ML20217P635
ML20217P635 | |
Person / Time | |
---|---|
Site: | Byron, Braidwood ![]() |
Issue date: | 09/29/1999 |
From: | HOLTEC INTERNATIONAL |
To: | |
Shared Package | |
ML20138H521 | List: |
References | |
HI-982083, HI-982083-R06, NUDOCS 9910290345 | |
Download: ML20217P635 (84) | |
Text
,
Holtec Center,555 Lincoln Drive West, Marlton, NJ 08053 LTEC ONAL
'*"" "" (6 ') 7- 9 Fax (609) 797-0909 LICENSING REPORT for SPENT FUEL RACK INSTALLATION at BYRON 4ND BRAIDWOOD N'UCL/ EAR STATIONS Holtec Report HI-982083 (Non-Proprietary Version)
Report Category: A l:
Prepared for Commonwealth Edison Co.
Purchase Order No. 367585 l Holtec Project 80944 c
l COMPANY PRIVATE O This document version has all proprietary information removed and has replaced those sections, figures, and tables with highlighting and/or notes to designate the removal of such information. This document is to be used only in connection with the performance k of work by Holtec International or its designated subcontractors. Reproduction, j publication or presentation, in whole or in part, for any other purpose by any party !
other than the Client is expressly forbidden. l 9910290345 991021 l PDR ADOCK 05000454 i P PDR ;
l
))Pp"j j d b
.i J (,J t
j Holtic Centar,555 Lincoln Drive W st, Marlton, NJ 08053 l HOLT' e V
'*'*P" "* (* *) '*7-
- Fax (609) 797 - 090f INTERNATIONAL i REVIEW AND CERTIFICATION LOG ,
l DOCUMENT NAME LICENSING REPORT FOR BYRON /BRAIDWOOD NUCLEAR STATIONS l
l HOLTEC DOCUMENT l.D. NUMBER 982083 1
HOLTEC PROJECT NUMBER - 80944 CUSTOMER / CLIENT: COMMONWEALTH EDISON REVISION BLOCP
!! !! !! l x REVISION, AUTHOR & REVIEWER & QA & DATE APPROVED & DIST.
NWBER DATE .DATE DATE ORIGINAL j
\ ,c$Y REVISION 1
_w REVISION 2 [
REVISION 3 g REVISION 4 REVISION 5
. n REVISION 6 M W MMW4 c,
*~%/n/r mluftr MMir c w +/n
!! This document cdnforrds to the requiremdnts of the design specification and (he ' applicable sections of the goveming codes.By signing on this page, you are confirming that you have filled out the DVC questionnaire stored in Holtec's network directory n:\pdoxwin\ working \dvc.
Note : Signatures and printed names required in the review block.
- A revision of this document will be ordered by the Project Manager and carried out if any of its contents is materially affected during evolution of this project. The determination as to the need for revision will be made by the Project Manager with input from others, as deemed necessary by i Must be Project Manager or his Sesignee x . Distnbution : C: Client M: Designated Manufactures F: Florida Office
- " Report category on the cover page indicates the contractual status of this document as "*
A = to be submitted to client for approval I = for client's information N = not submitted externally THE REVISION CONTROL OF THIS DOCUMENT IS BY A "
SUMMARY
OF REVISIONS LOG" PLACED BEFORE THE TEXT OF THE REPORT.
7
~
SUMMARY
OF REVISIONS Revision 0: fInititial issue.
L Revision 1: Figures'11.1 through 11.13, which showed'a preliminary rack change-out l
sequence, were removed from Chapter 11 (Installation) and the List of Figures. I Minor corrections were also made to the Table of Contents.
. Revision 2: Text was added to Sections 11.5 and 11.6, which describe the fuel shuffling and
'. e new rack installation. Minor changes were also made to the Table of Loatents.
l Revision 3: - Minor changes were made to Paragraph 11.1 h. (ALARA Procedure), Subsection 11.7.2, and Subsection 11.7.3. <
Revision 4: Editorial changes were made on pages 1-1,5-2,5-17,8-7, and 8-11.
- Revision 5: . Chapter 7 was reorganized to include a section on construction accidents (i.e.,
rack drop accident).. Figure 6.5.1 'and Table 8.2 were also corrected. Table 6.5.3 was added to the report.1These change: also affected the Table of Contents and
- the List of Figures.
Revision 6: A new paragraph was added to Section 6.4 (Synthetic Time' Histories) on page 6-
- 6. Minor editorial changes were also made on pages 6-12 and 6-14. ;
't a
i l
l
[
1>
I l
- Holtec International; i Report HI-982083 bi;
6.0 STRUCTURAL SEISMIC CONSIDERATIONS
)
6.1 . Introduction . J l
I This chapter considers the structural adequacy of the new maximum density spent fuel racks i
]
under all loads postulated for normal, seismic, and accident conditions at Byron and Braidwood j Nuclear stations. l l
As discussed in Chapter 1, the reracking of the Byron and Braidwood pools involves replacement j of existing high-density storage racks with new racks with a slight increase in the total capacity.
The reracking is being undertaken to remove the BoraDex neutron material from the two pools because ofits ongoing degradation and loss in neutron attenuation ability. The new racks, like the existing racks, will be installed in a free-standing configuration. At the time of the previous rerack, however, the state-of-the-art limited the seismic evaluation to single rack 3-D l simulations. As we discuss in this chapter, it is now possible to model the entire assemblage of rack modules in one comprehensive simulation known as the 3-D Whole Pool Multi-Rack (WPMR) analysis. In order to maintain continuity with the previous analysis methods, both l single rack and WPMR analyses have been performed to establish the structural margins of safety in the Byron /Braidwood racks.
The analyses undertaken to confirm the structural integrity of the racks are in full compliance 1
with the USNRC Standard Review Plan [6.1.1] and the OT Position Paper [6.1.2]. For each of '
the analyses, an abstract of the methodology, modeling assumptions, key results, and summary of parametric evaluations are presented. Delineation of the relevant criteria are discussed in the text associated with each analysis.
6.2 Overview of Rack Structural Analysis Methodolony The response of a free-standing rack module to seismic inputs is highly nonlinear and involves a
(
complex combination of motions (sliding, rocking, twisting, and turning), resulting in impacts and friction effects. Some of the unique attributes of the rack dynamic behavior include a large fraction of the total structural mass in a confined rattling motion, friction support of rack pedestals against lateral motion, and large fluid coupling effects due to deep submergence and motion of closely spaced prismatic structures.
SilADED TEXT CONTAINS PROPRIETARY INFORMATION lloltec International 6-1 Report 111-982083
l
)
1 Linear methods, such as modal analysis and response spectrum techniques, cannot accurately simulate the structural response of such a highly nonlinear structure to seismic excitation. An accurate simulation is obtained only by direct integration of the nonlinear equations of motion with the three pool slab acceleration time-histories applied as the forcing functions acting simultaneously.
I
' Both Whole Pool Multi-Rack (WPMR) and Single Rack (SR) analysis are used in this project to I simulate the dynamic behavior of the high density rack structures described in Chapter 2 of this report. The following sections provide the basis for the selection of the appropriate methodology and discussion on its development.
6.2.1 Backaround of Analysis Methodoloav i
Reliable assessment of the stress field and kinematic behavior of the rack modules calls for a l
conservative dynamic model incorporating all key attributes of the actual structure. This means that the model must feature the ability to execute the concurrent motion forms compatible with the free-standing configuration of the modules.
The model must possess the capability to effect momentum transfers which occur due to rattling of fuel assemblies inside storage cells and the capability to simulate lift-off and subsequent impact of support pedestals with the pool liner (or bearing pad). The contribution of the water mass in the interstitial spaces around the rack modules and within the storage cells must be modeled in an accurate manner since erring in quantification of fluid coupling on either side of the actual value is no guarantee of conservatism.
The Coulomb friction coefficient at the pedestal-to-pool liner (or bearing pad) interface may lie in a rather wide range and a conservative value of friction cannot be prescribed apriori. In fact, a perusal of results of rack dynamic analyses in numerous dockets (Table 6.2.1) indicates that an upper bound value of the coefficient of friction often maximizes the computed rack displacements as well as the equivalent clastostatic stresses.
In short, there are a large number of parameters with potential influence on the rack kinematics.
The comprehensive structural evaluation must deal with all of these without sacrificing conservatism. l t
SilADED TEXT CONTAINS PROPRIETARY INFORMATION llottec International 6-2 Report 111-982083 I
i l
The three-dimensional single rack dynamic model introduced by Holtec International personnel in the Enrico Fermi Unit 2 rack project (ca.1980) and used in some 50 rerack projects since that time, including Byron and Braidwood in the late 80s (Table 6.2.1) addresses most of the above mentioned array of parameters. The details of this methodology are also published in the permanent literature [6.2.1]. Despite the versatility of the 3-D seismic model, the accuracy of the single rack simulations has been suspect'due to one key element; namely, hydrodynamic participation of water around the racks. During dynamic rack motion, hydraulic energy M.ither drawn from or added to the moving rack, modifying its submerged motion in a significant manner. Therefore, the dynamics of one rack affects the motion of all others in the pool.
However, Single Rack analysis is still a valuable tool to examine the behavior of a rack under different load conditions. It is used here as a first step in evaluating the racks. WPMR analysis builds upon the Sing'e Rack model. The worst-case loads and stresses that result from these two models are used to determine the structural adequacy of the racks.
The 3-D rack model dynamic simulation, involving one or more spent fuel racks, handles the ,
array.of variables as follows:
Interface Coefficient of Friction Parametric runs are made with upper bound and lower bound values of the coefficient of friction (COF). The limiting values are based on experimental data that have been found to be bounded by the values 0.2 and 0.8. Simulations are also performed with the array of pedestals having randomly chosen coefficients of friction in a Gaussian distribution with a mean of 0.5 and lower and upper limits of 0.2 and 0.8, respectively. In the fuel rack simulations, the Coulomb friction interface between rack support pedestal and liner is simulated by piecewise linear (friction) elements. These elements function only when the pedestal is physically in contact with the pool liner.
Rack Elastic Behavior Rack elasticity, relative to the rack base, is included in the model by introducing linear springs to represent the clastic bending action, twisting, and extensions.
l Impact Phenomena Compression-only gap elements are used to provide for opening and closn 3 ofinterfaces such as the pedestal-to-bearing pad interface, and the fuel assembly-to-cell wall interface. These interface gaps are modeled using nonlinear spring elements. The term l
" nonlinear spring" is a generic term used to denote the mathematical representation of the j condition where a restoring force is not linearly proportional to displacement. !
I SIIADED TEXT CONTAINS PROPRIETARY INFORMATION )
lloltec International 6-3 Report 111-982083 J;
s M
L l
- Fuel Loadinn Scenarios The fuel assemblies are conservatively assumed to rattle in unison, which'obvio'usly exaggerates the effect of fuel impacts against the cell walls. Partial fuel loadings (e.gia rack that has fuel assemblies in only half ofits cells) are simulated by offsetting
~ the center of gravity of the stored fuel mass with respect to the rack center of gravity, as ;
appropriate.
t Fluid Couclinn Holtec International extended Fritz's classical two-body fluid coupling model to multiple bodies and utilized it to perform the first two-dimensional multi-rack analysis (Diablo l
. Canyon, ca.1987). Subsequently, laboratory experiments were conducted to validate the multi-rack fluid coupling theory. - This technology was incorporated in the computer code .
I : DYNARACK [6.2.4] which handles simultaneous simulation of all racks in the pool as a Whole l Pool Multi-Rack 3-D analysis. This development was first utilized in Chinshan, Oyster Creek, and Shearon Harris plants in the 80's [6.2.1,;6.2.3] and, subsequently, in numerous other rerack 2
projects. The WPMR analyses have corroborated the accu cy of the single rack 3-D solutions'in predicting the maximum structural stresses, and also serve to improve predictions of rack
- kinematics.-
The Whole Pool Multi-Rack (WPMR) model used to predict the dynamic behavior of the storage racks contains elements specifically designed to represent the attributes necessary to simulate rack motions during earthquakes. These elements include non-linear springs to develop the
- interaction between racks, between racks and walls, and between fuel assemblies and rack
. internal cell walls.- Hydrodynamic effects within these interstitial spaces are accounted for using I Fritz's classical method which relates the fluid kinetic energy in the annulus due to relative motion to an equivalent hydrodynamic mass.
The modeling technique used was chosen based on the applicable Codes, Standards and i Specifications given in Section IV (2) of the NRC guidance on spent fuel pool modifications L entitled," Review and Acceptance of Spent Fuel Storage and Handling Applications," dated April
- 14,- 1978, which states that " Design...may be performed based upon the AISC specification or
. Subsection NF requirements of Section III of the ASME B&PV Code for Class 3 component supports." The rack modeling technique is consistent with the linear support beam-element type members covered by these codes.
S11ADED TEXT CONTAINS PROPRIETARY INFORMATION t 11oltec International; 6-4 Report 111-982083
Although it is acknowledged that finite element models could be developed using plate and fluid elements which may also provide satisfactory simulated behavior for a single rack, there is no known commercial finite element code which can treat multi-body fluid interaction correctly and sufficiently account for near and far field fluid effects involving many bodies (racks) in a closed pool. It is for this reason that the global dynamic analysis uses the formulation specifically developed and contained within DYNARACK.
The computer software validation of the DYNARACK program is documented in validation manual HI-91700. The validation manual demonstrates that the DYNARACK code verification is adequate for engineering applications without further experimental verification.
For closely spaced racks, demonstration of kinematic compliance is verified by including all modules in one comprehensive simulation using a WPMR model. In WPMR analysis, all rack ruodules are modeled simultaneously and the coupling effect due to this multi-body motion is included in the analysis. Due to the superiority of this technique in predicting the dynamic behavior of closely spaced submerged storage racks, the Whole Pool Multi-Rack analysis methodology is used as the principal vehicle for seismic qualification in the Byron /Braidwood project.
6.3 Description of Racks and Fuel As discussed in Chapter 3, the proposed rack layouts for Byron and Braidwood are identical. A total of twenty-four racks is proposed to be installed in each pool. Four new racks use a flux-trap design and are referred to as Region I racks. The remaining racks do not utilize flux-trrps and are referred to as Region II racks. The dynamic rack models include all twenty-four racks. For dynamic simulations, the dry fuel weight is conservatively taken to be 1600 lbs.
6.4 Synthetic Time-Histories Synthetic time-histories in three orthogonal directions (N-S, E-W, and vertical) are generated in accordance with the provisions of SRP 3.7.1 [6.4.1]. In order to prepare an acceptable set of acceleration time-histories, Holtec International's proprietary code GENEQ [6.4.2] is utilized.
A prelerred criterion for the synthetic time-histories in SRP 3.7.1 calls for both the response spectrum and the power spectral density (PSD) corresponding to the generated acceleration time-SIIADED TEXT CONTAINS PROPRIETARY INFORMATION lioltec International 6-5 Report 111-982083
F history to envelope their target (design basis) counterparts with only finite enveloping infractions.
The time-histories for the pools have been generated to satisfy this preferred (and more rigorous) criterion.
The design response spectra used to develop the synthetic time histories is the envelope of the Byron spectra and Braidwood spectra obtained from the building seismic models at the fuel pool elevation. This is believed to be conservative because be Byron foundation is on rock whereas the Braidwood foundation is comprised of soil. Furthermore, the broadened design spectra of the envelope curves are used to make the comparison with the response spectra of the synthetic time histories. l The target PSD is generated using a program called GENEQ. GENEQ is a Q.A. validated synthetic time-history generator and has been used by Holtec International to generate statistically independent artificial acceleration time histories in over 40 reracking projects.
GENEQ accepts an initial digitized response spectra as input and generates a new bounding response spectra along with a PSD corresponding to both the target and generated spectra along ,
with an acceleration time history.
To prepare the PSD, the digitized design basis response spectra fr.t the elevation ofinterest (at the floor of the pool) was initially input to obtain the target respc? c spectra and target PSD and a generated spectra and PSD. If the generated spectra and PSD does not bound the initially input target, then an iterative process begins which involves revising the input spectra (by broadening, increasing peak values, etc.) to establish a bounding spectra and PSD. At the completion of each iterative step the newly generated spectra and PSD are compared against the target spectra and PSD developed from the initially input digitzed design basis response spectra. If necessary, the response spectra data is smoothed to prepare comparable results.
The synthetic time-histeries in the three directions also satisfy the requirements of statistical independence mandated by SRP 3.7.1. I Figures 6.4.1 through 6.4.3 and 6.4.4 through 6.4.6 provide plots of the time-history )
accelerograms that were generated over a 20-second duration for SSE and OBE events, !
respectively. These artificial time-histories are used in all non-linear dynamic simulations of the racks.
SilADED TEXT CONTAINS PRODRIETARY INFOIWATION IIoltec International 6-6 Report 111-982083 i
Results of the correlation function of the three time-histories are given in Table 6.4.1. It is noted that the absolute values of all correlation coefficients are well below 0.15_ indicating that the desired statistical independence of the three data sets has been met.
6.5 3-D Nonlinear Rack Model for Dynamic Analysis 6.5.1 General Remarks The single rack 3-D model of the Byron /Braidwood racks has been prepared with due consideration of the following characteristics, which are typical of high-density modules designed by Holtec International.
- i. As a continuous structure, the rack possesses an infinite number of degrees-of-freedom (DOF), of which the cantilever beam type modes are most pronounced under seismic excitation if the rack is of the honeycomb construction genre. (The Byron /Braidwood racks, like all prior Holtec designs, are of the honeycomb type.)
ii. The fuel assemblies are " nimble" structures with a relatively low beam mode fimdamental frequency.
iii. The interstitial gap between the storage cells and the stored fuel assemblies leads to a rattling condition in the storage cells during a seismic event.
iv. The lateral motion of the rack due to seismic input is resisted by the pedestal-to-pool slab interfacial friction and is abetted or retarded by the fluid coupling forces produced by the proximity of the rack to other structures. (The fluid coupling forces are distinct from the nonconservative forces such as fluid " drag" which are, by NRC regulations, excluded from the analysis). The construction of a 3-D single rack dynamic model consists of modeling the rack as a multi-degree-of-freedom structure in such a manner that the selected DOFs capture all macro-motion modes of the rack, such as twisting, overturning, lift-off, sliding, flexing, and combinations thereof. Particular attention must be paid to incorporating the potential for the friction-resisted sliding of the rack on the liner, lift-off and subsequent impact of the pedestals on the slab, collision of the rack with adjacent structures, and most important, rattling of the fuel in the storage cells. The SHADED TEXT CONTAINS PROPRIETARY INFORMATION Holtec International 6-7 Report 111-982083
dynamic model must also provide for the ability to simulate the scenarios of partially loaded racks with arbitrary loading patterns.
I As the name implies, the Single Rad (SR) dynamic model is a 3-D structural model for one rack in the pool. The rack selected for the SR analysis in this project is the one with the most mass, or most non-square cross section (i.e., pool aspect ratio). The dynamic model of this rack, i.e., its structural stiffness characteristics, rattling effect of the stored fuel, etc., can be prepared with extreme diligence in the manner described in the following, resulting in an excellent articulation of the rack structure. Even the fluid coupling effects between the fuel assemblies and the storage cell can be modeled with acceptable accuracy [6.5.2]. If the rack is adjacent to a wall, the fluid coupling effects between the rack and the wall can also be set down deterministically because the wall is a fixed structure. Such a definitive situation does not exist, however, when the neighboring structure to the subject rack is another free-standing rack. During a seismic event, the subject rack and the neighboring rack will both undergo dynamic motions which will be governed by the interaction among the inertia, fluid, friction, and rattling forces for each rack.
The fluid coupling forces between two racks, however, depend on their relative motions.
Because the motion of the neighboring rack is undefined, it is not possible to characterize the hydrodynamic forces arising from the fluid coupling between the neighboring rack and the subject rack. This inability to accurately model the inter-rack fluid coupling effects is a central limitation in the single rack analysis.
To overcome this limitation intrinsic to the single rack solutions, an artificial boundary condition, j referred to as the "out-of-phase" assumption, has been historically made to bound the problem. l
)
In the out-of phase assumption, it is assumed that all racks adjacent to the subject rack are vibrating 180 out-of-phase, resulting in a plane of symmetry between the subject rack and the neighboring rack, across which water will not flow. Thus, the subject rack is essentially surrounded by a fictitious box with walls that are midway to the adjacent racks. Impact with the adjacent rack is assumed to have occurred if the subject rack contacts the " box wall" l l
In summary, in the out-of-phase motion analysis the analyst makes the election that the adjacent l racks are moving at 180 out-of-phase from the subject rack at all times during the seismic event.
This is an artificial technical construct, albeit one that is known to predict rack-to-rack impact ,
conservatively.
SHADED TEXT CONTAINS PROPRIETARY INFORMATION llottec International 6-8 Report 111-982083
However, this assumption also increases the relative contribution of fluid coupling, which depends on fluid gaps and relative movements of bodies, making overall conservatism a less certain assertion. As is well known, the fluid forces between adjacent rack modules can reach rather large values in closely spaced rack geometries. It is, therefore, essential that the contribution of the fluid forces be included in a comprehensive manner. This is possible only if all racks in the pool are allowed to execute 3-D motion in the mathematical model. For this reason single rack, or even multi-rack models involving only a portion of the racks in the pool, are inherently inaccurate. The Whole Pool Multi-Rack model removes this intrinsic limitation of tia rack dynamic models by simulating the 3-D : notion of all modules simultaneously. The fluid coupling ef fect, therefore, encompasses interaction between every set of racks in the pool, i.e., ,
the motion of one rack produces fluid forces on all other racks and on the pool walls. Stated more formally, both near-field and far-field fluid coupling effects are included in the WPMR analysis.
Therefore, to maintain consistency with past analyses, an array of single rack 3-D simulations were carried out, principally to compare the results (viz., rack-to-rack impact, maximum primary stress levels, pedestal loads, etc.) with the more definitive WPMR analysis. The description below provides the essentials of the 22 DOF model for a single rack. This model is used in both 3-D single rack simulations and as the building block for the more complicated WPMR analyses, which are described later in this chapter.
The dynamic modeling of the rack structure is prepared with special consideration of all nonlinearities and parametric variations. Particulars of modeling details and assumptions for the rack analysis are given in the following:
- a. The fuel rack structure motion is captured by modeling the rack as a 12 degree-of-freedom structure. Movement of the rack cross-section at any height is described by six degrees-of-freedom of the rack base and six degrees-of-freedom at the rack top. In this manner, the response of the module, relative to the baseplate, is captured in the dynamic analyses once suitable springs are introduced to couple the rack degrees-of-freedom and simulate rack stiffness.
i
- b. Rattling fuel m,semblies within the rack are modeled by five lumped masses located at H, .75H, .5H, .25H, and at the rack base (H is the rack height measured i I
above the baseplate). Each lumped fuel mass has two horizontal displacement degrees-of-freedom. Vertical motion of the fuel assembly mass is assumed equal l to rack vertical motion at the baseplate level. The centroid of each fuel assembly mass can be located off-center, relative to the rack structure centroid at that level, to simulate a partially loaded rack.
SIIADED TEXT CONTAINS PROPRIETARY INFORMATION 11oltec International 6-9 Report 111-982083 l
- c. Seismic motion of a fuel rack is characterized by random rattling of fuel assemblies in their individual storage locations. An upper bound on the effective cumulative fuel assembly mass is established using the previously described artificial time histories.
' d. - Fluid coupling between rack and fuel assemblies, and between rack and wall, is simulated by appropriate inertial coupling in the system kinetic energy. Inclusion of these effects uses the methods of[6.5.2,6.5.3] for rack / assembly coupling and for rack-to-rack coupling.
- e. ' Fluid damping and form' drag are conservatively neglected.
- f. Sloshing is found to be negligible at the top of the rack and is, therefore, neglected in the analysis of the rack.
~
- g. Potential impacts between the cell walls of the new racks and the contained fuel assemblies are accounted for by appropriate compression-only gap elements
. between masses involved. The possible incidence of rack-to-wall or rack-to-rack impact is simulated by gap elements at the top and bottom of the rack in two
- horizontal directions. Bottom gap elements are located at the baseplate elevation.
~ The initial gaps reflect the presence of baseplate extensions, and the rack
- stiffnesses are chosen to simulate local structural detail.' <
- h. Pedestals are modeled by gap elements in the vertical direction and as " rigid D
links" for transferring horizontal stress. Each pedestal support is mathematically -
linked to the pool liner (or bearing pad) by two friction springs. The spring rate for the friction springs includes any lateral elasticity of the stub pedestals. Local
- pedestal vertical spring stiffness accounts for floor elasticity and for local rack elasticityjust above the pedestal. ,
1 I' !
i.- Rattling of fuel assemblies inside the storage locations causes the gap between l- fuel assemblies and cell wall to change from a maximum of twice the nominal gap to a theoretical zero gap. Fluid coupling coefficients are based on the nominal gap
'in order to provide a conservative measure of fluid resistance to gap closure.
l
' j. The model for the rack is considered supported, at the base level, on four pedestals modeled as non-linear compression only gap spring elements and eight piecewise linear friction spring elements; these elements are properly located with respect to the centerline of the rack beam, and allow for arbitrary rocking and sliding motions.
SHADED TEXT CONTAINS PROPRIETARY INFORMATION Holtec International 6-10 Report HI-982083
l 6.5.2 Element Details l
Figure 6.5.1 shows a schematic of the dynamic model of a single rack. The schematic depicts many of the characteristics of the model including all of the degrees-of-freedom and some of the spring restraint elements.
Table 6.5.1 provides a complete listing of each of the 22 degrees-of-freedom for a rack model.
Six transitional and six rotational degrees-of-freedom (three of each type on each end) describe the motion of the rack structure. Rattling fuel mass motions (shown at nodes 1*,2*,3*,4*, and 5*
in Figure 6.5.1) are described by ten horizontal transitional degrees-of-freedom (two at each of the five fuel masses). The vertical fuel mass motion is assumed (and modeled) to be the sc is I that of the rack baseplate.
l Figure 6.5.2 depicts the fuel to rack impact springs (used to develop potential impact loads !
between the fuel assembly mass and rack cell inner walls) in a schematic isometric. Only one of the five fuel masses is shown in this figure. Four compression only springs, acting in the horizontal direction, are provided at each fuel mass. Figure 6.5.3 provides a 2-D schematic elevation of the storage rack model, discussed in more detail in Section 6.5.3. This view shows the vertical location of the five storage masses and some of the support pedestal spring members.
Figure 6.5.4 shows the modeling technique and degrees-of-freedom associated with rack elasticity. In each bending plane a shear and bending spring simulate clastic effects [6.5.4]. l Linear elastic springs coupling rack vertical and torsional degrees-of-freedom are also illustrated in this figure. !
Figure 6.5.5 depicts the inter-rack impact springs (used to develop potential impact loads between racks or between rack and wall). The approximate spring contact location and 1 1
numbering of each impact spring used in the model are shown it. Figure 6.8.1 and Figure 6.8.2.
I 6.5.3 Fluid Coupline Effect ii In its simplest form, the so-called " fluid coupling effect" [6.5.2,6.5.3] can be explained by i
considering the proximate motion of two bodies under water. If one body (mass mi) vibrates adjacent to a second body (mass m2), and both bodies are submerged in frictionless fluid, then Newton's equations of motion for the two bodies are:
SII ADED TEXT CONTAINS PROPRIETARY INFORMATION llottec International 6-11 Report 111-982083
2 (mi + M ) X i+ Mi2 X 2= applied forces on mass mi + O(Xi )
ii M 2X +i (m2 + M )22X = 2applied forces on mass m2 + O(X2 2 )
X iand X denote 2 absolute accelerations of masses mi and m2, respectively, and the notation l 2
O(X ) denotes nonlinear terms. The hydrodynamic coupling effect is shown to be composed of an added nonlinear term which varies with geometry and a component which varies with the square of velocity. This is easily shown by considering a typical example where fluid coupling plays a significant role. Consider two long beams oflength "1" width "h" and a distance "s" apart:
/ // / /h/ /
/ S /
/ /
/ / h i
/ /
B / / A
/ / / / V/
It i.: assumed that s<<h<<l which is always the case for spent fuel racks. It is shown by Levy
{d6.1] that the force exerted by the fluid on " beam" A is given by F,.,,,,,,
= pih' ' F - s '-
m naves to the right) 12s ( 2s, i
F,.,,,, = # -Y+ (A moves to the left) ;
12s y s, Thc above solution is valid at each instant in time so that as the beams (racks) approach each i other larger forces result which tend to reduce rack motion and preclude rack-to-rack impact. For l l
conservative results, the rack analyses are based only on the nominal gap that exists prior to any :
seismic event. Therefore the forces exerted have the form. .
I SHADED TEXT CONTAINS PROPRIETARY INFORMATION IIoltec International 6-12 Report HI-982083
s ,
2 F = C(-- s t as ) - ]
The non-linear term for the case of a prismatic fuel assembly in a square cell has been derived by j Soler & Singh (1982). " Dynamic Coupling in a Closely Spaced Two-Body System Vibrating in
. Liquid Medium: The Case of Fuel Racks".
In the actual spent fuel rack analyses, the geometry is more complex but the resulting non- I linearities have the same character of an added mass multiplier by the acceleration of the rack plus a velocity squared fluid damping term. As the interstitial gap changes, the resulting fluid 2 2 mass changes also result in non-linear terms. The non-linear terms, O (X i ) and O (X2 ), been !
consistently neglected in order to maintain the requirement that no credit be taken for fluid damping in the seismic analysis. )
l Mn, M ,i2M , and21 M22 are fluid coupling coefficients which depend on body shape, relative
' disposition, etc. Fritz [6.5.3] gives data for My for various body shapes and arrangements. The fluid adds mass to the body (Mn to mass mi), and an inertial force proportional to acceleration of the adjacent body (mass m2). Thus, acceleration of one body affects the force field on another. j This force field is a function ofinter-body gap, reaching large values for small gaps. Lateral i motion of a fuel assembl inside a storage location encounters this effect. For example, fluid
- coupling behavior will be experienced between nodes 2 and 2* in Figure 6.5.1. The rack analysis also contains inertial fluid coupling terms that model the effect of fluid in the gaps between adjacent racks. l l
l These terms are usually computed assuming that all racks adjacent to the rack being r Jyzed are
. vibrating in-phase or 180 degrees out of phase. The WPMR analyses do not require any !
assumptions with regard to phase.'
Rack-to-rack gap elements have initial gaps set to 100% of the physical gap between the racks or between outermost racks and the adjacent pool walls.
'6.5.4 Stiffness Element Details -
)
Table 6.5.2 lists all spring elements used in the 3-D 22-DOF single rack model. It helps to explain the stiffness details. Byron and Braidwood are mirror images about the E-W direction.
SHADED TEXT CONTAINS PROPRIETARY INFORM ATION
' Holtec International - 6-13 Report HI-982083
- The analysis for Braidwood serves for Byron as well. In the table, the following coordinate system applies:
x= Horizontal axis along Braidwood plant North (Byron, South) l y= Horizontal axis along Braidwood plant West (Byron, East) z= Vertical axis upward from the rack base If the simulation model is restricted to two dimensions (one horizontal motion plus one vertical motion, for example), for the purposes of model clarification only, then Figure 6.5.3 describes 4 the configuration. This simpler model is used to elaborate on the various stiffness modeling elements.
Type 3 gap elements modeling impacts between fuel assemblies and racks have local stiffness Ki in Figure 6.5.3. In Table 6.5.2, for example, type 3 gap elements 5 through 8 act on the rattling
- fuel mass at the rack top. Support pedestal spring rates Ks are modeled by type 3 gap elements 1 through 4, as listed in Table 6.5.2. Local compliance of the concrete floor is included in Ks. The type 2 friction elements listed in Table 6.5.2 are shown in Figure 6.5.3 as Kr. The spring elements depicted in Figure 6.5.4 represent type 1 elements.
Friction at support / liner interface is modeled by the piecewise linear friction springs with suitably large stiffness Kr p u to the limiting lateral lead pN, where N is the current compression load at
- the interface between support and liner. At every time-step during transient analysis, the current
- value of N (either zero if the pedestal has lifted off the liner, or a compressive finite value) is computed.
The gap element Ks, modeling the effective compression stiffness of the structure in the vicinity of the support, includes stiffness of the pedestal, local stiffness of the underlying pool slab, and local stiffness of the rack cellular structure above the pedestal.
The previous discussion is limited to a 2-D model solely for simplicity. Actual analyses incorporate 3-D motions and include all stiffness elements listed in Table 6.5.2. Table 6.5.3 provides a list of typical stiffness values, which are used to model the spent fuel racks for Byron and Braidwood.
SIIADED TEXT CONTAINS PROPRIETARY INFORMATION lioltec International 6-14 Report 111-982083
g s ,
- 6.6 ' ,Whole Pool Muhi-Rack Method lony-
[6i6'E " General Remarks-The single rack 3-D (22-DOF) models for the new racks outlin' ed in the preceding subsection are u
' sed as a first step to evaluate the structural integrity and physical stability of the rack modules.
i However, prescribing the motion of the racks adjacent to the module being analyzed is an assumption in the single rack simulations that cannot be_ defended on the grounds of conservatism. lFor closely spaced racks, demonstration of the kinematic compliance is further verified by including all modules in one comprehensive simulation using a Whole Pool . Multi-Rack (WPMR)'model.1The WPMR analysis builds on the Single Rack model by simultaneously'
- modeling~all racks with full con s~ ideration of the multi-body' fluid coupling effects (discussed in the next subsection $
- Recognizing that the analysis work effort must' deal with both stress and displacement criteria, the sequence of model development and analysis steps that are undertaken are summarized in the following:
- a. : Prepare 3-D' dynamic models suitable for a time-history analysis of the new
- maximum density racks. These models include the assemblage of all rack modules in the pool.' Include all fluid coupling interactions and mechanical coupling appropriate to performing an accurate non-linear simulation. This 3-D !
[ ^ simulation is referred to as'a Whole Pool Multi-Rack model. j i
b.' : Perform 3-D _ dynamic analyses on various physical conditions (such as coefficient of friction and extent of cells containing fuel assemblies).' Archive appropriate .
4 displacement and load outputs from the dynamic model for post-proessmg. j i
- c. > Perform stress analysis of high stress areas for the limiting case of all, ,, rack )
dynamic analyses. Demonstrate compliance with ASME Code Sectiot fi, Subsection NF limits on stress and displacement.
E6;6.2. Multi-Body Fluid Couplina Phenomena l
During the seismic event, all racks m the pool are subject to the input excitation simultaneously. !
. The motion of each free-standing module would be autonomous and independent of others as !
long as they do not impact each other and no water is present in the pool. While the scenario of i SHADED TEXT CONTAINS PROPRIETARY INFORMATION Holtec International ' 6-15 Report HI-982083
1 3
1 inter-rack impact is not a common occurrence and depends on rack spacing, the effect of water-the so-called fluid coupling effect -is a universal factor. As noted in Ref. [6.5.2,6.5.3], the
~ fluid forces can reach rather large values in closely spaced rack geometries. It is, therefo:e, essential that the contribution of the fluid forces be included in a comprehensive manner. This is possible only if all racks in the pool are allowed to execute 3-D motion in the mathematical model. For this reason, single rack or even multi-rack models involving only a portion of the j racks in the pool, are inherently inaccurate. The Whole Pool Multi-Rack model removes this l intrinsic limitation of the rack dynamic models by simulating the 3-D motion of all modules simultaneously. The fluid coupling effect, therefore, encompasses interaction between every set of racks in the pool, i.e., the motion of one rack produces fluid forces on all other racks and on the pool walls. Stated more formally, both near-field and far-field fluid coupling effects are l included in the analysis.
The derivation of the fluid coupling matrix [6.6.2] relies on the classical inviscid fluid mechanics principles, namely the principle of continuity and Kelvin's recirculation theorem. While the derivation of the fluid coupling matrix is based on no artificial construct, it has been nevertheless verified by an extensive set of shake table experiments [6.6.2].
6.6.3 Coefficients of Friction To eliminate the last significant element of uncertainty in rack dynamic analyses, multiple simulations are performed to adjust the friction coefficient ascribed to the support pedestal / pool-bearing pad interface. These friction coefficients are chosen consistent with the two bounding extremes from Rabinowicz's data [6.5.1]. Simulations are also perfomied by imposing intermediate value friction coefficients developed by a random number generator with Gaussian normal distribution characteristics. The assigned values are then held constant during the entire simulation in order to obtain reproducible results. i Thus, in this manner, the WPMR anasysis results are brought closer to the realistic structural conditio is.
The coefficient of friction ( ) between the pedestal supports and the pool floor is in& terminate.
According to Rabinowicz [6.5.1], results of 199 tests performed on austenitic stainless teel plates submerged in water show a mean value of to be 0.503 with standard deviation of 0.125.
l i
It is noted that DYNARACK has the capability to change the coefficient of friction at any pedestal at each instant of contact based on a random reading of the computer clock cycle. Ilowever. exercising this option would yield results that could not be reproduced. Therefore, the random choice of coefficients is made only once per run.
SIIADED TEXT CONTAINS PROPRIETARY INFORMATION lloitec International 6-16 Report 111-982083
p i
1 L
Upper and lower bounds (based on twice standard deviation) are 0.753 and 0.253, respectively.-
Analyses are therefore performed for coefficient of friction values of 0.2 (lower limit),0.8 (upper limit), and for random friction values clustered about a mean of 0.5. The bounding values of p =
. 0.2 and 0.8 have been found to envelope the upper limit of module response in previous rerack ;
projects.
i
~ The bearing pads, which are inserted between the support pedestals and the pool liner, may L require additional shim plates in order to span liner weld seams. These shims will be welded to - :
. the bearing pads prior to final installation in the spent fuel pool.~ The presence of these shims
> does not affect the range of friction coefficients that is used in the dynamic rack simulations. If i sliding does' occur, the bearing pad is expected to remain' stationary as the support pedestal-
~ moves on its surface. This is because the interface between the bearing pad and the support pedestal is the only friction surface that involves two different materials. The liner plate, the bearing pads, and the shim plates are all fabricated with SA240-304 stainless steel, whereas the 3
- support pedestal is fabricated with SA564-630 stainless steel. I i
6.6.4' . Governine Eauations of Motion -
l i
LUsing the structural model discussed in the foregoing, equatior.s of motion corresponding to each l
- degree-of-freedom are obtained using Lagrange's FormuMon [6.6.1]. The system kinetic energy includes contributions from solid structures and from trapped and surrounding fluid. The final
' system'of equations obtained have the~ matrix form:
4
.'d'q.
.[M] _-g,, = M W[
- where:
[M]. -.
total mass matrix (including structural and fluid mass 1 contributions). The size of this matrix will be 22n x22n for a l WPMR analysis (n = number of racks in the model). )
> q1 -
the nodal displacement vector relative to the pool slab ;
displacement (the term with q indicates the second derivative with l respect to time, i.e., acceleration)
- [G) -
a vector dependent on the given ground acceleration SliADED TEXT CONTAINS PROPRIETARY INFORMATION lioltec International 6-17 Report 111-982083 A
J
[Q] -
a vector dependent on the spring forces (linear and nonlinear) and the coupling between degrees-of-freedom The above column vectors have length 22n. The equations can be rewritten as follows:
i
)
= pfJ' [Q] + pf J' [G]
_ l_
This equation set is mass uncoupled, displacement coupled at each instant in time. The numerical solution uses a central difference scheme built into the proprietary computer program DYNARACK [6.2.4].
No convergence problems were experienced during any of the simulations. As demonstrated during an intensive NRC review of DYNARACK in the V.C. Summer station rerack license (ca.
1983) the central difference iteration scheme used in DYNARACK ensures that the solution will be unconditionally convergent. DYNARACK has been used to perform over 2000 seismic simulations of fuel racks in more than 40 dockets over the past 20 years. Stability is achieved after a certain sninimum time step (interval between solutions) is established. The time step is chosen based on previous experience with the solver and is adjusted during the initial runs if solutions are not obtained.
6.7 Structural Evaluation of Spent Fuel Rack Desien 6.7.1 Kinematic and Stress Acceptance Criteria There are two sets of criteria to be satisfied by the rack modules:
- a. Kinematic Criteria An isolated fuel rack situated in the middle of the storage cavity is most vulnerable to overturning because such a rack would be hydrodynamically )
uncoupled from any adjacent structures. Therefore, to assess the margin against overturning, a single rack module is evaluated. According to the O.T. Position paper (USNRC, ca 1978), the minimum required safety margic.s under the OBE j and SSE events is 1.5 and 1.1, respectively. The maximum rotation of the rack (about its two principal axes) is obtained from a post processing of the rack time !
history response output. The ratio of the rotation required to produce incipient !
tipping in either principal plane to the actual maximum rotation in that plane from l SHADED TEXT CONTAINS PROPRIETARY INFORMATION l Holtec International 6-18 Report HI-982083 l l
i-
, the time history solution is the margin of safety. All ratios available for the OBE and SSE events should be greater than 1.5 and 1.1, respectively to satisfy the regulatory acceptance criteria.
- b. Stress Limit Criteria Stress limits must not be exceeded under the postulated load combinations provided herein.
f6.7.2 l Stress Limit Evaluations The stress limits presented below apply to the rack structure and are derived from the ASME Code,Section III, Subsection NF [6.7.1]. Parameters and terminology are in accordance with'the ASME Code. Material properties are obtained from the ASME Code Appendices [6.7.2], and are listed in Table 6.3.1.
(i)' Normal and Upset Conditions (Level A or Level B)
- a. Allowable stress in tension on a net section is:
Fi = 0.6 S y-Where, Sy = yield stress at temperature, and Fi is equivalent to primary membrane stress.
- b. ' Allowable stress in shear on a net section is:
Fy = .4 S _-
y
- c. Allowable stress in compression on a net section '
=
ki' Fa Sy
<.47 444 rs kA/r for the main rack body is based on the full height and cross section of the !
honeycomb region and does not exceed 120 for all sections.
SHADED TEXT CONTAINS PROPRIETARY INFORMATION Holtec h ternational - 6-19 Report HI-982083 k
e l
J A= unsupported length of component k= length coefficient which gives influence of boundary conditions. The - i following values are appropriate for the described end conditions: l
= 1 (simple support both ends) j
=. 2 (cantilever beam)
= % (clamped at both ends) r'.=. radius of gyration of component
- d. Maximum' allowable bending stress at the outermost liber of a net section, due to flexure about one plane of symmetry is:
1 Fe = . 0.60 S y : (equivalent to primary bending)
- e. Combined bending and compression on a net section satisfies:
_h C, fs, ' C., fs, F,' . D, Fs, .- D, Fe, where:
l f, = - Direct compressive stress in the section . I fx b
= Maximum bending stress along x-axis
-fby. = Maximum bending stress along y-axis j
= 0.85' Cnix ;
Co,y - = - 0.85 Dx .
= 1 - (f,/F,x) '
~
D; y
=
1 - (f,/F,y) .
2
' F,(,y- =. (x E)/(2.15 (kl/r)2x y)
.EL = Young's Modulus and subscripts x,y reflect the particular bending plane,
- f. Combined flexure and compression (or tension) on a net section:
l" ^+ +
< 1. 0 0.6 S, Fs, . Fs, SHADED TEXT CONTAINS PROPRIL s'ARY INFORMATION 11oltec International 6-20 Report HI-982083
The above requirements are to be met for both direct tension or compression.
- g. Welds Allowable maximum shear stress on the net section of a weld is given by:
F = 0.3 So where So is the weld material ultimate strength at temperature. For fillet weld legs in contact with base metal, the shear stress on the gross section is limited to 0.4Sy, where Syis the base material yield strength at temperature.
(ii) Level D Service Limits Section F-1334 (ASME Section III, Appendix F) [6.7.2], states that the limits for the Level D condition are the minimum of 1.2 (S /F y t) or 0.7 (So/F t) times the corresponding limits for the Level A condition. Su is ultimate tensile stress at the specified rack design temperature. Examination of material properties for 304L stainless steel demonstrates that 1.2 times the yield strength is less than the 0.7 times the ultimate strength.
Exceptions to the above general multiplier are the following:
a) Stresses in shear shall not exceed the lesser of 0.72S yor 0.42So. In the case of the austenitic stainless steel material used here,0.72Sy governs.
b) Axial Compression Loads shall be limited to 2/3 of the calculated buckling load.
c) Combined Axial Compression and Bending - The equations for Level A conditions shall apply except that:
F, = 0.667 x Buckling Load / Gross Section Area, and the terms Fn and Fey may be increased by the factor 1.65.
d) For we!ds, the Level D allowable maximum weld stress is not specified in Appendix F of the ASME Code. An appropriate limit for weld throat stress is conservatively set here as:
SHADED TEXT CONTAINS PROPRIETARY INFORMATION Holtec Internationa! 6-21 Report HI-982083
1 U m -
- 4 r y
- Fe (0.3 S.) x facto' rl
. where:'-
_j.
i factor = .(Level D shear stress limit)/(Level A shear stress limit) -
- y -
6.-7.3- Dimensionless Stress Factors g ,
" For convenience l the stress results are presented in dimensionless form. ' Dimensionless stress factor's are defined as the ratio of the actual developed stress to the specified limiting value. The
, limiting value of each stress factor is 1.0, based on the allowable strengths for each level, for
- ' Levels A, B, and D (where 1.2Sy < .7S ). Stress factors reported are
. Ri = x Ratio of' direct tensile or compressive stress on a net section to its allowable value (note pedestals only resist compression) -
I
-R2 = L Ratio.of gross shear on a net section in the x-direction to its allowable value {
L R3 1=~ Ratio of maximum x-axis bending stress to its allowable value foi the'section .
1 R4 =, Ratio of maximum y-axis bending stress to its allowable value'for the section R 3 =1 Combined flexure and compressive factor (as defined in the foregoing).
Rd =' Combined flexure and tensiod (or compression) factor (as defined in the foregoing) ,
- R7 .=: ' Ratio of gross shear on a net section in the y-direction to its allowable value
'6.7.4 Loads and Loadine Combinations for Spent Fuel Racks l
~ The applicable loads and their combinations'which must be ' considered in the seismic analysis of
~
~
- rack modules is excerpted from Refs. [6.1.2) and [6.7.3]. The load combinations considered are
!-> identified below: y
{ !
L SHADED TEXT CONTAINS PROPRIETARY INFORMATION 1
- Holtec International 6-22 Report HI-982083
]
i L ,
4 j
Loading Combination Service Level D+L Level A D+L+To D + L + To + E D + L + T, + E Level B D + L + To + Pr D + L + T. + E' Level D D + L + To + Fa The functional capability of the fuel racks must be demonstrated.
where:
D= Dead weight-induced loads (including fuel assembly weight)
L =
Live Load (not applicable for the fuel rack, since there are no moving objects in the rack load path)
Pr = Upward force on the racks caused by postulated stuck fuel assembly Fd= Impact force from accidental drop of the heaviest load from the maximum possible height.
E= Operating Basis Earthquake (OBE)
E' = Safe Shutdown Earthquake (SSE)
To = Differential temperature induced loads (normal operating or shutdown condition based on the most critical transient or steady state condition)
T= Differential temperature induced loads (the highest temperature associated with the postulated abnormal design conditions) t T, and To produce local thermal stresses. The worst thermal stress field in a fuel t ack is obtained when an isolated storage location has a fuel assembly generating heat at maximum postulated I rate and surrounding storage locations contain no fuel. Heated water makes unobstructed contact with the inside of the storage walls, thereby producing maximum possible temperature difference )
between adjacent cells. Secondary stresses produced are limited to the body of the rack; that is, support pedestals do not experience secondary (thermal) stresses.
l SIIADED TEXT CONTAINS PROPRIETARY INFORMATION 11oltec International 6-23 Report 111-982083
r l
6.8 Parametric Simulations l
Consideration of the parameters described above results in a number of scenarios for both the WPMR and the Single Rack analyses. The single rack analysis considers only one rack in the analysis model whereas the WPMR' analysis considers all racks in the model. Since the proposed l
. pool layout and rack modules for both plants are exactly alike, the set of WPMR and Single Rack models developed for one plant is equally valid for the other plant. The pool layout is shown in Figure 2.1. The rack numbering scheme used in the dynarack model for WPMR simulation is introduced in Figure 6.8.1 along with the gap spring numbering scheme.
l The Single Rack analysis considers the heavie'st rack module and the rack moduie with highest ;
aspect ratio (i.e. the ratio oflength to width of a rack). Rack K (one of the heaviest racks) and
]'
Rack J (featuring the highest aspect ratio) are selected for single rack analysis. The single rack model with the heaviest rack is most likely to produce the largest pedestal loads while the model with the highest aspect ratio rack is highly susceptible to large displacements using the Single Rack method. In addition to these Single Rack simulations, a Single Rack run that exhibits the greatest displacement is re-run solely for the purpose of checking the potential for overturning.
I The table below presents a complete listing of the simulations discussed herein.
LIST OF RACK SIMULATIONS Run Model Load Case COF Event 1 WPMR Full Pool 0.2 SSE 2 WPMR' Full Pool 0.8 SSE 3 WPMR Full Pool Random SSE 4 WPMR Full Pool 0.2 OBE 5 WPMR Full Pool 0.8- OBE 6 WPMR Full Pool Random OBE '
7; Single Rack (J) Full Rack 0.8 SSE 8- Single Rack (J) Full Rack 0.2 SSE 9 Single Rack (J) Half-Full Rack (synunetric ' O.8 SSE about short axis) 10 Single ' Rack (J) Half-Full Rack (synunetric 0.2 SSE about short axis) 11 Single Rack (J) Nearly Empty 0.8 SSE SHADED TEXT CONTAINS PROPRIETARY INFORMATION lloltec International 6-24 Report 111-982083 m
I LIST OF RACK SIMULATIONS I Run Model Load Case COF Event l 12 Single Rack (J) Nearly Empty 0.2 SSE 13 Single Rack (K) Full Rack 0.8 SSE 14 Single Rack (K) Full Rack 0.2 SSE q 15 Single Rack (K). Half-Full Rack (symmetric 0.8 SSE about short axis) -
16 Single Rack (K) Half-Full Rack (symmetric 0.2 SSE l about short axis) 17 Single Rack (K) Nearly Empty 0.8 SSE l8 Single Rack (K) Nearly Empty 0.2 SSE 19 Single Rack (J) Full Rack ' O.8 OBE 20- Single Rack (J) . Full Rack 0.2 OBE 21 Single Rack (J) Half-Full Rack (symmetric 0.8 OBE about short axis) 22 Single Rack (J) Half-Full Rack (symmetric 0.2 OBE about short axis) l 23 Single Rack (J) Nearly Empty 0.8 OBE 24 Single Rack (J) Nearly Empty 0.2 OBE 25 Single Rack (K) Full Rack 0.8 OBE
-26 Single Rack (K) Full Rack 0.2 OBE 27 Single Rack (K) Half-Full Rack (symmetric 0.8 OBE about short axis) 28 Single Rack (K) Half-Full Rack (symmetric 0.2 OBE about short axis) 29 Single Rack (K) Nearly Empty - 0.8 OBE 30 . Single Rack (K) Nearly Empty 0.2 OBE 31 Single Rack Full Rack 0.8 SSE ,
Overturning Check (K) p where:
Random = Gaussian distribution with a mean coefficient of friction (COF) of 0.5 and a standard
- deviation equal to 0.15.
SHADED TEXT CONTAINS PROPRIETARY INFORMATION j Holtee International 6-25 Report HI-982083 i
i-l b .
. Note that Run No. 31 is a re-run of Run No.13 except that the rack module in this run is l' i simulated as an isolated rack in the pool as ' described earlier in subsection 6.7.1.
L t-6.9 :- Time History Simulation Results.
l The results from the DYNARACK runs may be seen in the raw data output files. However, due to the huge quantity of output data, a post-processor is used to scan for worst case conditions and =
- develop l the stress' factors. ; Further reduction in this bulk of information is provided in this .
- section' by extracting the.-- worst.: case 1 values. from the parameters of interest; namely -
. displacements, support pedestal forces, impact loads,' and stress factors. This section also summarizes other analyses performed to develop and evaluate structural member stresses which l
. are not determined by the post processor. For each table, the Pool Condition /COF column refers L to whether the pool is full, half full or nearly. empty (a few cells loaded). COF is the interface
-coefficient.of friction discussed in subsection 6.2.1. The " Rack" column denotes racks by l number. (applicable to the DYNARACK model) and by letter (applicable to the pool layout )
drawing). l 6.9.1 Rack' Disolacements - l l
. A tabulated summary of the maximum displacement for each simulation is provided below. It is j noted that all of the maximum displacements -occurred at the tops of the storage racks, as-expected, from swaying, bending and tipping. The location / direction terms are defined as !
- follows: l l
)
uxt, nyt'. = displacement of top corner of rack, relative to the slab, in the North-South and East-West directions, respectively, in the Braidwood pool. The maximum displacements for every simulation, including the single rack tipover run, occurred ;
at the top of the racks shown in the last table column.
SHADED TEXT CONTAINS PROPRIETARY INFORMATION Holtec International - 6-26 Report HI-982083 l
i RACK DISPLACEMENT RESULTS Run Model Pool Event ' Max. Direction Rack Condition / Disp. (in)
COF l' -WPMR Full /0.2 SSE 0.688 uyt 14(P) 2 WPMR Full /0.8 SSE 0.865 uxt 12(M) 3 WPMR Full / Rand.' SSE 0.823 uxt 12(M) 4 WPMR Full /0.2 OBE 0.468 uyt 12(M) 5 WPMR Full /0.8 OBE 0.467 uyt 12(M)
WPMR OBE 0,467 12(M) 6 Full / Rand. uyt 7' Single Rack Full /0.8 SSE 10.234 uyt 9(J) 8 Single Rack Full /0.2 SSE 0.207 uyt 9(J) 9 Single Rack Half /0.8 SSE 0.124 uyt 9(J) 10 Single Rack Half /0.2 SSE 0.116 uyt 9(J) 11 Single Rack Empty /0.8 SSE 0.0256 uyt 9(J) 12 Single Rack Empty /0.2 SSE 0.0232 uyt 9(J)
Single Rack Full /0.8 SSE 0.162 uxt' 13 10(K) 14 Single Rack Full /0.2 SSE 0.143 uxt 10(K) 15 Single Rack Half /0.8 SSE 0.0766 uxt 10(K) 16 Single Rack Half /0.2 SSE 0.0682 uyt 10(K) 17 Single Rack Empty /0.8 SSE 0.0262 uyt 10(K)
Single Rack Empty /0.2 SSE 0.0265 uyt 18 10(K) 19 Single Rack Full /0.8 OBE 0.09 uxt 9(J) 20 Single Rack Full /0.2 OBE 0.0897 uxt 9(J)
- 21. Single Rack Half /0.8 OBE 0.0665 uxt 9(J) 22' Single Rack Half /0.2 'OBE C.0634 uxt 9(J) 23 Single Rack Empty /0.8 OBE 0.0118 uxt 9(J) 24 Single Rack Empty /0.2 OBE 0.0121 uxt 9(J) 2* Single Rack Full /0.8 OBE 0.0915 uyt 10(K) 26 Single Rack Full /0.'2 OBE 0.072 uyt 10(K) 27 Single Rack Half /0.8 OBE 0.058 uyt 10(K)
SIIADED TEXT CONTAINS PROPRIETARY INFORMATION llottec Intemational 6-27 Report 111-982083
RACK DISPLACEMENT RESULTS Run Model Pool Event Max. Direction Rack Condition / Disp. (in)
COF 28 Single Rack Half /0.2 OBE 0.050 uyt 10(K) 29 Single Rack Empty /0.8 OBE 0.0148 uyt 10(K) 30 Single Rack Empty /0.2 OBE 0.0145 uyt 10(K) 31 Single Rack Single Rack SSE 1.500 uyt 10(K)
Overturning Check The table shows that the maximum rack displacement is 1.50 inches (Run 31). With this given value, an evaluation of rack overturning is performed. The factor of safety obtained from this evaluation is 61, which is much higher than the prescribed limit of 1.5 for OBE conditions. This indicates that tipover is not a concern. Table 6.9.1 shows the maximum calculated and maximum allowed rotation of the rack.
The maximum rack displacements at the baseplate elevation are 0.5975 inches (Run No.1) and 0.2358 inches (Run No. 4) for the SSE event and the OBE event, respectively.
The displacement shape of each rack, from the baseplate elevation to the top of the rack, is nearly linear. This indicates that the primary displacement mode of the fuel racks is rigid body motion (i.e., sliding and tilting). In other words, the clastic deformation of the cell structure due to bending is negligible compared to the rigid body displacements.
6.9.2 Pedestal Vertical Load Pedestal number 1 for each rack is located in the northeast and the southwest comer of racks of the Braidwood and Byron station, respectively. Numbering increases counterclockwise around the periphery of each rack. The following bounding vertical pedestal forces are obtained for each run:
SIIADED TEXT CONTAINS PROPRIETARY INFORMATIO'N Iloitec International 6-28 Report HI-982083
MAXIMUM VERTICAL LOADS Run Model Pool Event Max. Vertical Rack l Condition / Load (th) !
COF 1 WPMR Full /0.2 SSE 198000 18(T) i l 2 WPMR' Full /0.8 SSE 229000 18(T) 3 WPMR Full / Rand. SSE 238000 2(B) 4 WPMR Full /0.2 OBE 131000 23(Y) 5 WPMR Full /0.8 OBE 144000 11(L) 6 WPMR Full / Rand. OBE 144000 11(L) 7 Single Rack Full /0.8 SSE 169000 9(y)
L 8 Single Rack Full /0.2 SSE 131000 9(y) 9 Single Rack Half /0.8 SSE 78200 9(y) 10 Single Rack Half /0.2 SSE 68700 9(y) 11 Single Rack Empty /0.8 SSE 15000 9(y)
^2 Single Rack Empty /0.2 SSE 13900 9(y) 13 Single Rack Full /0.8 SSE 109000 l 10(K) 14 Single Rack Full /0.2 SSE 101000 10(K) 15 Single Rack Half /0.8 SSE 52900 10(K) !
16 Single Rack Half /0.2 SSE 56500 10(K) 17 Single Rack Empty /0.8 SSE 20600 10(K) 18 Single Rack Empty /0.2 SSE 19600 10(K) 19 Single Rack Full /0.8 OBE 92300 9(y) 20 Single Rack Full /0.2 OBE 92500 9(y) 21 Single Rack Half /0.8 OBE 51400 9(y) 22 Single Rack Half /0.2 OBE 51300 9(y) 23 Single Rack Empty /0.8 OBE 10800 9(y) 24 Single Rack Empty /0.2 OBE 10900 9(y) 25 Single Rack Full /0.8 OBE 81400 10(K) 26 Single Rack Full /0.2 OBE 79700 10(K) 27 Single Rack Half /0.8 OBE 42800 10(K)
SHADED TEXT CONTAINS PROPRIETARY INFORMATION lloltec International 6-29 Report HI-982083
i MAXIMUM VERTICAL LOADS Run Model Pool Event Max. Vertical Rack Condition / Load (Ib)
COF 28 Single Rack Half /0.2 OBE 42800 10(K) 29 Single Rack Empty /0.8 OBE 14500 10(K) 30 Single Rack Empty /0.2 OBE 14300 to(g) J The highest maximum vertical pedestal loads from all simulations for the SSE and OBE conditions are 238,000 lbs and 144,000 lbs, respectively. The effect of these loads is evaluated in bearing pad and rack fatigue analyses.
6.9.3 Pedestal Friction Forces The maximum (x or y direction) shear load bounding all pedestals in the simulation are reported below and are obtained by compilation of the complete tabular data produced by time history solution of whole pool multirack and single rack simulations.
MAXIMUM HORIZONTAL LOADS Run Model Pool Event Max. Shear Rack Condition / Load (lb) i COF 1 WPMR Full /0.2 SSE 33200 20(V) 2 WPMR Full /0.8 SSE 101000 11(L) 3 WPMR Full / Rand. SSE 83300 3(C) 4 WPMR Full /0.2 OBE 24200 16(R) 5 WPMR Full /0.8 OBE 50100 3(C) 6 WPMR Full / Rand. OBE 48500 4(D) 7 Single Rack Full /0.8 SSE 55200 9(y) 8 Single Rack Full /0.2 SSE 24600 9(y) 9 Single Rack Half /0.8 SSE 33800 9(y) 10 Single Rack Half /0.2 SSE 13200 9(y) 11 Single Rack Empty /0.8 SSE 4990 9(y)
SliADED TEXT CONTAINS PROPRIETARY INFORMATION Holtec Intemational 6-30 Report 111-982083 1 i
m MAXIMUM HORIZONTAL LOADS Run Model Pool Event Max. Shear Rack Condition / Load (th)
COF 12 Single Rack Empty /0.2 SSE 2740 9(y) 13 Single Rack Full /0.8 SSE 50000 10(K) 14 Single Rack Full /0.2 SSE 20000 10(K) 15 Single Rack Half /0.8 SSE 19200 10(K) 16 Single Rack Half /0.2 SSE 9720 10(K) 17 Single Rack Empty /0.8 SSE 6270 10(K) 18 Single Rack Empty /0.2 SSE 3600 10(K) 19 Single Rack Full /0.8 OBE 10000 9(y) 20 Single Rack Full /0.2 OBE 16000 9(y) 21 Single Rack Half /0.8 OBE 7540 9(y) 22 Single Rack Half /0.2 OBE 10100 9(y) 23 Single Rack Empty /0.8 OBE 1380 9(y) 24 Single Rack Empty /0.2 OBE 1940 9(y) 25 Single Rack Full /0.8 OBE 8660 10(K) l 26 Single Rack Full /0.2 OBE 11800 10(K) 27 Single Rack Half /0.8 OBE 14600 10(K) 28 Single Rack Half /0.2 OBE 8170 10(K) 29 Single Rack Empty /0.8 OBE 2130 10(K) j 30 Single Rack - Empty /0.2 OBE 2390 to(g)
The largest horizontal pedestal load of 101,000 lbs occurs in run 2. The effect of this load is '
evaluated in the liner fatigue analysis.
6.9.4 Rack Impact Loads l A freestanding rack, by definition, is a structure subject to potential impacts during a seismic ]
event. Impacts arise from rattling of the fuel assemblies in the storage rack locations and, in some instances, from localized impacts between the racks, or between a peripheral rack and the pool wall.
SHADED TEXT CONTAINS PROPRIETARY INFORMATION !
Holtec International 6-31 Report HI-982083 1
l l
1 Results of simulations predict no impact (in the rack cellular region) between racks, or with walls under any simulation. The time history solution does indicate the occurrence oflocalized impacts at rack baseplate location. The maximum local rack-bottom impact force from each set are reported as fo!!ows:
Run Impact Force,Ibf Analysis 2 80,070 WPMR 8 13,980 SINGLE RACK The rack baseplates, which are manufactured from a continuous steel plate, which is 0.75 inch thick, can sustain impacts at the baseplate level that are greater than the forces listed above. The compressive stress in the baseplate due to the maximum impact load of 80,070 lbs is 8,897 psi, which is less than the yield stress of the baseplate material (21,300 psi). Therefore, the calculated rack-to-rack impact loads at the baseplate are acceptable, f 6.9.4.1 Fuel to Cell WallImpact Loads As discussed in subsection 6.5.1 the fuel assemblies are modelled using five lumped masses.
Each of these masses interacts with the rack via four nonlinear compression only spring elements. These elements are termed nonlinear, since they have the capability of being inactive (non-load bearing) until the fuel assembly mass comes in contact with the cell wall. The loads developed by these elements are conservative, since the actual assembly to cell wall impacts will be experienced at the assembly spacer grids. Since the number of spacer grids are greater than the number oflumped masses used in the model, the impact loadings will actually occur at more locations, resulting in lower loads at each point of contact.
The DYNARACK program produces a complete time history of the loadings within these non-linear compression only gap / spring elements and archives the results for later review or post-processing. Post-processors enable scanning of the large number of time steps for instants where loads are actually present and allows for easy retrieval of the bounding values.
Fuel assembly integrity is assured by comparing the calculated impact load against manufacturers test data for assembly grid side loadings. Additionally, it should be noted that the impact lo4s experienced by the fuel assemblies from postulated seismic event during storage in SIIADED TEXT CONTAINS PROPRIETARY INFORMATION lloltec Intemational 6-32 Report 111-982083
the racks is expected to be exceeded by the loadings experienced during service within the reactor under similar events as previously analyzed and accepted during original plant licensing.
The cell wall integrity is determined by cornparison of the calculated load with an allowable impact load developed using plastic analysis of the local cell wall impact zone.
The fluid coupling between the fuct assembly and the cell wall is treated by inertial coupling in the system Ju. tic energy. The methodology is taken from classical mechanics as described by Fritz in, "The Effects of Liquids on the Dynamic Motions ofimmersed Solids" [6.5.3]and by j Singh and Soler in " Dynamic Coupling in a Closely Spaced Two-Body System Vibrating in )
4 Liquid Medium: The Case of Fuel Racks" [6.5.2].
A review of all simulations performed allows determination of the maximum instantaneous ;
impact load between fuel assembly and fuel cell wall at any modeled impact site. The maximum fuel to cell wall impact load values are reported in the following table.
FUEL-TO-CELL WALL IMPACT Run Model Pool Event Impact Load Rack l Condition / (lh)
COF l
[, I WPMR Full /0.2 SSE 822 4(D) 2 WPMR Full /0.8 SSE 830 4(D) 3 WPMR Full / Rand. SSE 828 4(D)
! 4 WPMR Full /0.2 OBE 42R 10(K) ,
5 WPMR Full /0.8 OBE 4 5 4(D) i 6 WPMR Full / Rand. G3E $ 4(D) 7 Single Rack Full /0.8 SSE 55L 9(y) 8 Single Rack Full /0.2 SSE 570 9(y) 9 Single Rack 11alf/0.8 SSE 529 9(j) 10 Single Rack llalf/0.2 SSE 514 9(j) 11 Single Rack Empty /0.8 SSE 48 9(y) 12 Single Rack Empty /0.2 SSE 565 9(j)
Single Rack Full /0.8 SSE 829 13 10(K)
SilADED TEXT CONTAINS PROPRIETARY INFORMATION lloltec International 6-33 Report 111-982083
1 l
l l
FUEL-TO-CELL WALL IMPACT Run Model Pool Event Impact Load Rack Condition / fjlb)
COF 14' Single Rack Full /0,2 SSE 822 10(K) m 15 Single Rack Half /0.8 yr 780 1 10(K) 16 Single Rack Half /0.2 i StL 559 10(K) 17 Single Rack Empty /0.8 SSE 808 io(g) 18 . Single Rack Empty /0.2 SSE 448 10(K).
19 Single Rack Full /0.8 CBE 292 9(y)
I 20 Single Rack - Full /0.2 OBE '291 9(y) 21 Single Rack Half /0.8 OBE 329 9(y) i 22 Single Rack Half /0.2 OBE 341 9(J) 23 Single Rack Empty /0.8 OBE I148 9(J) 24 Single Rack Empty /0.2 OBE 1113 9(y) 25 Single Rack Full /0.8 'OBE 387 10(K) 26 Single Rack Full /0.2 OBE 389 10(K) 27 . Single Rack' Half /0.8 OBE 893 10(K) 28 Single Rack Half /0.2 OBE 893 10(K)
-29 Single Rack - Empty /0.8 'OBE 1118 10(K) ,
30 Single Rack Empty /0.2 'OBE I120 jo(g) j The maximum fuel-to-cell t .dl impact load is recorded to be 1,148 lbs during run no. 23. The structural integrity of the cell wall under the impact of this load is evaluated. The discussion of
' this evaluation is provided in section 6.10.3 of this report.
The permissible lateral load on an irradiated spent fuel assembly has been studied by the Lawrence Livermore National Laboratory. The LLNL report [6.10.1] states that "...for the most vulnerable fuel assembly, axial buckling varies from 82g's at initial storage to 95g's after 20 !
- years' storage. In a side drop, no yielding is expected below 63g's at initial storage to 74g's after I 20 years' [ dry] storage". The most significant load on the fuel assembly arises from rattling during the seismic event. For the five lumped mass model, the limiting lateral load, therefore, is equal to F , where SilADED TEXT CONTAINS PROPRIETARY INFORMATIO?
Iloltec International 6-34 Repon 111-982083
l l
F, = (w x a)/5 l-- where:
.w=
weight of one fuel assembly (upper bound value = 1600 lbs) a= permissible lateral acceleration in g's (a = 63)
Therefore, F, = 20,160 lbs.-
I The maximum fuel-to-cell wall impact force from the array of parametric runs listed in the above table is 1,148 lbs. Therefore, the nominal factor of safety against Spent Nuclear Fuel (SNF) failure is computed to be 17.
6.9.5 Rack Vertical Displacement 4
A tabulated summary of the maximum vertic I displacement for each simulation is provided b'elow. Note that these displacements represent the rack lift-off during the seismic event, as a result of rocking, sliding, swaying, bending, and tipping behavior of the rack module.
l RACK VERTICAL DISPLACEMENT RESULTS Run Model Pool Condition /- Event Max. direction Rack COF Vertical Disp. (in)
-1 WPMR Full /0.2 SSE 0.0998 upward 23/Y 2 WPMR Full /0.8 SSE I '334 upward 2/B 3 WPMR Full / Rand. SSE 0.1301 upwe d ll/L 4 WPMR . Full /0.2 OPE 0.0754 upward 21/W 5' WPMR Full /0.8 OBE 0.0766 upward 21/W 6 WPMR Full / Rand. OBE 0.0761 upward 21/W i
7 Single Rack Full /0.8 SSE 0.087 upward 9(y) 8 Single Rack Full /0.2 SSE 0.087 upward 9(y) 9- Single Rack IIalf/0.8 SSE 0.043 upward 9(y) j 10 Single Rack lialf/0.2 SSE 0.0426 upward 9(y)
SIIADED TEXT CONTAINS PROPRIETARY INFORMATION Holtec International 6-35 Report 111-982083 I'
E
I l
l RACK VERTICAL DISPLACEMENT RESULTS Run Model Pool Condition / Event Max. direction Rack COF Vertical l Disp. (in) 11 Single Rack Empty /0.8 SSE 0.0084 upward 9(y) 12 Single Rack Empty /0.2 SSE 0.0084 upward 9(j) 13 Single Rack Full /0.8 SSE 0.0443 upward 10(K) 14 Single Rack Full /0.2 SSE 0.0431 upward 10(K) 15 Single Rack Half /0.8 SSE 0.020 upward 10(K) 16 Single Rack Half /0.2 SSE 0.0219 upward 10(K) 17 Single Rack Empty /0.8 SSE 0.0070 upward jo(g) 18 Single Rack Empty /0.2 SSE 0.0069 upward 10(K) 19 Single Rack Full /0.8 OBE 0.0675 upward 9(y) 20 Single Rack Full /0.2 OBE 0.0675 upward 9(y) 21 Single Rack Half /0.8 OBE 0.0343 upward 9(y) 22 Single Rack Half /0.2 OBE 0.0343 upward 9(y) 23 Single Rack Empty /0.8 OBE 0.0069 upward 9(y) 0.0069 upward I 24 Single Rack Empty /0.2 OBE 9(y) 25 Single Rack Full /0.8 OBE 0.0344 upward 10(K) 26 Single Rack Full /0.2 OBE 0.0344 upward 10(K) 27 Single Rack Half /0.8 OBE 0.0176 upward 10(K) ;
28 Single Rack Half /0.2 OBE 0.0176 upward 10(K) 29 Single Rack Empty /0.8 OBE 0.0058 upward !
10(K) 30 Single Rack Empty /0.2 OBE 0.0058 upward to(g)
Single Rack Single Rack SSE 0.3199 upward 31_ 10(K)
Overturning Check The maximum vertical displacement is 0.3199 inches which occurs in run No. 31.
l >
SilADED TEXT CONTAINS PROPRIETARY INFORMATION lloitec International 6-36 Repon 111-982083 t
s 6.10 Rack Structural Evaluation 6.10.1 Rack Stress Factors l l
With time history results available for pedestal normal and lateral interface forces, the maximum I values for the previously defined stress factors can be determined for every pedestal in the array of racks. With this infonnation available, the structural integrity of the pedestal can be assessed j
- and reported. The net section maximum (in time) bending moments and shear forces can also be determined at the bottom casting-rack cellular structure interface for each spent fuel rack in the pool. With this information in hand, the maximum stress in the most stressed rack cell (box) can I be evaluated.
An evaluation of the stress factors for all of the simulations performed leads to the conclusion that all stress factors are less than the mandated limit of 1.0 for the load cases examined. From all of the simulations, the bounding stress factors for each run, in either the cellular or the pedestal region, are summarized below :
l MAXIMUM STRESS FACTORS ,
1
'Run Model Pool Event Stress Factor Rack / Factor l Condition / Tyne COF 1 WPMR Full /0.2 SSE 0.252 18(T)/R5 2 WPMR Full /0.8 SSE 0.443 3(C)/R6 3 WPMR Full / Rand. SSE 0.407 11(L)/R6 4 WPMR Full /0.2 OBE 0.428 12(M)/R6 5 WPMR Full /0.8 OBE 0.494 12(M)/R6 6 WPMR Full / Rand. OBE 0.459 12(M)/R6 7 Single Rack Full /0.8 SSE 0.302 9(J)/R6 8 Single Rack Full /0.2 SSE 0.173 9(J)/R6 9 Single Rack Half /0.8 SSE 0.140 9(J)/R6 10 Single Rack Half /0.2 SSE 0.091 9(J)/R5 11 Single Rack Empty /0.8 SSE 0.025 9(J)/R6 12 Single Rack Empty /0.2 SSE 0.018 9(J)/R5 SHADED TEXT CONTAINS PROPRIETAPY INFORMATION lloitec International 6-3"/ Report 111-982083 I
MAXIMUM STRESS FACTORS Run Model Pool Event Stress Factor Rack / Factor ;
Condition / Type l COF 13 Single Rack Full /0.8 SSE 0.248 10(K)/R6 j 14 Single Rack Full /0.2 SSE 0.125 10(K)/R5 15 Single Rack Half /0.8 SSE 0.087 10(K)/R6 f 16 Single Rack Half /0.2 SSE 0.067 10(K)/R5,R6 !
17 Single Rack Empty /0.8 SSE 0.028 10(K)/R5 18 Single Rack Empty /0.2 SSE 0.023 10(K)/R5 19 Single Rack Full /0.8 OBE 0.228 9(J)/R5 20 Single Rack Full /0.2 OBE 0.229 9(J)/R5 21 Single Rack Half /0.8 OBE 0.131 9(J)/R5 22 S:ngle Rack Half /0,2 OBE 0.130 9(J)/R5 23 Single Rack Empty /0.8 OBE 0.028 9(J)/R5 24 Single Rack Empty /0.2 OBE 0.029 9(J)/R5 25 . Single Rack Full /0.8 OBE 0.170 10(K)/R5 26- Single Rack Full /0.2 OBE 0.169 10(K)/R5 l 27 Single Rack Half /0.8 OBE 0.137 10(K)/R6 l 28 Single Rack Half /0.2 OBE 0.103 10(K)/R5 29 Single Rack Empty /0.8 OBE 0.031 10(K)/R5,R6 30 Single Rack Empty /0.2 OBE 0.032 10(K)/R5 The maximum stress factor scanned from above table is 0.443 for SSE and 0.494 for OBE, which is below the prescribed Code limit of 1.0. Therefore, the stress allowables are indeed satisfied for the load levels considered for every limiting location in every rack in the array.
6.10.2 Pedestal Thread Shear Stress ,
l The complete post-processor results give thread stresses under faulted conditions for every pedestal for every rack in the pool. The average shear stress in the engagement region is given below for the limiting pedestal in each simulation.
SHADED TEXT CONTAINS PROPRIETARY INFORMATION Iloltec International 6-38 Report HI-982083
THREAD SIIEAR STRESS Run Model - Pool Event Stress (psi) Rack Condition / Numher t
COF 1 WPMR Full /0.2 SSE 8404 18(T) 2 WPMR Full /0.8 SSE 9719 18(T) 3 WPMR Full / Rand. SSE 10102 2(B) 4 WPMR Full /0.2 OBE 5560 23(Y 5 WPMR Full /0.8 OBE 6112 11(L)
~6 WPMR Full / Rand. OBE 6112 11(L) 7 Single Rack Full /0.8 SSE 7173 9(y) 8 Single Rack Full /0.2 SSE 5560 9(y) 9 Single Rack Half /0.8 SSE 3319 9(y) 10 Single Rack Half /0.2 SSE 2916 9(y) 11 ' Single Rack Empty /0.8 SSE 637 9(y) 12 Single Rack Empty /0.2 -SSE 590 9(y) i 13 Single Rack Full /0.8 SSE 4626 10(K) 14 Single Rack Fall /0,2 SSE 4287 10(K) l Singk Rack Half /0.8 SSE 2245 15 10(K) 2398 l 16 Single Rack Half /0.2 SSE 10(K)
Single Rack Empty /0.8 SSE 874 17 10(K) 18 Single Rack Empty /0.2 SSE 832 to(g) 19 Single Rack Full /0.8 OBE 3918 9(y) 20 Single Rack Full /0.2 OBE 3926 9(y) 21 Single Rack Half /0.8 OBE 2181 9(y) 22 Single Rack Half /0.2 OBE 2177 9(y) 23 Single Rack Empty /0.8 OBE 458 9(y) 24 Single Rack Empty /0.2 OBE 462 9(y) 25 Single Rack Full /0.8 OBE 3455 jo(g) 26 Single Rack Full /0.2 OBE 3383 10(K)
Single Rack Half /0.8 OBE 1817 l 27 10(K)
SliADED TEXT CONTAINS PROPRIETARY INFORMATION llottec Intemational 6-39 . Report 11I-982083
o i
THREAD SHEAR STRESS Run Model Pool Event Stress (psi) Rack ,
Condition / Number )
COF i 28 . Single Rack Half /0.2 OBE 1817 10(K) 29 Single Rack Empty /0.8 OBE 615 10(K) 30 Single Rack Empty /0.2 OBE 607 10(K)
I The ultimate strength of the internally threaded part of the pedestal is 66,200 psi. The yield stress for this material is 21,300 psi. The allowable shear stress for Level B (OBE) conditions is 0.4 times the yield stress which gives 8,520 psi and the allowable shear stress for level D is 0.72 times the yield stress which gives 15,336 psi. The maximum calculated shear stress value for the !
SSE 'is 10,102 psi and 6112 pJ. for the OBE which are less than their respective allowables.
]
Therefore, thread shear stresses are acceptable under all conditions. ;
I 6.10.3 Local Stresses Due to impacts Impact loads at the pedestal base (discussed in subsection 6.9.2) produce stresses in tne pedestal for which explicit stress limits are prescribed in the Code. However, impact loads on the cellular region of the racks, as discussed in subsection 6.9.4.1 above, produce stresses which attenuate i rapidly away from the loaded region. This behavior is characteristic of secondary stresses. ;
Even though limits on secondary stresses are not prescribed in the Code for Class 3 NF structures, evaluations must be made to ensure that the localized impacts do not lead to plastic l deformations in the storage cells which affect the subcriticality of the stored fuel array.
L.ocal cell wall integrity is conservatively estimated from peak impact loads. Plastic analysis is !
used to obtain the limiting impact load which would lead to gross permanent deformation. Table 6.9.1 indicates that the limiting impact load (3,698 lbf, including a safety factor of 2.0) is much greater than the highest calculated impact load value (1,148 lbf, see subsection 6.9.4.1) obtained 3 from any of the rack analyses. Therefore, fuel impacts do not rer, resent a concern with respect to fuel rack cell deformation.
SilADED TEXT CONTAINS PROPRIETARY INFORMATION lloltec International 6-40 Report 111-982083 1
)
6.10.4 Assessment of Rack Faticue Marcin I
Deeply submerged high density spent fuel storage racks arrayed in close proximity to each other I in a free-standing configuration behave primarily as a nonlinear cantilevered structure when subjected to 3-D seismic excitations. In addition to the pulsations in the vertical load at each I pedestal, lateral friction forces at the pedestal / bearing pad-liner interface, which help prevent or mitigate lateral sliding of the rack, also exert a time-varying moment in the baseplate region of the rack. The friction-induced lateral forces act simultaneously in x and y directions with the requirement that their vectorial sum does not exceed N where is the limiting interface l coefficient of friction and N is the concomitant vertical thrust on the liner (at the given time instant). As the vertical thrust at a pedestal location changes, so does the maximum friction force, F, that the interface can exert.- In other words, the lateral force at the pedestal / liner interface, F, is given by l
F s y N(r)
I where N (vertical thrust) is the time-varying function of T. F does not always equal N; rather, N is the maximum value it can attain at any time. The actual value, of course, is determined by the dynamic equilibrium of the rack structure. In summary, the horizontal friction force at the pedestal / liner interface is a function of time. Its magnitude and direction of action varies during the earthquake event.
The time-varying lateral (horizontal) and vertical forces on the extremities of the support pedestals produce stresses at the root of the pedestals in the manner of an end-loaded cantilever.
The stress field in the cellular region of the rack is quite complex, with its maximum values located in the region closest to the pedestal. The maximum magnitude of the stresses depends on the severity of the pedestal end loads and on the geometry of the pedestal / rack baseplate region.
- Alternating stresses in metals produce metal fatigue if the amplitude of the stress cycles is sufficiently large. In high density racks designed for sites with moderate to high postulated seismic action, the stress intensity amplitudes frequently reach values above the material endurance limit, leading to expenditure of the fatigue " usage" reserve in the material.
Because .he locations of maximum stress (viz., the pedestal / rack baseplate junction) and the close placement of racks, a post-earthquake inspection of the high stressed regions in the racks is SliADED TEXT CONTAINS PROPRIETARY INFORMATION 11oltec International 6-41 Report 111-982083
t not feasible. Therefore, the racks must be engineered to withstand multiple earthquakes without reliance on nondestructive inspections for post-earthquake integrity assessment. The fatigue life evaluation of racks is an integral aspect of a sound design.
l The time-history method of analysis employed in this report provides the means to obtain a l
complete cycle history of the stress inteusities in the highly stressed regions of the rack. Havmg determined the amplitude of the stress intensity cycles and their number, the cumulative damage factor, U, can be determined using the classical Miner's rule ;
.U = I E i Ni where ni is the number of stress intensity cycles cf amplitude oi, and N i is the permissible number of cycles corresponding to cifrom the ASME fatigue curve for Ge material of construction. U must be less than or equal to 1.0.
To evaluate the cumulative damage factor, a conservative model of a portion of the spent fuel rack in the vicinity of a support pedestal is constructed. Using the archived results of the spent fuel rack dynamic analyses (pedestal load histories versus time) enables a time-history of stress intensity to be established at the most limiting location. This permits establishing a set of alternating stress intensity ranges versus cycles for an SSE and an OBE event. Based on ASME Code Subsection NF guidelines, the cumulative damage factor (U) is conservatively calculated to be 0.950 due to the combined effect of one SSE and twenty OBE events. This value is below the ASME Code limit of 1.0.
6.10.5 Weld Stresses Weld locations subjected to significant seismic loading are at the bottom of the rack at the baseplate-to-cell connection, at the pedestal-to-baseplate connection, and at cell-to-cell connections. Bounding values of resultant loads are used to qualify these connections. The paragraphs below summarize each of the weld evaluations. The numerical results are also summarized in Table 6.9.1.
SilADED TEXT CONTAINS PROPRIETARY INFORMATION Holtec International 6-42 Report 111-982083
(
i l
l
- a. Basenlate-to-Cell Welds The highest predicted weld stress for SSE is calculated from the highest R6 value (provided in 6.10.1 above). The ratio of 2.15 is developed from the differences in material thickness and length versus weld throat dimension and length ;
I RATIO = ( M in
- MN in) /(hi$ in
- 0.7071
- lin) = 2.15 i R6 (obe) * [(0.6) Fy]
- RATIO = 0.494 * [0.6
- 21300 psi]
- 2.15 = 13,574 psi R6 (sse) * [(1.2) Fy]
- RATIO = 0.443 * [1.2
- 21300 psi)
- 2.15 = 24,345 psi The above calculated values are less than the OBE allowable weld stress value of 19,860 psi and weld stress allowable va'ue of 35,748 psi. Therefore, weld stresses between the ;
baseplate and cell wall base are acceptable.
I
- b. Baseplate-to-Pedestal Welds The maximum weld stress between the baseplate and the support pedestal,15,380 psi under an SSE event and 13,600 psi under an OBE event, is verified to be less than the allowable values of 35,748 psi and 19,860 psi, respectively.
- c. Cell-to-Cell Welds Cell-to-cell connections are formed by a series of connecting welds along the cell height.
Stresses in storage cell to cell welds develop due to fuel assembly impacts with the cell waH These weld stresses are conservatively calculated 1,7 assuming that fuel assemblies in au,;acent cells are moving out of phase with one another so that impact loads in two adjacent cells are in opposite directions; this tends to separate the two cells from each other at the weld.
Table 6.9.1 gives results for the maximum allowable load that can be transferred by these welds based on the available weld area. An upper bound of the transferred load is also given in Table 6.9.1, and it is much lower than the allowable load. This upper bound value is conservatively obtained by applying the maximum rack-to-fuel impact load from any simulation in two orthogonal directions simultaneously and multiplying the result by SIIADED TEXT CONTAINS PROPRIETARY INFORM ATION lloltec International 6-43 Report 111-982083
(I i
2 to account for the simultaneous impact of two assemblies. An equilibrium analysis at l the connection then yields an upper bound of the transferred load. It is seen from the -
result in Table 6.9.1 that the calculated load is well below the allowable.
'6.11 Level A Evaluation The Level A condition is not a governing condition for spent fuel racks since the general level of <
loading is far less than Level B loading. To illustrate this, the heaviest spent fuel rack is considered under the dead weight load. It is shown below that the maximum pedestal load is low I
and that further stress evaluations are unnecessary.
i LEVEL A MAXIMUM PEDESTAL LOAD -
Dry Weight of a 14X11 Rack = 23,679 lbf Dry Weight of 154 Fuel Assemblies = 246,400 lbf Total Dry Weight = 270,079 lbf Total Buoyant Weight (0.87 X Total Dry Weight) = 234,969 lbf Load per Pedestal = 58,742 lbf The stress allowables for the normal condition is the same as for the upset condition, which resulted in a maximum pedestal load of 144,000 lbs. Since this load (and the corresponding stress throughout the rack members) is greater than the 58,742 lb load calculated above, the seismic condition controls over normal (Gravity) condition.
6.12 Hydrodynamic Loads on Pool Walls l The maximum hydrodynamic pressure (in psi) that develop between the fuel racks and the spent fuel pool walls correspond to the case in which the racks exhibit the largest displacements. The maximum pressure is computed for both the SSE and OBE cases. The results for these worst case conditions are shown in the table below.
Case Maximum Pressure (psi)
SSE 10.5 OBE- 7.5 SilADED TEXT CONTAINS PROPRIETARY INFORMATION lloltec International 6-44 Report 111-982083
r: 1 These hydrodynamic pressures must be considered in the evaluation of the Spent Fuel Pool )
structure.
6.13' Thermal Stresses From Asymmetric Her.t Generation l
~ Inter-cell welded joints are examined under the loading conditions arising from thermal effects due to an isolated hot cell. A thermal gradient between cells will develop when an isolated storage location contains a fuel assembly emitting maximum postulated heat, while the surrounding locations are empty. The temperature difference between these cells can be obtained !
from section 5.8.3 of this report. The maximum W.aperature difference between the local water temperature and the bulk SFP temperature is 38.3' F.
l We can obtain a conservative estimate of weld stresses along the length of an isolated heated cell by using a finite element model (illustrated in figure 6.13.1) which is based on the following assumptions:
(a) The cell walls surrounding the " hot" cell are assumed to be at the exit temperature of the coolant. In actuality, the water temperature rises monotonically from the bulk temperature value at the base plate elevation to its maximum value at exit. By assuming the cell wall to be at exit temperature of water, the state of computed thermal stress will be maximized.
(b) The cell walls contiguous to the hot cell are assumed to be at the pool bulk temperature.
(c) . The connectivity between adjacent cells is through six discrete lineal welds which is explicitly modeled in the finite element model.
(d) . The bottom edges of all cell walls are assumed to be fixed.
(e) The top edges of all cell walls are assumed to be free.
The finite element solution exploits the symmetry of the problem about the two vertical planes to permit a quarter symmetric model (figure 6.13.1). It is clear from the physics of the problem and
' from the above finite element model that the locaiions of sharp temperature change, namely the longitudinal welds, are locations of maximum stress which arise from restraint of thermal expansion between adjoining cells.
SliADED TEXT CONTAINS PROPRIETARY INFORMATION
. IIoltec International 6-45 Report 111-982083 l
I The finite element solution confirms this expected result. The maximum weld shear stress, however, is limited to 8,384 psi. Thermal stresses, which are " secondary stresses" in the stress hierarchy of the ASME Code, have no prescribed limit for NF Class 3 structures. Since the maximum shear stress in weld material is less than the code allowable limit of 19,860 (Table 6.9.1), it is concluded that the " isolated cell" scenario will not lead to any primary yielding in the cell connectivity.
6.14 Overhead Storace The spent fuel racks for Byron and Braidwood are also qualified for two additional storage functions.
The Region Il racks are designed to acconunodate an Overhead Platform, which has a capacity of 3 tons (dry). The platform is movable, and it can be installed on top of any Region Il rack by inserting its four support legs into empty storage cells. The surface of the platfonn is elevated 13 inches above the top of the rack and measures 52 inches square, which covers a six by six area of cells. Multiple items can be stored oMhe platform as long as (i) the total dry weight is less than 6,000 lb and (ii) each item completely rests e 6 storage surface (i.e.,52 inch square area). The stored objects are protected from falling off of the platform by 14 inch high side walls. 4 Both the Region I and Region 11 racks, when they are completely empty, are also capable of supporting miscellaneous equipment (e.g., tools) directly on top of the storage cells. The object must weigh less than 2,000 lb (dry), and it must span a minimum of three storage cells.
6.15 Conclusion Thirty discrete freestanding dynamic simulations of maximum density spent fuel storage racks have been performed to establish the structural margins of safety. Of the thirty parametric analyses, six simulations consisted of modeling all 24 fuel racks in the pool in one comprehensive Whole Pool Multi Rack (WPMR) model. The remaining twenty-six runs were carried out with the classical single rack 3-D model. The parameters varied in the different runs consistni of the rack / pool liner interface coefficient of friction, extent of storage locations occupied by spent nuclear fuel (ranging from nearly empty to full) and the type of seismic input (SSE or OBE). Maximum (maximum in time and space) values of pedestal vertical, shear forces, SHADED TEXT CONTAINS PROPRIETARY INFORMATION Holtec International 6-46 Report HI-982083 I I
F l displacements and stress factors (normalized stresses for NF Class 3 linear type structures) have been post-processed from the array of runs and summarized in tables in this chapter. The results i show that- I (i) - All stresses are well below their corresponding "NF" limits.
(ii) There is no rack-to-rack or rack-to-wall impact anywhere in the cellular region of I the rack modules i i
(iii) The factor of safety against overturning of a rack is in excess of 60.
An evaluation of the fatigue expenditure in the most stressed location in the most heavily loaded rack module under the combined effect of one SSE and twenty OBE events shows that the :
Cumulative Damage Factor (using Miner's rule) is 0.950, which is less than the permissible value of 1.0.
An evaluation of the thermal (secondary) stress produced by the condition of maximum thermal gradient (obtained when a maximum heat emitting fuel assembly is stored in a cell surrounded by ,
empty storage locations wherein no heat is generated) was performed. The thermal stresses for which no statutory limit in the code (Section III, Subsection NF, Class 3 Structures) exists, is found to be limited to 8,384 psi, which is well below the allowable limit of 19,860 psi.
In conclusion, all evaluations of structural safety, mandated by the OT Position Paper (6.1.2] and the contemporary fuel rck structural analysis practice have been carried out. They demonstrate consistently large margins of safety in all new storage modules.
As a final note, the continued compliance of the installed rack arrays with the licensing basis is an essential part of a plant's safety considerations. Since the fuel racks are free-standing structures, the inter-body spacings in the Byron and Braidwood pools, after a site seismic event, may be different from the as-installed values. A plant's procedures would require a comprehensive survey of the inter-module and module-to-wall gaps subsequent to a seismic event. If the gaps are found to have changed, then a re-evaluation of the acceptability of the module layout configuration (using the WPMR model) will be performed to complete a no-significant-hazards evaluation pursuant to 10CFR50.59. The rack modules will be restored to
' their original installed locations (pre-seismic) if the conclusion of the 50.59 evaluation is non-
' SliADED TEXT CONTAINS PROPRIETARY INFORMATION l Iloitec International 6-47 Report HI-982083 1
i
L pf ,
4 4
> [aflirmative, or if the plant elects to skip the analytical evaluation (Q50.59) step and move directly Lto reposition the modules.- ,
l
( 6.161 References for Section'6i
\
, ;[6.1.1]i USNRC NUREG-0800, Standard Review Plan, June 1987.
,n ,
[6.1.2] - (USNRC Office of Technology) "OT Position for Review and Acceptance of-
. Spent Fuel Storage and Handling Applications", dated April 14,1978, and January 18,1979 amendment thereto.
[6.2.1]: Soler,'A.I. and Singh, K.P., " Seismic Responses of Free Standing Fuel Rack -
i : Constructions to 3-D Motions", Nuclear Engineering and Design, Vol. 80,'pp.
4V N
?315-329 (1984)..
-[6.2.2]f Soler, A.I. and Singh, K.P., "Some Results from Simultaneous Seismic '
- Simulations of All Racks in a Fuel Pool" INNM Spent Fuel Management Seminar L January,1993.
'_[6.2.3] Singh,'K.P. and Soler, A.I., " Seismic Qualification of Free Standing Nuclear Fuel Storage Racks - the Chin Shan Experience, Nuclear Engineering International, UK
. (March 1991).-
l[6.2.4]= JH.olt'ec Proprietary Report HI-961465 - WPMR Analysis User Manual for Pre & Post Proces' sors & Solver, August,1997.
.[6.2.5] 1 Holtec Proprietary Report HI-91700 - Validation Manual for DYNARACK.
1 I
-[6.4.1] . USNRC Standard Review Plan, NUREG-0800 (Section 3.7.1, Rev. 2,1989).
[6.4.2] Holtec Proprietary Report HI-89364;- Verification and User's Manual for l L Computer Cods GENEQ, January,1990.'
' [6.5.1] . Rabinowicz, E., " Friction. Coefficients of Water Lubricated Stainless Steels for a ;
' Spent Fuel Rack Facility," MIT, a report for Boston Edison Company,1976.
[65.2].. 'Singh,' K.P. and Soler, A.I., " Dynamic Coupling in a Closely Spaced Two-Body System Vibrating in Liquid Medium: The Case'of Fuel Racks," 3rd International Conference on Nuclear Po'wer Safety, Keswick, England, May 1982.
L .
-[6.5.3)) l Frith R.J., "The Effects of LiqvMs on the Dynamic Motions ofImmersed Solids,"
Journal of Engineering fbr Industry, Trans. of the ASME, February 1972, pp 167-
.172.
SHADED TEXT CONTAINS PROPRIETARY INFORMATION -
Holtec International . 6-48' Report 111-982083 t.
2 >
[6.6.1] Levy, S. and Wilkinson, J.P.D., "The Component Element Method in Dynamics with Application to Earthquake and Vehicle Engineering," McGraw Hill,1976.
[6.6.2] Paul, B., " Fluid Coupling in Fuel Racks: Correlation of Theory and Experiment", l (Proprietary), NUSCO/Holtec Report HI-88243. l
[6.7.1] . ASME Boiler & Pressure Vessel Code,Section III, Subsection NF,1989 Edition.
i
~[6.7.2]' ' ASME Boiler & Pressure Vessel Code,Section III, Appendices,1985 Edition. j
. [6.7.3] . USNRC Standard Review Plan, NUREG-0800 (Section 3.8.4, Rev. 2,1989). l
[6.10.1] - Chun, R., Witte, M. and Schwartz, M., " Dynamic Impact Effects on Spent Fuel Assemblies," UCID-21246, Lawrence Livermore National Laboratory, October '
1987.
I i
t j-l l
i i
SHADED TEXT CONTAINS PROPRIETARY INFORMATION 4 llottec International 6-49 Report 111-982083 s
j l
i Holtec Center,555 Lincoln Drive West, Marlton, NJ 08053 Telephone (609) 797-0900 Fax (609) 797-0909 )
lNTERNATIONAL '
l
'~
LICENSING REPORT l for SPENT FUEL RACK INSTALLATION l at BYRON 4ND BRAIDWOOD NUCL/ EAR STATIONS Holtec Report HI-982083 (Non-Proprietary Version)
Report Category: A Prepared for Commonwealth Edison Co.
Purchase Order No. 367585 Holtec Project 80944 COMPANY PRWATE This document version has all proprietary information removed and has replaced those sections, figures, and tables with highlighting and/or notes to designate the removal of such information. This document is to be used only in connection with the performance of work by Holtec International or its designated subcontractors. Reproduction, publication or presentation, in whole or in part, for any other purpose by any party other than the Client is expressly forbidden.
e 7 1 o n t
a 9 w o D - 9 t t
o is i
v
_ & 4 t
- p er r
e ,~ - u. r 2 f o
C w -
7-i e
v p
f 3
P Q
d e
e N
o
. R e
J o
w y
g g
'g i
l P
f r
e ht t
n e
o s 5 M N O
I S e ; N
/
7 f
d o
e a
n o
2 3
L n
i 1 - C
- g ta I t V a 4 1 -
n : -
E D A p A l is i m
g s m r R & l f f
N, /r i g
o e
r F
o N
t t C o r
N 4 r
g g ,
- l d
e S d ht is e T
N M C'
A u
6
- N- 1 -
ht e H.
T R
E -
b e- dM fe r
o t
jk c . 6QEP e
M b oc U
C O
W e ~
7 9
li f
f i
e y t r o pl e b h we f i ove R
E i
i T
D -
t a -
J l
a nr F
O R E D ' u o
[- q "n.
P e T -
y it O Y & - -
n ut h X l
T r
^- 9 9 a p lon E i
e ,
I 1 vi T T T 4 w - I9 m o
ed gr e E U R ie m # - c 1 A O v 3- M '
b 3- e ni nqu du re 1
T .
r e 7 ht E E P ,
1 L
P E
R N o
. R
, n 4 b
i d
n e
d e te ar c
R O
F I
T L
N d v a
s f
f e es a a m E
B D
U 4 M 4 O
I S
I e n 1 1
% d n
a, lyn iad l
e E
C A
V t a /p - )
s rt en L R E p (
r t i a r P 9 R D A A Y e mp O
F 0 & /s, f
/s ie w
ins d G'
8 r o
Jp J e p " '
v e
s a s O
L G
t r ne ht T e r S O
b5e e u ht .y t
nt u N L :
A -
d n oa c ng O ba I
N R na ap si S O E B
b r o m iS f
t I V
I h oc ot E T M t yan R A f g u e ah ne F U f c aT O C N q et f
[p ht ro i t e r c Y I
F T t
a - yl f e go e
j R
D bt e I
T C v a r nP A R
E C
E J
O R
r e
w n.
1 uo on d r le o i e hh t
a cm t
M M
U S
P 3 e f-:
l iene f
w j e o of r D i v b r A N R e s Pt u Y
shi t
ep A o u c ht ni B N mih yl S W a t s u b ar I T
E 5 N t li t k c d o e N I
O c u rh t E V
E R
8 I S
I V e
% P ed h or Cp ei dr w or ege U
M M E t
a l G n k or b a C 8 O 9 R D
& I A p p f
A [
p l
- V u
- it aw c
o l liwh !
n a D S
- r o J, gp g J t
g d H V f i
irfo r
r t
nt c l l
l h M e d e e mjeo i
T d t u Ve b m ur F nd g e O A e u n
cP o
- i. n e e n d he L Db o ist O R d y it h y R c t b T E e a e f N
B t ta m s o de nt 1 2 3 4 5 6 7 8 r na O L ics o M et mn r
e r
e e r r e
r e
r e
r e
r e & A or s e r imo C U ui o s
t p
t p
t p
t p
t p
t p
t p
t p ) V s we a te p
ie vb N
N ct r oo h a
h a
h a a h h a
h a
h a
h a E OE ei v h a el rl O
I S
T DP C C C C C C C C E RT nr C e Awi R N PA VI C PD E O :
R P I AD s e E E
R .' 1 2
3
- 4. 6 7' 8 AN QA t t
t N t
o l i
T
e o n t
a t w o i
D t o is v
' p
& u, e
r r r e Y 2 f o
S i w
e 9
3 P
d e
v Q e
. R e y /
c g l f
I I
n e
o /
r e
ht N 7 P t o
- f. s 5 N f o a 2 O M n N d 3 I
e io L S e n t a C I t a g V D p in E
R A p a M is s m r
m r
o
& p S' i g t e F r 4 g o e t
o n b d l
t A s S
T c ht u
i ht h T
e T
A N
E M
m
'4 C
5 fe b
e r
o t
jk c .
e oc R
O P
E r o R U E e pl E
7 e b l
C i
7 f y h w I I
O e f
i t
f ie T ove F D
R E D
t a UC P9 I
l a
u q
nr o e O P T O
H Y
T r
" -3 y n
a p
itut h lon vi X
E T
T T 4 e 0 - m ed gr e E U R w e *b~ o c
I ini r u I
T A O P
i v
e " ht e
u e q E
E E . R n d r d er R
L R o i d
e a O P N t F
Ad e c s E I v f ee B T N s a f a a m L O D U
M 4 I
S I
V e t
b-
)
s n
a lyn iad l
r t e
E C
A a ten a n L
R 4 E D p
(
r mp P O 9 R p p N e w s d G"
F 0 p p 3 i e
is a n
O G 8 r
o d 9 v e
r t
ne s L h S O t e te u r u ht . nt N L :
A d n y o an c O na g I N R ap t si i S S
I E
O r m f V I B ht o oc o . E ya m T M 7 9 u e ne R
F A U y 9-ah et faT O C N e f ht r o it r c Y I t 1 y f ee F
I T D a i b le gm a
j R
A T C f 2 t v nP E & /y uo on a M R J r
i 2
d r Mhe M E O e a le o t
U cm t
C D P R 3 i w
e v
g t p
l i
f en w
e joe of r
r S
A e b Pt u N R o t i s
e p Y A sh c
_ u ht n B G
o M mih i S
W 7 yl N t b ar I E 8 is w d o T N
I V 2#
E IO S
g V l
k c c u ed h or t
eh r
ei dr w t
N E
M R 8 I
e . Cp ore U
_ V t n k e g C
_ 9 E
R D a
a p =A M d, ioro ta w b
l l
a n
a O
D
/ & f ic f r iwM s y nf o t l r .
nt c
- o g j e d r e
l i
ht t )
Ve b e
mje T A
u Q nd ge i e u
n m um cP o
F O
, s n d e L f e n h O
R Dbe o it is t h y R E
d y c t b T 0 te a e s
f o de N B t nt 9 1 I
1 2
1 am r na O M et r r r r L ic rs o imo C mn e e e e & A me r ui e s
U t p tp t
p t
p ) V sa we e t
p ie vb N
N ct r oo h a
h a
h a
h a E OE ei v a h
el rl IO T DP C C C C E RT h T re C Aw i S R PA VI E
O PD :
R P I AD s te E E
~
0 1
2 AN t o 1 R 9 1 1 1 I QA t t N 1
T
F l
l 1
SUMMARY
OF REVISIONS Revision 0: Inititial issue.
. Revision 1: Figures 11.1 through 11.13, which showed a preliminary rack change-out sequence, were removed from Chapter 11 (Installation) and the List of Figures.
Minor corrections were also made to the Table of Contents.
Revision 2: Text was added to Sections 11.5 and 11.6, which describe the fuel shuffling and i the new rack installation. Minor changes were also made to the Table of Contents.
Revision 3: Minor changes were made to Paragraph 11.1 h. (ALARA Procedure), Subsection 11.7.2, and Subsection 11.7.3.
Revision 4: Editorial changes were made on pages 1-1,5-2,5-17,8-7, and 8-11.
Revision 5: Chapter 7 was reorganized to include a section on construction accidents (i.e.,
rack drop accident). Figure 6.5.1 and Table 8.2 were also corrected. Table 6.5.3 was added to the report. These changes also affected the Table of Contents and the List of Figures.
1
! lioltec International i Report HI-982083 Q_
l
\
I TABLE OF CONTENTS 1
1 1.0 I NTR OD UCTI ON . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 - 1 I 1.1 References for Section 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 -4 2.0 HIGH DENSITY SPENT FUEL RACKS . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . 2-1 )
2.1 General Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2- 1 2.2 Summary of Principal Design Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-2 , ;
2.3 Applicable Codes and Standards . . . . . . . . . . . .. .................... 2-4 '
2.4 Quality Assurance Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2- 1 1 2.5 Mechanical Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2- 12 l 2.6 Rack Fabrication . . . . . . . . . . . . . . ..............................2-12 l 1
2.6.1 Fabrication Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-13 !
2.6.2 Byron and Braidwood Rack Modules . . . . . . . . . . .............. 2-13 I 3.0 MATERIAL AND HEAVY LOAD CONSIDERATIONS . . . . . . . . . . . . . . . . . . . . . 3-1 ;
3.1 Introd u cti on . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 - 1 l 3.2 S tructural Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1 1 3.3 Neutron Absorbing Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1 f
3.4 Compatibility with Coolant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-3 3.5 J
Heavy Load Considerations for the Proposed Reracking Operation . . . . . . . . 3-3 3.6 References for Section 3 ................ . ..................... 3-7 4.0 CRITICALITY S AFETY EVALUATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-1 4.1 Design Bases . . . . . . . . . . . . . . . . . . ............... .... ........ 4-1 4.2 Summary of Criticality Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-4 4.2.1 Normal Operating Conditions . . . . . . . . . . . . . . . . , , . . . . . . . . . . . . 4-4 4.2.1.1 Regi o n I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-4 4.2.1.2 Region II . . . . . . . ................. .......... ... 4-4 4.2.2 Abnormal and Accident Conditions . . . . . . . . . . . ... ... ....... 4-5 4.3 Reference Fuel Storage Cells . . . . . . . . . ......................... .. 4-7 4.3.1 Re ference Fuel Assembly . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-7 4.3.2 Region 1 Fuel Storage Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-7 4.3.3 Region H Fuel Storage Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-8 l 4.4 Analytical Methodology . . . . . . . . . . . . . . . . . . ..... ............. 4-9 I 4.4.1 Reference Design Calculations . . . . . . . . . . ..... ............. 4-9 l 4.4.2 Fuel Bumup Calculations and Uncertainties . . . . . . . . . . . . . . . . 4- 10 4.4.3 Effect of Axial Burnup Distribution . . . . . . . . . . . . . . . . . . . . . . 4- 1 1 4.4.4 Long-Term Changes in Reactivity . . . . . . . . . . . . . . . . . . . . . . . . . 4-12 Holtcc Intemational ii Report HI-982083 l
r TABLE OF CONTENTS 4.5 Region I Criticality Analyses and Tolerances . . . . . . . . . . . . . . ..... 4-13 4.5.1 Nominal Design Case . . . . . . . . . ........................4-13 4.5.2 Uncertainties Due to Burnup . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-13 4.5.3 Uncertainties Due To Tolerances . . ........................4-13 J 4.5.4 Eccentric Fuel Positioning . . . . . . . . ......................4-13 1 4.5.5 Water-Gap Spacing Between Racks . . ..................4-14, 4.6 Region II Criticality Analyses and Tolerances . . . . . . . . . . . . . . . . . . . . . 4- 15
)
l 4.6.1 Nominal Design Case . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-15 4.6.2 Uncertainties Due to Bumup . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-15 4.6.3 Uncertainties Due to Tolerances . . . . . . . . . . . . . . . . . . . . . . . . . 4-15 4.6.4 Eccentric Fuel Positioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-16 4.6.5 Water-Gap Spacing Between Racks . . . . . . . . . . ........ . . . 4-16 4.7 Abnormal and Accident Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-17 4.7.1 Temperature and Water Density Effects . . . . . . . . . . . . . . . . . . . 4-17 4.7.2 Lateral Rack Movement . . . . . . . . . . . ......................4-17 4.7.3 Abnormal Location of a Fuel Assembly ...................4-18 4.7.4 Dropped Fuel Assembly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-19 4.8 References for Section 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-20 Appendix 4 A: Benchmark Calculations 5.0 THERMAL-HYDRAULIC CONSIDERATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . 5-1 5.1 Introduction . ........... .... ...... ........... ............. 5-1 5.2 Spent Fuel Pool and Cooling System Descriptions . . ................. 5-2 5.3 Decay Heat Load Calculations . . . . . . . . . . ................ . ...... 5-4 5.4 Discharge Scenarios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .......... 5-5 5.5 Bulk Pool Temperatures ...... ........... .. ................. 5-6 5.6 Local Pool Water Temperature . . . . . . . . . . ........ ......... . 5-10 5.6.1 B as i s . . . . . . . . . . . . . . . . . . . . . . . ..... .. . . . . . . . . . . . 5- 10 5.6.2 Local Temperature Evaluation Model . . . . . . . . . . . . . . . . . . . . . . . 5-11 5.7 Cladding Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-13 5.8 R esu lts . . . . . . . . . . . . . . . . . . . . . . . . . ..........................5-15 5.8.1 Bulk Pool Temperature . . . . . ......... ...... . ....... . 5-15 5.8.2 Time-to-Boil . . . . . . . . . . . ...... .....................5-16 5.8.3 Local Water and Cladding Temperature ..................5-16 5.9 References for Section 5 . . . . . . . . . . . . . . . ... .... .... . ... 5-18 L
I Holtec International iii Report HI-982083
1 i
L TABLE OF CONTENTS 6.0 STRUCTURAL SEISMIC CONSIDERATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . 6-1 6.1 Introd u ctio n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6- 1 6.2 Overview of Rack Structural Analysis Methodology . . . . . . . . . . . . . . . . . . 6-1 l 6.2.1 Background of Analysis Methodology . . . . . . . . . . . . . . . . . . . 6-2 1 6.3 Description of Racks and Fuel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-5
) 6.4 Synthetic Time-Histories . . . . . . . . . . . . . ......... ... ........... 6-5 ,
6.5 3-D Nonlinear Rack Model for Dynamic Analysis . . . . . . . . . . . . . . . . . . . 6-7 6.5.1 General Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-7 6.5.2 Element Details . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6- 11 l 6.5.3 Fluid Coupling Effect ..... ....... . ..... .............. 6-11 6.5.4 Stiffness Element Details . . . . . .. ... .......... . . . . . . . 6 13 l 6.6 Whole Pool Multi-Rack Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-15 6.6.1 General Remarks . . ..................................6-15 6.6.2 Multi-Body Fluid Coupling Phenomena . . . . . . . . . . . . . . . . . . . 6-15 I 6.6.3 Coefficients of Friction . ................................6-16 6.6.4 Governing Equations of Motion . . . . . . . . . . . . . . . . . . . . . . .... 6-17 I 6.7 Structural Evaluation of Spent Fuel Rack Design , . . . . . . . . . . . . . . . . . . . . 6-18 6.7.1 Kinematic and Stress Acceptance Criteria . . . . . . . . . . . . . . . . . . . 6-18 6.7.2 Stress Limit Evaluations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-19 !
6.7.3 Dimensionless Stress Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0-2 2 l 6.7.4 Loads and Loading Combinations for Spent Fuel Racks . . . . . . . . . 6-22 6.8 Parametric Simulations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-24 6.9 Time History Simulation Results . . . . . . . . . . . . . . . . . . .. . . . 6-26 6.9.1 Rack Displacements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-26 6.9.2 Pedestal Vertical Load . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-28 6.9.3 Pedestal Friction Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-30 6.9.4 Rack Impact Loads ...................................6-31 6.9.4.1 Fuel to Cell Wall Impact Loads . . . . . . . . . . . . . . . . . . . . . . 6-32 6.9.5 Rack Vertical Displacement . . . . . . ................ . . . . . . . 6-3 5 6.10 - Rack Structural Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ....... 6-37 6.10.1 Rack Stress Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-3 7 6.10.2 Pedestal Thread Shear Stress . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-3 8 6.10.3 Local Stresses Due to Impacts . . . . ... ..... ... ....... . 6-40 6.10.4 Assessment of Rack Fatigue Margin . . . ..... ... ... ....... 6-41 l 6.10.5 Weld Stresses . . . . . . . . . . . . . . . . . . . . . . . . . ... ...... .... 6-42 6.11 Level A Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-44 1- 6.12 Hydrodynamic Loads on Pool Walls ......... .. .... .......... . 6-44 6.13 Thermal Stresses From Asymmetric Heat Generation . . . . . . . . . . . . . . . 6-45 l 6.14 Overhead Storage . .. . . . . . . . . . . .............................6-46 l 6.15 Concl usion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . 6-4 6 1
i Holtec International iv Report HI-982083 1
l
v 1 L
l f
m -
4 l
c ,
- TABLE OF CONTENTS I
l f,
6.16 References for Section 6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-48 L7.0 - L FUEL HANDLING AND. CONSTRUCTION ACCIDENTS . . . . . . . . . . . '. . . . . . . . 7-1 .
-7.14 Introduction' -
7-1:
' ' ~
o 7.2 Fuel Handling Accidents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-1 -
.7.2. l f Description . . . . . . . . . . . . . . . . . . . . . . . . .. . . . .. . . . . . . . . . . . . . . . . . 7- 1 ,
7.2.2 Incident Fuel Assembly Velocity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-3 i 7.2.3 K Mathematical Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-4 7.2.4 - Resu l ts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-5 7.3 ' ' Construction Accidents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-6 7.4 ' Conc l u s ion . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-6
, 7.5 ' References for Section 7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-7 .
8.0 . FUEL POOL STRUCTURAL INTEGRITY CONSIDERATIONS . . . . . . . . . . . . . . 8-1 F
'8.1 Introduction . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8- 1 18.2' _ Description of the Spent Fuel Pool Structure . . . . . . . . . . . . . ., . . . . . . . . . . . 8-2 8.3 Material Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-3 8.4 : " Load Combinations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-4 l' 8.5 ' Analysis Methodology'. . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-6 ,
8.6 : Pool Liner Integrity Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-9 H
- 8.7 ' Bearing Pad Analysis . . . . . . . . . . . . .. . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-1 1 HI '8.8 Results and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-12 8.9 References for Section 8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 8-13 L 9.0! RADIOLOGICAL EVALUATION ' . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-1
, 9.1 ' Fuel Handling Accident . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-1 9.2 . Solid Radwast e . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-2 9.3 Gaseous and Liquid Releases .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-2 9.4 - . Personnel Exposures . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-2 L 1 9.5 : - Anticipated Exposures During Reracking ............................ 9-3 L % - 10.0 : BORAL SURVEILLANCE PROGRAM . . . . . . . . . .. . . . .. . . . . . . . . .. . . . .. . . . . . . 10-1 i
- . 10.1 - P urpo s e '. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 0- 1 )
f 10.2 Coupon Surveillance Progrcm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-2 l U L 10.2.1 Coupon Description s. . . . . . . . . _ . . . .. . . . . . . . . . . . . . . . . . . . . . . . . 10-2
- 10.2.2 Surveillance Coupon Testing Schedule '. . . . . . . . . . . . . . . . . . . . . . . 10-3 10.2.3 Measurement. Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . 10-4 10.2.4 Surveillance Coupon Acceptance Criteria . . . . . . . . . . . . . . . . . . . . . 10-4 L10.3. In-Service Inspection (Blackness Tests) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-5 10.4 ~ References for Section 10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-7 l
Holtec International v - Report HI-982083 I l
r
4 LTABLE OF CONTENTS i
4 11.0 ' IN STALLATI ON . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 - 1 11.1: . Introd uction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 - 1
! , 1 1.2 . Rack Arrangement .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 -4 11.3 Pool Survey and Inspection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-5 11.4' Pool Cooling and Purification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-5 -
i . l l .4.1 Pool Cooling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 -5 ,
- 1 1.4.2 P uri fication . . . . . . . . . . . . . . . . . . . . . . . . . . . -. . . . . . . . . . . . . . . . . 1 1-5 11.5 Fuel Shuffling . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . 1 1 -5 11.6 Installation of New Racks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-6 11.7 Safety, Radiation Protection, and ALARA Methods . . . . . . . . . . . . . . . . . . . . Il-7 1 1.7.1 ' S a fety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 -7 11.7.2 Radiation Protection' . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 .
' 1 1.7.3 ALARA' . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 - 8 11.8: Radwaste Material Content ~ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 'l 1-9 12.d , : ENVIRONMENTAL COST / BENEFIT ASSESSMENT . . . . . . . . . . . . . . . . . . . . . 12-1
.12.1 Introduction . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . 12- 1 12.2 . . Imperative for Rack Replacement - . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-1 12.3' Appraisal of Alternative Options .. . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 12-1
- 12.43 Cost Estimat e . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . 12-6
.12.5 . Resource Commitment ; . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-6 12.6: Environmental Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-7
.12.7 l References for Section 12 . . .. . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-7 i
r l l
ru 4
2 Holtec International vi Repon HI-982083 1
I 7
j lc u )
LIST OF FIGURES l
l l-1. New Rack Layout for Byron and Braidwood Nuclear Stations 2-1 Pictorial View of Typical Region I Rack 2-2 - Pictorial View of Typical Region II Rack 2-3 Seam Welding Precision Formed Channels 2-4 A Cross Sectional View of an Array of Region I Storage Cells 2-5 A Cross Sectional View of an Array of Region II Storage Cells 2-6 Elevation View of Region I Cells 2-7 Elevation View of Region II Cells 2-8 Sheathing Shown Installed on the Box 1
2-9 Adjustable Support Leg 4.1.1 Minimum Required Fuel Assembly Burnup As A Function of NominalInitial Enrichment to Permit Storage in Region II (Fuel assemblies with enrichments less than 2.0 wt%2 "U 2
will conservatively be required to meet the bumup requirements of 2.0 wt% "U assemblies) 4.3.1 A Cross-Sectional View of the Calculational Model Used for the Region I Rack Analysis ,
(NOT TO SCALE) !
l 4.3.2 A Cross-Sectional View of the Calculational Model Used for the Region II Rack Analysis (NOT TO SCALE) 4A.1 Calculated k-eff Values for Various Values of the Spectral Index 4A.2 Calculated k-eff Values for Various Values of the Spectral Index !
l 4A.3 MCNP Calculated k-eff Values at Various U-235 Enrichments f
. 4 A.4 KENO Calculate k-eff Values at Various U-235 Enrichments 1
4A.5 Comparison of MCNP and KEN 05A Calculations for Various Fuel Enrichments I Holtec International vii Report HI-982083 l
I
( __
l 1
]
LIST OF FIGURES 4A.6 Comparison of MCNP and KENO 5a Calculations for Various Boron-10 Areal Densities 5.4.1 Byron and Braidwood Spent Fuel Pools Discharge Scenario Case (i) 5.4.2 Byron and Braidwood Spent Fuel Pools Discharge Scenario Case (ii) 5.4.3 Byron and Braidwood Spent Fuel Pools Discharge Scenario Case (iii) 5.5.1 Spent Fuel Pool Cooling Model 5,8.1 Bulk Pool Transient Temperature Plot for Case (i) Normal Discharge Scenario 5.8.2 Bulk Pool Transient Temperature Plot for Case (ii) Full Core Discharge Scenario 5.8.3 Bulk Pool Transient Temperature Plot for Case (iii) Back-to-Back Discharge Scenario 5.8.4 Fuel Pool Decay Heat Load for Case (i) Normal Discharge Scenario 5.8.5 Fuel Pool Decay Heat Load for Case (ii) Full Core Discharge Scenario 5.8.6 Fuel Pool Decay Heat Load for Case (iii) Back-to-Back Discharge Scenario 5.8.7 Post Loss of Forced Cooling Transient Pool Depth Plot 5.8.8 Byron & Braidwood Pool Local Temperature Plot 5.8.9 Byron & Braidwood Pool Velocity Vectors Plot 6.4.1 Acceleration Time-History SSE x-direction (4% damping) 6.4.2 Acceleration Time-History SSE y-direction (4% damping) 6.4.3 Acceleration Time-History SSE z-direction (4% damping) 6.4.4 Acceleration Time-History OBE x-direction (2% damping) 6.4.5 Acceleration Time-History OBE y-direction (2% damping) 6.4.6 Acceleration Time-History OBE z-direction (2% damping)
Holtec International viii Report HI-982083
r i
I LIST OF FIGURES 6.5.1 Schematic of the Dynamic Model of a Single Rack Module Used in DYNARACK
- 6.5.2 ' Fuel-to-Ruck Gap / Impact Elements at Level of Rattling Mass 6.5.3 Two Dimensional View of the Spring-Mass Simulation )
l 6.5.4 Rack Degrees-of-Freedom for X-Z Plane Bending with Shear and Bending Spring 6.5.5 Rack Periphery Gap / Impact Elements 6.8.1 Gap Spring Identification Scheme (At Rack Bottom) 6.8.2 Gap Spring Identification Scheme (At Rack Top) 6.10.1 Rack Fatigue Analysis 6.13.1 Quarter Symmetric.Model for " Hot Cell" Thermoelastic Analysis 7.2.1 Shallow Drop on a Peripheral Cell 7.2.2 Deep Drop on a Support Leg Location 7.2.3 Deep Drop on a Center Cell Location 7.2.4 Plan View of Shallow Drop Scenario 7.2.5 Maximum Cell Deformation for Shallow Drop 7.2.6 Plan View of Deep Drop Scenarios 7.2.7 Maximum Von Mises Stress of the Liner for Deep Drop Scenario 2 l 7.2.8 Maximum Compressive Stress of the Concrete Slab for Deep Drop Scenario SC2 7.2.9 Maximum Baseplate Deformation for Deep Drop Scenario SCI 8.2.1 Isometric (Pictorial) View of Byron /Braidwood Nuclear l'lant Spent Fuel Storage Pool d.2.2 Plan View of Byron /Braidwood Nuclear Station Spent Fuel Storage Pool Holtec Intemational ix Report HI-982083
I
- LIST OF FIGURES l
8.2.3 Section View (A-A) of Byron /Braidwood Nuclear Station Spent Fuel Storage Pool 8.2.4 Section View (B-B) of Byron /Braidwood Nuclear Station Spent Fuel Storage Pool l
Holtec Intemational x Report HI-982083
e 2
(mi + M ) X i+ Mi2 X 2= applied forces on mass mi + O(Xi )
ii M 2i X +i (m2 + M )22X =2 applied forces on mass m2 + O(X2 ')
- X iand X denote 2 abholute accelerations of masses mi and m2, respectively, and the notation 2
O(X ) denotes nonlinear terms. The hydrodynamic coupling effect is shown to be composed of an added nonlinear term which varies with geometry and a component which varies with the . , _
square of velocity. This is easily shown by considering a typical example where fluid coupling plays a significant rolef Consider two long beams oflength "1" width "h" and a distance "s" apart:
/ // / /uf /
/ .
S
/
/ /
/ / h
/ /
'B / / A
///
'/ V/
It is assumed that s h I which is always the case for spent fuel racks. It is shown by Levy
. [6.6.1] that the force exerted by the fluid on " beam" A is given by E
Ey,,, = . plh'-J-' (A moves to th right) 2s < .si l
l 3
Ey,,, = pih' -' J + -- (A moves to the left)
The above solution is' valid at each instant in time so that as the beams (racks) approach each other larger forces result which tend to reduce rack motion and preclude rack-to-rack impact. For ]
conservative results, the rack analyses are based only on the nominal gap that exists prior to any j seismic event. Therefore the forces exerted have the form.
SilADED TEXT CONTAINS PROPRIETARY INFORMATION lloitec International 6-12 Report 111982083 1
)
p i
The analysis for Braidwood serves for Byron as well. In the table, the following coc.dinate system applies:
x.= Ilorizontal axis along Braidwood plant North (Byron, South) y= Ilonzontal axis along Braidwood plant West z= Vertical axis upward from the rack base
'~
If the simulation m6 del is restricted to two dimensions (one horizontal motion plus one vertical motion, for example), for the purposes of model clarification only, then Figure 6.5.3 describes the configuration. This simpler model is used to elaborate on the various stiffness modeling elements.
Type 3 gap elements modeling impacts between fuel assemblies and racks have local stiffness Ki in Figure 6.5.3. In Table 6.5.2, for example, type 3 gap elements 5 through 8 act on the rattling fuel mass at the rack top. Support pedestal spring rates Ks are modeled by type 3 gap elements 1 through 4, as listed in Table 6.5.2. Local compliance of the concrete floor is included in Ks. The type 2 friction elements listed in Table 6.5.2 are shown in Figure 6.5.3 as Kr. The spring elements depicted in Figure 6.5.4 represent type 1 elements.
Friction at support / liner interface is modeled by the piecewise linear friction springs with suitably large stiffness Kr up to the limiting lateral load :N, where N is the current compression load at the interface between support and liner. At every time-step during transient analysis, the current '
value of N (either zero if the pedestal has lifted off the liner, or a compressive finite value) is computed.
The gap element Ks, modeling the effective compression stiffness of the structure in the vicinity of the support, includes stiffness of the pedestal, local stiffness of the underlying pool slab, and local stiffness of the rack cellular stmeture above the pedestal.
The previous discussion is limited to a 2-D model solely for simplicity. Actual analyses incorporate 3-D motions and include all stiffness elements listed in Table 6.5.2. Table 6.5.3 provides a list of typical stiffness values, which are used to model the spent fuel racks for Byron and Braidwood.
SilADED TEXT CONTAINS PROPRIETARY INFORMATION lloltec International 6-14 Report 111-982083
e 1
(
- a. Baseolate-to-Cell Welds The highest predicted weld stress for SSE is calculated from the highest R6 value (provided in '6.10.1 above). The ratio of 2.15 is developed from the differences in material thickness and length versus weld throat dimension and length:
RATIO = ( . in
- in)'/ (E in
- 0.7071 * [in) = 2.15 R6 (obe) * [(0.6) Fy]
- RATIO = 0.494 * [0.6
- 21300 psi]
- 2.15 = 13,574 psi l R6 (sse) * [(1.2) Fy]
- RATIO = 0.443 * [1.2
- 21300 psi)
- 2.15 = 24,345 psi i The above calculated values are less than the OBE allowable weld stress value of 19,860 j psi and weld stress allowable value of 35,748 psi. Therefore, weld stresses between the I baseplate and cell wall base are acceptable. !
- b. Baseolate-to-Pedestal Welds The maximum weld stress between the baseplate and the support pedestal,15,380 psi ;
under an SSE event and 13,600 psi under an OBE event, is verified to be less than the )
allowable values of 35,748 psi and 19,860 psi, respectively. l
- c. Cell-to-Cell Welds !
l Cell-to-cell connections are formed by a series of connecting welds along the cell height.
Stresses in storage cell to cell welds develop due to fuel assembly impacts with the cell wall. These weld stresses are conservatively calculated by assuming that fuel assemblies in adjacent cells are moving out of phase with one another so that impact loads in two ;
adjacent cells are in opposite directions; this tends to separate the two cells from each other at the weld.
l Table 6.9.1 gives results for the maximum allowable load that can be transferred by these l welds based on the available weld area. .An upper bound of the transferred load is also j
I given in Table 6.9.1, and it is much lower than the allowable load. This upper bound value is conservatively obtained by applying the maximum rack-to-fuel impact load from any simulation in two orthogonal directions simultaneously and multiplying the result by I' SilADED TEXT CONTAINS PROPRIETARY INFORMATION llottec International 6-43 Report HI-982083
I Table 6.5.3 TYPICAL STlFFNESS VALUES FOR BYRON AND BRAIDWOOD SPENT FUEL RACKSt )
i item Stiffness Value i I
Pedestal Compression Spring 1.010 x 108 lbf/in {
8 Pedestal / Liner Friction Spring 1.010 x 10 lbf/in Rack Bending Spring in the X-Z Plane 4.672 x 10 lbf-in/ rad Rack Bending Spring in the Y-Z Plane 2.458 x 10' lbf-in/ rad 1 Rack Shear Spring in X Direction 3.315 x 108 lbf/in Rack Shear Spring in Y Direction 2.775 x 108 lbf/in 7
Rack Extension Spring 3.298 x 10 lbf/in 8
Rack Torsional Spring 3.596 x 10 lbf-in/ rad Rack-to-Rack Impact Spring at Baseplate 1.380 x 108 lbf/in Rack-to-Rack Irnpact Spring at Top Corner 2.760 x 108 lbf/in i
The values listed correspond to Rack J. The spring constants for other racks vary slightly i depending on the shape of the rack and the number of storage cells.
l I
l 1
SilADED TEXT CONTAINS PROPRIETARY INFORMATION i i
llottec International 6-57 Report HI-982083
Table 6.9.1 COMPARISON OF BOUNDING CALCULATED LOADS / STRESSES VS CODE ALLOWABLES AT IMPACT LOCATIONS AND WELDS
' Item / Location Calculated Allowable Fuel assembly / cell wall impact,lbf 1,148 3,698*
Cell-to - baseplate weld stress, psi 13,574 (OBE) 19,860 (OBE) 24,345 (SSE) 35,748 (SSE)
Pedestal - to - baseplate weld stress, 15,380 (SSE) 35,748 (SSE) psi 13,600 (OBE) 19,860 (OBE)
Cell-to - cell weld load, Ibf 15,680* * (SSE) 35,748 (SSE) -l Maximum rack rotation, degrees 0.478 29.21 I l
i l
Based on the limit load for a cell wall. The allowable load on the fuel assembly itself may be less than this value but is greater than 1,148 lbs.
- Based on the fuel assembly to cell wall impact load simultaneously applied in two orthogonal directions.
' SIIADED TEXT CONTAINS PROPRIETARY INFORMATION lloltec International 6-58 Report HI-982083
(I 1
f uq12 RA TTLING
, ~ make up water can control the postulated damage. 4 . 7.5 ! References for Section 7 [7.1] '"OT Position for Review and. Acceptance of Spent Fuel Storage and Handling ( Applications," dated April 14,1978. '[7.2]- " Analysis of the Mechanical Accidents for Byron /Braidwood Nuclear Station," ' Holtec Report No. HI-982086, Rev.1. l l l l> l .5. l lt . .
- SilADED TEXT CONTAINS PROPRIETARY INFORMATION lloitec International , 7- Report HI-982083
l l liiPACTOR i i I 1 ) t - - . j s FIGURE 7.2.4 PLAN VIEW OF SHA'LLOW DROP SCENARIO ! l l i 1 111-982083' 1" "C 2 ~ 0 0 _ + E 0 0 - 0 0 P 1 O = R R O D T C W A F O E L L A L C A S H S R O . F N O I T A M R O 0 0 F 5 3 E D E L D L O N E T A C 0 0 M _ E + U 4 1 7 M I 0 5 X 1 8 A. 0 0-L P M E S I 5 4 D 6 0 . 3 _ 0 X 7 _ 0 A 0 3 M E 1 R E = U M G I Zh I T F 8 3 P E : T S 3 8 0 2 8 9 I H BASEPLATE O O O O O' ' L 0 0 0 0 0 ~~ O O O O O. OQOOpt ( a ) SCENARIO SC1 1 BASEPLATE i tilPACTOR 2 MO O O O 0 0 0 0 0 O O O O O. OQOQO t l ( b ) SCENARIO SC2 FIGURE 7.2.6 PLAN VIEW OF DEEP DROP SCENARIOS 111-982083 ;
- O 4 4 4 4 4 4 3 3 3 3 1 1
~_ = 0 0 0 0 0 0 0 0 0 0 0 0 m0 EE E E + +0 +0 0 0 + 0 + 0 0+ 0+ 0+ 0+0 + + E+ E E E E E E E E 1~ 5 6 65 61 7 68 9 1 06 62 40 3 21 3 3 2 3 8 8 ~ 5 5 4 2 1 9 8 7 2 7 2 0 0 1 1 9 7 5 2 0 6 5 3 2 4 4 2 2 1 1 1 1 1 8 6 4 2 7 7 R E N I L E H T F O S2 S EO I RR S A T N S E E S C S O I MP N R O O V D P ME UED 2 MR I O A I R X AF O N E C M S 7 2 P" 7 O 3 R0 D0 E P E R E 9 5 U E 3 D7 G ' 9 I D5 F O 25 X O S W=E ID E AI MS zsi R TMI N B /6O N1 '- OP V_ RE X YT A BSM 3 8 0 2 8 9 i H
- O 2
0 02 020 30 3 0 03 P30 J0 3 0 03 03 4 4 0 0 0 0 0 0 0 0 0 0 0 0 0 + + + + + + + + + + + + + EE E E E E E E E E E 1 E E 8 8 7 8 7 5 4 2 9 8 9 9 . 8 8 9 7 1 5 9 3 71 0 4 0 0 9 9 3 1 4 6 8 3 6 8 1 1 2 2 9 9 9 9 9 0 0 0 0 0 0 1 1 8- 1 2- 3 -4 6- 7 -8 9- - - 1 1 B A L S _ E T E R C N O C E H T2 FC S O _ S O I SR E _ RA T N S E E C S V P I SO SR E D R PP ME E 2 OD C O R I R MO A N UF E C M I S X P" O 3 A R0 D0 - M 8 PE 2 E8 7 E 9 7 D9 ' 9 E' D9 X R O 23 O U W= DE - G I AI M F I R B T )D I zb N1 5M( / Z OPZ RE YTI G 3 BSS 8 0 2 8 9 1 1 i i ~ 0 . m . Pue 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 D 0 D 0 0 0 1 . ms a m0+ E E 0 + +0 +0 +0 -0 C 0 C 0 0 0 E E E E E E E E E E +E + , Dy . S L m n ~ -Btg 2 2 9 5 2 9 85 72 59 36 23 00 00 3 9 9 3 3 43 55 5 ~N 7 7 6 4 2 0 97 6 7 7 83 95 07 0 1 1 0 0 1 1 1 1 1 1 8 7 5 3 1 0 0 R O F N O I T A M R . OI FC E S D O EI TR AA LN PE EC S S A P - BO MR UD MP I E - XE AD _ M - 9 2 7 2 0 E 0- R E U 9 7 8 G 9 T I 9 4 N E F
- 0. M 1
X = E C E A ML T P I 1 S I D ZA 2 L PA ET TO 3 ST 8 0 2 8 9 i H rc +. I L Table 8.2.1: BYRON /BRAIDWOOD POOL STRUCTURE DATA l Iteni Thickness l Base Mat 6'-0" l North and South Walls 5 '-0" East Wall (Byron) 5 '-6" ' West Wall (Braidwood) ,_ West Wall (Byron) 6'-0" East Wall (Braidwood) Ileight (four walls) 4 l '.0" 8.3 ' Material Properties The design basis evaluation [8.1.5] utilized conservative material properties so as to minimize the computed ultimate strength of the pool structure. Table 8.3.1 provides a summary of the key material properties. Table 8.3.1: MATERIAL PROPERTIES Parameter Value Concrete Compressive Strength (psi) 3.500E+03 ' ! Un-Cracked Concrete Elastic Modulus (psi) 3.372E+06 Concrete Poisson's Ratio 0.167 Concrete Weight Density (Ib/ft') 150.0 , Concrete Thermal Expansion Coefficient 5.500E-06 Reinforcement Yield Strength (psi) ' 6.000E+04 i ' Reinforcement Elastic Modulus (psi) 2.900E+07 j p l SilADED TEXT CONTAINS PROPRIETARY INFORM ATION lloitec International 8-3 Report 111-982083 j .