PLATE x Y Q l

==- t \\ IiI ) i 6 DIA. HOLE TYP. TYPICAL CELL ELEVAT1CN FIGUP}~r 4 i 10 i J

1 7.75 POISON 10.4 0 PITCH .0 7 5 THK, BOX M l l I l i. s l' l a l f C ~ d I l t l 8 e es l l vo I i i i a z [ -..l-I s Q U

m. s.

a + o a. c 8.7f 0 3" { i l l I, SQ ~ j . 26 6d' y COVER s k i s j i ? i t w-l \\ ~\\.0 90 T H K. .0 96 TH K. I'2 STRIP 7.75" WIDF ( 13 G A ) NOM. ANGLE GAP 1 i i FIGURE 5 11

CELL ";.[ j' BASEPLATE .1 1 l' L 1 5 I I \\ A 1 ' L ___m b li l! h / n n-n I,', GUSSET ~ x 5 i i z = ,,. i. 'I i i / / L g g 3 4 2 n 2 15 " h d \\ n ~ '7 N 5/24UNc CLASS 1A y 1 I ', r i, e i 1" DIA. HOLE TERP 9" oo = l FIGURE 6 ADJUSTABLE SUPPORT 1 1 12 J ________--D

o gi 1 TS$4 l l l l SEISMIC ANALYSES FOR V0GTLE ELECTRIC GENERATING PLANT SPENT FUEL STORAGE RACKS i l l [ ---___-____a

l ,v~ p Rev 1 - 12/10/87 PREFACE The seismic qualification of t h e. V o g t l e. racks is carried out using the computer code DYNARACK with full consideration of non-linear fluid coupling effects. The object of the analysis is to ensure that the racks do not collide with each other or with the pool walls. Moreover, it is necessary to demonstrate that all dimensionless stress factors remain below the Code postulated limits. I This report is a summary capsulization of the seismic analyses { p'erformed to demonstrate the structural integrity of Vogtle racks. The time histories supplied by Bechtel Power Corporation were used in production runs. Independently produced time histories by Holtec International were used to make a completely independent evaluation. In addition to stress

factors, this report also contains information on weld

stresses, support foot

loads, rack displacements and temperature induced stresses.

The seismic. _mulat. tuns indicate that all dimensionless stress factors for the SSE loading condition satisfy the OBE limits. Therefore, separate runs for OBE were not required. l This issue (Revision 1) of this report deals with normal fuel stored in wet or dry pov.. A subsequent revision to this issue (Rev. 2) will incorporate consolidated fuel storage scenarios. i

,s i / 1.0 OBJECTIVE The purpose of this section is to demonstrate the structural adequacy of the vogtle Unit 2 spent fuel rack design under norma ~. and accident loading conditions. The method of analysis presented herein uses a time-history integration method similar to that previously used in the Licensing Reports on High Density Spent Fuel Racks for Fermi 2 (Docket No. 50-341), Quad Cities 1 and 2 (Docket Nos. 50-254 and 50-265), Rancho Seco (Docket No. 50-312), Grand Gulf Unit 1 (Docket No. 50-416), Oyster Creek (Docket No. 50-219), V.C. Summer (Docket No. 50-395), and Diablo Canyon 1 and 2 (Docket Nos. 50-275 and 50-323). The results show that the high density spent fuel racks are structurally adequate to resist the postulated stress combinations associated with level A, B, C, and D conditions as defined in References 1, 2 and 12. 2.0 ANALYSIS OUTLINE 1 The spent fuel storage racks are Seismic Category I equipment. Thus, they are required to remain functional during and after a Safe Shutdown Earthquake (Ref. 3). The design is also required to satisfy all load combinations pertaining to the OBE condition. ] These racks are neither anchored to the pool floor nor attached I 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 and maximum rattling mass), partially filled or it may be completely empty (minimum inertia). The coefficient of friction, y, between the supports and pool floor is another indeterminate -- l -

1 ,s 1 p-factor. In reality, the coefficient of friction varies from the incipient sliding to the sliding interface condition.

However, according to Rabinowicz (Ref.

4) the results of 199 tests performed on austenitic stainless 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. Two separate analyses are performed for the rack assemblies with values of the coefficient of friction equal to 0.2 (lower limit) and 0.8 (upper limit), respectively. Analyses performed for the geometrically { limiting rack modules focus on limiting values of the coefficient ) f of friction, and the number of fuel assemblies stored. Cases studied are listed'in Appendix A to this report. i The simulations were performed using Westinghouse 17x17 fuel (weight of fuel: 1647# per cell for the unconsolidaNd l condition). l 1 l The method of analysis employed is the time-history method. The three orthogonal pool slab acceleration data were developed by Holtec from the response spectra provided by the Bechtel Power Cc=pany. These time histories are shown to be statistically independent by Holtec International. Bechtel Western Power l Company also generated statistically independent time histories for SSE and OBE (Figs. 1 through 6) which were used in the time history analysis productica runs, with Holtec generated ground motion set serving as an independent check on certain runs. Bechtel Western Power Cc=pany has documented the statistical independence of their time histories. -) - w--- - -

r The objective of the seismic analysis is to determine the structural response (stresses, deformation, rigid body motion, etc.) due to simultaneous application of the three statistically independent, orthogonal excitations.

Thus, recourse to approximate statistical summation techniques such as the " Square-Root-of-the-Sum-of-the-Squares" method (Ref. 5) is avoided. For nonlinear analysis, the only practical method is simultaneous application.

Pool slab acceleration data are provided for two earthquakes: Operating Basis Earthquake (OBE) and Safe Shutdown Earthquake (SSE). Figures 1 through 6 show the time-histories corresponding l l to the SSE and OBE conditions (Bechtel's sets). The seismic analysis is performed in three steps, namely: 1. Development of a nonlinear dynamic model consisting of inertial mass elements and gap and friction elements. l 2. Generation of the equations of motion and inertial i coupling and solution of the equations using the " component element time integration scheme" { (References 6 and 7) to determine nodal forces and displacements. 3. Computation of the detailed stress field in the rack (at the critical location) and in the support legs using the nodal forces calculated in the previous step. These stresses are checked against the design limits given in Section 6. A brief description of the dynamic model follows. i 3.0 FUEL RACK - FUEL ASSEMBLY MODEL 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 -?- ..__..___________.__a

? of rigid body motions. For example, the rack may slide on the pool floor (so-called " 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 structural model should permit simulation of these kinematic events with inherent built-in conservatism. Since the vogtle modules are designed to preclude the incidence of inter-rack

impact, it is also necessary to_ include the potential 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 elements, and Coulomb friction between the rack and the pool liner must be simulated by appropriate piecewise linear springs. These special attributes of the rack dynamics require a strong emphasis on the modeling.of the nonlinear elements, dampers, and stop elements. The model outline in the remainder of this section, and the model description in the following section describe the detailed modeling technique to simulate these effects, with emphasis placed on the nonlinearity l of the rack seismic response. 3.1 Outline of Model for Comnuter Code DYNARACK

a. The fuel rack structure is a folded metal plate asemblage welded to a baseplate and supported on a minimum of four legs.

The rack structure itself is a very rigid structure. Dynamic analysis of typical multicell racks has shown that the motion of the structure is captured almost completely by the behavior of a six degrees-of-freedom structure; therefore, the movement of the rack cross-section. at any height is described in terms of the six degrees-of-freedom of the rack base. The rattling fuel is modelled by five lumped masses located at H, .75H, .5H, .25H, ansd at the rack base. l _t-l-E-_-----_____--_--------- 1

l P i l b. The seismic motion of a fuel rack is characterized by l random rattling of fuel assemblies in their individual storage locations. Assuming an equivalent dynamic rattling mass greatly reduces the required degrees-of-freedom I needed to model the fuel assemblies which are i represented by five lumped masses located at different I levels of the rack. The horizontal rattling inertia is computed by utilizing the square-root-of-the-sum-of-squares { method with a factor of safety incorporated in it. The I hydrodynamic mass ascribed to the rattling inertia is i consistent with the effective rattling mass. The centroid of 1 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 rack-support interface is l

modeled conservatively in the analysis. i ,d. The rack base support may slide or lift off the pool floor. l l l

e. The pool floor has a specified time-history of seismic-l accelerations along the three orthogonal directions.

l

f. Fluid coupling between rack and assemblies, and between rack and adjacent racks, is simulated by introducing a l

inertial coupling into the system kinetic energy. ppropriate Inclusion j of these effects uses the methods of References 6 and 7 for rack / assembly coupling and for rack / rack coupling.

g. Potential impacts between rack and assemblies are accounted for by appropriate " compression only" gap elements between masses involved.

h. Fluid damping between rack and assemblies, and between rack and adjacent rack, is conservatively neglected.

i. The supports are modeled as " compression only" elements for the vertical direction and as " rigid links" for dynamic analysis. The bottom of a support leg is attached to a frictional spring as described in Section 4.

The cross-section inertial properties of the support legs are computed and used in the final computations to determine support leg stresses. f/ J

i r

j. The effect of sloshing can be shown to be negligible at the bottom of a pool and is hence neglected.

k. The possible incidence of inter-rack impact is simulated by a series of. gap elements at the top and bottom of the rack in the two horizontal directions.

The most conservative case of adjacent rack movement is assumed; each adjacent rack is assumed to move completely out of phase with. the rack being analyzed.

1. The form drag opposing the motion of the fuel assemblies in the storage locations is conservatively neglected in the j

results reported herein. m. The form drag opposing the motion of the fuel rack in the water is also conservatively neglected in the results reported herein. 'n. The rattling of the fuel assemblies inside the storage locations causes the " gap" between the fuel assemblies and the cell wall to change from a maximum of twice the nominal gap to a theoretical zero gap. Therefore, the fluid coupling coefficients (Refs. 8,9) utilized are based on non-linear vibration theory. Studies in the literature show that inclusion of the nonlinear effect (viz. vibration amplitude of the same order of magnitude as the gap) provides a more 4 accurate characterization of the equipment response (Ref. 10).

o. The cross coupling effects due to the movement of fluid from one interstitial (inter-rack) space to the adjacent one is medelled.using potential flow 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 (Ref. 13). Figure 7 shows a schematic of the model. Six degrees-of-freedom are used to track the motion of the rack structure. Figures 8 and 9, respectively, show the inter-rack impact springs and fuel assembly / storage cell impact springs at a particular level. -h-

As shown in Figure 7, the model ~for simulating

• uel assembly motion incorporates five lumped masses. The five rattling masses are located at the baseplate, at quarter height, at half height, at three quarter height, and at the top of the rack. Two degrees-of-freedom are used to track the motion of each rattling mass in the horizontal plane. The vertical motion of each rattling mass is assumed to be the same as the rack base.

3.2 Model Description The absolute degrees-of-freedom associated with each of the mass locations are identified in Figure 7 and Table 1. The rattling masses (nodes 1*, 2*, 3*, 4*, 5*) are described by transnational degrees-of-freedom q7-q16 Ui(t) is the pool floor slab displacement seismic time-history. Thus, there are sixteen degrees-of-freedom in the system. 3.3 Fluid Couplina An effect of some significance requiring careful modeling is the so-called " fluid coupling effect". If one body of mass (ml) 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 form: 5-M N = applied forces on mass mi (m1 + M11) 1 12 2 E + (m2 + M22) E = applied forces on mass m2 -M21 l 2 5< 1 2 denote absolute accelerations of mass mi and m2r respectively. M11, M12, M21, and M22 are fluid coupling coefficients which depend on the shape of the two

bodies, their relative disposition, etc. Fritz (Ref. 9) gives data for Mij for various

i r body shapes and arrangements. It is to be noted that the above equation indicates that the effect of the fluid is to add a certain amount of mass to the body (M11 to body 1), and on 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 l fuel assembly inside the storage location will encounter this effect. So will the motion of a rack adjacent to another rack. These effects are included in the equations of motion. For example, the fluid coupling is between nodes 2 and 2* in Figure 7,. 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 f 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 considered surrounded by a j i 1 1 hydrodynamic mass computed as if there were a plane of symmetry { l located in the middle of the gap region. 1 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). 3.4 Damoina In reality, damping of the rack motion arises from material hysteresis (material damping), relative intercomponent motion in structures (structural damping), and fluid drag effects (fluid damping). In the analysis, the damping of 4% for the SSE and 2% 1 0l 6

O P i f .for the OBE have been considered.'This is in accordance~.with the FSAR and NRC guidelines (Ref. 11). . Material and fluid damping -are conservatively neglected. The dynamic model has the provision to incorporate. fluid damping effects; however, no fluid damping has been used for this analysis. 3.5 Imeact Any fuel assembly ' node (e.g. 2*) may impact the corresponding. structural mass node 2. To simulate this

impact, four.

compression-only gap elements around each rattling. fuel' assembly node are provided (see Figure 9)..As noted previously, fluid dampers may also be provided in parallel with ' the springs. 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 8 shows the location of the impact springs used to simulates any . potential for inter-rack impacts. 4.0 ASSEMBLY OF THE DWAMIC MODEL The cartesian coordinate system associated with the. rack has the following nomenclature: O x= Horizontal coordinate along the short direction of rack rectangular platform O y = Horizontal coordinate along the long direction of .the rack rectangular platform O z = Vertically-upward i l 1 ,C /

Table 1 DEGREES OF FREEDOM Displacement Rotation I Location Ux U U 6 6 6 y 2 x y 2 (Node) 1 1 P1 P2 P3 94 95 96 2 Point 2 is assumed attached to rigid rack at the top most point. i 2* p7 98 3* pg pio l 4* pli p12 i 5* pl3 P14 1* P15 P16 where: pi qi(t) +Ul(t) i = 1,7,9,11,13,15 = qi(t) + U (t) i = 2,8,10,12,14,16 = 2 l qi(t) + U (t) i=3 = 3 Ui(t) are the 3 known earthquake displacements (denoted by VH1, VH2 ansd VV in Figures 1 thru 6). 0'

i ? I I Table 2 NUMBERING SYSTEM FOR GAP ELEMENTS AND FRICTION ELEMENTS I. Nonlinear Serinas (Gao Elements) (64 Total) Number Node Location Description 1 Support S1 Z compression only element i 2 Support S2 Z compression only element l 3 Support S3 Z compression only element 4 Support S4 Z compression only element 5 2,2* X rack / fuel assembly impact element 6 2,2* X rack / fuel assembly impact element 7 2,2* Y rack / fuel assembly impact element 8 2,2* Y rack / fuel assembly impact element l 9-24 other rattling masses 25 Bottom cross-Inter-rack impact elements section of rack (around edge) Inter-rack impact elements Inter-rack impact elements l Inter-rack impact elements Inter-rack impact elements Inter-rack impact elements Inter-rack impact elements l 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 elements Inter-rack impact elements Inter-rack impact elements 64 Inter-rack impact elements __ / / -

f i o p. i l l Table 2 (continued) i NUMBERING SYSTEM FOR GAP ELEMENTS AND FRICTION EIEMENTS II. Friction Elements (16 total) Number Fode Location Description 1 Support S1 X direction support. friction 2 Support S1 Y direction' friction 3 . Support S2 X direction friction 1 4 Support S2 Y direction friction 5 Support S3 X direction friction 6 Support S3 Y direction friction 7 Support S4 X-direction friction 8 Support S4 Y direction friction

• 9 S1 X Slab moment 10 S1 Y Slab moment.

11 S2 X Slab moment-12 S2 Y Slab moment 13 S3 X Slab moment 14 S3 Y Slab moment 15 S4 X Slab moment 16 S4 Y Slab moment _. [ y[ -

? As described in the preceding section, the rack, along with the base, supports, and stored fuel assemblies, is modeled for the i general three-dimensional (3-D) motion simulation by a sixteen l degree-of-freedom model. To simulate the impact and sliding phenomena expected, 64 nonlinear gap elements and 16 nonlinear friction elements are used. Gap and friction elements, with their connectivity and purpose, are presented in Table 2. 1 If the simulation model is restricted to two dimensions (one horizontal motion plus vertical motion, for example) for the ourcoses of model clarification onlv, then a descriptive model of { the simulated structure which includes gap and friction elements 1 1 is shown in Figure 10. i 1 e The impacts between fuel assemblies and rack show up in the gap elements, having local stiffness K, in Figure 10. In Table 2, I 1 gap elements 5 through 8 are for the vibrating mass at the top of the rack. The support leg spring rates Ks are modeled by I elements 1 through 4 in Table 2. 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 2) are shown in Figure 10. The friction of the support / liner interface is modeled by a piecewise linear spring with a suitably large stiffness Kf up to the limiting lateral load, pN, where N is the current compression load at the interf ace. between support and liner. At every time step during the transient analysis, the i current value of N (either zero for liftoff condition, or a compressive finite value) is computed. Finally, the support J 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 (Ref. 4) and is included to simulate the resistive moment of the support to A-

l \\ counteract rotation of the rack leg in a vertical plane. This rotation spring is also nonlinear, with a1zero spring constant value assigned after a certain limiting condition of slab moment loading is reached. The nonlinearity of these springs (friction elements 9, 11, 13, and 15 in Table 2) reflects the edging limitation imposed on the base of the rack support legs. If this effect is neglected; any support leg bending, induced by liner / baseplate friction' forces, is resisted by the leg acting as a beam cantilevered from the l rack baseplate. This-leads to higher predicted loads-at the support leg - baseplate junction. T,he spring rate Ks modeling the effective compression stiffness of the structure in-the vicinity of the support, is computed from the equation: 1 1 1 1 _+ + KS Ki K2 K3 where: Ki= spring rate of the support leg treated as a tension-compression member'= ESUPPORT

• ASUPPORT/h (h = length of support leg)

K2 = 1.05EcB/ (1 1/2 ) = local spring rate of pool slab (Ee = Young's modulus of concrete, and B =_ length of bearing surface) K3 = spring rate of folded plate cell structure above support leg (same form as K2 with E chosen to reflect the local stiffness of the honeycomb structure above the leg) lf = Poisson ratio of the underlying concrete material. -l4-

1 ) 1 For the 3-D simulation, all support elements (listed in Table 2) are included in the model. Coupling between the two j horizontal seismic motions is provided both by the offset of the fuel assembly group centroid which causes the rotation of the entire rack and by the possibility of liftoff of one or more j support legs. The potential exists for the rack to be supported on one or more support legs or to liftoff completely during any instant of a complex 3-D seismic event. All of these potential events may be simulated during a 3-D motion and have been observed in the results. 5.0 TIME INTEGRATION CF THE EOUATIONS OF MOTION l 5.1 Time-Historv Analysis Usina Multi-Decree of Freedom Rack Model i Having assembled the structural model, the-dynamic equations of l l motion corresponding to each degree-of-freedom are written by l using Lagrange's Formulation. The system kinetic energy can be l constructed including contributions from 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 l matrix notation as: '~ [M] {q} = {Q} + {G} where the vector {Q} is a function of nodal displacements - and velocities, and {G} depends on the coupling inertia and the ground acceleration. Premultiplying the above equations by (M]-1 renders the resulting equation uncoupled in mass. We have: {q} = (M]-1 {Q} + (M]-1 {G} j I l }V 1

\\ 1 i This equation set is mass uncoupled, displacement coupled, and is 1 ideally suited for numerical solution'using a central difference scheme. The computer program "DYNARACK"* is utilized for this purpose. i Stresses in various portions of the structure are computed from known element forces at each instant of time. Dynamic analysis of typical multicell racks has shown.that the 1 motion of the structure is captured almost completely by the behavior of a six-degree-of-freedom structure; therefore, in this analysis model, the movement of the rack cross-section at 'any 1 height is described in terms of the rack base degrees-of-freedom (q1(t),...q6(t). The remaining degrees-of-freedom are associated with horizontal movements of the fuel assembly masses. In this dynamic model, five rattling masses are used to represent fuel ) assembly movement in the horizontal plane. Therefore, the final dynamic model consists of six degrees-of-freedom for the rack plus ten additional mass degrees-of-freedom for the five rattling nasses. The totality of fuel mass is included in the simulation and is distributed among the five rattling masses. i This code has been previously utilized in licensing of similar racks for Fermi 2 (Docket No. 50-341), Quad Cities 1 and 2 (Docket Mos. 50-254 and 265), Rancho Seco (Docket No. 50-312), Oyster Creek (Docket No. 50-219), V.C. Summer (Docket No. 50-395), and Diablo Canyon 1 and 2 (Docket Nos. 50-275 and 50-323), St. Lucie Unit I (Docket No. 50-335) and a Byron Units I and II (Docket Nos. 50-454,.50-455). DYNARACK is a. Q.A. validated computer code under Holtec International's nuclear quality assurance program which .] conforms to 10CFR50 - Appendix B. l I ____.___________a

1 .,; 5, y 8 ) I 5.2 Evaluation of Potential for Tyter-Rack I7neact Since the racks are closely spaced, the simulatio/n includes impact springs to model the potential for inter-rack impact, U' especially for low values of the friction coefficient between the i support and the pool liner. To account for this3potentiaf, yet still retain the simplicity of simulating only a single rackt gap 1 elements were located on the rack at the top And at the baseplate ? level. Figure 8 shows the location of these(gap elements. LoeMs in these

elements, if computed by dynamic

analysis, would indicate inter-rack impact which is not an acenptable design 1

condition. Inter-rack impacts were found to not' ' occur by wide margins of safety as described in the "results" section of this report. l 6.0 STRUCTURAL ACCEPTANCE CPI'1ERIA l There are two sets of criteria to be satisfied by the rack ) modules: a. Kinematic Criterion This criterion seeks to ensure that the rack is a i physically stable structure. Vogtle racks are designed to preclude inter-rack impacts. Therefore, physical i stability of the rack is considered along 'with the criterion that inter-rack impact or rack-ro-wall f impacts do not occur. g 'q i c b. Stress Limits The stress limits of the ASME

Code, Section

III, 3

Subsection NF, 1983 Edition through Summer 1984

Addenda, are used since.this code provides the most I

appropriate and consistent set o f. limits for stress types and various loading conditi,cus.ylrious The following loading combinations are applicabl6 'Ref. 1). l s l l f i l 1

]y _,3 _s fj n ,1 i .4 - <L o 8' kl ' I f. i Loadina Combination. Stress Limit: 1

e,.

D + L-([ Level'IA service: limits D + L + To- .i. + D + L +.To + E1 J' 'd-. v' ......-...mg .( D +~L + Ta +-E ' Level' B service -limitr y ' ~ D + L + To'+ P _ f r = hp D + L + Ta + E ' Level D seriice.L. limits: 'jd D + L +~Fd The functionti 'c'apability., i t ^" of the'fu61.Isckereshould bedemonstratc{.{}; } g, j. t.\\ ' ,. 7: + where: 1

• w.

}; .// D =- ~ Dead weight-induced stresses (includinci fuel assembly weight) ~ d Live Load. ( 0.- for the structare,tvinceINe<re l .L' = are 'no moving. objects in the rac)EloadSynth). n, . ~ caused by -' '\\.:he ' accidental * ' drop- @tthe., a. li.. Fa Force = heaviest load. fromL t;he ' maximum -possible' height. ..3 e .) Pf Upward. force on the-. racks Ecaused by- = postulated stuck fuel' assembly Operating Basis. Earthquake [ l 5., E = l E' Safe Shutdown Eaithquake = To Differential temperature inducsd -loads = .(normal or upset condition)' f,. s. Ta Differential tempsrathte-induced loads L., = (abnormal-design conditions) i . 7. y 'f t !- ' L j. l_l l li 1 /' ~'3 4e i

j i em u

e t.3 + u, L' __;-_

i 1 l I i The conditions T and To cause local thermal stresses to be 2 produced. The verst 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 ara ~ asntmed to contain no fuel. The heated water makes unchettructed contact with the inside of the storage walls, themby producing the maximum possible temperature difference betwean 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. a 7.0 MATERIALLPROPEKhiS_ l l l The data on the physical properties of the rack and support materials, obtained from the ASME Boiler & Pressure Vessel Code, { t Sebtlon III, appendices, and supplier's catalog, are listed in j i '2 ables 3 and 4..Since the maximum pool bulk temperature (except for the full core discharge case) is less than 1500, this is used I 1 as the reference design temperature for evaluation of material p:operties. 8.O FTFES3 IIMITR FCR yARIOUS CONDITIONS The following atress limits are derived from the guidelines of the KSFf Code, Section III, Subsection NF, in conjunction with the matur;ial properties data of the preceding section. 8.1 l'ormal and Unset Conditions (Level A or Level B) a. Allowable stress in tension on a net section =Ft =.0.6 Sy or Ft (0.6) (23,150) = 13,890 psi (rack material) l Vt = is equivalent to primary membrane stresses l 1 l I

d Table 3 RACK MATERIAL DATA Young's Yield Ultimate Modulus Strength Strength Material ' El(psi) Sy (psi) Su (Psi) 304L S.S. 27.9_x 106 23150 -68100 Section III Table Table Table Reference I-6.0 I-2.2 I-3.2-t Table-4 SUPPORT MATERIAL DATA-Material 1 ASTM-240,. Type 304L-27.9 x 106 23,150 68,100 (upper part of support psi-Ps i.- psi feet) 2 ASTM 564-630 27.9 x 106 110,650 140,000 psi psi psi 4 e

d. Maximum. allowable bending stress at-the outermost i fiber.due to flexure about one plane of symmetry: Fb = 0.60 Sy = 13,890 psi (rack body). Fb = 13,890 psi (upper part of support feet) = 72,360 psi (lower part of support feet) Combined flexure and compression: e. fa Cmx f x Cmy by b f + + <1 Fa DxF x DFy by b where: i fa Direct compressive stress in the = section fx Maximum flexural stress along x-b = axis fby Maximum flexural stress along y- = axis Cmx Cmy = 0.85 = Dx = 1 - I'ex fa Dy=1- ) F'ey j where: 2 12 n g F'ex ey = i p 23 ( I) rbxty and the subscripts x,y reflect the particular bending plane of interest. l' o __

7 Ft = (.6) (23',150) = 13,890 psi (upper part of support feet) = (.6) (110,650).= 66,390 psi (lower part of. support feet)' b. On the gross section, allowable stress 'in shear is: Fy =.4 S (.4 )y(23,150). = 9,260 psi (main rack body) ( '. 4 ) (23,150) = 9,260. psi'(upper part.of Fy = support feet) Fy = (.'4) (110,650) = 44,260 psi (lower part'of support feet) Allowable stress in. compression, Fa: c. [1 -/ ki g 2 2 ( ) 2c ) s. c y Fa " 5 kl kl 3 3 [( ) + [3 ( ) 8Cc] ,[( ) 8C -)) c 3 r r where: (2n2 g) i2 f cc= ( ) Sy k1/r for the main rack body is based on the full height and cross section of the honeycomb region. Substituting numbers, we obtain, for both support leg and honeycomb region: Fa =,13,890 psi (main rack body) Fa = 13,890 psi (upper part of support feet) = 66,390 psi (lower part of support feet)

w-f. Combined flexure and compression (or tension): fa fx fby b -+ + < 1.0 0.6S Fx Fby b y The above requirement should be met for both the direct tension or compression case. 8.2 Level D Service Limits F-1370 (Section III, Appendix F), states that the limits for the Level D condition are the minimum of 1.2 (S /F ) or (0.7Su/F ) y t t times the corresponding limits for Level A condition. Since 1.2 S is less than 0.7 Su for the lower part of the support feet, y the multiplying factor for the limits is 2.0 for the SSE condition for the upper section. The factor is 1.48 for the lower section under SSE conditions. Instead.of tabulating the results of these six different stresses as dimensioned values, they are presented in a dimensionless form. These so-called 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 OBE and 2.0 or 1.48 for the SSE condition. i 8.3 Bearine Lead on Concrete Floor In accordance with GPC/Bechtel specifications, the fuel racks are limited to a bearing pressure of 4760 psi on the concrete under liner plate in areas which are over 5" from the centerline of any leak chase, and 2380 psi over areas which are within 5" of the leak chase centerline. No credit for bearing area can be taken for any bearing surface over a 1.25" width on either side of the leak chase centerline. Supports over the centerline of the leak chase should be designed to preclude liner plate bending. h t

s Suitable bearing plates are incorporated in the design to meet the above surface pressure criterion. As would-be expected, the governing criterion for si::ing the bearing plate follows from the consolidated storage condition. 9.0 RESULTS Numerical results are abstracted here for all cases described in Appendix A. The following terms of stress factors are needed to interpret the output data. R1 Ratio of direct tensile or compressive stress on a = net section to its allowable value (note support feet only support compression) Ratio of gross shear on a net section to its-R2 = allowable value Ratio of maximum bending stress due to bending R3 = about the x-axis to its allowable value for the section l R4 Ratio. of maximum bending stress due to bending { = l about the y-axis to~its allowable value i R5 Combined flexure and compressive factor = Combined flexure and tension (or compression) R6 = factor 1 As stated before, the allowable value of Ri (i =1,2,3,4,5,6) is l 1 for the OBE condition, and 2 for the SSE (exceot for the lower section of the sunoort where the factor is 1.48) l l The dynamic analysis gives the maximax (maximum in time and in space) values of the stress factors at critical locations in the rack module. Values are also obtained for maximum rack l l ,G- 'V i

s displacements and for critical impact loads. ~ Table 5 presents the dimensionless stress factors for all " wet rack" dynamic analyses. Tables 6 and 7 give the horizontal and vertical displacements, and vertical loads for all cases. It is found that the results corresponding to SSE are most critical vis-a-vis the corresponding allowable limits. The results given herein are for the SSE. The maximum stress factors (Ri) are below the limiting value for the OBE condition for all sections. It is noted that the critical load factors reported for the support feet are all for the upper segment of the foot and are to be compared with the limiting value of 2.0. Analyses (not included here) have been carried out to show that significant margins of safety exist against local deformation of the fuel storage cell due to rattling impact of fuel assemblies. 10.0 IMPACT ANALYSES 1 10.1 Incact Loadina Between Fuel Assembly and Cell Wall The local stress in a cell wall is estimated from peak impact leads obtained from the dynamic simulations. Plastic analysis is used to obtain the limiting impact load that can be tolerated. We find that the maximum fuel assembly to cell impact load is 2576 pounds (Table 5) which is a small fraction of the allowable limit load of 7442 pounds (calculated with a facter of safety of two over the collapse load). 10.2 Impacts Between Adiacent Rocks l All of the dynamic analyses assume, conservatively, that adjacent racks move completely out of phase. Thus, the highest potential

4 TABLE 5 RACK

SUMMARY

NORMAL FUEL - FLOODED POOL, STRESS FACTOR VALUES STRESS FACTORS (DIMENSIONLESS) NET RUN # FUEL ASSEMBLY (Upper values for rack base TO CELL. IMPACT Lower values for Support Feet) LOAD (#) R1 R_ R3 R4 R5 R-2 6 (PER ASSEMBLY) 1 5.372 x 104 .121 .059 .310 .135 .391 .445 (120 Cells) (2311 lb/ cell) .343 .281 .396' .138 .654 .714 2 5.336 x 104 .120 .033 .110 .106 .248 .271 (120 Cells) (2296 lb/ cell) .301 .088 .121 .084 .399 .417 3 5.277 x 104 .105 .045 .283 .293 .443 .506 (99 Cells) (2500) .297 .227 .265 .204 .480 .520 4 5.048 x 104 .108 .023 .094 .093 .232 .254 (99 Cells) (2391) .232 .059 .065 .082 .276 .287 5 5.437 x 104 .110 .044 .272 .166 .386 .441 (99 Cells) (2576) .32 .187 .200 .103 .439 .459

1 8 9 Table 5 (continued) STRESS FACTORS (DIMENSIONLESS) NET RUN $ ' FUEL ASSEMBLY (Upper vclues for rack base TO CELL IMPACT Lower values for Support Feet) LOAD (#) R1 R2 R3 R4 R5 R6 (PER ASSEMBLY) 6 5.237 x 104 .106 .026 .093 .121 .200 .219 (99 Cells) .240 .049 .079 .105 .331 .347 i 7 5.093 x 104 .103 .054 .329 .168 .405 .459 (110 cells) (2292) .345 .208 .283 .130 .514 .545 l 8 5.069 x 104 .114 .027 .096 .120 .226 .246 ) (110 cells) (2281) .236 .063 .095 .078 .331 .343 9 N/A .007 .009 .049 .049 .083 .097 (empty rack) .068 .057 .072 .066 .115 .126 l 10 N/A .005 .006 .029 .025 .034 .040 (empty rack) .030 .009 .014 .014 .033 .034 1 i i l l -al' I I e L

9 for inter-rack impact is achieved. The displacements obtained l from the dynamic analyses are less than 50% of the rack-to-rack spacing or rack-to-wall spacing. Therefore, we conclude that no impacts between racks or between racks and walls occur during the seismic events. 11.0 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 l simulation codes are surveyed and the maximum loading is used to qualify the welds on these locations. 11.1 Baseelate to Rack Welds and Cell-to-Cell Welds ( Section NF permits, for the SSE condition, an allowable weld stress I = .42 Su 27,300 psi. The calculated weld stress, = based on the highest load factor, is 10370 psi is well-below the limiting value for the baseplate to rack welds. The critical area that must be considered for cell-to-cell' welds is the weld between gap channels and tubes. Where skip welding is used, this weld is continuous near the baseplate but is a skip weld (4.5" inch per 30 inch length) as we move up the tube. The critical shear stress in this area for the SSE condition is less than 6980 psi where account is taken of the skip welds used in this region. Near the bottom of the rack where this shear stress dominates, there are no other stresses on the weld which need be considered. S e

e 4 i l 1 \\ l Stresses in the channel-to-cell welds may also develop due to l fuel assembly impact with the cell wall near the top of the rack. This will occur if fuel assemblies in adjacent tubes are moving out of phase with one another so that impact loads in two j adjacent cells are in opposite directions which would tend to separate the channel from the tube at the weld. Our analysis shows that the maximum weld shear stress in this area is less l than 2071 psi. 1 11.2 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 above the value associated with all surrounding cells. The thermal gradient in the vertical direction is assumed to be IJnear. Using a 0 differential temperature of 50 F we show that the maximum weld stress near the top of the cell is less than 70% of the material yield stress. It is noted that this stress occurs near the top of the rack when the seismic induced stresses are negligible. i 11.3 Suneert Leo to Rack Baseolate Weld j The most critical combination of shear and moment is considered to act on the weld at this location. The weld available, { exclusive of the gusset bracing, is two groove welds and an outside fillet weld. An elastic stress calculation shows a weld j stress of 18955 psi which is less than 27,300

psi, the NF allowable for the material.

Calculations in the archival report, based on ultimate strength, show an even larger margin. l l 7k WI' l _______________J t

4 i 12.0

SUMMARY

OF POSTULATED ACCIDENT CASES 1. 5000A Uolift on Corner of Rack The stress at the rack base is less than 160 psi. a. i b. No yielding will occur in the cell wall at the load point if the load is spread over a distance greater than 2.88". If. any local yielding does occur it will be confined to a region well above the top of the active fuel. 2. Drooned Fuel Assembly (16474) Impactino the Too of the rack after a 36" droo (in air) The maximum normal local stress above the active fuel i region is

23844, which is below the dynamic yield stress estimated as 115%.of the static yield stress.

If local buckling does occur, the shear area.to support the impact loads such that high local shear stresses will exist only in a depth of 7.53". There is no safety concern since this is above the active fuel region. l 3. Drocoed Fuel Assembly (16478) Imoactina the j Baseolate after a 205" dron (in water) We show that the baseplate will not be punctured, and that baseplate lateral deformation, if it occurs, will not exceed 1.59". Since this is less than the minimum distance from baseplate to liner (4.75"), no damage to the liner will result. This is the primary safety concern in the wet pool environment. 13.0 ANALYSIS FOR STORING NEW FUEL IN THE HIGH DENSITY RACKS IN A DRY POOL Referring to Table 9, the maximum horizontal displacements of the Al rack is calculated for the bounding conditions of empty, full, checkerboard and half checkerboard. We show that under dry fuel conditions, the rack stress factors do not exceed specified requirements, and that the fuel rack and fuel assemblies maintain integrity during a seismic event. Local bearing stresses on the pool floor are also well within the code allowables. Tables 8, 9 and 10 contain the detailed output data. -?O'

i TABLE 6 RACK

SUMMARY

NORMAL FUEL.- FLOODED, MAXIMUM' HORIZONTAL DISPLACEMENTS 1 RUN # REMARKS MAX. DISP.

MAX. DISP.

DX-(IN.), DY.(IN.)' (rack short (rack long-direction)' direction) 1 D-2, SSE, .2525 .4492-(D290) Full Fuel Load-Cof =.8 (Regular Fuel) 2 D-2, SSE, .2581 .2293 (D291) Full Fuel Load Cof = .2' Regular Fuel 3 A-1, SSE .4597 .3351 (A196) Full Fuel Load Cof = .8-Regular Fuel 2' Racks in Pool 4 A-1, SSE .3978 .4024 (A197) Full Fuel Load 'i Cof =.2 Regular Fuel 2 Racks in' Pool 5 A-1, SSE .2196 .4765 (A199) Full Fuel Load Cof =.8 4 Racks in Pool 6 A-1, SSE .1554 .2549 (A199) Full Fuel Load Cof = 0.2 4 Racks in Pool - 3 l '"

r ;l 3 e ' 1 Table 6-(continued) RUN # REMARKS MAX.-DISP. MAX. DISP. DX (IN.) DY (IN.) (rack short (rack long direction) direction). 7 B-3, SSE .2815 .4612 B390 Full Fuel Load Cof =.8 '8 Racks in Pool 8 B-3, SSE .2361' .1852 B391 Full Fuel Load l Cof .2 8 Racks in Pool 9 A-1, SSE 0.258 .353 Empty Cof - 0.8 41 Racks in Pool 10 A-1, SSE 0.101 .142 Empty Cof = 0.2 4 Racks in Pool I 1 l i l j ~3 1 - i

i. t .j g 1 TABLE 7 NORMAL FUEL, FLOODED POOL, VERTICAL REACTIONS & DISPLACEMENTS MAX. FLOOR MAX. VERT. LOAD (#) MAX. FLOOR LOAD-FT. VERTICAL {#) RUN # DISP. (IN.) 4 FEET. l 1 .1369 6.439 x 105 1. 1.929 x 105 (D290) 2. -2.305 x 105 3, 2.118 x 105 4. 2.282 at 105 2 .0557 6.403 x 105 1. 1.718 x 105 (,D291) 2. .1.660 x 105 3. 2.023 x 105 4. 1.747 x 105 3 .0702 4.702 x 105 1. 1.785 x 105 ^(A196) 2. 1.995 x 105 3. 1.674 x.105 4. 1.636 x'105 4 .0454 4.78 x 105 1. 1.389 x 105 (A197) 2. 1.448 x 10 ' 5 3. '1.336 x 105-4. '1.562 x 105 5 .0562 - 4.879 x 105 1. 2.149 x 105 (A199) 2. 1.68 x 105 3. 1.561 x 105 4. 1.579 x 105 1 6 .0376 4.74 x 105 1. 1.611 x 105 j (A199) 2. 1.311 x 105 3. 1.271 x 105 i 4. 1.313 x 105 .$b'

e 1 l l Table 7 (continued) 1 MAX. FLOOR MAX. VERT. LOAD (#) MAX. FLOOR LOAD {#) RUN # DISP. (IN.) 4 FEET FT. VERTICAL I 7 .0869 5.076 x 105 1. 2.318 x 105 2. 1.846 x 105 3. 1.965 x 105 4. 1.687 x 105 8 .0417 5.617 x 105 1. 1.786 x 105 2. 1.468 x 105 3. 1.588 x 105 4 4. 1.381 x 105 9 .072 5.797 x 104 1. 3.933 x 104 l 2. 3.524 x 104 l 3. 4.581 x 104 l 4. 3.841 x 104 j 10 0 4.761 x 104 1. 2.039 x 104 2. 1.630 x 104 l 3. 1.688 x 104 4. 1.978 x 104 ) l l l ? ' d L__________.__

j TABLE 8 j RACK

SUMMARY

, Al RACK NORMAL FUEL - DRY, STRESS FACTORS STRESS FACTORS RUN # FUEL ASSEMBLY (Upper values for rack base TO CELL IMPACT Lower values for Support Feet) LOAD (#) R1 R2 R3 R4 R5 R6 1 4.275 x 104 .117 .036 .179 .217 .302 .341 ('.A191) (99 Cells) (2025) .245 .126 .135 .138 .294 .305 2 2.974 x 104 .089 .021 .121 .146 .214 .239 (A192) (99 Cells) (1409) .183 .052 .053 .073 .221 .228 3 N/A .043 .030 .155 .159 .277 .320 ] (A195) ] i .216 .187 .225 .212 .478 .525 l 4 N/A .012 .004 .021 .025 .038 .043 (A194) .030 .009 .010 .014 .036 .037 5 1.312 x 104 l .031 .012 .059 .067 .103 .116 (A193) (32 Cells) f (3333) i .093 .036 .033 .043 .106 .110 t - ]$ $5 ' i _J

4 Table 8 (continued) STRESS FACTORS-RUN # FUEL ASSEMBLY (Upper values for rack base TO CELL' IMPACT . Lower values for Support Feet) LOAD (#) R1 R2 R3 .R4 R5 R6 6 1.012 x 104 .032 .009 .044 .053 .090 .101 j (A190) (32 Cells) .l (2571) .078 .021 .020 .023 .098 .1021 ] 7 6657 .026 .015 .079 .075 .114 .130 ('a189d) (4854) .083 .058 .081 .065 .136 .145 ) 8 6.657 x 103 .022 .006 .067 .036 .095 .109 3 '(a188d) (16 cells) (4854) .063 .017 .024 .021 .080 .083 l l )

TABLE 9 RACK

SUMMARY

NORMAL FUEL - DRY (A-1 RACK),_ HORIZONTAL DISPLACEMENTS 1 RUN i REMARKS MAX. DISP. MAX. DISP. DX:(IN.) DY (IN.); 1 Cof =.8 .3596 .3142. (A191) SSE,.99 Cells Filled '2 Cof =.2 .6759 .4649 (A192) SSE, 99 Cells Filled 3 Cof =.8 2.5966 1.4577 ~ (a195) SSE, Empty 4 Cof =.2 2.5994 3.0246 (A194) SSE, Empty I 5 Cof =.8 .0722' .0698 (A193) SSE, 32 Cells-with Fuel 6 Cof =.2 .4154- - 3073 (A190) SSE, 32 Cells Filled 7 Cof =.8 .247 .245 (a189d) 16 Cells in Pos. y Quad 8 Cof =.2 3.5946 2.4014 (a188d) 16 Cells in Pos. y Quad. )'

3 e TABLE 10 Al RACK, DRY CONDITION NORMAL FUEL, VERTICAL LOADS-&-DISPLACEMENTS q J MAX. FLOOR MAX. VERT. LOAD (#) MAX. FLOOR LOAD.{#', RUN # DISP. (IN.) 4 FEET FT. VERTICAL l j 1 .0391 4'.89 x 105 1. 1.653 x 105 l (A191) 2. 1.406 x 105 .) 3. 1.292.x 105. / 4. 1.512:x 105 2 .0027 3.719 x 105 li 1.23 x 105 ('192) 2.

9. 8 69 ]c 104 A

3. 1.110 x 105 1 4. 1.178 x 105 .) l 3 .4569 1.792 x 105 1. -1.095 x 105 (a195) 2. 1.206 x 104 3. 1.454 x 105 4. 1.341 x 105 4 0. 4.979 x 104 1. 2.051 x 104 (A194) 2. 1.699 x 104 l 3. 1.801-x 104 l 4. 1.932 x 104 5 .0004 1.288 x 105 1. 5.613 x 104 (A195) 2. 4.443 x 104 3. 4.543 x 104 4. 5.078 x 104 6 0. 1.327 x 105 1. 5.27 x 104 (A190) 2. 4.012 x 104 3. 4.248 x 104 4. 4.714 x 104 i j -i 21' __m__.._..._

r 1 Table 10 (continued) l 1 MAX. FLOOR MAX. VERT. LOAD (#) MAX. FLOOR LOAD {#) RUN #- DISP. (IN.) 4 FEET FT. VERTICAL 7 .055 1.093 x 105 1. 4.237 x-104 (a189d) 2, 5.352 x-104 3. 5.598 x 104 1 4. 3.836.x 104 8 .0169 9.042 x 104 1. 2.419 x 104 (a188d) 2. 3.383 x 104 3. 4.322 x 104 4. 2.901 x'104 1 3

l 1 13.1 Results-The Al rack is assumed leaded with regular fuel (new) and subjected to the SSE seismic event. The cases 1 thru 8 in Appendix A describe the various loading scenarios (horizontal earthquakes

VH1, VH2, and vertical quake VV).

. Cases for coefficients of friction p .8, . 2 _ are run and the results = summarized in Tables 8 and 10. It is seen that the critical' load factors fo2/ separate. and combined loading are all below the limiting values for the SSE condition as well as the. OBE i condition. Note that the maximum vertical and shear loads reported for a single foot do not necessarily occur'at the.same instant of time. 14.0 CONSOLIDATED STORAGE CONDITION In the next revision of this report. { 1 1 15.0 DEFINITION OF TERMS USED IN SECTION 6 ) S1, S2, S3, S4 Support designations Pi Absolute degree-of-freedom number i i qi Relative degree-of-freedom number i y Coefficient of friction Ui Pool floor slab displacement time history in the i-th direction x,y coordinates horizontal direction z coordinate vertical direction K I Impact spring between fuel assemblies and cell 0'

c. K Linear component of friction spring'.. f KS Axial. spring of support. -leg -locations N Compression' load in a support foot K. R Rotational spring provided by the pool slab' Subscript 1-W h e n.u s e d ' w i t h' U or - X indicates direction ~(i = 1 x. direction, i-= 2 y-direction, i'= 3 z-direction) l I l I i l 1 L 1 -_-_-___ - - _ O

s e j REFERENCES i 1. NRC 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 of January 18, 1979. 2. ASME Boiler & Pressure Vessel Code, Section III, Subsection NF (1983 with Summer 1984 Addenda. l l 3. USNRC Regulatory Guide 1.29, " Seismic Design Classification," Rev. 3, 1978. l 4. " Friction Coefficients of Water Lubricated Stainless Steels for a Spent Fuel Rack Facility," Prof. Ernest Rabinowicz, MIT, a report for Boston Edison Compa'ny, 1976. 5. USNRC Regulatory Guide 1.92, " Combining Modal Responses and l Spatial Components in Seismic Response Analysis," Rev. 1, February, 1976. 6. "The Component Element Method in Dynamics with Application to Earthquake and Vehicle Engineering," S. Levy and J.P.D. Wilkinson, McGraw Hill, 1976. 7. " Dynamics of Structures," R.W. Clough and J. Penzien, McGraw { Hill (1975). 1 L 8. " Mechanical Design of Heat Exchangers and Pressure Vessel Components," Chapter 16, K.P. Singh and A.I. Soler, Arcturus Publishers, Inc., 1984. 9. 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. 10. " 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.

9A'

8 ) e l References (continued) 'I l q 11. USNRC Regulatory Guide 1.61, " Damping Values for Seismic Design of Nuclear Power Plants," 1973. 12. USNRC Standard Review Plan, NUREG-0800, SRP-3.8.4'(1981). 13. " Design Basis, Planar Motion ansd 3-D Analysis with Cross l Coupling Coefficients",'Holtec International Report.No. BI- .j 87117 (April 1987). i 14. Archival Reports, Holtec HI-87188, 87189-(1987). i s 4 l f

f, ^ y f [y ,y d[v ig E a B ON O ITA f R i E L 1 9 E f 7 C s 1 C A 1 gL 6 E)7 ( 9 e 1 G H E D (VN L I B i L LA \\ ET UN FO Z EI L R TO l GH OV t f j h i 6 4 2 0 2 4 O 0 0 0 0 0 0 zO aeo m ~ if

hL' ft.v l" h T %1 .V F nn il i) E i\\ B ,M i ON l O (I TI b '( A a1 R E ,y L l 9 E I 7 C J 1 C I A 1 L \\ 5 &E )7 l e 2 9 G H E D L (V N B L LA ET ci UN 1 FO 1 Z EI L R l TO J GH O V l \\I \\ ./ w \\ t R y V l \\ A 4 3 2 9_ 1 2 3 4 1 C 0 0 0 0 o 0 0 0 0 nov ZO x,._ d O t ?g -

i r ? ^y E d*y B O ?'4 Y ?*y R O T I y f S b, H I 9 7 l 1 E M \\ Q ET l L I y ~ f G )V j D ) L V B( f I L L E A l i C UI l FT R EE y LV T G I O l V I7 l [ f, N"y 6 4 2 0 2 4 6 r. 0 0 0 0 0 0 0 0 Z OOC 2" ~ ll l

.r! l l -i. g ~ 4 1. _ CO 6 .a [ s 5. e, cR / t.~ -c:::= _O 2::=. e-i i W 7 U*i -sc

• m t/f y

%y <x 4 'IC Ig pH 1E ik 2 n _N m a' i

h. W g

v e2 x Z ~ ~ J bY O i H O (.. W _.O y m c' N g _J OI v Q_J v> W L_- cc 2 J -O J< w-DZ LO N -e LY. J1-O 3 O "C ""~ ) O -+ y N E ) 1 s -N 6 i i i i i 4 O O t. N. O. N. t. O. (' O O O O O O O i l I (0) NOI.1.VB313OOV _u,- Figure 4 L______------

0 r' 2 a W A Y A V 8 x' a 1 T m A W n Y 6 A. 1 i i 7 E A y S 1 S g y 4 M i Y R k 1 p O I T l j l L S j 2 \\ ) S I 9 i H l' l 1 D 7 N I 1 O E 1 LM C I I 0 E E T i e l 1 S G) ( 1 DV i LV E B( I M 8 i I L T LE A jj U C I FT s I R 6 EE j LV T l G \\ h O 1 a j 4 V ] ny A y A y 2 i A y n y O 6 4 2 0 2 4 6 0 0 0 0 0 0 0 O '" O ~IyU"oW ~; ?"g w je'

^ ^ "" A' N" A'J"W" E '1 S SN \\ l O I h T j - A f j i - R E ' / L \\ 9 E 1 7 C 1 C / A L l @ E)5 1 8 1 G H E DL (VN B L LA ET UN I l FO l' l Z EI L R I TO GH O V l l l 1 { U 6 4 2 0 2 4 6 0 0 0 0 0 0 0 C ^85 r U.O t 5$ m

L j 43 / / 2* 4

p8 p7 H/4 I

g6 3* a RACK GEOMETRIC 9 CENTER LINE N H/4 H 4 >P12 n H/4 ll r5* -- p 14 b / b + S4 p13 "I# A5 i

1. Y*Q2 p

j A X y B / / t Y 6 B ~ je Ur LONG DIRECTION / / SUPP'T 8' 81 2 j 44 TYP. FRICTION ELEMENT 4 41 SCHEMATIC MODEL FOR DYNRACK -DD' FIGURE '7 '

i TYPICAL TOP IMP ACT ELEMEjBT I h JJ I;~ M, N W$ \\ s 1, :N J a 4 i H i h~ M / 7 \\ v t i 1 RACK STRUCTURE TYP. BOTTOM IMP ACT l ELEMENT h 1 ^ w a',: /) s $N M r l jM t~. R y< #77 r l, /R7 p l R ACK TO.R ACK IMPACT SPRINGS l 5j ' FIGURE 8

t s + 4 Y CELL WALL F '4 'M ASS i l Xg g. __7 / [ FUEL ASSEMBLY / CELL i / IMPACT SPRING / / ~ / / YB / / / / / / \\ > x 'i I ~I l l I l ll l l IMPACT SPRING ARRANGEMENT g. AT NODE i 'e , FIGURE 9-

j FUEL ASSY/ CELL IMPACT ' SPRING, K -.4 h A 1 0.25H H/2 -view V l0.258 h l RACK \\ j it s q p 1f \\ ' h h I I 0.25H j H/2 ~ W TYPICAL RATILING MASS 0.25H 6 l 7 y SUPPORT LEG FRICTION E SPRING, K INTERFACE 8 SPRING, Kg ? l M f FOUNDATION ROTATIONAL COMPLIANCE SPRING, K R FIGURE 10 TJO DIMENSIONAL VIEW OF THE SPRING-MASS SIMULATION t_________.__-.

I i Seismic Analysis Matrix . Appendix A 1 " Dry" rack runs: All runs were made'on Module A-1 with normal fuel. The following conditions were. analyzed. l 1 I -Run No. Coefficient of Friction Loading Pattern j 1 0.8 Fully Loaded 2 0.2 Fully Loaded 3 0.8 Empty. 4 0.2 Empty 5 0.8 Modified Checkerboard Pattern-6 0.2 Modified Checkerboard Pattern 1 7 0.8 One half of rack loaded with modified I checkerboard 8 0.2 One half of rack loaded with modified ) checkerboard Flooded Pool Condition The following set of runs were made. Run Module Coefficient of l No. I.D. Friction Loading Condition 1 0-2 .0.8 Fully Loaded 1 2 D-2 0.2 Fully Loaded 3 A-1 0.8 One side wall, one side rack, 2 sides open (two racks in pool condition)

Run Module Coefficient of No. I.D. Friction Loading Condition 4 A-1 0.2 One side wall, one side rack, 2 sides open (two racks in pool condition). 5 A-1 0.8 One side wall, 2 sides rack, one side open (four racks in pool condition) 6 A-1 0.2 One side wall, 2 sides rack, one side open (four racks in pool condition). 7 B-3 0.8 One side wall, 2 sides racks, I side open (eight racks in pool condition) 8 B-3 0.2 One side wall, 2 sides rack, 1 side open (eight racks in pool condition) Special Note: Run the worst of the above cases (max. displacement case) with rack empty (one run). l l I l u__________

X Z X X X X Z Z X XR X Z X X XD X X X X X ,EI l X X X X XE Z X X X XM i l

X X X X X X X X X X E,X X X X ,I i t

-e o i i a CRITICALITY SAFETY ANALYSES FOR V0GTLE ELECTRIC GENERATING PLANT SPENT FUEL STORAGE RACKS-i s 9 ___m._m__-_--

i 1 1 \\ l TABLE OF CONTENTS Pace i 4.0 CRITICALITY SAFETY ANALYSES............................ 4-1 4.1 DESIGN BASES...................................... 4-1 i 4.2

SUMMARY

OF CRITICALITY ANALYSES................... 4-3 I 4.2.1 Normal Operating Conditions............. 4-3 i 4.2.2 Abnormal and Accident Conditions........ 4-5 I 4.2.3 New Fuel Storage in Dry Condition....... 4-6 4.3 REFERENCE FUEL STORAGE CELL....................... 4-7 4.3.1 Reference Fuel Assembly................. 4-7 4.3.2 Fuel Storage Cells...................... 4-7 4.4 ANALYTICAL METHODOLOGY 4-9 1 f 4.5 CRITICALITY ANALYSES AND TOLERANCE 4-12 VARIATIONS........................................ i 4.5.1 Nominal Design Case..................... 4-12 4.5.2 Uncertainties Due to Manufacturing 4-12 Tolerances.......................g 4.5.2.1 Boron Loading Variation................. 4-12 4.5.2.2 Storage Cell Lattice Pitch Variation.... 4-13 1 4.5.2.3 Boraflex Width Tolerance variation...... 4-14 4.5.2.4 Boraflex Integrity...................... 4 4.5.2.5 Stainless Steel Thickness Tolerances.... 4-15 i 4.5.2.6 Fuel Enrichment and Density Variation... 4-15 4.5.2.7 Eccentric Positioning of Fuel Assembly 4-15 in Storage Rack......................... 4.5.3 Reactivity Effects of Boraflex Axial 4-16 Length 4.6 ABUORMAL AND ACCIDENT CONDITIONS 4-17 4.6.1 Temperature and Water Density Effects... 4-17 4.6.2 Dropped Fuel-Assembly................... 4-18 4.6.3 Abnormal Location of a Fuel Assembly.... 4-18 4.6.4 Seismic Event........................... 4-18 4.7 FUEL STORAGE UNDER DRY CONDITIONS................. 4-20 REFERENCES 4-21 APPENDIX A - Benchmark Calculations

LIST OF TABLES h EASLt 4.1

SUMMARY

OF CRITICALITY SAFETY ANALYSES' 4-4 4'. 2 REACTIVITY EFFECTS.OF ABNORMAL.AND. ACCIDENT 4-5 CONDITIONS 4'.'3 FUEL ASSEMBLY DESIGN SPECIFICATIONS 4-8 4.4 EFFECT:OF TEMPERATURE'AND VOID ON CALCULATED-4-17 REACTIVITY OF STORAGE RACK LIST OF FIGURES M* Pacte 4.1 VOGTLE ELECTRIC GENERATING PLANT 4-8 / q

e 4.0 CRITICALITY SAFETY ANALYSES 4.1 DESIGN BASES The high density spent fuel storage racks for the Vogtle Electric Generating Plant are designed to assure that the neutron multiplication factor (keff) is equal to or less than 0.95 with the racks fully loaded with fuel of the highest anticipated reactivity, and flooded with unborated water at a temperature corresponding to the highest reactivity. The maximum i i calculated reactivity includes a calculational bias, a margin for uncertainty in reactivity calculations and in mechanical t'olerances. The uncertainties in reactivity calculations and mechanical tolerances are statistically combined such that the true keff 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 Criterion 62, Prevention of Criticality in Fuel Storage and Handling. O USNRC Standard Review Plan, NUREG-0800, Section 9.1.1, New Fuel

Storage, and Section 9.1.2, Spent Fuel Storage.

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. O USNRC Regulatory Guide 1.13, Spent Fuel Storage Facility Design Basis, Rev. 2 (proposed),- December, 1981. 1 l -/' j b

F 1 O USNRC Regulatory Guide 3.41, Validation of Calculational Methods for Nuclear Criticality Safety (and related ANSI N16.9-1975). O ANSI /ANS-57.2-1983, Design Requirements for Light Water Reactor Spent Fuel Storage Facilities at Nuclear Power Plants. O ANSI N210-1976, Design Objectives for. Light Water Reactor Spent Fuel Storage Facilities at Nuclear Power Plants. O ANS-8.17-1984, Criticality Safety Criteria for the

Handling, Storage and I' transportation of LWR Fuel Outside Reactors.

To assure the true reactivity will always be less than the calculated reactivity, the following conservative assumptions were made: I O Moderator is pure, unborated water at a temperature within the design basis range corresponding to the highest reactivity. O Lattice of storage racks is assumed infinite in all directions, i.e., no credit is taken for axial or radial neutron leakage (except in the assessment of certain abnormal / accident conditions). O Neutron absorption in minor structural members is neglected, i.e., spacer grids are replaced by water. O The fuel of highest anticipated reactivity is assumed, i.e., fresh un-burned fuel of 4.55% enrichment, in the Westinghouse 17x17 optimized fuel assembly geometry. 1

1 4.2

SUMMARY

OF CRITICALITY SAFETY ANALYSES 4.2.1 Normal Ooeratine Conditions Based upon the Westinghouse optimized 17x17 fuel assembly design at 4.55% enrichment (52.70 grams U-235 per axial centimeter), the maximum infinite multiplication f actor (b) is O.936, including all known uncertainties (95% probability at the 95% confidence level). Table 4.1 summarizes the results of the criticality safety analyses. Independent calculations for the Westinghouse standard 17x17 fuel assembly resulted in a lower b and the optimized fuel assembly at the most reactive point in life was therefore used for the design basis calculation. I I

.z Table 4.1,

SUMMARY

OF CRITICALITY SAFETY ANALYSES 0 Temperature assumed for' analysis 20 C' Reference km _ ( ANTX-KENO ).

0.9162' Calculationalibias"

0.0106 Uncertainties Calculational 10.0044 Bias 10.0048-

=B-10 concentration 10.0020 Boraflex thickness 't0.0023-Boraflex width 10.0017' Inner box dimension 10.0012 Water gap thickness 't0.0034 SS thickness t0.0005 Fuel enrichment 10.0020 Fuel density 10.0024 Eccentric assembly position: <0.0001 s Statistical combination of 10.0087 uncertainties (1) Total 0.9268 i 0.0089 Maximum reactivity 0.936 (1). Square root of sum of squares of all uncertainties. _4-

j 4.2.2 Abnormal and Accident Conditions Although credit for the, soluble poison normally present in the spent fuel pool water is permitted under abnormal or accident conditions,* most abnormal or accident conditions will .not result in exceeding the limiting reactivity (keff of'0.95) even in the absence of soluble poison. The effects on reactivity of ' credible abnormal and accident conditions are summarized in Table 4.2 below. Of these abnormal / accident conditions, only two. have the potential for a more than negligible positive reactivity effect. Table 4.2 REACTIVITY EFFECTS OF ABNORMAL AND ACCIDENT CONDITIONS ' Accident / Abnormal Conditions Reactivity Effect Temperature increase Negative Void (boiling) Negative Assembly dropped on top of rack Negligible i 1 Lateral rack module movement (seismic) Positive Misplacement of a fuel assembly Positive i Double contingency principle of ANSI N16.1 -19 7 5, as specified in the April 14, 1978 NRC letter.(Section 1.2) and implied in the proposed revision (draft) to Reg. Guide 1.13 (Section 1.4, Appendix A). t a

i Either a major seismic event or the misplacement of a fresh fuel assembly outside and adjacent to a rack module have the potential for exceeding the limiting reactivity should there be a 2 concurrent and independent accident condition resulting in the loss of all soluble poison. Administrative procedures will assure the presence of soluble poison after an earthquake and I during fuel handling operations and preclude the possibility of l the simultaneous occurrence of two independent accident conditions. 4.2.3 New Fuel Storace in Drv Condition i I New fuel is normally stored in the dry condition with a very low multiplication factor. The misplacement of a fuel i l assembly during fuel handling operations has no criticality j consequences under these normally dry conditions. NRC guidelines (SRP 9.1.1) require the evaluation of criticality safety under flooded conditions and for hypothetical low-density moderation. The double contingency principle of ANSI N16.1-1975 (invoked by the NRC April 14, 1978 letter) eliminates the necessity of considering independent and concurrent accident conditions. I I Therefore, the conditions specified in SRP 9.1.1 are the accident conditions evaluated. The storage rack in the flooded condition is the reference design basis case previously evaluated. Witn Boraflex absorber between asserblies, conditions do not exist for the appearance of a peak in reactivity at low moderator densities, and the fully flooded condition corresponds to the highest reactivity (optimum moderation). Calculations at low moderator densities (5% to 15%) confirm the very low reactivities. At the 10% water density where low density " optimum" moderation might otherwise be expected, the infinite multiplication factor is only 0.56. N' L____________._____

4.3 REFERENCE FUEL STORAGE CELL 4.3.1 Reference Fuel Assembly The design basis fuel assembly, illustrated in Fig. 4.1, is a 17 x 17 array of fuel rods with 25 rods replaced by 24 control rod guide tubes and 1 instrument thimble. Table 4.3 summarizes the design specifications and the range of significant variations. A separate calculation, with the Westinghouse standard fuel assembly (see Table 4.3), confirmed that the optimized design exhibited the highest reactivity and was therefore used as the design basis. 4.3.2 Fuel Storace Cells The nominal spent fuel storage cell used for the . criticality analyses is shown in Fig. 4.1. The rack is composed of Boraflex absorber material sandwiched between an 8.75-inch I.D., 0.075-inch thick inner stainless steel box, and a 0.020-inch outer stainless steel coverplate. The fuel assemblies are centrally located in each storage cell on a nominal lattice spacing of 10.40 0.02 inches in one direction and 10.58 t 0.02 inches in the other direction. Stainless steel tabs connect one storage cell box to another in a rigid structure and define an outer water space between boxes. This outer water space constitutes a flux-trap between the two Boraflex absorber sheets that are each essentially opaque (black) to thermal neutrons. The Boraflex absorber has a thickness of 0.075 1 0.007 inch and a-nominal B-10 areal density of 0.0238 gram per em, l 2 _9'

Table 4.3 FUEL ASSEMBLY DESIGN SPECIFICATIONS FUEL ROD DATA QEA STANDARD Outside diameter, in. 0.360 0.374 Cladding thickness, in. 0.0225 0.0225 Cladding inside diameter, in. 0.315 0.329 Cladding material Zr-4 Zr-4 Pellet density, % T.D. 95.0 95.0 P,ellet diameter, in. 0.3088 0.3225 i Enrichment, wt % U-235 4.55 t 0.05 4.55 i 0.05 Stack density, gUO /cc 10.30 i 0.22 10.30 C.22 2 FUEL ASSEMBLY DATA Fuel red array 17 x 17 Number of fuel rods 264 Fuel rod pitch, in. 0.496 Number of control rod guide tubes 24 Guide thimbles, O.D., in. 0.474 Guide thimbles, I.D., in. 0.442 Number of instrument thimbles 1 Instrument thimble, O.D., in 0.474 Instrument thimble, I.D., in 0.442

i 4.4 ANALYTICAL METHODOLOGY In the fuel rack analyses, criticality analyses of the high density spent fuel storage racks were performed with the AMPX-KENO computer package (Refs. 1 and 2), using the 27-group SCALE

• cross-section library (Ref. 3) with the NITAWL (ref. 1) subroutine for U-238 resonance shielding effects (Nordheim integral treatment).

Benchmark calculations are presented in Appendix A and indicate a bias of 0.0106 i 0.0048 (95%/95%). In the geometric model used in KENO, each fuel rod and its cladding were described explicitly and reflecting boundary conditions (zero neutron current) were used in the axial direction and at ] t,he centerline of the water-gap between storage cells. These boundary conditions have the effect of creating an infinite array of storage cells in all directions. The CASMO-2E computer code (Refs. 4, 5 and 6), a two-dimensional multigroup transport theory code for fuel assemblies, has also been benchmarked (Appendix A) and was used both for verification calculations and as a means of evaluating small l reactivity increments associated with manufacturing tolerances. CASMO-2E benchmarking resulted in a calculational bias of 0.0013 t 0.0018 (95%/95%). However, limitations in the gecmetry options ( available in CASMO-2E required minor approximations in the 'i geonetric description (e.g. in the description of the Boraflex absorber and in the use of an average water-gap thickness) which apparently contributes to a small over-prediction in the absolute SCALE is an acronym for Standardized _Q.omputer Analysis for Licensing Evaluation, a standard cross-section set developed by ORNL for the USNRC. i -}' ) d

. - - -. _ _ _ _ _ - - - - - -. - - - - - - - - - - - - - - - - - - - - - - - - - - - ~ - - -, e a value of the CASMO cell infinite multiplication factor. Two group diffusion theory constants were edited in the output of CASMO-2E and used in the one dimensional SNEID* diffusion theory routine to evaluate reactivity effects of the Boraflex axial length. l l A third independent method of criticality analysis, utilizing diffusion / blackness

theory, was also used for additional confidence in results of the primary calculational

) methods, although no reliance for criticality safety is placed on the reactivity value from the diffusion / blackness theory j t,echnique. This technique,

however, is used for auxiliary calculations of the small incremental reactivity effect of eccentric fuel positioning that would otherwise be lost in normal KENO statistical variations, or would be inconsistent with CASMO-2E geometry limitations.

Cross sections for the diffusion / blackness theory calculations were derived from the NULIF l computer code (Ref. 7), supplemented by a blackness theory routine that effectively imposes a transport theory boundary j condition at the surface of the Boraflex neutron absorber. l Shielded cross-sections were then used in the spatial diffusion j theory code, PDQ-7 (Ref. 8), in two dimensions, to calculate reactivities. 1 Comparison of the three independent methods of analysis for { the reference design resulted in the following data which I confirms the validity of the analytical methodology. SNEID is a one-dimensional diffusion theory program for the microcomputer, benchmarked against one-dimensional PDQ-7 calculations. 0-ar

i - i Maximum kco - . I Analvtical Method Bias-corrected kao -(95%/95%) AMPX-KENO 0.9268 1 0.0065 0.933' CASMO-2E 0.9420 t.0.0018 0.944-' Diffusion-blackness 0.939 0.939 theory.(PDQ-7) These ' data' indicate that the geometric' approximations necessary in CASMO-2E contributed to. an over-prediction ' in kan f of. 'abouti,1%. 1 Ak. - l l In this. compariso', 'the calculational uncertainties. listed.were j n ' derived from benchmark analyses (Appendix.'A)

and, for ' KENO,

'l ihcludes the. additional statistical, variation' assoc.iated with' ) Monte' Carlo calculations. Manufacturing uncertainties.. are L. not included in this comparison. 1 t ll - 1 Lt-L - _ -. _

1 \\ l 4.5 CRITICALITY ANALYSES AND TOLERANCE VARIATIONS 4.5.1 Nominal Desian Case i i For the reference design, the AMPX-KENO calculation resulted in a b of 0.9162 1 0.0024 (200,000 histories) wh'ich, I when corrected for the 0.0106 Ak bias and using a one-sided i tolerance factor for 95% probability at a 95% confidence level, resulted in a b of 0.9268 0.0065. Combining this with all known manuf acturing uncertainty factors results in a maximum b of 0.936. Independent calculations with CASMO-2E resulted in a b o'f 0.9420 0.0018, which, when combined with manuf acturing uncertainties gives a maximum kco of 0.948. This value is conservatively high largely because of approximations necessary in CASMO, but is .still less than the 0.95 design bases limit. 4.5.2 Uncertainties Due to Manufacturing Tolerances 4.5.2.1 Boron Loadine Variation The Boraflex absorber sheets used in the storage cells are nominally 0.075-inch thick, with a B-10 areal density of 2 0.0238 g/cm. Independent manufacturing tolerance limits are 0.007 inch in thickness and 10.0089 g/cm3 in B-10 content. This assures that at any point where the minimum boron 3 l concentration (0.1160 gram B-10/cm ) and minimum Boraflex thickness (0.068 inch) may coincide, the boron-10 areal density will not be less than 0.020 gram /cm. Differential CASMO-2E 2 calculations indicate that these tolerance limits result in an incremental reactivity uncertainty of 10.0020 Ak for boron concentration and 10.0023 for Boraflex thickness variations. -lb' lL__

7 4.5.2.2 Storace Cell Lattice Pitch variation l The design storage cell lattice spacing betwaen fuel assemblies is 10.40 t 0.02 in one direction and 10.58 1 0.02 inches in the other direction. A decrease in storage cell lattice spacing may or may not increase reactivity depending upon other dimensional changes that may be associated with the j decrease in lattice spacing. Increasing the water thichness i between the fuel and the inner stainless steel box results in a small increase in reactivity. The reactivity effect of the flux-trap water thickness,

however, is more significant, and 1

decreasing the flux-trap water thickness increases reactivity. Both of these effects have been evaluated for independent design tolerances. l The inner stainless steel box dimension, 8.750 ! 0.03 . inches, defines the inner water thickness between the fuel and l l the inside box. For the tolerance limit, the uncertainty in I reactivity is 0.0012 Ak as determined by differential CASMO-2Tl l calculations, eith km increasing as the inner stainless steel box' dimension (and derivative lattice spacing) increases. l { The design flux-trap water thickness is 1.28 0.04 i inches in one direction and 1.46 0.04 inches in the other l direction, which results in an uncertainty of 10.0034 Ak due to the tolerance in flux-trap water thickness, assuming the water thickness is simultaneously reduced on all four sides. Since the manufacturing tolerances on each of the four sides are statistically independent, the actual reactivity uncertainties l would be less than 10.0034, although the more conservative value has been used in the criticality evaluation. 1 - J L________

l i 4.5.2.3 Boraflex Width Tolerance Variation l 1 The reference storage cell design uses a Boraflex blade width of 7.75 2 0.063 inches. A positive increment in reactivity occurs for a decrease in Boraflex absorber width. For a j reduction in width of the maximum tolerance, 0.063 inch, the ] calculated positive reactivity increment is +0.0004 Ak. However, { to allow for radiation-induced shrinkage in width of the Boraflex I and for possible small edge affects, the width tolerance was increased to 0.25 inches corresponding to an uncertainty of t0.0017Ak. 4,.5.2.4 Boraflex Intecritv_ 1 The stability and integrity of the Boraflex absorber { material under irradiation has recently been investigated (11) and further irradiation testing is currently underway. Available j .information confirms there is no loss of' boron during irradiation { although there is some radiation induced -shrinkage. Under l irradiation, Boraflex becomes a hard ceramic material and apparen.ly shrinks 2 to '2-1/2 percent. At a very high radiation ] dose, there is evidence of a small edge deterioration. In the Vogtle racks, the Boraflex sheets are installed in a gap of sufficient size to allow unimpeded shrinkage and thereby preclude any mechanism that might cause gaps to develop. To allow for shrinkage, the Boraflex sheets are initially 4 inches longer (2.8%) than would otherwise be necessary. Width shrinkage is accommodated by increasing the tolesrance to 0.25 inches from the nominal 0.063 inches. In both cases, shrinkage would increase the boron concentration in l the Boraflex although no credit is taken for this increased i l

J loading. Shrinkage in thickness would not change the B-10 areal-1 density. 4.5.2.5 Stainless Steel Thickness Tolerances The nominal stainless si. eel thicknessL-is 0.075 t 0.005. inch for the inner = stainless steel. box and 0.020 2.0.003 inch for the Boraflex coverplate. The maximum positive reactivity effect of the expected stainless steel thickness tolerance variations, statistically combined, was calculated l(CASMO-2E) to.be 10.0005 hk. 4.5.2.6 Fuel-Enrichment and Density Variation ~ The design maximum enrichment is 4.55'i 0.05 wt% U-235. Calculations of the sensitivity _-to small ' enrichment f variations by CASMO-2E yielded a coefficientLof 0.0040 hk per 0.1 wt% U-235 at the design enrichment. For a tolerance on U-235 enrichment of 0.05 in wt%, the uncertainty on k.'is t0.0020 hk. Calculations were also made with the UO2 fuel density increased to the maximum expected value of ' 10.5 3 g/cm.(smeared I density). For the reference design calculations, the uncertainty l in reactivity is t 0.0024 Ak over the maximum expected range of UO2 densities. 4.5.2.7 Eccentric Positioning of Fuel Assembly in Storace Rack The fuel assembly in assumed to be normally located in the center of the storage rack cell. Calculations with the fuel assemblies assumed to be eccentrically located in the corner of -lh i l ____--___________________J

the storage rack cell (four-assembly cluster at closest approach), indicated a nogi.tgible change in reactivity as determined by differential PDJ-7 calculations. 4.5.3 Reactivity Effects of Boraflex Axial Lencth Based upon diffusion theory constants. edited in the CASMO-2E output (reference design and a special case with. water replacing the Boraflex), one-dimensional axial calculations were. made to evaluate the reactivity effect of reduced Boraflex axial lengths. Reduced length of the Boraflex' leaves small regions of active fuel without poison at each end of the fuel assembly. The i unpoisoned region at each end is referred to as " cutback". l The axial calculations used a thick (30 cm.) water reflector, neglecting the higher absorption of the stainless-steel structural material at the ends of the fuel assembly. .Results of the calculations showed that the keff remains less than the reference km of the storage cells until the axial f reduction in Boraflex length (cutback) exceeds four inches top I and bottom.

Thus, the axial neutron leakage more than compensates for the 4-inch design cutback and the reference km remains a conservative over-estimate of the true k gf.

A 4-inch e cutback is used at the bottom of the rack. However, no cutback is initially used at the top of the rack which provides an I allowance of 4 inches (2.8%) to accommodate radiation-induced shrinkage of the Boraflex without exceeding the allowable cutback. I Y'

I 4.6 ABNORMAL AND ACCIDENT CONDITIONS 4.6.1 Temperature and Water Density Effects The moderator temperature coefficient of reactivity is 0 negative and a conservative moderator temperature of 20 C was assumed for the reference design which assures that the true reactivity will always be lower. Temperature effects on reactivity have been calculated and the results are shown in Table 4.4 Introducing voids in the l water internal to the storage cell (to simulate boiling) j l decreased reactivity, as shown in the table. Voids due to j boiling will not occur in the outer (flux-trap) water region. Table 4.4 j i EFFECT OF TEMPERATURE AND VOID ON CALCUIATED I REACTIVITY OF STORAGE RACK j Case Incremental Reactivity Change, Ak 1 0 i 20 C Reference j 0 50 C -0.005 0 80 C -0.012 0 120 C -0.024 I 0 120 C + 20% void -0.085 1 With soluble poison present, the temperature coefficients of reactivity would differ from those inferred from the data in l Table 4.4 However, the reactivities would also be.substantially l lower at all temperatures with soluble boron present, and the l data in Tab _e 4.4 is pertinent to the higher reactivity unborated Case. I - - _ - -

I l 4.6.2 Droceed Fuel Assembly Accident For a drop on top of the rack, the fuel assembly will come to rest horizontally on top of the rack with a minimum separation distance (Report HI-87182) from the fuel of more than I the 12 inches sufficient to preclude neutron coupling (i.e., an effectively infinite separation). Maximum expected deformation under seismic or accident conditions will not reduce the minimum spacing between fuel assemblies to less than 12 inches. 1 Consequently, a fuel assembly drop accident will result in only j an insignificant increase in reactivity (<0.0001) due to the { 1 separation distance. Furthermore, soluble boron in the pool water d i whuld substantially reduce the reactivity and assure that the ] true reactivity is always less than the limiting value for any conceivable dropped fuel accident. I 1 1

• 4.6.3 Abnormal Location of a Fuel Assembly l

l l The abnormal location of a fresh unirradiated fuel assembly of 4.55% enrichment could, in the absence of soluble poison, result in exceeding the design reactivity limitation (km of 0.95). This could occur if the assembly were to be positioned outside and adjacent to a storage rack module. Soluble poison, however, is normally present in the spent fuel pool water (for which credit is permitted under these conditions) and would l maintain the reactivity substantially less than the design limitation. 4.6.4 During and after the design basis earthquake it is possible (Holtec Report HI-87180) for rack modules to move so as to reduce the spacing between modules to less thau the design 1 -/?-

.) water-gap. For.this condition, credit-for soluble-poison is permitted and will~ assure a k.ff:less than-the design limit'of: 0.95. I 9 i I 1 -) I j -Ll9-

)

a 4 c' ) L__ --__--__--_____-_-_-_-____---__---_-_--------l--------------^- - - - - - - - - - - - - - - - - - - - - - - - - = - - - 2---'----------

1 I 4.7 FUEL STORAGE UNDER, DRY CONDITIONS For storage of new fuel in the dry condition, SRP9.1.1 requires that the maximum reactivity shall not exceed a keff of 0.98 with fuel of the highest anticipated reactivity in place and assuming the optimum hypothetical low density moderation -(i.e., fog or foam). With even relatively weak absorbers between fuel assemblies, however, conditions do not exist (reference'10) for a significant maxima in reactivity. at low moderator densities. AMPX-KENO calculations for an infinite array (no leakage) gave j the following values: l l i 5% water density, b 0.566 i 0.004 1 = 10% water density, b 0.564 t 0.004 = 15% water density, b 0.582 1 0.006 = i A Under the accident condition of - flooding, the k-infinite would be the 0.936 (maximum) as previously defined (Table 4.1). Reducing water density reduces reactivity monotonically. At 95% water density, k-infinity is reduced by A 0.002 Ak and reduced by l NO.080 Ak at 75% water density. Leakage, if included, would i substantially further reduce these values. These very low values confirm the results of Cano, et al., (reference 10) and demonstrate the criticality safety of the Vogtle fuel storage racks under hypothetical low density i moderation in conformance with SRP 9.1.1. Therefore, fresh fuel assemblies of 4.55% enrichment may be safely stored without any j criticality restrictions other than the limitation on enrichment. i ~ 30 "

REFERENCES 1. Green, Lucious, Petrie, Ford, White, Wright, "PSR-63/AMPX-1 (code package), AMPX Modular Code System for Generating Coupled Multigroup Neutron-Gamma Libraries' from ENDF/B," ORNL-TM-3706, Oak Ridge National Laboratory, March 1976. 2. L.M. Petrie and N.F.

Cross,

" KENO-IV, An Improved Monte Carlo Criticality Program," ORNL-4938, Oak Ridge National Laboratory, November 1975. 3. R.M. Westfall et al., " SCALE: A Modular Code System for Performing Standardized Computer Analyses for Licensing Evaluation," NUREG/CR-0200, 1979. 4. A. Ahlin, M. Edenius, H. Haggblom, "CASMO - A Fuel Assembly i Burnup Program," AE-RF-76-4158,- Studsvik report (proprietary). 5. A. Ahlin and M. Edenius, "CASMO - A Fast Transport Theory Depletion Code for LWR Analysis," ANS Transactions, Vol. 26,

p. 604, 1977.

6. M. Edenius et al., "CASMC Benchmark Report," Studsvik/RF 6293, Aktiebolaget Atomenergi, March 1978. 7. W.A.

Wittkopf, "NULIF Neutron Spectrum Generator, Few-Group Constant Generator and Fuel Depletion Code," BAW-426, The Babcock and Wilcox Company, August 1976.

8. W.R. Cadwell, PDQ107 Reference Manual, WAPD-TM-678, Bettis Atomic Power Laboratory, January 1967. 1 9. M.G. Natrella, Experimental Statistics National Bureau of Standards, Handbook 91, August 1963. 10. J.M. Cano et al., "Supercriticality Through Optimum Moderation in Nuclear Fuel Storage," Nuclear Technoloav, Vol. 48, pp. 251-260, May 1980. 11. S.E.

Turner,

" Irradiation Tests of Boraflex", Nusurtec Incorporated, NST-87-107 (Preliminary), November 1987. -2l- _a hA

~ -. -. _ - - -. _ l I l l t i APPENDIX A l BENCHMARK CALCULATIONS I 4 A-1 1

l l. INTRCCUCTION AND SUMMARv l l The cbjective of th is benchmarking study is to verify both the AMPX (NITANL)-KENO (Refs. 1 and 2) me:hodology with the 27-group SCALE cross-section library (Refs. 3 and 4) and the CASMo-I 2E code (Refs. 5, 6, 7, and 8) for use in criticality calcula-tions of high density spent fuel storage ra ck s'. Both calcu-I ~ i lational methods are based on transport theory and have been I benchmarked against critical experiments that simulate typical I spent fuel storage rack designs as realistically as possible. Results of these benchmark calculations with both methodologies are consistent with corresponding calculations report'd in the I e literature and with the requirements of Regulatory Guide 3.41, i Rev. 1, May 197/. Results of these benchmark calculations show that the 27-group (SCALE) AMPX-KENO calculations consistently underpredict the critical eigenvalue by 0.0106 0.0048 Ak (with a 95% proba-bility at a 95% confidence level) for critical experiments selected to be representative of realistic spent fuel storage rack configurations and poisen worths. Similar calculations by Westinghouse suggest a bias cf 0.012 0.00:3, and the results of ORNL analyses cf 54 r e '.a t ive ly " clean" critical experiments show a bias of 0.0100 0.00'.3. Similar ca_:ula:icns with CASMC-2I for clean critical exnerime- :

ss.'.:2d in a :i2s cf 0. 0 0 '.

0.1013 '?54 '? 5 T ). CAS".C-2E an; A".E' ~.::7 in:3 :::::a ric an esicula:itns ci indini 3 arr27s cf ic ne cal _ condicur2 icns antu ve J gnc; ac ?e en: and succe

na:

1

22 Of 1.10'3 = 0 M I. 3 i: the reas nt:1, expec ad :i:2 an unes :2;n:. 5:r CA5:!C ^F :a l cu '.a :i c n s.

" Validation of Calculational Methods f:: Nuclear Criticality Safety. (See also A:1SI N16.9-1975.) A-2 (

=The benchmark. :calcul'tions reported - [here Lindicatel tha t. a

• .either the 2 -group (SCALE)l AMPX-KENO 'or CASMC-2E. calculations r

.j are faccectable for criticality analysis of high ' density spent - s fuel storage racks. -i l 2. AMPX (NITAWL)'-KENO BENCHMARK CALCULATIONS ~ i Analysis of a ' series of. Babcock & Wilcox '(B&W) critical j expe riments ; ' (Ref. 9), which include some with absorber. sheets typical 'of a poisoned spent fuel rack, is summarizedz in ' Table 1, as calculated with AMPX-KENO using the ' 27-group SCALE - cross-section -library and the Nordheim resonance Integral treatment in. NITAWte. The mean for. these calculations. is 0.9894 't 0.0'019, conservatively assuming the larger standard deviation calculated frem the kegg values. With a one-sided.. telerance ' factor i (K e 2.502), corresponding to 95% probability ati a 95% confidence - level (Ref. 10), the calculational bias is +0.0106 with an uncar-tainty of *0.0048. I Similar calculational deviations rep rted by Westinghouse (Ref. 11) are aisc shcun in 'Ta:le 1.and suggest a bias,cf 0.012

• f 0.0022 (95i/951;.

In addi:icn, ORML (Ref. 12 )~ has analy:ed sc=e 54 critical en eri=ents using the sa:e me:hedol gy, cb sining.a 'l 1 bias of 0.0100 = 0.0012 '951/951). These gublished resu;ts -==- 1 i are in g Od ac r:-eme n t wi:n :he results c::ained in tne c resen:- analysis and le.d further :redance t: the val tity of the 27-1

rcu AN ?

'.- ?.I"C calcula: ion 21 medel for use in crt:ic2.ity ana'.7-sis of hign densi:y scen: fuel s:Orace racks.

7ariance ana;7 sis of the da:2 in Ta la i suggeits the possibility thac an uhr.newn fac: r may be causinc. a slign:ly larger varianc2 than.migh: bei expected fr m the Monte Caric statistics alone; However, such a A-3

Table 1 RESULTS OF 27-GROUP (SCALE) AMPX-KINO CALCULATIONS OF B&W CRITICAL EXPERIMENTS Westinghouse Experiment Calculated Calculated-meas. Number keff a keff I 0.9889 0.0049 -0.008 i II 1.0040 0.0037 -0.012 III 0.9985 0.0046 -0.008 IX(1) 0.9924 .00.0 46 -0.016 X 0.9907 . 0.0039 -0.008 XI 0.9989 0.0044 +0.002 XII O.9932 0.0046 -0.013 XIII 0.9890 0.0054 -0.007 XIV 0.9830 0.0038 -0.013 XV 0.9852 0.0044 -0.016 XVI 0.9875 0.0042 -0.015 XVII 0.9811 0.0041 -0.015 X7I:: 0.9734 0.0050 -0.015 XIX 0.9833 0.0033 -0.016 XX 0.9922 0.0048 -0.011 1 XX: 0.9733 0.0039 -0.017 g.. Mean 0.9894 0.0011 '- -0.01:1 0.3010 z.4 2.. 9. g. n :. ' O. 9 0 '. c. - 4 0. 0 ' - *. - 313s ( 9 3 ^!. '3 3 3 ) 3.0136 d0.0048 0.0123 , O. ) '.', 2 3 "a::inu:- 3 as 0.015 0.11 ~. \\ 4 II' E::perimen s -I thr ucn *,C:: used 8 C pin a scr:er. and aa;3 l 3 nct censidered regrasentative of poiscned storage r i m'. s. (,,Calcula:2d fr : individua standard deviations. (3)Calcula:ed from k gg values and used as ref erence. e ~. A-4 i

f actor,:-if one truly exis ts, is. too 'smail to be resolved ' on - tho'- basis of critical-experiment data presently available. - Noitrends in: k ~with. intra-assembly water gap,- with absorber sheet-

)

egg 5 reactivity worth, or with' soluble poisen concentration were identified. ~ 3. CASMC-2E BENCHMARK CALCULATIONS t 3.1 GENERAL 4 q 3 i -The CASMO-2E. code is a multigroup transport theory code

{

utili:~ing transmission probabilities to accompl,ish.two-dimen- .i sional calculations of reactivity and ' depletion for BWR and PWR fuel assemblies. As such, CASMO-2E is well-suited to the criti-k cality analysis of spent fuel storage racks,; since general q practice is to treat the racks as an infinite medium of ' storage d cells, neglecting leakage effects. CASMC-2E is closely. analegous to the E?RI-CPM code : (Ref. 13) .and has ' been extensively benchmarked against, hot and cold crit-l ical experiments by Studsvik Energitekni% (Refs. 5, 6, 7, and 8). Reper:ed ana'.yses of 26 cricical experiments indicate a mean. 1 k,g g. of 1.000 = 0.0037 (le). Yankee At:mic ' (Ref. 14 F has also i reported results cf ext ens ive benchmark calcula:icns with CASMC-2E. Their analysis of 54 S traw0 ridge and Barry critical'ex eri-ments (Ref. 15; using the re crted bucklingL indicates a mean of 1 1 0.9987 0.00C? (le', or a :ias cf 0. 0 ". ~. 2 0.0013 (wi:n 955 p recar ility a: a 9F condifence l e 'r e '. Calcula:icns were repea:ed f;r seten Of the 3:r2woridge and 3arry exterimenta l (, . Significantly 'arge trends in k reactivity worth have.g with ' vater gao and with ab-eg sorber sheet ceen re:Orted (Ref. 16) for AMPX-KEMC calculations with the 123-grou: G.M-TH ERMOS library. I A-5

1 silocted at randem, L yielding a mean kogg.of 0.9987

• 0.0021 (le),

a .thereby confirming that the cross-section library and analytical.- methodology being used for the present calculations. ara the same as those used in the Yankee -analyses.- Thus,. the expected bias L for CASMO-2E 'in the analysis of " clean" critical experiments is 0.0013 t 0.0018 (95%/95% ). s 1 i 3.2 BENCHMARK CALCULATIONS- ) i CASMO 2E benchmark calculations have also been made for the-B&W series of critical experiments with absorber shee ts, 'simu-- j lating high' density: spent fuel storage - racks. However, CASMO-2E, as an assembly : code, ' cannot directly ' represent an entire core - configuration

• without -introducing uncertainty due,to ref1'ector constants and the appropriateness of their spectral weighting.

For this reason, the poisoned cell. configurations of the ~ central

assembly, as calculated by CASMO-2E, were benchmarked. against corresponding calculations with the 27-group (SCALE) AMPX-KENO' cod'e package.

Results of this.ccmparison are shown in' Table 2. { .Since the differences are well within the. normal KENO statistical. ) ) variation, these calculations confirm the validity of CAS:4C-2 E calculations f:: the tycical high de ns ity pcisoned spent fuel rack configura:icns. The differences sh:wn. in Table 2 are also l consisten: with a bias of 0.3013 -0.0013, determined in Section I i 3.1 as the ex;s :ed bias and uncertainty of CASMC-25 calcula-l tions. i l i-i l l 4 8 ?an<a-a has a :1 ::ed. such calculations '?sf. '. 4 us ing CAS:tC '22-l generated cens:Ents in a twc-dimensional, Ecur-group ?CQ medal, cocaining a mean ke,, of 1. 00 5 for 11

isened cases and 1.009 5or 5 un;oiscned c55es.

Thus, Yankee

enchmark calcula: ions I

suggest that CA3:40-2E tends to slightly crerpredict reactivity. A-6 1 4 ) L

l Table 2 RESULTS CF CA5:10-2E BENCHMARK (INTERCOM?ARISCN) CALCULATIONS l l l l k.III B&W Exceriment No.(1) AMPX-KEN 0(2) CASMO-2E Ak XIX 1.1203 0.0032 1.1193 0.0010 XVII 1.1149 0.0039 1.1129 0.0020 1 XV 1.1059. 0.0038 1.1052 0.0007 Interpolated (3) 1.1024 0.0042 1.1011 0.0013 XIV 1.0983 0.0041 1.0979 0.0004 l XIII. 1.0992 0.0034 1.0979 0.0013 p Mean 0.0038 0.0011 Uncertaint? 0.0006 1 BWR fuel rack 0.9212 0.0027 0.9218 -0.006 ) l (1) Infinite array of central assemblies of 9-assembly S&W criti-cal confitur2:ica (Ref. 9). (,f )I, k fr:r A.'t?X 7.ENO corrected for bias cf 0.0106 ak. (# Inter;cla:ed from Fig. 29 cf Ref. 9 f: soluble bcron cencen-tratica a: critical conditicn. l l l l A-7 1 f 1 L

ggpERENCES TO AP9ENDIX A I 1. Green, Lucious, Petrie, Ford, White, Wright, " P S R-6 3 / AMpy,- 1 (code package), AMPX Modular Code System for Generating Coucled Multigroup Neutron-Gamma Libraries from END F/ 3, " ORNL-TM-3706, Oak Ridge National Laboratory, March 1976. 2. L. M. Petrie and N. F.

Cross,

" KENO-IV, An Improved Monte Carlo ' Criticality Program," ORNL-4938, Oak Ridge National Laboratory, November 1975. 3. R. M. Westfall et al., " SCALE: A Modular Code System for Performing Standardized Comouter Analyses for Licensing Evaluation," NURG /CR-0200, 1979. 4. W. E. F.ord, III et al., "A 218-Neutron Group Master Cross-section Library for Criticality Safety Studies," ORNL/TM-4, 1976. 5. A. Ahlin, M.

Edenius, H. Haggblom, "CASMO - A Fuel Assembly Burnup P rog ram, "

AE-RF-76-4158, Studsvik report (proprietary). 6. A. Ahlin and M.

Edenius, "CASMO -A Fast Transport Thec ry Depletion Code for LWR Analysis," ANS Transactions, Vol. 26,

p. 604, 1977.

4 7. M. Edenius et al., "CASMO Benchmark Re p o r t, Studsvik/RF-78/6293, Aktiebolaget Atomenergi, March 1978. 8. " CA5 ".0 - 2 E Muclear Fuel Assem:ly Ana_' / sis, Application Lise rs Manual," Rev. A, Control Data Corporation, 1982. 9. M. M. Baldwin et al., " Critical Ex:eriments Su;;;rting Cicse Pr:ximity Water Storage of Power Reactor Fuel," 3 AW- ; 4 8 4 ~, The Babe::k & Wilcox C:mpany, July 1979. 10. M. C. Ma:rella, E::c e rire n t al S P. a n i s t i c s, Ma t i o n a '. Eureau :f S t anda rds, Mandb:T< i_, Au us-i

0..

11. 3. F.

ene'r et al.,

"C:rtaris s of E:::s rire.:s and Cal:ulanicis 50: LUE 3: r2c? Gecr9 ries, We31'.ntncusi NES, A::3 Crans = :icns, Vcl. 39, 31,

vem:9r 1981.

12. R. 1 Westfall and J. F. 'ni n, " 3 :: '. 3 ~!:35 Cr ss-Est ;:n Validatic.- with ShiCinc-cask Cr tical. I::ce ri. e nts, " A: S l Transse:ic s, Vcl. 33, c. 3 6 d., McVer:er 19~9. 13. "The EPR:-CPM Data Library," ARMF Concuter C:ie Manuals, l Par II, Chaoter 4, CCN3, Electric ? we r F.esearen Ins:::ute, l November 1975. A-8

b C ' REFERENCES TO' APPENDIX A'(Continued) 14. E. E. Pilat,- " Methods for - the ' Analysis. of Boilinc: Wat=* Reactors (Lattice-Physics),"' YAEC-1232', _ Yankee Atomic j Electric Co., December 1980. . u

15..

L. E. Strawbridge and R. F. Barry, " Criticality ' Calculations-for Uniform, Water-moderated Lattices," Nuclear' Science and Encineerinc, Vol'. 23, p. 58, Septem'cer.1965. 16. S. E.. Turner and M. K.-

Gurley,

" Evaluation 'of.' AMPX-KENO ' Benchmark Calculations for High ' Density Spent Fuel Storage Racks," Nuclear Science and Encinee rine, 80(2): 230-237, j February 1982. 4 9 l l 1 A-9 s}}