ML20043A381

Evaluation of High Density Spent Fuel Rack Structural Analysis for Southern California Edison Co,San Onofre Nuclear Generating Station,Units 2 & 3

ML20043A381
Person / Time
Site:	San Onofre
Issue date:	03/31/1990
From:	Degrassi G BROOKHAVEN NATIONAL LABORATORY
To:
Shared Package
ML20042G265	List: ML20042G264 ML20042G266 ML20042G290 ML20043A381
References
NUDOCS 9005210278
Download: ML20043A381 (68)

Text

-

-,r m

/99

,,,.u#r,$- 841 qM ~.

a. _

m 1

.O I

W."..

.t.

,c or,,,.

W.<

l z e Am##hr canuturen.

y ','

l

.. < i t.w 31

- s.

~

W.Pamn i

.c

1. wV m7

?k W, :$, g,h A,-

y 1.+5;b..

'i g{

ih' ' ^

bh kk]c

.k,y...,. e4

'NI

.d,.f.

g'" EVALUATION OF NIGH DENSITY.$ PENT

, EL' RACK 7.4pMM4

,,r,

. %,,.. s.r.

..~.

4,_

'g.h

.g7,gg q,.3 W % $TRUCTURAL' ANALYSIS TOR, M

<h SOUTHERNCALIFORNIAEDISONCOMPANYdyj c:

e ra.,.

i SAN ONOFRE NUCLEAR GENERATING STATION - G

\\

%p,c,.

eh l

d.A J..s.

NRC DOCKET N05."50-361 AND 50-362 S.i..6s 3 6 yi.@%

I b :?i 1, % &@M. h Qb**Y. Q WyDy} $ c N W M W W h]) %

$lQQf?

' h&O?v

=

. +. M 'r, & :'n..,

kJ-l,.?,.h y a ?.., c C 4M.@.!!.l4.W.pS,. 9M.6,%'$.

4?.$k Ka iy'N

\\

i

• c.J..[

y., :

G. DEGRASSI cpges MpW.(NtW'jri@%

f.L..

-:KN i

.,s-

'X4 y.

i.,

f( s 7;.q:g,'.g.Qbbr@.;.qMTKfMwCM'MRR.Ridt '4W

(

,s

~y.

. c.m y,n y p v&h MVM TY & t.fi'% fL" s 1 *W' J

@ W M,'.Q.(',lR'Q%fdkdMhNf,'*,NOhMbNhd M"'

l f,, z v.g..

r.;-

9, ; p a n y.Gus+4 gfl}&W&.!f!'.Q W. ff. ty'%f 4

. g. g~' /g~g g..

b 1

g.

i DEPARTMENT OF NUCLEAR ENERGY yth STRUCTURAL ANALYSIS DIVISION.

I

+...

i

~

,jpWee. m?,-

F 4W%. %$49#t%&..Y'

b. s.ev e

ex BROOKHAVEN NATIONAL LABORATORY

. %,;.. 'i' Q 4 6.,n.

.. M'.*&$. 2 /

i D",... 1.

r UPTON, NEW YORK 11973.

+

p y

<u,

~

.r.w

.%n '..

.w. @ m - 40.n$ f l

. '. f. ;;s m.f. f '..i ?,l,' ;

.,.~(,

Q e

. k

" ;,3" @Q.~

Q mv.m R_

.l

',. ',,. c,

.g.,,, 4,

MARCH 1990

.... v ' '.'

=

y

)

s 3

a w v l% : S;..

l

<..p., - 4 r tre{.q d.

r w.

'r.Y*.r.

. +.,..

\\

w E

U DEPARTMENT Of E(LEAR INERGY, BROOKHAVEN hail 0HAL LABORATOR V'JK r.

UPTON, NEW YORK 11973

'... i y',(.:, x

~

j.

..q...

.q e.S

,,.v:? _bc.1. ?

.,%,,.,A, p,,...,..

ec

.e v -

m l

s

(

,j

+ N(;

.g'",. ['; O '

'. E I ' ' ',,

{

l s

[,

< !, t ' d,*@

WW $ /*, N ', / 'I'f'. ( ' #'i.'. ~ M' ',". '.f.,

(hYd # NN ' # )

N'

~ *

'Q s

..,, A n,.

?: %.. :. J-

,,e

.' ' f '.p ',

y.g, g,,' g.

g I

y L,

(

M Prept'ed for the U S NvMeer Reg / story Commission j j 3 y $ fs,;j:p -

k' Office of Nuc6 ear Reactor Regulation v... e g.%"

^

i-. s Contra,ct No DE ACO2 76CH00016.* '*71..I,t.n. 9.M.M. * <"

• /

i s:#

/.A C

C g

,,. l y, x I

.t

.t

..-4

(

f ; *,

q 's s.", '.

b.S '. 0 I h ' N 3 D.I

'M l -

N*

p 1

I i

e. m

. c.

9005210070 900DU1

.E.., '.

C v., -

(

ppo ADDtn 05000361

b..

.;;' ' '). '

PDn w

- ~

\\

.,e o

.. y

's w ~. d.-.

~ w G.-.a. L -.

.. G w,,.,, i x L'. c.1

jW

+

7

' '86 C 4 841 I

J

.m....

,,4,

2 y

. UsenatWert q

.,Q m wy ll

. wg' ystw

.g n

n --

7,<

.r.c Mr EL,, ~

,6 D;

4 p u.. X V. A L U..T I O..N.

.~IG.N.D.E.N. $ TTY PENT y

~

m

~

+~

ofix tpfg h;p%si... - :u, 4-y.y i.. m.g&,SOUTNERN CALIFO.RNIAl. S!$.;1.FOR,,

s

..T,.RUC.T.U.RA.L.. 'RNALY y,,.

r...,

..f, g EDI, SON. COMPANY.

<'m r

.. ~

m.

g

. S.AN. O.NOFRE NUCLEAR SENERATING STAT 10N.C4NIJS 2 AND 8.i e. :. -..x~.

f

,J..

R.,g%.,$KIdy,NRC00CKETjkOS.s.-50-361'AND 50-362

, 7.

~

a a

%ph,%

ad g 3 tw e.

..u

.+

i %

f

?@

-y d i di W tgla'f g A,8. DEGRAS$1 i

C hY

. h

%$ki

,. m.. g.T!NT

... ; y @f m

ers. w w al4 @

kv.t k.et P

ba

,h:k'hM$h,N(,"

3 N'P

$w@' -

w: g,g' g :t g.i[ f 8 M { $ f*v M r.i.n h 4,v v,

@.f M l 5 %s M?q.r..

N a

e wp s

. y.

w6 o 4; fty 9 @t"k STRUCTURAL ANALYSIS. DIVISION s.

W'-& W" s,

-y 9.p. y.~:.

.d

~a:

OEPARTHENT OF NUCLEAR ENER8Y-3-

. < ' ^ '."

v ygg4 -

Ms!Mi@p BR00KHAVEN' NATIONAL D BORAf0EY.

.4

$W

.t#Mid%@b!f ~

2 y

??ge % UPTON, NEIYORf11973,' Ph1 W

.ae v

.- 4.eean M

&r..k. ?M r~;-

,esm;ow w.v4m w A:e@ A W M #.t 4..m, et#9W h, f

. h;t

. ~

$d

)

^

t i$6W.),&YkkW&g {f{y.h,..ypi

. hN

?

JI:l;

y

' ws

x. a e>11 h 9

.c4

,;., 4 e n %,%,. w%ubetyp94.%:AC-%My?l %ihh..,

Me

.v m.,

4 1%,@ :eVQh%@ ', M?w&

d^y.~:;

f.c.p..

W ht k

.b@::l4 : ~

%&l-WE'Y'jdb;;"

{*

c.f.p'p-DEPARTMENT OF NUMA't EWERCf, St00KHAVEN NATIONAL LABORATORY, M

Wu;rl,g..gg.:,.

1..

UPTO NW YORK nFN#Ff'-@V4 fh.}

7f>f'dfhd yhk i;@.M Q.

,k h-h.

N

.y.

6 i

Y '*

,edWs

.gpyg..p(']yy:

j pg' Y

Mb.51 e :

O lgfM(yyM

,..v%* M'.

&'lG g,J.3 3 h_dt' ?/9.edh,.

bNH -

s g' ' g,Np Office of Nuclear Reactor Regutetten iI ~ pared for

. Pre

.g y1;,' i'. Lp'~'. *t p

af,. *"

=

k.

gla,T4

.c y

<c Cont M25,4.w.ract N.o DE,-A00N6CH00018',

r.. u....

., a:n y Y

i

$!Yh.

• Y bk fth.yi.;

v. 3, M.

.I,e d o :, W iJ. M G. M ? / M l W b 4hd ~' M ['pify, f pa.R,.p,, i:M " 9) yMPk 3

1A 9005210278 900501

~ %,8c.'

l'h),,(,"Tb.

r W juu,$?/.

4 '.q y. pA, Q.e..~ ~., pQ q'., p.4, y'.

9)

U PDR ADOCK 05000361 e 9.,y,m x. m,3,9;;.h.14 M$j s.y y f..

. ) ?.p. Q V 9

PDR ef y k

p

.t T.,

.Q lQ y

g..

s...

. ~.,,, c m:

1

. w.

w.

h

~

• J

i A-3841-3/90 j

EVALUATION OF HIGH DENSITY SPENT FUEL RACK STRUCTURAL ANALYSIS FOR j

SOUTHERN CALIFORNIA EDISON COMPANY j

SAN ONOFRE NUCLEAR GENERATING STATION - UNITS 2 AND 3 NRC DOCKET NOS. 50-361 AND 50-362, G. DEGRASSI i

r STRUCTURAL ANALYSIS DIVISION DEPARTMENT OF NUCLEAR ENERGY BROOKHAVEN NATIONAL IABORATORY UPTON, NEW YORK 11973 Y

MARCH 1990 t

--a

.-.s-

... -.,.,,., +.

,m..

- _.. ~

.--,..,-.v-7...,m,,-,.,.,,...,,.,,,_

r-

4 I

EXECUTIVE

SUMMARY

l This report describes and presents the results of the l

Brookhaven National Laboratory (BNL) technical evaluation of the structural analysis submitted by southern California Edison Company in support of their licensing submittal on the use of high density spent fuel racks at the San onofre Nuclear Generating station (SONGS) Units 2 and 3.

The review was conducted to ensure that the racks meet the structural requirements defined in the NRC Standard i

~

Review Plan and the NRC OT Position for Review and Acceptance of Spent Fuel Pool Storage and Handling Applications.

The ; proposed reracking of the spent fuel pool involves the installathon of eight free-standing, self-supporting modules of varying sizes arranged within close proximity of each other and the pool walls.

Each rack wodule consists of individual cells of square cross-section, each designed to accommodate one fuel i

1 assembly.

Since the racks are neither anchored to the pool floor or walls nor connected to each other, during an earthpake, the racks would be free to slide and tilt with the possibility that the racks could impact each other or the pool walls.

Because of l

the nonlinear nature of this design, a time history analysis was required to characterize the seismic response of the fuel racks.

The BNL review focused primarily on the seismic analysis of the fuel rack modules because of the compicxity of the analysis method and the number of simplifying assumptions that were required in devoleping the dynamic models.

BNL also reviewed other analyses submitted by the Licenseg including fuel handling accident hnalyr.us, thermal stress-analyses, and spent fuel poo). structure and linar analysas.

The primary document reviewed was the Licensing Report which provided a detailed summa.ry of the struc-tural ana]ynes and evaluations.

In addition, BNL participated in NRC audits at the plant site, the fuel vendor's fabrication facility, and the offices of the Architect Engineer.

During these audits, detailod calculations were reviewed on a sampling basis and additional questions were discussed.

The fuel rack seismic analysis was based on state-of-the-art methodology.

The development of the simplified nonlinear dynamic fuel rack model effective structural properties was based on finite element analysis of more detailed structural models of both Region I and Region II racks.

Careful consideration was given to geometric and material nonlinearities which affect rack sliding, tilting, and potential impacts.

The effects of rack submersion in water including three dimensional multiple rack interaction effects were accounted for.

A sufficient number of bounding load cases were investigated.

To address uncertainties, additional sensi-tivity studies were performed to demonstrate the conservatism of the model.

The results of the analysis showed that the rack lii

__ r modules would not impact each other or the pool walls and that code allowables are met with adequate margins.

Based on the BNL review of the Licensee's analyses, it was 1

concluded that the proposed BONGS 2 & 3 high density spent fuel i

racks-and spent fuel pool are designed with sufficient structural i

capacity to withstand the offacts of the required environmental and abnormal loads.

i t

4 t

t iv s

4

..v~--

TABLE OF CONTENTS PAGE

1.0 INTRODUCTION

1 1.1 Purpose........................................

1 1.2 Background.....................................

1 1.3 Scope of Review................................

1 2.0 ACCEPTANCE CRITERIA.................................

3 3.0 FUEL RACK DESCRIPTION.........

3....................

5

4.0 TECHNICAL EVALUATION

7 4.1 Fuel Rack Seismic Analysis.....................

7 4.1.1 Dynamic Model...........................

7 4.1.2 Fluid Coupling Effects..................

10 4.1.3 Friction Effects........................

11 4.1.4 Damping.................................

11 4.1.5 Seismic Loads...........................

12 4.1.6 Load Caues..............................

12 4.1.6.1 Primary Analysis...............

13 4.1.6.2 Confirmatory Analysis..........

13 4.1.7 Analysis Method.......m................

14 4.1.8 Analysis Resulta........................

14 4el.8.1 Displacement Resultr............

14 4.1.C.2 Stress Results.................

15 4.2 Puel Rack ThorLal Stress Aaelytis........,.....

17 4.3 Drop Accident and Uplift Analyses..............

17 4.4 Spent Fuel Pool Analysis.......................

18-4.4.1 Loads and Load Combinations.............

18 4.4.2 Pool Structure Evaluation...............

19 4.4.3 Pool Liner and Anchorage Evaluation.....

21 4.4.4 Pool Floor Bearing Load Evaluation......

22 4.4.5 Foundation Stability and Soil Bearing...

23

5.0 CONCLUSION

S.........................................

24

6.0 REFERENCES

25 Y

-I

I I

LIST OF TABLES

]

anza num nas 1

RACK DATA (EACH UNIT)..........................

26 2

LISTING OF SEISMIC ANALYSIS BOUNDING CASES (PRIMARY ANALYSIS).............................

27 3

LISTING OF. SEISMIC ANALYSIS BOUNDING CASES (CONFIRMATORY ANALYSIS)........................

28 i

4 SAN ONOFRE RACK DISPIACEMENT

SUMMARY

( FI NAL IAY0UT).................................

29

)

5

SUMMARY

OF RACK DISPIACEMENT VERSUS GAP

( FI N AL IAYOUT ).................................

30 6

LOADS AND ICAD COMBINATIONS FOR SPENT FUEL RACKS 31 7

MINIMUM MARGIN To ALIDWABLE REGION I...........

32 8

MINIMUM MARGIN To ALIcWABLE REGION II...~.......

33 9

LOAD COMBINATIONS FOR SPENT FUEL P00L..........

34 10 CURRENT EVALUATION RESULTS FOR THE SPENT FUEL POOL WALLS AND BASEMAT.........................

35 11 COMPARISON OF GOVERNING RESULTS FOR THE ORIGINAL DESIGN VERSUS THE CURRENT EVALUATION FOR THE SPENT FUEL P00L........................

36 t

vi

--.4,

,-----,e u

. ~ _ _ _ _ _.. _.

b i

LIST OF FIGURES FIGURE TITLE PAGE 1

SPENT FUEL STORAGE RACK ARRANGEMENT.............

37 4

2 REGION I CELL IAYOUT............................

38 3

FUEL STORAGE RACK (REGION I)....................

39 4

REGION I RACK CROSS-SECTION.....................

40 5

REGION II CELL IAYOUT...........................

41 6

REGION II FUEL STORAGE RACK.....................

42 7

REGION II RACK TOP-VIEW.........................

43 l

8 REGION II RACK CROSS-SECTION....................

44 9

STRUCTURAL MODELS REGION I & II.................

45 10 EFFECTIVE STRUCTURAL MODEIA REGIONS I & II......

46 11 3-D NONLINEAR SEISMIC MODEL REGION I &

II.......

47 12 NONLINEAR SEISMIC MODEL (2-D VIEW OF 3-D MODEL)

REGION I........................................

48 I

13 NONLINEAR SEISMIC MODEL (2-D VIEW OF 3-D MODEL) l REGION II.......................................

49 l

14 MULTIPLE RACK MODEL FULL / FULL REGION I..........

50 l

15 MULTIPLE RACK MODEL EMPTY / FULL REGION I.........

51 1

16 MULTIPLE RACK MODEL FULL / FULL REGION II.........

52 17 MULTIPLE RACK MODEL EMPTY / FULL REGION II........

53 18 MULTIPLE RACK MODEL (FIAN VIEW) REGION II.......

54

, 19 SONGS 2 & 3 SPENT FUEL POOL FLOOR ACCELERATION TIME HISTORY HORIZONTAL N-S DBE.................

55 20 SONGS 2 & 3 SPENT FUEL POOL FLOOR ACCELERATION TIME HISTORY HORIZONTAL E-W DBE.................

56 21 SONGS 2 & 3 SPENT FUEL POOL FLOOR ACCELERATION TIME HISTORY VERTICAL DBE.......................

57 vii

,,_,,.mv v

e^v

^

LIST OF FIGURES (Continued)

FIGURE TITLE g

22 SONGS 2 & 3 FUEL BUILDING POOL FIDOR HORIZONTAL NS (NE368 C4) FOR 4% DAMPING DBE SPECTRA........

58 23 SONGS 2 & 3 FUEL BUILDING POOL FIDOR HORIEONTAL EW (NE368 C4) FOR 4% DAMPING DBE SPECTRA........

59 24 SONGS 2 & 3 FUEL BUILDING POOL FLOOR VERTICAL (NE3 60 C4) FOR 4 % DAMPING DBE SPECTRA...........

60 25 PRELIMINARY SPENT FUEL STORAGE RACK ARRANGEMENT 61 26 FINITE ELEMENT MODEL OF FUEL NANDLING BUILDING 62 27 LOADING ON SUPPORT PAD..........................

63 9

l l

l l

viii l

m

]

... s

1.0 INTRODUCTION

1.1 Purpose This technical evaluation report (TER) describes and presents the results of the BNL review of Southern California Edison company's (SCE) licensing submittal on the use of high density fuel racks at San Onofre Nuclear Generating Station (SONGS) Units 1 and 2 with respect to their structural adequacy.

1.2 Background

The spent fuel racks are located in the spent fuel pool (SFP) in the fuel handling building (FHB) which is a Seismic Category I reinforced concrete structure. Separate FHBs are provided for Unit 2 and for Unit 3.

Each FHB contains a SFP in which 15 spent fuel storage racks are currently located.

Each of the racks is bolted to beams which are anchored to the SFP floor.

The racks in each pool provide underwater storage locations for up to 800 fuel assemblies (3 2/3 cores) per unit.

With the existing capacity, Unit 2 and Unit 3 will each retain its full-core reserve storage capacity through cycle 6 which is scheduled to begin in 1991 and 1992, respectively. To increase the capacity, SCE plans to replace the existing storage racks with free standing high density spent fuel storage racks.

This will expand the capacity of each pool to 1542 fuel assemblies and extend the fu21-core reserve storage capability for each unit through cycle 11 which is scheduled to begin in 2001 and 2002, respectively.

The Licensee provided a summary of his safety analysis and evaluation of the proposed racks in a Licensing Report (Ref. 1).

The original report was issued in March, 1989.

The report was revised several times to incorporate design changes, reanalysis results, and responses to NRC questions.

The latest version (Revision 6) was issued in February 1990.

The Licensing Report provided a detailed description of the new fuel rack designs, the structural analysis of the racks, and the structural reevaluation of - the exit ting SFPs.

BNL reviewed the Licensing Report and participated in NRC audits at the SONGS plant

site, the Westinghouse rack fabrication facility in Pensacola, Florida and at Bechtel offices in Norwalk, California.

During these audits, BNL and NRC personnel inspected the spent fuel pool, witnessed the rack fabricatior,

process, and reviewed sample calculations.

Additional questions pertaining to the rack and pool evaluation were raised and discussed.

Final questions were resolved in a series of telephone conference calls.

1.3 Scope of Review The objective of the BNL technical review was to evaluate the adequacy of the Licensee's structural analysis and design of the proposed high density spent fuel racks and spent fuel pool.

Due 1

_____________m_____.________.-__.____

to the complex nature of the fuel rack seismic analysis,, the primary focus of the review was on the adequacy of the non-linear fuel rack models and their dynamic analysis.

The structural evaluation of fuel racks subjected to the dropped fuel and jammed fuel handling accidents described in the Licensee's report (Ref.

1) were included in this review.

However, the definition of these postulated accidents and their parameters (drop height, uplift force, etc.) were beyond the scope of this review.

A limited review of the spent fuel pool was conducted to ensure that appro-priate loads, methodology and acceptance criteria were applied.

i l

9 2

l

~ - - -.. -

u

,.n,.._,,

g,

.2.0. ACCEPTANCE CRITERIA The acceptance criteria for the evaluation of the spent fuel rack applications are provided in the NRC OT Position for Review and Acceptance of Spent Fuel Storage and Handling Applications position).

Structural requirements and criteria given in this (Ref. 2 paper were updated and included as Appendix D to Standard Review Plan 3.8.4, " Technical Position on Spent Fuel pool Racks,"

(Ref. 3).

These documents state that the main safety function of the spent fuel pool and fuel racks is to maintain the spent fuel assemblies in a safe configuration through all environmental and abnormal loadings, such as earthquakes, and impact due to spent fuel cask drop, drop of a spent fuel assembly, or drop of any other heavy object during routine spent fuel handling.

1 Section 2 of SRP 3.8.4, Appendix D gives the applicable Codes, Standards and Specifications.

Construction materials should conform to Section III, Subsection NF of the ASME Codes.

Design, fabrication and installation of stainless steel spent fuel racks may be performed based upon the ASME Code Subsection NF require-ments for Class 3 component supports.

Requirements for seismic and impact loads are discussed in Section 3 of Appendix D.

It states that seismic excitation along three orthogonal directions should be imposed simultaneously for the design of the new rack system.

Submergence in water may be taken into account.

The effects of submergence are considered on a case-by-caea basis.

Impact Loads generated by the closing of fuel assembly to fuel rack gaps during a seismic excitation should be considered for local as well as overall effects.

It should also be demonstrated that the consequent loads on the fuel assemblies do not lead to fuel damage.

Loads generated from other postulated events may be acceptable if sufficient analytical parameters are provided for review.

Load knd load combination requirsments are provided in Section 4.

Specific loads and load combinations are acceptable if they are in conformance with Section 3.8.4-II.3 and Table 1, Appendix D of the Standard Review Plan.

Changes in temperature gradients across the rack structure due to differential heating effect between a full and'am empty cell should be incorporated in the rack design.

Maximum uplift forces from the crane should be considered in the design.

Section 5 discusses design and analysis procedures. It states that design and analysis procedures in accordance with Section 3.8.4-II.4 of the Standard Review Plan are acceptable. The effects of gaps, sloshing water, and increase of effective mass and damping due to submergence in water should be quantified.

Details of the mathematical model including a description of how the important parameters are obtained should be provided.

3 i

L____

i l

Structural acceptance criteria are provided in Section 6.

The acceptance criteria are given in Table 1 of Appendix D.

For impact loading, the ductility ratios utilized to absorb kinetic energy should be quantified.

When considering seismic loads, factors of aafety against gross sliding and overturning of the racks shall be in accordance with section 3.8.5-II.5 of the standard Review Plan l

unless it can be shown that either (a) sliding motions are minimal, impacts between adjacent racks and between racks and walls are prevented and the factors of safety against tilting are met, or (b) sliding and tilting motions will be contained within geometric constraints and any impact due to the clearances is incorporated.

4 l

r e

9 4

9 J

4

1 i

3.0, FUEL RACK DESCRIPTION The new high density spent fuel storage racks will provide for l

storage of new and spent fuel assemblies.

Westinghouse 14 x 14 fuel assemblies from Unit 1 and C-E 16 x 16 fuel assemblies from i

Units 2 & 3 may be stored in either Unit 2 or Unit 3 racks.

The fuel will be stored in two regions within each pool. Region I (312 l

locations) consists of two high density fuel racks, each with 12 (1230 locati(nominally 125.5 inches by 135.9 inches).

Region II x 13 cells l

ons) consists of six high density fuel racks, four with 14 x 15 cells (nominally 124.82 inches by 133.67 ir,ches) inches).

and two with 13 x 15 cells (nominally 115.97 inches by 133.67 Region I will be used to store non-irradiated, 4.1 w/o or less U235.

enriched fuel and fuel which has not achieved a pre-determined burnup.

Region II is designed to accommodate irradiated fuel which meets the predetermined burnup.

Fuel which does not meet the burnup criterion may be placed in Region II in a checkerboard or alternating row array.

A plan view of the arrangement of the eight fuel rack modules i

in the Unit 2 pool is shown in Figure 1.

The Unit 3 rack arrange-ment is a mirror image of Unit 2.

As indicated, the rack modules i

are installed with minimum nominal gaps cf 3.38 inches between I

adjacent racks and 5.63 inches to the pool walls.

Physical data l'

for each rack is given in Table 1.

The two Region I racks are composed of individual storage cells with an inner square dimension of s.64 inches.

Each cell is fabricated from one sheet of 0.110 inch thick stainless steel I

welded together at one corner.

A neutron absorbing material, l

Boraflex, surrounds the cell walls and is held in place by a thin i

stainless steel wrapper which is spot we:tded to the walls along its l

edges.

The cells are interconnected by grid assemblies and stiffener clips to form an integral structure as shown in Figures I

2 and 3.

Each rack is provided with 26 leveling pads which contact the SFP floor or pN1 floor plates and are remotely adjustable from above.

The racks are neither anchored to the floor nor braced to the pool walls.

They also are not connected to each other.

The pool floor plates are also not attached to the floor.

Figure 4 illustrates the basic sections of the fuel rack assembly which are the storage cell, the neutron absorber material, the Boraflex i

t wrapper, the top and bottom grid assemblies, the base plate, and I

the leveling pad assembly.

All rack components are made fro:n Type 304LN stainless steel except the leveling screws which are SA-564 Type 630 stainless steel.

The six Region II storage racks consist of stainless steel cells assembled in a checkerboard pattern, producing a honeycomb type structure as shown in Figures 5 and 6.

The cells are located in every other location and are welded together at the cell corners without the use of grids.

Each cell is of the same basic design as described for Region I and includes the Boraflex neutron 5

\\

absorbing material held in place by the wrapper.

The checkerboard configuration results in open "non-cella locations atound. the periphery of the rack.

Stannless steel cover plates are used to close off the openings and form storage locations.

These plates i

are welded to the adjacent cells as shown in Figure 7.

The cells are welded to a base support assembly which includes 33 leveling s

pads that are remotely adjustable from above. Stiffener plates are welded against all cells and cover plates around the periphery of each rack adjacent'to the base plate.

The racks are not anchored to the pool floor nor braced to the pool walls or each other.

The racks rest on floor plates which are not attached to the pool 4

floor.

The basic sections of tna Region II fuel rack assembly are l

illustrated in Figures 7 and 8.

They are the storage cell, the cover plate, the stiffener plate, the neutron absorber material, i

the Boraflex wrapper, the base plate, and the leveling pad assembly.

All components are made from Type 304LN stainless steel 1

except for the leveling screws which are SA-C64 Type 630 stainless steel.

l 1

L l

1

(

l 6

l l

Ea

,4.0, TECHNICAL EVALUATION l'

4.1 Puel Rack Seismic Analysis The spent fuel storage racks are seismic Category I equipment required to remain functional during and after a safe shutdown earthquake.

As described in Section 3.0, the proposed racks consist of 8 distinct free-standing modules which are neither anchored to the pool floor, attached to the side walls, nor connected to each other.

Any rack may be completely loaded with fuel assemblies, partially loaded, or completely empty.

The fuel assemblies arc free to rattle within their storage cells.

Seismic forces are transmitted to the racks through friction at the support leg to pool floor interface.

If seismic displace-ments are large enough, the racks can impact against each other or the pool walls and the support legs can lift off and impact the pool floor.

Because of these non-linearities, a time history analysis of nonlinear rack models vas required to charactorize the seismic rasponsa of the fuel racks.

BNL's revieu of the details of the modeling technique and analysis method is descr Ded in the following sections.

4.1.1 Dynamic Model In order to perform a seismic time history analysis, the Licensee developed a three dimensional ncnlinear finite element model of multiple racks in the pool.

The model included linear r-elements representing thu fuel-and rack stif fnesses and geometric properties with sufficient dynamic degrees of freedom to capture the dynamic response of the combined fuel and rack system.

The model nonlinear elements needed to represent impacting between the e

fuel assembly grids and cell included the impact stiffness, impact damping and gap size.

The nonlinear elements that represent the support pads which may lift off and impact the floor, or may slide relative to the floor included the impact stiffness, impact damping and Coulomb friction.

To compile the nonlinear dynamic model, the linear effective structural properties were obtained from a detailed 3-D finite element structural model of the fuel rack.

Two difforent struc-tural models were developed as shown in Figure 9 to represent a Region I and a Region II rack.

Rack cells are represented as beams with cross sectional properties of the individual cells.

Support pads are modeled as beams with cross sectional properties of the pad components. The base plate is modeled as effective beams which connect the support pads with the cells.

Effective properties are based on a cross section of the plate and a portion of the bottom grid (Region I) or a portion of the cell wall (Region II).

The cell to cell connections are produced by stiffener clips in Region and by cell to cell welds in Region II.

These connections are modeled by effective beams which connect between cells.

The 7

---__.--.------,---,----------._.----------------~----~~.m--"

_.7 4

~

properties of these beams are obtained from a separate finite element analysis of a detailed model of a section of call, wall, and stiffener clip or weld.

By performing analyses of the detailed structural models, simpler effective structural models were developed.

The effective l

models shown in Figure 10 are composed of elements which represent l

the cell assembly, call to cell connections, and the support I

pad / base plate assembly.

Beam elements representing the cell assembly have the same properties as those in the structural model.

Rotational stiffness elements represent the cell to cell connection average rotational stiffness at each elevation.

The rotational i

stiffness between the bottom of the cell and the base plate is included in the stiffness matrix element, (K).

The support pad / base plate assembly is represented by a combination of rigid beams with vertical spring elements on the ends.

The vertical stiffness of the effective corner pads is calculated by equating the rotational stiffness of the vertical spring / rigid beam assembly to the base plate rotational stiffness derived from the structural

.model.

To ensure that the effective structural model is an edequate dynamic representation of the structural modal, frequen-cies and mode shapes from both models were obtained and compared.

l The ncnlinear model, shown in Figura 11. :.

$s a 3-D model i

consisting of the effective structural rack modal with additional l

elencnts to account for the fuel assembly, fusi to c4,11 gap, faal j

hydrodynamic mass, support pad boundary conditions, and hydro-1 dynamic mass between the nck and poel wall or between racks. More detailed two dimensional illustrations of the Region I and Region II models are shown in Figures 12 and 13.

In this model, the fuel i

assembly is represtnted by a combination of beam elements and rotational stiffness elements.

The beam elements have cross l

sectional properties of the fuel rods and fuel skeleton.

The rotational stiffness elements account for the connection of the fuel rods at the fuel spacer grid locations.

The dynamic proper-ties of the fuel assembly were verified by comparison to frequency i

and mode shape values supplied by the fuel vendor.

The fuel to cell gaps are modeled by 3-D dynamic elements with gaps, impact j

stiffness, and impact damping.

The impact stiffness considered 1

i both the local stiffness of the cell wall and the spacer grid i

I impact stiffness supplied by the vendors.

There are 12 gap elements along the length of the fuel assembly representing 11 spacer grids and one top and fitting.

The effective support pad / base plate vertical stiffness is represented by a 3-D friction j

element which has the capability to slide on the horizontal plane or lose contact in the vertical direction and impact upon contact.

l l

The hydrodynamic mass between the fuel assembly and cell and between fuel rack and pool wall or adjacent racks is modeled by 3-D mass matrix elements.

The elements are located at 12 elevations along the length of the fuel and cell as shown in Figures 12 and 13.

Hydrodynamic mass effects are discussed in more detail in Section 4.1.2.

j 8

i Additional nonlinear models were developed for partial fuel

' loading.

They included quadrant fuel loading, tour row fuel loading and empty rack.

The displacement results of theso models and the full rack model were used to determine which two racks produce the maximum relative motion, and also to justify the placement of fuel in the rack in any configuration.

These models were similar to the full rack model modified so that the loaded fuel cells are shifted to the center of gravity of the fuel loading with stiffness and mass properties equivalent to the number of j

cells or fuel assemblies an the partial loading condition.

i In order to address rack to rack interaction, a nonlinear model of multiple fuel racks was developed.

The details of each rack in this model are the same as those of the single rack model.

There are two differont multiple rack models, full / full and empty / full, for each region as shown in Figures 14 thrcugh 17.

The number of dynamic degrees of freedom for these models are 190 1

full / full and 129 empty / full for Region I and 180 full / full and 123 empty / full for Region II.

The full / full model was used to produce the maximum absolute cisplacement of the reeks in order to address the relative motion between the rack _ and 1 mot ve.11.

The empty / full model was used to produce the maximum relative displace ment between racks.

Results cf rack 2o141 analysis for full partial fuc2 loading in quadrant, partici foal loading in four rows, and empty rack demonstrated that maximum relative displace-ment was produced by the combination of a fully loaded rack and an empty rack.

I A plan view of the multiple rack model is shown in Figure 18.

It shows the position of the two rack models with solid lines and l

the position of two " effective" rack models with dotted lines. The l

effective racks are not actual models, but are shown there because they represent the effect of adjacent racks on the two actual racks.

Thus the two actual racks respond in a manner which accounts for the effect of adjacent racks in both the N-S and E-W directions.

As a result, the Licenses refers to the multiple rack model as a four rack model.

During the NRC audit at the Westinghouse plant in Pensacola in April 1989, BNL reviewed the methods used to develop the nonlinear multiple rack models.

Westinghouse personnel explained the modeling techniques and made available proprietary calculations for BNL review.

Based on a review of sample calculations and further discussions with Westinghouse, BNL found the modeling process acceptable.

The nonlinear dynamic rack models were judged to be reasonably conservative and contained sufficient detail to l

provide realistic results.

The methods used to define some of the key parameters including fluid coupling, friction, and damping.are discussed in more detail below.

l l

9

4.1.2 Fluid Coupling Effects The effect of submergence on the fuel racks in a pool of water has a significant effect on their seismic response.

If one body of mass m vibrates adjacent to another body of mass m, and both 3

e bodies are submerged in a frictionless fluid medium, then Newton's equations of motion for the two bodies have the forms (m3 + Mu) E - M : W = applied forces on mass m3 3

3

-Mn M + (m + Mu) En = applied forces on mass n 3

a 2,

We denote absolute accelerations of mass m3 and m,

3 e

respectively.

Mue Mat e M and M, are fluid coupling coefficients n

which depend on the shape of the bodies and their relative disposition.

The equations indicate that the effect of the fluid is to add a certain amount of vass to t.ach body and an external force which is propertional te the accelechtion of ths adjacent i

body.

Thus the ;tcceleration of one body Mfects the force on the adjacent body.

The forco te a wrong function of the interbody gap, reaching large values for very shall gaps.

This inertial coupling is called fluid or hydrodynamic coupling and has an important effect on the dynamic rerponsa of the racks as well as the fuel assemblies inside che storage onlis.

The hydrodynamic coupling between the fuel and cell and between the rack and pool wall is represented in the models by mass i

matrix elements.

The hydrodynamic coupling between the racks and between the racks and pool wall was calculated using a method based upon potential flow theory outlined in a paper by R.J. Fritz (Ref.

4).

To account for flow in three dimensions, the hydrodynamic mass for flow in the horizontal direction around the racks and for flow in the vertical direction up over the top of the racks and down i

below the bottom of the racks were calculated independently, and combined to produce an overall hydrodynamic mass value, f'

The hydrodynamic mass between standard fuel assemblies and cell walls is based upon the fuel rod array size and cell inside dimensions using the technique of potential flow and kinetic energy.

The concept is discussed in a paper by D.F. DeSanto (Ref.

5).

For consideration of consolidated fuel which has an enclosure around the fuel rods, the hydrodynamic mass model is based on a square body inside a square body using the methodology of Fritz.

BNL reviewed the detailed fluid coupling parameter calcula-tions and concluded that the analytical models were reasonable and the methods were consistent with current state-of-the-art analysis-methods for fuel rack seismic response. The methods addressed both the virtual mass and the fluid coupling of fuel assemblies and fuel racks.

Flow models for standard fuel assemblies properly consid-ered flow through the open array of fuel rods while models for consolidated fuel considered only flow in the channels around the 10 l

l

, fuel,. For the rack hydrodynamic effects, both rack to pool wall' and rack to rack hydrodynamic masses were calculated.

Proper consider-ation was given to three dimensional flow effects.

A limitation of the potential flow theory used in the hydrodynamic mass calculation is the underlying assumption that the vibratory deflections are small relative to the mise of the gaps.

This is generally not the case for seismic response of fuel racks.

The hydrodynamic masses are calculated on the basis that the gaps between bodies remain constant.

In reality, the problem is nonlinear with hydrodynamic masses increasing as gaps decrease.

The licensee has stated that the constant gap assumption is conservative because the restoring forces as the gaps decrease are larger than predicted.

BNL has investigated this issue during reviews of other fuel rack analyses and has judged the methodology acceptable as long as the overall model ar.d analysis method is conservative.

4.1.3 Priction Effects Friction alonents were used at support pad to pool floor and fuel to rack base interface locations.

The values of frict).on coefficient used in the analysis were based on tests performed by Rabinowicz (Raf.

6).

The results of 399 tests perforned on austenitic stainless steel plates submerged in water shownd a mean value of friction coefficient to be 0.503 with a standard deviation of 0.125.

Therefore, the upper and lower bounds based on two standard deviations are 0.753 and 0.253 respectively. The Licensee performed separate analyses for bounding values of 0.2 and 0.8 at the support pad to pool floor locations.

At the fuel to rack interface, the Licensee indicated that preliminary studies using upper and lower bounds showed no significant difference in response and therefore, the lower bound was used in all load cases.

The use of upper and lower bound friction coefficients at the rack to floor interface is consistent with current industry practica. -The results of the analyses show that different dynamic behavior was produced by each coefficient.

Maximum sliding occurs with the lower bound and maximum tilting occurs with the upper bound.' In order to demonstrate that the results are bounding, an additional analysis was performed using a mean coefficient of friction of 0.5.

That study demonstrated that the maximum rack displacement for the mean friction coefficient case was bounded by the 0.2 and 0.8 cases.

4.1.4 Damping

+

The model incorporated structural damping and impact damping.

The model did not consider fluid damping.

The structural damping values were 2% for OBE and 4% for SSE. These values are in accordance with NRC Regulatory Guide 1.61 for welded steel structures and the SONGS 2 and 3 UFSAR.

Impact damping was used 11

at the fuel to rack cell and at the support pad to pool floor.

The damping value used for the fuel impact was 4.4% which was supplied by the fuel vendor.

The damping value for support pad impact was based on the coefficient of restitution for steel on steel impact.

Reference 7 gives a range of 0.5 to 0.8.

Conservatively using a value of 0.85 coefficient of restitution produced an effective impact damping of 44 for the support pads.

The damping values used j

in the analysis were judged to be reasonable and acceptable.

4.1.5 Seismic Loads Seismic floor response spectra for the SFP floor were developed using the methods described in the SONGS 2 and 3 UFSAR.

The parameters of the original lumpsd mass model of the FHB were adjusted to reflect the increaned mans corresponding to the now high density spent tual storage rachu completely filled with Units 2 and 3 spent fuel. e.ssetblies.

The resulting SFP floor response spectra were used to generats statictically independent 80 second duration time hiat. cry records, one for each of the three orthogonal i

directions as shown in Figures 19 through 21.

Each pair of time histories has a correlation coefficient of less than 0.1.

The l

response spectrum from each corresponding $SE time history record

[

is plotted in Figures 22 through 24 to show their enveloping of the tioor response spectra.

The pool floor time hintories were uuhd as input to the dynamic analysis of the rz.eks.

Pool wall time hiratories were not conalderad since the fuel racks have no l

connections to the vall.

It was noted that rack action is tied to pool wall motion by fluid coupling but based on a review of the pool design, differences between floor motion and wall motion are not expected to be significant.

The seismic input notion was judged to be acceptable.

l 4.1.6 Load Cases The Licensee ran a number of seismic load cases to account for the parameter variations including friction coefficient, Region I and Region II rack designs, and fuel loading in order to predict the bounding rack responses.

In addition, the original analyses considered a preliminary pool layout which included two identical 12 x 13 Region I racks and six identical 14 x 15 Region II racks as shown in Figure 25.

During the course of the design, other considerations dictated the need to remove one row of cells in the east-west direction.

The final pool layout shown in Figure 1 includes two 13 x 15 and four 14 x 15 Region II racks as well as small differences in gaps between racks.

The Licensee raran a limited number of load cases to determine bounding responses for the final rack configurations.

The Licensee referred to the original _ analysis as the " primary" analysis and the additional analysis of the final configuration as the confirmatory analysis.

. Load cases for the two sets of analyses are discussed below.

12 m-,.-

-e

- - - -, - ~ ~

• m.

I

,4.1.,6.1 Primary Analysis The load cases analyzed in the primary analysis for the preliminary pool layout are summarized in Table 2.

The cases covered both rack types, upper and lower bound friction coeffi-cients, and various fuel load conditions.

Both standard fuel and consolidated fuel with twice the standard fuel weight were considered although the license amendment will not allow for the use of consolidated fuel at this time.

Fully loaded racks, empty racks and partially loaded racks (quadrant and four cuter rows) were analyzed.

Studies using single rack models with various fuel load conditions demonstrated that the bounding load cases for multiple rack analysis, ars full / full and empty / full fuel load conditions. A review of the results and the studies indicated that nearly all conceivabla conditions and parameter variations were addressed and that the 'multip.le rack model analyses covered the bounding cases.

4.1.6.2 Confirmatory Analysis The load caves analyzed in the confirmatory knalysis for too finsi pool layout are summarized in Table 3.

The non!inear dyntah:

model used in the confirnatory analysis was identical to the model used An the primary analysis with the exception of changes in the hydrodynamic mass retrix values to reflect the changes in rack to pool wall and rack to rack gaps.

However, only a limited number of Region II racks were included in the confirectory analyses. The Licensee provided the following justification for selecting only Region II rack cases for the confirmatory analyses.

l 1.

Region I and Region II racks have similar stiffness and fundamental frequency and for comparable fuel loading will respond to seismic loading in a similar manner.

2.

The rack to pool wall gaps for the' Region II racks changed by a greater percentage than the ga ps for Region I racks.

Therefore, the change in response wt11 be greater for Region II racks than for Region I racks.

3.

The Region II cases included in the confirmatory analysis are the conditions of maximum absolute and relative displacements and loads so that the limiting conditions were evaluated.

A review of the primary analysis results confirmed that the load cases in Table 3 provided large displacements and loads. Some Region I rack load cases had larger displacements but because of larger clearances were not limiting cases.

It was noted that three of the four confirmatory analysis load cases include consolidated rather than standard fuel because the consolidated fuel cases provided generally larger loads and displacements.

However, the license amendment will not permit the use of consolidated fuel at

.this time.-

In the confirmatory analysis, Region I rack responses 13 l

l l

l l

were estimated by applying the maximum ratio of Region II confir-matory to primary analysis results to the Region I primary analysis results.

Based on a review of the design changes and the detailed methods in which the bounding responses were determined, the Licensee's approach was judged to be acceptable.

4.1.7 Analysis Method i

The seismic analysis of the fuel racks was a time history analysis performed on the 3-D nonlinear finite element models subjected to the simultaneous input of three statistically independent acceleration time histories at the pool floor elevation. The analysis was performed on the Westinghouse Electric Computer Analycis WECAN.)sageneralpurposofinitesiementcodeCode using the (WZCAN position acthoi.

]

s-developed by Westinghouse (Ref. 8 and 9).

The nonlinear todal superposition. method was vsveloped to e,nalyse nonlinshr structural dynamics problens involving ir. pact between components and coulomb friction.

The finite ohnent method is used to express the equations al motion with the nonlinearities represented by a pseudo-force vector.

References 10 and 11 provide additional 1

information and application of the nonlinear modn1 superposition j

methol.

The WECAN program had been previously reviewed by NRC.

In addition to general reviews, the application of the nonitnear modal superposition method in the WECAN code for spent fuel rack seismic analysis had been specifically reviewed and accepted by NRC during the licensing review of several other fuel rack dockets.

l 4.1.8 Analysis Results In evaluating the structural adequacy of the fuel racks, the Licensee applied a displacement criteria and a stress criteria.

The displacement criteria seeks to ensure that the rack is a physically stable structure.

The racks were evaluated for margin against overturning and for rack displacement to ensure that rack to rack and rack to pool wall impact does not occur.

The stress criteria was based on the ASME Code,Section III, Subsection NF, 1986 Edition up to and including A-86 Addenda.

A summary of the results and their evaluation is presented and discussed below.

1 4.1.8.1 Displacement Results l

The displacement responses of the fuel racks were determined from the results of the load cases listed in Tables 2 and 3.

For the bounding Region II 14 x 15 racks, the displacements were taken directly from the confirmatory seismic analysis.

For the Region II 13 x 15 racks, an 8% increase in rocking displacement was applied for the 0.8 coefficient of friction case (the condition with significant rocking response).

The increase was calculated based on applying the same lift-off energy per location as the 14 e.~

m r

4 1

.14 w.15 rack to the smaller and lighter 13 x 15 rack.

For the 0.2

)

friction coefficient load case, no increase factor was applied since rack size does not have a significant effect on rack sliding displacement.

l l

For the Region I racks and the Region II racks that were not reanalyzed, the displacements were estimated by multiplying the r

i primary analysis displacements by the highest ratio of the Region i

II confirmatory to primary analysis displacements.

The maximum percentage change of lit was used even though a smaller change occurred for the limiting case.

The displacement results for all the load cases are summarized in Table 4.

Both absolute displace-ments (with respect to pool floor) and relative displacemente (with-respect to adjacent racks) are presented. Results for consolidated fuel load cases are for informatiort enly since the license amendment will not allow the use of consolidated fuel at this tirt.

Using the displacement recalts from the confirmatory analysis the remaining gaps were detarldned by startir.g with the nominal gap and subtracting the seismic displacement, thermal growth, and rack installation tolerance.

For rack to pool gaps, an additional i

adjustment was made to account for pool wall variation.

From the resuits prosented in Table 5, it was concluded that rack to rack and rack to pool vall impact will not occur.

It was also noted that the smallent rammining gaps result from consolidated fual load cases.

Standard fuel load cases have greater margins.

The maximum support pad vertical lift-off displacement was also extracted from the time history results and found to be 0.32 inches.

For that displacement, the factor of safety against rack overturning was determined to be greater than 39 which satisfies the requirements of Section 3.5.5.II.5 of the Standard Review Plan.

Based on the review of the detailed methodology used to calculate maximum absolute and relative displacements for both the preliminary and final pool layout configurations, BNL concurs wich the reasonableness of the approach and with the conclusion that rack to rack and rack to pool wall impacts should not occur during the SSE.

Furthermore, there is ample margin against rack over-turning to satisfy SRP requirements.

4.1.8.2 Stress Results The. structural analysis of the fuel racks was a limit analysis per ASME Code,Section III, Subsection NF, paragraph NF-3340 (Ref.

12).

The loads and load combinations of Appendix D of SRP Section 3.8.4 were used.

The " factored" loads were compared to " limit" loads to obtain the margin to allowable.

The limit load.was calculated using the yield stress from the ASME Code and assuming elastic perfectly plastic material.

Welds were proportioned to meet the requirements of NF-3342.2 (e) (4).

15

i The load combinations and' acceptance limits are shown in Table 6.

Seismic loads (E and E') include overall rack loads pl.us local loads due to fuel impact.

Load combination 7 included thermal loads even though they can be considered secondary, self-limiting stresses.

The Licensee demonstrated that the limiting load combinations are combinations 3,

5 and 7.

The results were presented in terms of minimum margin to allowable which was defined as the limit load divided by the factored load minus one.

The structural evaluation of the rack components and welds used seismic loads from both the primary analysis and the confir-matory analysis.

For the Region II 14 x 15 racks, the seismic loads used to calculate rack stresses were taken from the confir-matos.y analysis.

For the Regicu II 13 x 15 racks, the loads were determinef. by using the same loads per storaga location as the 14 x 15 ra::ks and than multiplying by the nuater of locations in the 13 % 15 rccks.

For the Region 3 racks, the percsintage chEngs of j

l the Region II 1 cads for the confirnatory analysis vueus tbn primary analysis loads was determined.

These load change factors were than applied to the Pogion I primary Miklysis Icads to calculate loads appropriate for the confirestory analysis.

1 The loads corresponding to the find rack confiepration vore used to calculate stresses er;d margins to allowable for the Region I and Region II rack components and welds.

Tables 'l and 8 provide the stress summary for tha critical rack components and welds in terms of margin to allowable for the OBE and DBE load combinations.

In addition, impact loads on the fuel assemblies due to rack interaction were determined. The maximum calculated seismic impact load at a spacer grid location was 2522 pounds which is less than the allowable spacer grid strength of both the Westinghouse and CE fuel assemblies.

During the Westinghouse audit, BNL reviewed a sample of stress calculations and found the methodology to be acceptable. Questions were raised regarding an assumption on transfer of loads through the cell seam velds. Westinghouse revised their calculations using more accurate methods based on finite element analysis.

Stress margins were shown to be lower than originally calculated but still within acceptance limits.

Questions regarding potential buckling of the cell valls from hydrodynamic loads were also raised.

Westinghouse provided an additional calculation which demonstrated that the cell wall is capable of carrying the loads.

Additional i

information on fuel assembly load limits was requested.

The Licensee provided a description of the Westinghouse dynamic impact test program on spacer grids and a test report from CE on a similar test program.

The Licences also provided additional information to justify the nethodology used to extrapolate the primary analysis results to the final pool layout configuration.

Based on a review of the Licensing report, backup documents, and additional material provided during.and following the audits, 16 4

m

..~

I

,. - ~...

BNL concurs with the adequacy of the methodology and results of the

'stra'ss evaluation. The test programs used to evaluate the strength of the fuel assemblies were appropriate for this application.

The methods used to extrapolate the loads and stresses to the final pool configuration were reasonable and adequate.

4.2 Fuel Rack Thermal Stress Analysis Two fuel rack thermal stress load cases were addressed.

The first case examined thermal stresses in the support structure resulting from the racks being installed at one temperature, followed by overall heating of the pool water.

This results in diffarential thermal growth between the rack and the pool floor.

The support pads are restrained laterally by friction.

Maximun the mal loads were calculated by assuming a friction coefficient of 0.8 times the deadweignt.

Stressas in the support pad com-ponents and wolds were calculated and ample margins were dunnstrated.

The second thermal load case evaluated stresces induced in a rack cell loaded with fuel with ad4,acent cells; empty.

This condition results in a hot cell surrounded by colder cells.

The maximum stress due to restrained thermal expansion of the hot cell occur in the cell to cell welds or clips.

The thermal stresses were calculated and combined with OBE stressas using the 1.3 load factor (load combination 5).

Adequate margins to allowable were demonstrated.

The Licensee argued that thermal stresses are secondary and self-limiting, and do not need to be combined with DBE stresses.

However, an evaluation of that load combination was also performed and margins were demonstrated.

4.3 Drop Accident and Uplift Analyses Several drop accident scenarios were postulated and analyzed.

Two cases of fuel assembly drop accidents were considered.

They-include a Westinghouse fuel assembly dropped from a height of 24.9 feet above the pool floor and a CE fuel assembly dropped from a 21.7 foot height.

Three drop orientations were considered:

1) drop of an assembly onto the top of a rack in a vertical position,

2) drop of an assembly onto the top of a rack in an inclined position, and'3) drop of a fuel assembly through an empty cell to the bottom of the pool. The acceptance criteria requires that fuel criticality does not occur, and perforation of the pool liner does not occur.

The Licensee used an energy balance approach and made several conservative assumptions.

The fuel assembly was assumed to fall freely in an infinite pool of static water.

No energy.was dissipated in the rack structure during the drop.

Pool liner and floor flexibilities were neglected.

When the fuel assembly falls through a cell and contacts the base plates, the base plate welds are assumed to fail allowing the fuel assembly to fall through to 17

1 and impact the pool floor. The kinetic energy of the fuel assembly is assumed to be absorbed by crushing the base plate, bottom fuel nozzle and fuel assembly guide tubes.

Using this approach, the Licensee demonstrated that the stresses imposed on the pool liner were well below the ASME Code faulted condition allowable limit and the liner will not be perforated.

The Licensee also stated that the presence of 1800 ppa Boron in the fuel pool water ensures that fuel criticality will not occur.

]

At the request of NRC, the Licensee performed an additional analysis to demonstrate that, by removing some of the conservatisms in the calculation method, the dropped fuel assembly will not impact the pool liner. The results of that calculation showed that the worst fuel drop accider.t will cause the rack base plate to deflect less than the minimum distance between the plate and the liner.

The Licentee also evaluated accidental drops of a spent fuel pool gata and of a 4500 ' pound piece of test equipment.

These analyses indicated that potential fuel damage may occur if these components fall en top of a rc;ck.

As a result, Licensee performed a radiological evaluation and established administrative controls to limit the consequences as discussed in the Licensing Report (Rei. 1).

The Licensee also performed an analysis to evaluate the effects of an uplift load of 6000 pounds produced by a jammed fuel assembly. Using worst geometry assumptions, the stresses resulting

)

from this load were calculated and compared to the acceptance limits.

This condition was determined to be enveloped by the results reported in Tables 7 and 8 for the limiting loading combinations.

4.4 Spent Fuel Pool Analysis The existing spent fuel pool was reanalyzed and reevaluated to ensure that it.can accommodate the increased loads associated with the new high density racks.

The following sections discuss J

the loads and load combinations, analysis methods and analysis results of the spent fuel pool reevaluation.

4.4.1 Loads and Load Combinations The following loads were considered in the SFP analysis a)

Structural Dead Load (D),

including the weight of the structure, hydrostatic loads, weight of new spent fuel racks filled with fuel, and a 50 lb/ft uniform load on floor slabs 8

to account for miscellaneous equipment, piping and electrical raceway loads.

l 18

b)

Live Load (L),

including weight of the cask in the cask storage pool and weight of the railroad car / cask in the railroad unloading area.

Hydrostatic loads were considered live loads in the. fuel transfer and cask storage pools since these pools do not require water at all times.

The SFP basemat has no floor occupancy, live loads.

l c)

Normal Operating Thermal Loads (To) due to temperature differences and temperature gradients through the concrete wall /basemat and stainless steel 1.iner plate system during normal operating or shutdown conditions. The followin temperature conditions were assumed in the analysis g normal j

Pool Water 150*F 1

Outside Ambient Air 4 0'F

-l Outside Soil-4 0*F Inside Ambient Air

' 6 5'F Initial Installation 65'F j

d)

Abnormal Thermal Load (Ta) due to temperature differences and temperature gradients through the wall /basemat and liner plate i-during a full core off-load and pool boiling including the

~ effects of concrete gamma heating.

The reinforced concrete temperature differences and gradients were determined based on an inside face temperature of 230*F (water temperature of L

212*F and gamma heating of 18'F).

For the liner plate, a lower temperature of 216'F was used since the gamma heating is less significant.

e)

Design Basis Earthquake (DBE) Load (E') - three directional seismic loads were applied based on the free field -DBE L

response spectrum from the UFSAR.

The resultant seismic load distributions were established by a three-component SRSS combination technique.

f)

Operhting Basis Earthquake (OBE) Load (E) - the free field OBE l

g spectrum acceleration values are one-half the DBE values.

Three directional. seismic loads were applied and combined in the same manner as the-DBE case.

1 l

The Load Combinations from the SONGS 2 & 3 UFSAR were con-L sidered and are shown in Table 9.

Loads due to tornado, pipe

-l l

rupture.and missile loads were not included because they are not L

applicable to the area being evaluated.

4.4.2 Pool Structure Evaluation I-The Licensee developed a new finite element model to reanalyze l,

the spent fuel pool structure.

The model was analyzed using the Bechtel Structural Analysis Program (BSAP).

This is a general t

purpose computer program which uses the direct stiffness approach L

to perform linear elastic analyses of three-dimensional structural L

19 l

models.

The finite element model of the FHB is shown in Figure 26.

It is consistent with the original lumped mass model'shown in the UFSAR but is refined in the area of the spent fuel pool to assure accurate results in the areas of concern.

The building is modelad in 3-D plate, beam, brick, and boundary elements at and below the pool deck, and is represented by a stick model for the portions above the pool deck.

Both static and dynamic analysis models were developed which were identical except for soil spring boundary elements and spring elements added to represent the hydrodynamic loads of the oscil-lating water in the dynamic model.

The methodology of AEC Report TID-7024 (Ref. 13) was used to develop the elements representing the oscillating water.

In addition, the hydrodynamic pressure loads'due to the motion of the racks relative to the pool walls were included in the pool wall evaluations.

The soil boundary elements were attached to the basemat master node at the center of gravity of the basemat.

The element properties were based on the FHB soil stiffness parameters listed in the UFSAR.

For the dynamic model, masses were lumped at the appropriate nodes and a modal analysis was performed in which enough nodes were extracted to achieve adequate mass participation. The results were used to perform OBE and DBE response spectrum analysis with damping values that are in agreement with the UFSAR.

The concrete section evaluations were performed using the OPTCON module of the BSAP program.

All evaluations were performed using the strength design method of ACI 318-71.

Moments due to thermal gradients are self relieving in nature and OPTCON accounts for the reduced moments of the cracked section using an iterative approach which considers equilibrium and compatibility equations for the cracked section as allowed by the applicable design code.

OPTCON also accounts for any additional moments induced by a hot liner plate and considers the effects of plate yielding (if it occurs) on the concrete.

The results of the SFP structure evaluation are summarized in Table 10.

The results are presented in terms of utilization factor which is defined as the percentage of resistance of reinforced concrete section that has been utilized relative to the zero curvature line.

Load combinations 6 and 7,

which have large temperature gradients, are the governing combinations. The largest utilization factor was 88.4% from load combination 7.

The north wall was found to be the critical wall for out-of-plane shear with a 78 kips /ft value compared with an allowable of 151 kips /ft.

The Licensee also compared the governing results for this evaluation against the governing UFSAR results.

The comparison, shown in Table 11, indicates that the current evaluation of the basemat and walls results in reduced section moments and membrane forces and increased c'esign margins even though the fuel rack loads 20

'incr' eased.

The Licensee attributed the load reduction to the refined analysis and design techniques used in the reevaluation.

It was noted.that the results presented in the Licensing Report were based on the preliminary rack layout shown in Figure 25.

The Licensee was asked to provide additional information on the effects on the deletion of one row of storage locations on the pool structure evaluation. The Licensee stated that the final rack layout had the following effects on pool loadst decreased hydrodynamic loads on the SFP walls, decreased rack dead loads on the SFP basemat, and increased rack seismic loads on the SFP basemat.

The SFP was reevaluated for the revised rack loads and the results of the wall evaluations showed that the resulting loads decrease sind the results of Tables 10 and 11 envelop the final reevaluation results.

The evaluation of the basemat showed that the resulting loads increased but the resulting membrane forces and moments increase by less than 0.5%.

The utilization factors would have a similar small increase which would have no impact on the conclusions.

Based on a review of the additional information provided by the Licensee, BNL concurs with the conclusion.

4.4.3 Pool Liner and Anchorage Evaluation The liner plate system was reevaluated for the new spent fuel.

rack floor loads, load drops, and thermal effects.

Local effects due to rack vertical loads are discussed in Section 4.4.4.

Overall rack horizontal loads due to friction were evaluated for both Region I and II racks:

Actual horizontal racks loads of 5252 Kips were shown to be less than the capacity of the liner and anchorage system.

The anchorage capacity (8240 Kips), which was controlling, was based on AISC allowables times a factor of 1.6 for

'DBE as allowed by UFSAR.

The original analysis for thermal effects on the liner was based on conservative parameters with a pool temperature of 220'F and an initial unstressed liner plate temperature of 60'F resulting in a temperature differential of 160'F.

The current design basis temperature is 216'F with an initial temperature of 6 5'F.

The temperature differential of 151*F is less severe than originally analyzed.

Therefore, the results of the original analysis which concluded that the liner plate will not buckle due to design thermal conditions remain valid.

Load drop evaluations were performed for the SFP gate, a test equipment load, and an empty fuel rack.

The analyses indicated that the liner plate could be perforated with the empty fuel rack drop giving the maximum penetration (5 3/4 inches deep by 63 square inches). The consequences of this perforation were evaluated. The Licensee concluded that since the concrete penetration is only 7%

of the basemat thickness, the water will be maintained within the pool / leak chase system.

The worst case rate of discharge to the 21

i

. leak detection sump was calculated to be 49 gal / min which is well i

within the makeup capacity of 150 gal / min. Therefore, the' Lice'nsee i

concluded that the Technical specification water level within the i

pool can be maintained, l

l 4.4.4 Pool Floor Bearing Load Evaluation Local effects due to the rack vertical loads were evaluated based on worst case single support pad loads for both Region I and R

Region II. The evaluation showed that single support pad loads are acceptable except when applied directly over or adjacent to the

,I leak chase channel weld seams or embedmont plates.

Consequently, the support pads over or adjacent to the leak chase channels were provided with load spreading floor plates to assure that the actual concrete bearing is within the ACI allowables.

The floor plates are fabricated from Type 304 stainless steel and sized to accom-zodate seismic displacements of the rack support pads.

The plates were not welded to the liner but were roughened on the bottom to L

permit rack sliding with minimal plate sliding during an l

earthquake.

The criteria used in the concrete bearing strength evaluation was taken from ACI-349, Paragraph 10.15.

The analysis considered both the vertical and horizontal components of loading.

The loading model assumed that the horizontal ( %, force is applied L

at the mid point of the length of the love.:

I r**ew, 4.37 inches above the liner as shown in Figure 27.

1 W 3 duces an over-turning moment in addition to the verticM wi which must be i

reacted at the pool floor. The sliding meismn action of the rack produces an eccentricity between the line of action of the vertical load and the centroidal axis of the floor plate which results in an additional moment.

Since the plate has no capability of transmitting a tensile load to the concrete, all forces and moments must be reacted through appropriate compressive bearing stress distributions as shown in Figure 27.

t To evaluate the limiting bearing stresses, a series of force-moment interaction diagrams were developed by representing the plate as an eccentrically loaded column without tension capability.

Limiting floor plate locations at leak chase and weld intersections i

were considered.

The results of the nonlinear seismic analysis were reviewed to identify the largest loads and load combinations.

l Conservative combinations of vertical and horizontal forces and i

eccentricities were taken to develop worst case force and moment l

combinations.

All combinations were shown to fall within the appropriate interaction diagram curves.

l l

BNL reviewed the details of the analysis method and concurs with the Licensee's approach and conclusion that the bearing stress criteria of ACI-349 have been satisfied.

22

_.__m..

'4.4.'5 Foundation Stability and Soil Bearing The original FHB seismic analysis documented in the UFSAR '

determined that the worst case foundation bearing pressure is 21 kips /ft" compared to an allowable baaring pressure of 44 kips /ft'.

The - total change in FHB mass due to the addition of the high density racks is less than 74.

The Licenses also determined that the building dynamic characteristics changed only slightly with frequencies shifting by less than 13%. Since the allowable is more than twice as large as the UFSAR actual and the changes are relatively small, the Licensee concluded that the bearing pressure will remain well within the allowable.

l L

i l-L l

\\

l

\\

i 23 L

t

~.- -..

i

5.0 CONCLUSION

S Based on the review and evaluation of the Spent Fuel Pool Raracking-Licensing Report and additional information provided by the Licensee during the course of this review, it is concluded that the proposed San Onofre Units 2 and 3 high density spent fuel racks have sufficient structural capacity to withstand the effects of all required normal,_

severe, and abnormal extreme environmental loadings.

The' Licensee submitted a thorough, detailed analysis based on state-of-the-art methods to demonstrate.the structural adequacy of the new fuel-racks and of the existing spont fuel pools.

The finite element modeling techniques provLded realistic, three dimensional nonlinear dynamic models for predicting seismic response.

Careful consideration was given to geometric and material nonlinearities which affect rack sliding, tilting, and potential impr. cts.

Rack submersion in water was adequa.tely considered.

A sufficient number of conditions with different combinations of parameters were analyzed to ensure that limiting responses were obtained.

Sensitivity studies were performed to address a number of uncertainties. The analytical modeling methods were generally. conservative and the results showed that code allowables are met with adequate margins.

The Licensee also demonstrated that seismic impact loads on the fuel assemblies would not damage the fuel.

The Licensee reanalyred the existing spent fuel pool to the revised loads resulting from storage of additional fuel assemblies in high density fuel racks.

The analysis demonstrated the fuel pool has adequate structural capacity to accommodate the revised loads with adequate margins.

?

e 24

~

.6.0. REFERENCES 1.

Southern California Edison

Company, San Onofre Nuclear Generating Station Units 2 and 3, Spent Fuel Pool Reracking.

Licensing Report," Revision 6, February 1990.

2.

U.S.

Nuclear Regulatory Commission, Letter to All Power Reactor Licensees, from B.L Grimes, April 14,

1978, "OT-Position for Roview and Acceptance of Spent Fuel. Storage and Handling Applications," as amended by the NRC letter dated January 18, 1979.

3.

-U.S.

Nuclear Regulatory Commission Standard Review Plan, Section 3.8.4 Appendix D,

" Technical Position on Spent Fuel Pool Racks," NUREG-0800, July 1981.-

4.

R.J. Fritz, "The Effects of Liquids on the Dynamic Motions of Immersed Solids," Journal of Engineering for Industry, Trans.

of the ASME, February 1972, pp 167-172.

5.

D.F.

DeSanto, "Added Mass and Hydrodynamic Damping of Perforated Plates Vibrating in Water,"

ASME Journal of Pressure Vessel Technology, May 1981.

6.

" Friction Coefficients of Water Lubricated Stainless Steels for a Spent Fuel Rack Facility," Prof. Ernest Rabinowicz, MIT, a report for Boston Edison Company, 1976.

7.

A.

Higdon and W.B.

Stiles, " Engineering Mechanics, Vector Edition," Prentice-Hall, Inc., Engelwood Cliffs, N.J.,

1962.

8.

WECAN

" Documentation of Selected Westinghouse Structural Analysis Computer' Codes," WCAP-8252.

9.

WECAN

" Benchmark Problem Solution Employed for Verification of the WECAN Computer Program," WCAP-8929, 10.

V.N. Shah, G.J. Bohm, and A.N. Nahavandi, " Modal Superposition Method for Computationally Economical Nonlinear Structural Analysis," ASME Journal of Pressure Vessel Technology, Vol.

101, May 1979, pp 134-141.

11.

V.N. Shah, C.B. Gilmore, " Dynamic Analysis of a Structure with Coulomb Friction," Submitted for Presentation at the 1982 ASME Pressure Vessel Piping Conference, Orlando, Florida, June 27 -

July 2, 1982.

12.

ASME Boiler & Pressure Vessel Code,Section III, Subsection NF (1986 Edition up to and including A-86 Addenda).

13.

United States AEC Report TID-7024,

" Nuclear Reactors and Earthquakes," August 1963.

25 l

l

e i

i Table 1 RACK DATA (Each Unit)

Reaion~1 Recion II Number of Storage

[

Locations 312 1230 f

.Ntunbar of Rack Arrays Two 12 x 13 Four 14 x 15 Two 13 x 15 Center-to-Center Spacing (inches) 10.40 8.85 Cell'Inside Width-(inches) 8.64 8.63 l

Type of Fuel Unit 2 & 3 Unit 2 & 3 16 x 16 and/or 16 x 16 and/or Unit-1 14 x 14 Unit 1 14 x 14 Rack Assembly outline 126 x 136 x 198.5 125 x 134 x 198.5 Dimensions (inches)

(14 x 15) 116 x 134 x 198.5 (13 x 15)

Dry Weights (lbs) 52,953 36,227 (14 x 15)

Per Rack Assembly 33,965 (13.x 15) i 9

26

Table 2 LISTING OF' SEISMIC ANALYSIS BOUNDING CASES.

(PRIMARY ANALYSIS)

RACK TYPE FUEL LOADING FRICTION COEFFICIENT Region I Partial, Quadrant 0.2 T<egion I Partial, Quadrant 0.8 Region I Partial, Four Rows 0.2 Region I Partial, Four Rows 0.8 Region I Full / Full 0.2 Region I Full / Full 0.8 Region I Empty / Full 0.2 Region I Empty / Full 0.8 Region II Full / Full 0.2 Region II Full / Full 0.8 Region II Empty / Full 0.2 Region II Empty / Full-0.8 27

'I l

Table 3 LISTING OF SEISMIC ANALYSIS BOUNDING CASES (CONFIRMATORY ANALYSIS)

RACK TYPE FUEL LOADING FRICTION COEFFICIENT Region II Full / Full (a) 0.2 Region II Full / Full (a) 0.8 Region II Empty / Full (a) 0.2 Region II Empty / Full (b)

O.8 a.

2 x Standard Case b.

Standard Case I'

h 28

\\

l

L Table 4 SAN ONOFRE RACK DISPLACEMENT

SUMMARY

(FINAL LAYOUT)

Absolute Dise (in)

Rel Diso (in)

Model.Tyne")

gg g

g g

g Reg.

1, Std. Full

.2 1.67 2.00

.37

.46 Reg.

1. Std. Full

.8 1.65 1.52

.93

.49 Reg. 1. Std. E/F

.2 1.44 1.68 1.54 1.92 Reg. 1, Std, E/F

.8 1.58 1.78 1.10 1.37

^

Rege 1, Consol, Full

.2 2.32 2.73

.37

.83 Reg.

1, Consol, Full

.8 1.17 1.38 1.31 1.63 Reg.

1, Consol, E/F

.2 2.19 2.99 1.86 2.64 Reg.

1, Consol, E/F

.8

.87

.88

.98 1.02

~ Reg.

2, Std. Full

.2 1.11 1.54

.29

.65 Reg.

2, Std. Full

.8 1.73 1.63

.82

.91-Reg.

2, Std. E/F

.2 1.21 1.17 1.10

.92 Reg.

2, Std.'E/F

.8 1.25 1.24 1.03

.80 Reg.

2, Consol, Full

.2 2.11 2.54

.42

. 44 Reg.

2, Consol, Full

.8 1.16 1.50

.90

.98 Reg.

2, Consol, E/F

.2 1.32 1.52 1.17 1.30 Reg.

2, Consol, E/F

.8 1.68 1.37 1.17

.95 4

")

Two terms are used to define the different loading conditions.

They are defined as follows:

Full is used to describe the fully loaded situation.

ELE describes a case where one rack is full and another is empty.

29

3-8 9

5 K

R c

a

-l g j W e

4 4

4 W

1.

Il a u" 2 2

3 3

2 2

c 3

j.,,,,,,,,

s s

e s

e s

Ig.

,1 la 1_

I-E a.

e a

l g,.

L t

,i L

g-2

=

=

=

l' m

.g

-, l E s i<

e 1 st n

st_

i-li e w5 s

a s-a a

a

_ aa a

a s e

e 4

a a

ri ng s

s s

s s

s s i g,-

g --

1 Ig I

m i

i 18!55555555 Is.

y

~

g

.i I E

I l

van i l l i

a l l vsmallaall I........

ig....---_

ig.

30

Table 6 LOADS AND LOAD COMBINATIONS FOR SPENT FUEL RACKS Lead Combination Accentance Limit 1.7 (D + L)

NF 3340 of ASME Code Section III 1.3 ( D + L + T.)

1. 7 - (D + L + E) 1.3 ( D + L + E + T.)

1.3 ( D + L + E + T.)

1.3 ( D + L + T. + P,)

1.1 ( D + L + T. + E ' )

1.

The abbreviations in the table above are those used in Section.

3.8.4 of the Standard Review Plan ~(SRP) where each term is defined except for T. and ' P,.

The term T. is defined here as the highest temperature associated with the postulated abnormal-design conditions.

The term P, is the upward force.on the racks caused by a postulated stuck fuel assembly.

2.

The provisions of:NF-3231.1 of ASME Section III, Division I, shall be amended by the requirements of Paragraph c.2.3 and 4 of Regulatory Guide 1.124, entitled " Design Limite and Load combinations for Class A Linear-Type Component Supports."

3.

For the faulted load combination, thermal loads will be neglected when they are secondary and self limiting in nature and the material is ductile.

31

~

Table 7 MINIMUM MARGIN TO ALicWABLE")

REGION I OBE(*) ~

. DBE")

support Pads 2.04

.1.21 l

- Cells 1.64 0.96 Grids 2.13 1.19 Cell,to Cell Clips.

2.01")

1.01 Welds Cell to Grid 0.46 0.21 l'

Cell to Clip 1.85 0.55 Grid to Grid 3.62 1.66 Grid to Base Plate 1.53 0.60-Cell Seam 1.40 0.41 l

Cell'to Wrapper 0.81 0.57 L

l-p a.

Load Combination 3 (1.7(C+L+E)) unless otherwise specified.

b.-

Load Combination 7 [1.1(D+L+T,+E') ).

c.

Load Combination 5 ' (1. 3 (D+L+E+T.) ).

^

i d.

The margins reflect the confirmatory analysis results for racks.

with standard fuel only.

.i 4

32 4

3

Table 8 MINIMUM MARGIN TO ALLOWABLE REGION II")

CBE(*)

DBE )

D Support Pads 1.65 0.84

'l Cells 1.14 0.55 Welds Call to Base Plate 1,38 0.46 Cell to Cell 1.03")

0.46

'l Cell Seam 1.66.

0.49 Cell to Wrapper 0.63 0.43 a-Load Combination'3 (1.7(D+L+E)] unless otherwise specified.

b.

Load combination 7 (1.1 (D+L+T +E ' ) ).

c.-

Load ' Combination 5 (1. 3 (D+L+E+T.) ).

i d.

The margins reflect the confirmatory analysis and are appli-cable to both the 14 x 15 and 13 x 15 Region II racks.with standard fuel only.

T 33

Table 9 LOAD COMBINATIONS FOR SPENT FUEL POOL Load Combination j

No.

1.

Normal Case 1.4D +~1.7L 4

2.

Severe Environmental Case i

'1.25D + 1.25L + 1.25E 3.

Severe Environmental Case 1.25D + 1.25L + 1.25E + 1.0T.

1 4.

Abnormal / Severe Environmental Case

1. 0D + 1. 0L + 1. 2 5E + '1. 0T, l-L

.5.

Abnormal / Severe Environmental 1 Case i

1.0D + 1.25E + 1.0T, i.

1 6.

Abnormal / Extreme Environmental Case L

1.0D + 1'.0L + 1.0E' + 1.0T.

l.

7.

-Abnormal / Extreme Environmental' Case i

1.0D + 1.0L + 1.0E' + 1.0T, 1'

IT l

\\

l l.

34 I

Table 10 CURRENT EVALUATIONM RESULTS FOR THE SPENT FUEL POOL WALLS'AND BASEMAT Governing j

UFSAR Load Utilization l

combinationM Factor (%)N North and South Walls:

Horizontal Reinforcement 7

88.4 Vertical Reinforcement 7

37.4 East Wall:'

Horizontal Reinforcement 7

23.1 L

. Vertical Reinforcement 7

47.0 West-Wall:

Horizontal Reinforcement 7

28.1 L

Vertical Reinforcement 6

79.5 Basemat:

North-South Reinforcement 7

51.7 East-West Reinforcement 6

81.4 I

a.

Refer to Table 9.

b.

The Utilization Factor is defined as the percentage of resistance of the reinforced concrete section that has been utilized relative to the zero curvature line.

The evaluation results shown are based on the original' rack g

c.

layout (Figure 25).. The final rack layout resulted in some i

rack interface loads ^ decreasing (rack dead and hydrodynamic loadu) and others increasing (rack seismic loads).-

These revised rack loads were evaluated and determined. to be enveloped in every caso.

35 1:

1 i-

i Table 11 COMPARISON OF GOVERNING RESULTS FOR THE ORIGINAL DESIGN VERSUS THE CURRENT EVALUATION FOR THE SPENT FUEL POOL i

Max Axial Flexural Flexural Location Governing Load Load Load in Spent Load Pu Mu Mu(Max)

Mu/Mu(Max)

Fuel Combination (kips)

(K-ft/ft)

(K-ft/ft) i Pool (a)

(b)

(c)

(d) 7-Foot Thick UFSAR 7

-527 2604 2660 0.98 Basemat in Pool Area CURRENT.

6 92 1465 1793 0.82 (E-W EVALUATION Reinf) 4-Foot Thick UFSAR 7

-404 445 947 0.47 (N or S)

Spent Fuel Pool Wall CURRENT 7

-67 208 554 0.38 (Vert EVALUATION Reint) 5-Foot Thick UFSAR 7

0 666 674 0.99

-(West)

Spent Fuel Pool Wall CURRENT 6

208 215 257 0.04 (Vert EVALUATION Reinf) a.

The UFSAR values are. from UFSAR table 3.8-10.

The ' current evaluations are the maximum values obtained and not'necessarily at the previous locations.

b.

Refer to Table 9.

c.

Sign convention for Pu:

Compression (-), Tension (+)

d.

Maximum flexural interaction capacity (Mu(Max)) given the axial

load shown (Pu).

e.

Refer to footnote c in Table 10 for discussion of rack interface loads.

36

Illil l

ji!I MN OW RFD 56 74(

0 3

0 0

6 a

0 6

f 5

7 7

7 8

1 a

6 7

6 4

r 1

1 3

3 4g

.t t

3 3

~

1 1

m

~

J' I

5 5 2 5

5 1

7 8 t

1 4

X t

g R

X gg 2

4 4

1 1

1 41 p

T S

N 0

E 0

M 5

5 E

4 2 1

t 1

G 8

X N

t f

0 4 R

X iR A

0 2 4

R 4

R 1

1 1

441 A

K 0

l 1

C 0

E A

8 e

R 25 7

r 5

u E

9 1

m g

G l

t 5

i A

i t

R X

R W

F R

1 0

1 3

O 0

1 L

4 O

T 4 1 O

S P

L

.4>-

EU N

0 F

89 0

3 9

T R

N 9

1 3

Dv t

1 E

E 5

4 x

R 3

X R

t W

P 9

2 O

S 1

1 2

L 2'

6 O

1 O

P a~ gs' k

6,

I 0

y d

5 kp

)

P d

t 5

i 9

7, 5

5 2

2 0

1 1

9 1

0 2

t 7

6 w*

I a

10.40 Ref.

Center to Center l.

8.640 Ret'._ Square Cell Inside Dimenaton a

f l

I

.110 Stock I

smmmmmmmmmmmmmmmmmmm.

Call Thicknese ~

l l

e

'l s

L

__ _ _ __. 1.. _ _ _..

__.. _ _ _ _ _ _1 i

l l

.020 Stock I

L Wrapper Thickness Seraflex Stiffener Clip

{ 10.40 Raf.

I l.110Stocka' Center

.127 -

to j -:

Cavity Center

[

Depth

[

g I.

I e

e

.i

,I

.+._ ___

i I

I, I.

l.

l 1

1.1.e Cell to cell (water) gap Mis (Typical)

Figure 2 REGION I CELL LAYOUT 38

- -~-~~ ~ ~ _

" ' " * - - ~ _

- g a g

1. g ag a g ag

- g

1. \\

- g

,g a g

,g ag

1. \\

1. g p g p g

- s

"\\

-g

- g I

"\\

- g ag p g

- g g

5 eg 4 y sg sg

\\

p g a g

1. g t

a

\\

ag ag eg

• g 4 eg

1. g "A

(

ag

1. \\

I ag "g

ag

~~

1. g

- ~,

ag 4

- ~a g

1. g A

-g

-~

alg l

-g 4

-g es 88

- s

,sa8" 5psa l

9 *%

1. *A M

,a k

4 p.g-

\\

4

, g g

s A

g h

da

- s f

'Ng I

j e==

i i

I i

l d

y I

l I

N f-l 1

%'g's'

.lf' j

g s

8 b

/

('

s a

p

• W

,,an

)

e h l

,/

I I

I s

y Figure shows 11 x 11 rack arre SONGS Region.1 racks are 12 x 13 arrays. Internal support pads are omitted. Botaflex is not required on rack sides that face the pool wall.

Figure 3 FUEL STORAGE RACK (REGION I)

I 39 l

i

\\

10.'40 R E F' CELL ASSEMBLY i

l j

UPPER i

• -- 8.6 4 SOU AR E -

ANACENT CELL TE i

.ASSEygty

\\

ID NT OP GR 10 7_

ASSE MBL Y i

h l

N 80RAFLEX 19 0.00 R E F l

l

% BORAFLEX l

1 WRAPPER' l

I s..

y l

--- h NE L

D 1-6-

0-l f//////////A OMIN

)

U RT BLOCK LEVEL IN G, P AD p,

[

- POOL,L INER L E VE L'ING SCREW NOTE

LINER BRIDGE PLATES

.NOT SHOWN Figure 4 REGION I RACK CROSS-SECTION 40

i l.

l.

8.630 Raf. Square-Call Inside Dimension typical I

t I

l l

.1?0 Stock j j

' -- Cell

=

Thichess l 8.630 Raf.

i I

Non-cell Insida "

I l

Dimension Typical

_____.9.___._

+

_.._.9._.,_._

I i

Boraflex I,

l o

.N

~-

.020 Stock-l

,o9e gag,_ ' l l

8.ts'Ref.

Wrapper 8

Thickness Cavity Center I

Depth l

to Wrapper j

Center i

Spacing

__.___..i___.___

4 i

.I l

I 8.85 Ref.

Center to Center Spacing e

Figure 5 i

REGION II CELL LAYOUT i

41 i

i l

l 1

c e

3

- 3 a 'a i

i i

J 3

a j

,sy

,l s'

n I

,g

,J s

/

~

[:.

3 i

1 3

3 a

/

3 a

a J

A 4

4

' J i

3

'Nf I

3 y f' J

l

}'

t s

- e p<

A s

Q'

/,/j.h s

by'

. r.

STIFFENER-PLATE e

g Ag e

e a

Figure shows 12 x 14 rack array.

SONGS Region II racks ~are 14 x 15 and 13 x 15 arrays. Boraflex is e

not required on the periphery cell walls.

b Figure 6 REGION II FUEL STORAGE RACK 42

...-s.

.w

._--_._a ema

>a

-.u.

a.n.-..

...a.n

...m_..

t CELL-TO-CELL WELD B.63 SQUARE CELL OPENING ~

-(TYPICAL)

CELL I

i I

I

[

k+ \\

f BASE PLATE V

(j L

y l.

l 8.85-T

\\

'i L

N O

i i

n #% n i n

V E* u V

j O

i.

I sTirrt-PLm N ov[,tm /

Figure 7 REGION II RACK TOP-VIEW l

43 l

- 8.85 RE F II

,,{ Eg

-#N y

a BORAFLEX

.x s** %

.. m'-

~

t I

f 90.co REF WRAPPER i

.ma i.

BASE PLATE 8

I w

a gm f ausman r// w m s,

$UPPORT BLOCX L EVELING P AQ

_.... _ = ~...

p M.

CC9aCRETE g PLA7E Ltvr u NG 4 CREW teor to staat NOTE: LINER BRIDGE PLATES NOT SHOWN Figure 8 44

I m

f

' REGION I REGION ll s

x x-n 32 1

a

~

lll n,r c---.

nntw

-a c= - y.

E g

g

--w-

.3 Top Grid

_...,g R.,

v : s = ;-

r cy -- - -

y sa r

gl,7

,y v-s i

W.w

-, '_ 4.

a u

dI!

' N

\\

ell to Cell Connection C

3

, 7 n

m s 1

e R

y a--

~

_ u,_,

Cell

~ : - -- ~- _ _

'$Lf_?._% J',

c*!

dl!

!h

[. -

,,,a

,,;i

3 w

. e = 3 p+ w; 7

a c

.K

-"__r

,'g w :l y

-._m Q.

gi, Basa Plate --

baa%

m-:Ee m

-- - - n, --

,5 Support Pad -

mumm r ; _'- ' ~ '

g -

- Z.. y l

~ ir mf Al Figure 9 S'i'RUCTURAL MODELS REGION I & II 45

,,_,,___._,m

. _ _ _ - - - - - - - - " - - - - - - - -- - - " - ' - - " - ^ - ~ ~ ' ^ ~ - ~ - - ^ ' - - ^ ' ^ ^ - ' - - ' ' '

....... ~.

-0 i

, j ji

. ).

j RECtON g REGION 11..

l

, e

=;gEIt:.. $k" 5:

l..

8........ q ~. !! :,p.g...

C t.. a..g we,, _4 -

}

t i

\\

~

i d

• I

1. y' i

n l#l'4

'i fl f,a IH l

ll

- i p.

- op

.f fly mijI

-,i M

I

\\\\

m

-q li:

I m '"

h e.'.

0

=

l lt ll 7

1 ylvtin.'.ta Jtvirni.ta l

c. u....

L..

c.;

j 3

t i

)

Figure 10 EFFECT2VE STRUCTURAL MODELS REGIONS I & II

't 9

46 i

s

2 p

S(

I [

[WN) e

[f 3" -[+,

s f

%.r!

/k f

A-4 4 h

1. , [f s s f g f x y" j'/l 4

% g p

we Jh N/

s i

A k Figure 11 3-D NONLINEAR SEISMIC MODEL REGION I & II 47

. ~. - -

%:4L;tt&2 Stilu!C weet;, st%3 g na r a

.v:a3 v%aw::* I * ' ' "w n g,

Ef'tkgh g;{' k C

g t

t.. alit wsu v, r.t. Allt wsLv h?;,hja

$:b

\\

z. p ruh

+.

W& $

4(4 A

m

+.

.=.

y u

ur

Mu'

^;,(Mb - l,

/

g.

+.

x IN,1 i

-+.

f"J"\\

0:'

/

rMO

/

w.

c:b:

/

0

/

T?

/

u rkb.

/

+

'W.tHc ~

W

/

EWENT 3

ocat 760-f/ // ////////.

Figure 12 NONLINEAR SEISMIC MODEL (2-D VIEW OF 3-D MODEL)

REGION I 48

4ChL igg an lg ll.ft.CDtL, lihSLg ascs 1bh its*W3

{~i!E'n*'"'

C.LL AlltWSLYq 4

rugL ag$t.3Ly

,,,rg.

"'*~

m,$

i&En b.

1. O n.

.g.M c

W& 8 W$L

~

emuusu 7

P-K.

/

w,w

=

M

.g,.L__

_/

mu.d

/

+

• s

'w x.6.

q

/

rh.

n.,

/

f u,*.

th' d'

/

~

~

/

~

L w-

/

/

v..,

c.,

n t w,',c - 2 L" l-

+

f/////////////

i Figure 13 l

l NONLINEAR SEISMIC MODEL (2-D VIEW OF 3-D MODEL) l REGION II l

l-49

"'"N.Wa% r.?hin'"va 7

T@

g C

/

/

e e

/

D SE G,c j/

,,, ~

tw

/

d C. -l O' '

N

/

+

+

/

9 9

7

/

e e aI 7

~

/

k e

Illll

/

/

Ilk @

l

((

!Il

[

7 l

/

!L 1 e JIIII 7:

@3

$f J IIll y

/

hL f l

• 0

/

/

/

l

,,,m

~

N V

/

/

l

"' a '.,,a )

g

/

hh

[

//////////////////////

nacs agri ru'L rutt i

i l

Figure 14 l

j MULTIPLE RACK MODEL l

FULL / FULL REGION I l

l 50 t

l

i s

s "u'*'h.W3 WhE""v" w

67

/r

/

/

25

,/

r 4.

/

/

/-

1

$?

/

/

$? U

/

/

$h

/

/

I~

gp

!I

/

/

L

$7 WI

/

e

/

l e.b.

I

/

/

L

@?

/

/

I

@?

/

/

O-

$7

,/

7

/

t

,/

,,,,.,...a, o u,,.-

/

w wa w

//////////////////////

L tt%!

Wa' I

Ll' 1

Figure 15 MULTIPLE RACK MODEL L

EMPTY / FULL REGION I l

51 1

l

4 e,-e--

%A26 -

.-s-6--

-*4 A.*-

-n_-uh--

---

.-m 4

4 a.a-a p

=-'-m---4--w

.-a 4

e

"" " N. M i:o# r. W o @ '"ua

~-

[

• h b

/-

/

1+2 se.

~

/

/

+

YN

(*h N

,/.

j l~ e e.

g x

m.

a.:

x

/

/

+

• Q

/

/

[~ 1?

62..

+

+.

4:

x

,/

/

4 A x

(e.'

/

D 5-[h E d g

/

llD @ j E d

/

/

Ik ey g-d

~

.a

/

/

E gg w

~

x

/

.,,. IF 83 h

/

w u

[

1,2f' -

%d*

[

/

w, ww e /.

//////////////u/// w /

-w FULL I

Figure 16 MULTIPLE RACK MODEL FULL / FULL REGION II l

l l

52 i

. ~. - -. - -. -... - - _ -. - - _

1 l

I Hr"o"v'rtw'Er50%6[itd'"U" e

u g

x N

@b

/

N

/

x N

/

[

X w;

x

.g

/

/

+-

@b x

~

l

/

+"

--tmC 9

w y

/

/

4.:

.e x

~

/

y-g g

/'

x x

j Ib

/

x

^..

@bdl

/

/

~

+-

x I

/

fh X

/

y G dll l

^,

/

k,,

[

/

ff///

/

/

EWPTV pagg g FULL l

Figure 17 l

MULTIPLE RACK MODEL f

EMPTY / FULL REGION II f

1 i

53

.~

L MULTIPLE RACK POOL POLEL (PLA1 VIEW)

Y I

l l

l 1

1 I

l l

.b.

.b.

I I

i l

'N' l

/ 's I

,e N

I y'

N

/

\\

\\

I Nl/

/

\\

l /

s,

.N

/

f N

A f

s

/

\\

/

N

/

N

/

s

/

M

\\

H; Ne/

h N

N RACK TO RACK RA0( TO RACK i

X-MRECTION X-DIRECTION I

H DI CT i.

X 4

o RACK 2 RACK 1 Y

Figure 18 MULTIPLE RACK MODEL (PLAN VIEW) i REGION II i

54

4 O

4

%M 0 75 -

~

0.30-J

'; j

'l I

(

ll g"C m')6pppgl,er) c nimy'i 7 q'

l 0

I 4

-0.30 -

- 0 75-O t0 b

$0 EO h

a k

g TIME (Seconds) i Figure 19 I

SONGS 2&3 SPENT FUEL POOL FIDOR ACCELERATION TIME HISTORY HORIZONTAL N-S DBE

l-L 1 00 eis-m E.

o.so -

i 0.25-l

}

i i

1, kN bl t I b lhklc E

'.1l1 d b lI I

l

.l l l1 1 I

l L

! U 4

N 1

I q

~

l g

( ') l$'

9] ll

];

l I

j 1

) [ l ji

> ! A t i

F l

r9 j'

i a

-o.2s -

l l

l U

j ui y

-o.so -

-ors-

-W

=,

i o

io 20 30 so so so yo ao i

TIME (Seconds) t Figure 20 SONCS 2&3 SPENT FUEL POOL FLOOR ACCELERATION TIME HISTORY HORI7.ONTAL E-W DBE l

1 m-m m

v w

e

. =. - - - - - - -

l 7

t f.00 05-1 E

l-o so-g o

o.n -

l j

E

.(1, l f I N i ah,[31 a ',

j h I

_I

?

H hylk6 L kht.lj diu d d

L i.

j l

o n,,-

_Tiyij.

e f r g ipg

D q p i

1 r

73 it pg, 'itq)"yr'yi-l q

g I

j i

g i

-o so-g

-on-f

-N' O

SO

.4 '

30 40 SO 60 70 90 TIME (Seconds) t i

Figure 21 i

SONGS 2&3 SPENT FUEL POOL FLOOR l

ACCELERATION TIME HISTORY VERTICAL DBE L

1

-y

5 i

g i

i l llll FLOCR RESPONSE SPECTRA

____ TIMEl# STORY ENVELOPE i

l f.

I m

I 2 3 9

p

<C E

~

ta

/

J

/

2 m

W

/

l' Q

/

.O

/

l

/

i i

/

1

/

l O'

1 l-FREQUENCY (CPS)10.

100.

4 i

Figure 22 I

SONGS 2 AND 3 FUEL BUILDING POOL FLOOR HORIZONTAL NS (NE368 C4)

FOR 4% DAMPING DBE SPECTRA l

1 l

+

4 5

I I

I I lill!

FLOOR RESPONSESPECTRA TIMEHISTORYENVELOPE

}.

4.,

i i

m U

2 3

1__

i 9

H i

4

~

E 121

, /

3 2

- f_

L i

Lil r

I E

O

]

l O

i

/

(

/

~

~

~

k e"

i i

/

I O

1 1-FREQUENCY (CPS)10.

100.

4 Figure 23 e

SONGS 2 AND 3 FUEL BUILDING POOL FLOOR HORIZONTAL EW-(NE368 C4) f FOR 4% DAMPING DBE SPECTRA l

i

{

[

~.

.i 5

g g

i i i i::

~

FLOOR RESPONSESPECTRA TNE HISTORY ENVELOPE r 1 i

4-t (3

{

e v

z 3-

=

/

t O

i F ~ -

1 4n' i

ta t

I

(

2 84 f

e Q

f O

4 s

~

l

~

y

~

~

i

/

1-e' N

/

[

i l

l O'

(

{

.1 1

FREQUENCY (CPS)10.

100.

l

'i t

1 Figure 24

~

i SONGS 2 AND 3 FUEL BUILDING POOL FLOOR VERTICAL (NE360 C4) i.

FOR 4% DAMPING DBE SPECTRA a

I

2D wN i

528.00 t

-.-> 4 _ 3.75

--p' 4-3.4 4 3.69 -- > 4-11.10 g j 135.90 --> D24.82 (3NN3.38 (2)

' ~

i 3r d

g 14 X 15 i

12 x 13 14 X 15 14 X 1;5 125.5-)

Rif 133.67 RN RM (2)'

t 276.00 7.00 kL h

l FF t

l 3.38 O

12 X 13 14 X 15 14 X 15 14 X 15 125.50 Rt Rif RN pg 0.00 I

.. 0

, r I

a k I

i a _ _ _ _

1 48.00 i

t i

i POCLOWE7599G POOLVWEL l

l Figure 25 l

PRELIMINARY SPENT FUEL STORAGE RACK ARRANGEMENT I

t

A %c 8

gG ya,l1 rah.r 's a

xw e

Figure 26 FINITE ELEMENT MODEL OF FUEL HANDLING BUILDING 62

r

(

j y

T, 1

e o

a 7,

b l

t l

i r

<<aa<

,,a>,,,

,3,,,

,,,,,, <>,,,,,'t,*f*,9

r oya r rf a

SRC Figure 27 LOADING ON SUPPORT PAD l

1

(

l I

63 ti

. -