Text

Rev 0 to Spent Fuel Pool Capacity Expansion Project Description & Sar
	ML20153F551
Person / Time
Site:	Millstone
Issue date:	05/31/1988
From:	; NORTHEAST NUCLEAR ENERGY CO.
To:	;
Shared Package
ML20153F505	List: B12844, Rev 0 to Spent Fuel Pool Capacity Expansion Project Description & Sar; ML20153F502;
References
	B12844, NUDOCS 8805100366
	Download: ML20153F551 (150)
	v • d • e

.~-

Docket No. 50-245 S12844 Attachment I Northeast Nuclear Energy Company Spent Fuci Pool Capacity Expansion Pro.iect Descriotion and Safety Analysis Report (Revision 0)

Si g f $ $$ p $5 P

TABLE OF CONTENTS

1.0 INTRODUCTION

1-1 2.0 GENERAL ARRANGEMENT 2-1 3.0 RACK FABRICATION 3-1 3.1 Fabrication Objective 1 3.2 Anatomy of the Rack Module 3-2 3.3 Rack Installation 3-5 3.4 Codes, Standards, and Practices for 3-6 the Spent Fuel Pool Modificati3n 4.0 CRITICALITY SAFETY ANALYSES 4-1 ,

4 .1 ' DESIGN BASES 4-1 1

4.2

SUMMARY

OF CRITICALITY SAFETY ANALYSES 4-4 4.2.1 Normal Operating Conditions 4-4 ;

4.2.2 Abnormal and Accident Conditions 7 4.3 REFERENCE FUEL STORAGE CELL 4-8 -

4.3.1 Fuel Assembly Design Specifications 4-8 4.3.2 Sturage Rack Cell Specifications 4-8 t 4.4 ANALYTICAL METHODOLOGY 4-11 4.5 CRITICALITY ANALYSES AND TOLERANCE VARIATIONS 4-13 ;

4.5.1 Nominal Design Case 4-13 4.5.2 Uncertainties Due to Manufacturing Tolerances 4-13 4.5.2.1 Boron Loading Variation 4-13 4.5.2.2 Boraflex Width Tolerance !

Variation 4-14 !

4.5.2.3 Storage Cell Lattice ;

Pitch Variation 4-14 4.5.2.4 Stainless Steel !

Thickness 4-14 4.5.2.5 Fuel Enrichment and Density Variation 4-15

,e ,-- , - - - - , , ~ , . - - . - . ,. - - - - , , , , - - , - - , . , - , , . , - . , , _ . - ,---,.c., , _ _ - - - -

4.5.2.6 Zirconium Flow Channel 4-15 4.5.3 Reactivity Effects of Boraflex 4-15 Axial Length 4.6 GADOLINIUM AND BURNUP CREDIT 4-17 4.7 ABNORMAL AND ACCIDENT CONDITIONS 4-20 4.7.1 Temperature and Water Density Effects 4-20 4.7.2 Abnormal Location of Fuel Assembly 4-20 4.7.3 Eccentric Fuel Assembly Positioning 4-21 4.7.4 Zirconium Fuel Channel Distortion 4-21 4.7.5 Dropped Fuel Assembly 4-21 4.7.6 Fuel Rack Lateral Movement 4-22 4.7.7 Lost or Missing Absorber Sheet 4-22

4.8 REFERENCES

TO SECTION 4 4-23 5.0 THERMAL-HYDRAULIC CONSIDERATIONS 5-1 5.1 DECAY HEAT CALCULATIONS FOR THE SPENT FUEL 5-2 5.1.1 Basis 5-2 5.1.2 Model Description 5-4 5.1.3 Bulk Pool Temperature Result 5-7 5.2 LOCAL POOL WATER TEMPERATURE 5-7 5.2.1 Basis 5-7 5.2.2 Model Description 5-8 5.3 CLADDING TEMPERATURE 5-10 5.4 TIME-TO-BOIL 5-11 5.5 SINGLE PUMP FAILURE SCENARIO 5-11

5.6 BLOCKED CELL SCENARIO 5-13

5.7 REFERENCES

TO SECTION FIVE 5-18 P

y _,e , , - - - - - -- - - - -+

i' 1.

i 6.0 STRUCTURAL ANALYSIS 6-1 !

6.1 ANALYSIS OUTLINE (FOR NEW PROPOSED RACK 6-1 ;

MODULES) 6.2 FUEL RACK - FUEL ASSEMBLY MODEL 3 i l

6.2.1 Outline of Model for Computer 6-4 ,

Code DYNARACK '

6.2.2 Model Description 6-7 6.2.3 Fluid Coupling 6-7 6.2.4 Damping 6-9 6.2.5 Impact 6-10 i 6.3 ASSEMBLY OF THE DYNAMIC MODEL 6-10 6.4 TIME INTEGRATION OF THE EQUATIONS OF MOTION 6-15 -

6.4.1 Time-History Analysis Using Multi- 6-15 :

Degree of Freedom Rack Model 6.4.2 Evaluation of Potential for Inter- 6-17 Rack Impact ,

6.5 STRUCTURAL ACCEPTANCE CRITERIA 6-17 i

6.6 MATERIAL PROPERTIES 6-19 ,

6.7 STRESS LIMITS FOR VARIOUS CONDITIONS 6-19 6.7.1 Normal and Upset conditions 6-19 6.7.2 Level D Service Limits 6-23 6.8 RESULTS FOR THE NEW SPENT FUEL RACKS 6-23 6.9 IMPACT ANALYSES 6-27 i

! 6.9.1 Impact Loading Between Fuel 6-27 Assembly and Cell Wall l 6.9.2 Impacts Between Adjacent Racks 6-27 6.10 WELD STRESSES 6.10.1 Baseplate to Rack Welds and 6-28 Cell to Cell Welds t 6.10.2 Heating of an Isolated Cell 6-29 l

6.11 RESULTS OF EXISTING RACK ANALYSIS 6-29

,.-.,...-..._,.,,,,,,.,...,~_--.~,,,-n, , . , - _ , . _ , , , . , . .- , . , , - - - . - . - - - , , ~ , -.

i 6.12 FLOOR SLAB AND BUILDING ANALYSIS - later - I 6.13 DEFINITIONS OF TERMS USED IN SECTION 6 .6-31 6.14 REFERENCES TO SECTION SIX 6-33 7.0- OTHER MECHANICAL LOADS 7-1 I

7.1 MECHANICAL LOADINGS 7-1 7.1.1 Fuel Handling 7-1 7.1.2 Dropped Fuel Accident I 7-1 7.1.3 Dropped Fuel Accident II 7-2 7.2 LOCAL BUCKLING OF FUEL CELL WALLS 7-2 7.3 ANALYSIS OF WELDED JOINTS IN RACK 7-3 7.4 CASK DROP ACCIDENT 7-4

7.5 REFERENCES

TO SECTION SEVEN 7-5 8.0 RADIOLOGICAL CONSIDERATIONS 8-1 8.1 OPERATING PERSONNEL EXPOSURE 8-1 f 8.2 ACCIDENT CONDITIONS 8-2 l 8.2.1 Shipping Cask Drop Accident 8-2 .

8.2.2 Dropped Fuel Accidents 8-2 i

8.3 REFERENCES

TO SECTION EIGHT 8-3 !

] 9.0 IN-SERVICE SURVEILLANCE PROGRAM FOR BORAFLEX 9-1 9.1 OVERVIEW 9-1

9.2 REFERENCES

TO SECTION HINE 9-3 10.0 COST / BENEFIT ASSESSMENT 10-1

10.1 INTRODUCTION

10-1 10.2 COST ASSESSMENT 11.1.1 Need for Increased Storage Capacity 10-1 11.1.2 Construction Costs 10-2 ;

10.3 ENVIRONMENTAL IMPACT 10-2

10.4 REFERENCES

TO SECTION TEN 10-4 APPENDIX A BENCHMARK CALCULATIONS A-1

)

l i

k 1-1 1.0 IllTRODUCTION Millstone Unit tio. 1 is a Boiling Water Reactor (BWR)

L nuclear power plant which was designed and constructed by the General Electric Company (GE). GE engaged Ebasco Services, Inc.,

New York, as their architect-engineer. Millstone Unit lio . 1 is owned and operated by the Northeast Nuclear Energy Company (UNECO) of Waterford, Connecticut. It is located five miles southwest of New London, Connecticut. The plant is rated at 2011 Mwt (thermal), and has a net electric output of approximately 652 l

megawatts. The plant went into commercial operation in 1970, and has undergone 11 refueling outages in the past seventeen years.

A total of 1732- fuel asserblies have been discharged in the Unit No. 1 spent fuel pool in that period (Table 1.1).

The present installed capacity for spent fuel storage in the Millstone Unit No. 1 spent fuel pool is 2184 locations. These storage locations are available in laterally restrained modules arranged in six groupings of the so-called "supermodules". As Table 1.1 shows, there are currently only 452 open (available) storage locations in the Millstone Unit No. I spent fuel pool, well below the reactor full core batch size of 580.

At the present time, the plant is able to maintain the full core reserve off-load capability by keeping a standby 152 cell rack module for irr.ediate emplacement in the cask laydown area.

The purpose of 11NECO's proposed spent fuel pool capacity i

)

1-2 !

expansion project is to eliminate the need for this temporary rack, and to extend the date of loss-of-full ' core discharge capability to approximtely the year 1999.

Specifically,1111ECO proposes to install new spent fuel racks to increase the number of storage cells in the Unit tio. 1 pool by 1045 locations to a total of 3229 cells. In addition, the

.reracked pool will contain 20 locations for storing failed fuel containers. Figure 1.1 shows the present and proposed pool storage capacities versus the anticipated fuel cycles.

The proposed racks are free standing and self-supporting.

The existing supermodule arrays which will all remain in the pool, also have been seismically requalified as free standing structures. The principal construction materials for the new racks are ASTM 240, Type 304 stainless steel for the structural members and shapes, and Boraflex, a patented product of BISCO (a division of Brand, Inc.) for neutron absorption. The support legs are of remotely adjustable jpe, consisting of an internally threaded, stationary member and a rotatable spindle. The latter

is made of SA564-630 precipitation hardened stainless steel to reduce tl.e probability of thread galling, and to build in a

relatively high bending and compressive strength capability in the foot pedestals.

lit!ECO has engaged CBI Services of Oak Brook, Illinois to

manufacture the new racks, and Holtec International of Mount I

Laurel,17ew Jersey, to design, analyse and qualify them for all I

'l l

t

1-3 loadings. Holtec International performed three-D time history, nuclear criticality safety, mechanical integrity, and thermal / hydraulic analyses on these racks. The results of Holtec's analyses and essentials of the rack module designs will be described in the following sections of this report. It will be shown that the new racks and requalified existing racks meet all relevant requirements, including those addressing the integrity of the rack structure under the specified combinations of inertial, seismic, and mechanical loads.

Also included in this report are descriptions of 1111ECO 's spent fuel rack in-service surveillance program, the contractors' quality assurance programs, and a comprehensive appraisal of the cost / benefit analysis for the proposed reracking.

~

1-4 ~

Table 1.1 Millstone Unit No. 1 Discharge Schedule Date of Discharce Number of Assembligg 9-72 28 9-74 '

208 10-75 143-11-75 1 10-76 124 3-78 124 5-79 148 10-80 168 9-82 192 5-84 200 11-85 200 7-87 jj.1 Total 1732 Number of open locations after 1987 outage = 2184 - 1732 = 452

1-5 6 o t i

50 4

O f<i! l A O C A C C

% C ,

C Ml o

3 8,l 2

a El O g

I M I w l a

_J l O 3

> l o l I

=

~

J l -

w I

~.

~

3 --

g__________ _____... _____.______ t =

1 *

- 5 l

o%l t C Cf2 e

~d Z > l z

l I P d< N

- - es>

Z o l c

p S I e.,

2 l

=

l 1

& Z m l 94 2 l

! Q, d Y9; I Q C l #@ A N

! - C l l E4 I k$d# I

~

l I i -

l l 5 5 5 5

Q

- R n s,36s%R .

5 gla 5 55C 5

'X2SSV#

2-1 2.0 GENERAL ARRANGEMENT The high density spent fuel storage racks consist of individual cells with 6.06" (nom.) inside square dimension, each of which accommodates a single GE (or equal) BWR fuel assembly.

The fuel assembly can be stored in the storage locations in channelled or unchannelled configuration. Table 2.1 gives the essential storage cell design data.

The rack modules currently in the pool are in four discrete sizes, labelled as modules A, B, C and D in Table 2.2. A total of 32 of such modules together provide 2184 storage locations.

The new modules proposed to be emplaced in the pool are in six distinct sizes, a total of ten in number, and cumulatively provide 1045 storage locations. Table 2.3 gives the cell count data for the new modules. Figure 2.1 shows the general arrangement of the modules in the Millstone Unit 1 pool.

The existing and new modules for the Millstone Unit no. 1 fuel pool are qualified as non-impacting free standing racks, i.e., each module is free standing and is shown to undergo minimal kinematic displacemente during the postulated seismic events. Thus, rack-to-rack or rack-to-wall impacts are precluded.

Figure 2.2 shows a pictorial rendering of a typical new Holtec/CBI rack module for the Millstone Unit No. 1 pool.

\-

2-2 Table 2.1 Design Data cell inside dimension (inch): 6.06 i .03 Cell pitch: 6.30 i .02 Min. B-10 loading (gms/sq.cm.)a ,

.01700

+ 0.5 Boraflex length (inch): 137 - 0 Table 2.2 Existing Module Data Number of Number of Cells Module Number of Cells Per E-W N-S Desianation Modulea Module Direction (indicated in Figure 2.1)

A 12 66 6 11 B '12 77 7 11 C 4 54 6 9 D 4 63 7 9 i

Total Cell Count = 2184 I

i i

I

2-3 Table 2.3

!!ew Module Data llumber of liumber of Cells Module 11 umber of Cells Per E-W 11 - S Desionation Modules Module Direction A 1 66 6 11 B 1 104 13 8 C 3 121 11 11 D 3 128 8 16 E 1 56* 8 7 F 1 72 9 8 Total Cell Count = 1045 Total 11 umber of Modules = 10 Plus 20 defective fuel cell containers.

6 < ,f1 ) l S

E L d U 1 D

O y on H -

D Ls~d IT .

8 L

O

=N _ " /

7 h h Y 3

m " 4' do- 4N 22, -

d'

~ II j

/ // _

/ 6 3

"6 I + /"2 2 1 D /

- x g 9 1 K A D 8 - -

2 4 0 l

/ 8 S E 1 A R 5

C A 1

- /

"6 f

l f 1 6

1 E7 u0 sl s) cl

. s 1

/

9 D x -

T

8 xl2i e 8 p( mc 0 5-l

l O

" ~

Y 6

1 3

l _ . , ~

~

= " A.

/

9 C xi t

r, _

_ A A* .

I 2 - " E

_ /2 L 9 l

6w ,*

,..l. .

. l 1

- D I

I

'* z . _

4 l O M

, - * /

C _. A* - 3 _

A -

3 D "6 _

E 1

/ 2 l

l _ *

- - - 9 2

S

" 9 C *I . .

I . -

,* 1 O 2 - _

- - - 4

/ P 0 I _

3 O

+ 6w C .

A 3

A

- 3 P

- R

= x _

l

.l m

" 1 "6

1 1 I

I ,

- e ,

= *2

/

9 C I

K _

. D _

3 g

- n -

n* E R

I 2 - - - I I

9

6* l.-

- ** - ' { .*

4

C I

, l I

,*I ,' l

/ 4 F

+ " * , il

, _ l 7

5 3_

6 D n -

n _ 3

/

1 8

B* 9

" 5 F x 2 1 l

/ 9 -

- * ,i : I1

,l

+

. ;l

=M 6w 5

), . _

w . i - n -

n l, I

8 -

n x - _

"6

/

3 2

1 lll/ 3 1

I l

'L ll n* : n

, 8 - D . t .

2 . I 9 3

1T, go - ._ _

w'Jf

> "6 -

.I

. )

l

.l y

6

[5 l i - I* _ A 4 A 1 x _

=_ +

?-

I

- 6 - .* _ '*

=

4 4 >-

/ p g

- U 3 h Y l

o. ~ 9$ =n 8,n~ hs a~

_ 2 N~ ] t_ ] f } =n .~~

I

,m28

. - - - -- . . __ _ . _ = _ _ . _ _ _ . _ _ _ _ _ ._ , .._ _ _

2-5 1~ TYPICAL DCUBLE PAggt i

TYPICAL - CONT AINING BORAFLEX

CILL CAVITY l Il lI lI l l

. + l l t

1lI I l

I l

l I

l l

lI llI I I

lI I I

l l

I l

l

. I I I I l l l .

l l

l

i A '

20LS/S

. ca.L w m i

! i i

i ;

i .

t, .. .

\ /

I \ /

i N / ' ADJUSTABLI N /

N / SUPPCRT LIGS i.

gg.E 2.2 TYPIC AL neb' Fl'EL R ACK MODEL i

I

3-1 3.0 RACK FADRICATIOli A11D Iti(TALLATIOtl 3.1 FABRICATIOli OBJECTIVE The central objective in manufacturing the high density storage racks for the Millstone Unit tio . 1 fuel pool may be stated in five interrelated points:

(1) The rack modules are fabricated in such a manner that there is Dn weld splatter inside the storage cell walls. Weld splatter on the lateral surface of the stcrage cell has been known to be a source of serious problems in several reracking projects.

(2) The storage locations are constructed in such a way that redundant flow paths for the coolant are available in case the main designated path is blocked.

(3) The fabrication process permits immediate and convenient verification by the inspection staff to ensure that the "poison" panels are correctly installed.

(4) The storage cells are connected to each other by autogenously produced oorner welds which lead to a honeycomb lattice construction. The extent of welding is selected to "detune" the racks from the ground motion (OBE and SSE).

i

3-2 (5) The poison material provides full surface lateral support; yet is not constrained from radiation induced dimensional changes.

3.2 ANATOt1Y OF THE RACK MODULE A complete description of the rack geometry is best presented by first introducing its constituent parts. The important parts can be denoted as (1) the storage cell box subassembly, (2) the base plates, (3) the neutron absorber material, (4) bottom and side spacer strips; (5) top strips, and (6) support legs.

Each part is briefly described below with the aid of sketches.

(1) Otorage cell box subassembly: The so-called "boxes" are fabricated from two precision tormed channels by seam welding them together in a seam welding machine equipped with copper chill bars and pnoumatic clamps to minimize distortion due to welding heat input. Figure 3.1 shows the "box".

The minimum weld penetration shall be 80% of the box metal gage which is 0.0751" (14 gage). The boxes are manufactured to 6.06 inch I.D. (inside dimension) i .03 inch.

3-3 Each box constitutes a storage location. As shown in Figure 3.2, each box has two lateral holes punched near its bottom edge to provide auxiliary coolant flow holes. Bottom, lateral and top spacer strips are attached to two walls of each box in the manner shown in Figure 3.2. The "poison" sheet is placed in the picture frame space on two side panels of the box.

An assemblage of box assemblies is welded edge to edge as shown in Figure 3.3. The numerical sequence of fabrication is also indicated in Figure 3.3. This method of fabrication results in a honeycomb structure with axial, flexural and torsional rigidity depending on the extent of intercell welding provided. It can be noted from Figure 3.3 that two edges of each interior box are connected to the contiguous boxes resulting in a well defined path for "shear flow".

(2) Base Plate: As shown in Figure 3.4 and Figure 3.5, the base plate provides a continuous horizontal surface for supporting boxes and the fuel arsemblies. The base plate has a concentric hole in each cell location. The seating surface of the fuel assembly, as shown in Figure 3.4, has a 450 countersink so as to provide a conformal contact surface for the "nose" of the BWR fuel assembly.

3-4 The base plate is attached to the box assemblage by fillet welds made by reaching in through the base plate holes using a "goose neck" welding head.

(3) The neutron absorber material: Boraflex is utilized as the neutron absorber material. Boraflex has been widely used in reracking projects in the U.S. and overseas. A partial list of spent fuel pocls in which Boraflex has been utilized is given in Table 3.1.

(A) Bottom and side spacer strips: These strips are shown in Figure 3.2 as a part of the box assembly.

This ensures unrestrained expansion or contraction and positive positioning of the poison material.

(4) The top strip is used as the filler strip to weld adjacent sides of the boxes to form a smooth "lead-in".

The corners of the boxes, ate, however, not welded to each other. Thus a natural vent for the interbox space at every cell corner is formed.

(5) Support Legs: As stated in Section 1.0, all support legs are of the adjustable type (Figure 3.5). The top (female) position is made of austenitic steel material.

The bottom part is made of SA564-630 precipitation hardened stainless steel to avoid galling problems.

i

3-5 The support leg is equipped with a hexagonal socket to enable remote leveling of the rack after.its placement in the pool. Lateral holes in the support provide the requisite coolant flow path.

3.3 RACK INSTALLATION Northeast Utilities'., CBI's and Holtec International's Quality Assurance Programs ensure that design, fabrication and installation activities conform to acceptable quality requirements throughout all areas of performance. These quality requirements are delineated in 10CFR50, Appendix B, Northeast Utilities Topical QA Report, Revision 10, CBI Nuclear QA Manual, Rev. 3, and Holtec QA Manual, Rev. 3.

The rack installation will begin with the removal of the existing control rod racks and lateral seismic support steel.

This steel will be shipped offsite to be decontaminated and salvaged. Once the support steel is removed the existing racks will be lifted and moved toward the Southwest corner of the pool.

Once the existing racks are in place the new racks will be located in the pool as shown in Figure 1.1. At no time will a rack Le lifted while it contains irradiated fuel, nor will a rack or any other heavy load be lifted over irradiated fuel. Fuel movements will be required to meet these requirements. These movements will also be used to ensure that personnel exposure is rainimized. In addition, remote tooling will be used in the place of divers whenever practical. The fuel movements and all'other l

l l

3-6 installation activities will be performed in accordance with Plant Operations ,

Review Committee approved QA Category 1 procedures.

3.4 CODES, STAllDARDS, A11D PRACTICES FOR THE SPEllT FUEL POOL MODIFICATIOli The fabrication of the 2.ack modules is performed under a strict quality assurance system suitable for ASME Section III, Class 1, 2 and 3 manufacturing which has been in place at CBI for over 25 years.

The following codes, standards and practices were used as applicable for the design, construction, and assembly of the spent fuel storage racks and analysis of the pool structure.

Additional specific references related to detailed analyses are given in each secticn.

a. Desian Cc> des (1) AISC Manual of Steel Construction, 8th Edition, 1980 (provides detailed structural criteria for linear type supports).

l (2) AllSI N210-1976, "Design Objectives for Light Water Reactor Spent Fuel Storage Facilities at liuclear l Power Stations" (contains guidelines for fuel rack l design).

(3) American Society of Mechanical Engineers (ASME),

t Boiler and Pressure Vessel Code,Section III, 1983 1

Edition up to and including Summer 1983 Addenda (Subsection liF) (governing structural design code).

l l

[

l i

l 3-7 (4) ASNT-TC-1A June, 1980 American Society for Mondestructive Testing (Recommended Practice for Personnel Qualifications).

b. Material Codes (1) American Society for Testing and Materials (ASTM)

Standards - A-240.

(2) American Society of Mechanical Engineers (ASME),

Boiler and Pressure Vessel Code, Section II-Parts A and C, 1983 Edition, up to and including Summer 1983 Addenda.

c. Weldina Codes ASME Boiler and Pressure Vessel Code, Section ' IX-Welding and Brazing Qualifications, 1983 Edition up to and including Summer, 1983 Addenda.

d. Ouality Assurance, Cleaoliness, Packacina, Shinoina, Receivina, Storace, and Handlina Requirements (1) ANSI N45.2.2 -

Packaging, Shipping, Receiving, Storage and Handling of Items for Nuclear Power Plants.

(2) ANSI 45.2.1 -

Cleaning of Fluid Systems and Associated Ccmponents during Construction Phase of Nuclear Power Plants.

(3) ASME Boiler and Pressure Vessel, Section V, Nondestructive Examination, 1983 Edition, including Summer and Winter 1983.

l (4) ANSI -

H16.1-75 Nuclear Criticality Safety

! Operations with Fissionable Materials Outside Reactors.

I (5) ANSI -

N16.9-75 Validation of Calculation Methods for duelear Criticality Safety.

l l

l

3-8

~(6) ANSI - N45.2.11, 1974 Quality- Assurance Requirements for the Design of Nuclear Power Plants,

e. Other References (1) NRC Regulatory Guides, Division 1, Regulatory Guides 1.13, Rev. 2 (proposed); 1.29, Rev. 3; 1.31, Rev. 3; 1.61, Rev. 0; 1.71, Rev. 0; 1.85, Rev. 22; 1.92, Rev. 1; 1.124, Rev. 1; and 3.41, Rev. 1.

(2) General Design Criteria for Nuclear Power Plants, Code of Federal Regulations, Title 10, Par'c 50, Appendix A (GDC Nos. 1, 2 , 61, 62, and 63).

(3) NUREG-0800, Standard Review Plan (1981).

(4) "OT Position for Review and Acceptance of Spent Fuel Storage and Handling Applications," dated April 14, 1978, and the modifications to this

! document of January 18, 1979.

3-9 Table 3.1 BORAFLEX EXPERIEllCE FOR HIGH DEllSITY RACKS Unit Plant NRC Licensing Site No. Type Docket No. Status Point Beach 1& 2 .PWR 50-226 & 301 Licensed liine Mile Point 1 BWR 50-220 Licensed Oconee 1& 2 PWR 50-269 & 270 Licensed Prairie Island 1& 2 PWR 50-282 & 306 Licensed Calvert Cliffs 2 PWR 50-318 Licensed Quad Cities 1& 2 BWR 50-254 & 265 Licensed Watts Bar 1&2 PWR 50-390 & 391 Pending Waterford 3 PWR 50-382 Pending Fermi 2 BWR 50-341 Licensed H.B. Robinson 2 PWR 50-261 Licensed River Bend 1 BWR 50-458 Licensed Rancho seco 1 PWR 50-312 Licensed liine Mile Point 2 BWR 50-410 Licensed Shearon Harris 1 PWR 50-400 Licensed Millstone 3 PWR 50-423 Licensed Grand Gulf 1 BWR 50-416 Pending Oyster Creek 1 BWR 50-219 Licensed

P -

P 3-10 Table 3.1.(continued)- ,

BORAFLEX EXPERIENCE FOR HIGH DENSITY RACKS

-Unit Plant NRC Licensing Site' No. Type Docket No. Status 50-395 Licensed V.C. Summer 1 PWR i Diablo Canyon 1&2 PWR 50-275 & 323 Licensed 50-335 St. Lucie 1 PWR In Licensing Byron 1 PWR 50-454 In Licensing Byron 2 PWR 50-455 In Licensing ,

Vogtle 1 PWR 50-424 License to be applied for -l I

s b

F l

2, . _ _ _ __ _ _ . , _ _ . _ . _ . _ _ _ . , _ , _ _ . _ . . , _ _ . . _ , _ .._,,..._ _ -._. _ ..._..._ .._ .,_ .._ ,_. ,_., _ . _ _ ,,, _ _ . . . - . . . ., . . , . _ _

0 3-11 s4 9

3 D

4 4 4 6

6 e

i 4

e

/ ;

3.[0 s 4

r,,,..

%i MELD O

e e

e O

FIGURE 3' gg g VELD 111G

. ,- - - - _ _ - - , - .__y. . . , - . . ..w, .._,.--_,__,_,._--,_,,,_,_-._,_._..._____,._____,___,,,___.,7_ - _ , . _ , _ . . _ , _ , _ . _ _ _ _ _ _ _ _ _ . , . _ _ . _ , . , , _ _ _ _ , . . , _ . . ,

s-n o ]

TOP STRIP f d

'\ ;

/

' i

- s TOP POISON y ,

STOP-STRIP \.. /

%,# c 4

7 A

4 LATERAL STRIP N, y n

?

3 N../

U I'

/ . ,

l v

~4

\- #

\./ AUXILIARY FLOW HOLE BOTTOM STRIP FIGURE 3.2 BOX WITH BOTTOM, SIDE AND TOP STRIP SHOWN VELDED IN-PLACE.

- - ~ - - - . , - - - , - - - --- . . . -

3-13 l OUTER COVER SHEET SIDE STRIP POISON SHEET 1

.s, 1 1 1 : x 1 1 n i Mb i 1 1 1 1 1 '%

= :

' a m n j m

( -

4

-t-

.;. + <

m -

M 4 8 y 4 6 ,

M 4 a n -

",L' -C *

, . , , , . 1, .L1, .

r' -

r-M M N g

W M '

M g

{_. -

, i M M

- ~

3 - 1 "

2 r .

W M .

H <

" pT ,,,,-

, 1 1 1 I ,- 1 1 =1 1 M 'F: -

_i[

s r '_r, _

M * "

t = , .

+ +

w a

+ 2 7 5 - 9 -

O I t

. t i 2n T

4 , y -

T I T J T I Y Y T Y I 1 -

1 g

6.32"

. 6,06" SQ. INSIDE CELL NOM FIGURE 3.3 INTER-CELL WELDINGPLAN

3--14

. 6.30" .

PITCH v- i I i

~'

~

g" l

i a 1 - I : l

!c [

%d %#

i

?yE i .

%s,# .

n I i l .

,i t % A -

l \

.- 6.06" '7 7 SECTION L 4.

~ SQ. TYP.' 5" LG

/

/ \ 4 SIDES r-'] , 1/16 "

h 2

il ! . ..ll. , p"

FUEL ASSEMBLY

? ? '

- ~

h : : -

[ l . , [ POISON

= 5 -

-/

n us -

n- - -

- C

c. .

l i f : * : i f ,, ,,

= _ _

e _ _

e ~7 _

L. _ _

t l

- ~_ ,

i 3/4" DIA. AUIILIARY FLOW h0LE

,f e-

~1, /-

e g ,., -. e 3 7, [.

e .. n .

- a

)'

't 4 lp \ -

s A w 1

~r 1 l .1,- l }

I BASEPLATE 1

FIGURE 3.4 ELEVATION 1 . _ _ -- _ - . __ . __ _ _. _ . _ _

3-15

- CELL

. / #

- ll' BASbPLATE l

.I 40 r i

. I L 3

3-i i s

a d Nf , ff

= / k t\ I /

)l 4

s 4 // fl ',

' 'i i !

s' : V

//

g /

v l l, ;' l\ i ---- J- -

/ A

, @ R g i n i r i'

T = ~.

e Y e e m

i

,;f[g, " __

I -4 i O M ' '#

N V -Y = k e ' ' 2 4-1/4" 4UNC CLASS 1A

. i 3/4" DIA. H0LE THR

6" 00 E s

FIGURE 3.5 ADJU 3 TABLE SUPPORT 1

1

~

4-1 4.0 CRITICALITY SAFETY ANALYSES 4.1 DESIGN BASES

-The high density spent fuel storage racks for Millstone Unit No. 1 are designed to assure that the neutron multiplication factor (keff) is equal to or less than 0.90 with the racks fully loaded with fuel of the highest anticipated reactivity and the pool flooded with unborated water at a temperature corresponding to the highest reactivity. The maximum calculated reactivity includes a margin for uncertainty in reactivity _ calculations and in mechanical tolerances, statistically combined, such that the true keff will-be equal to or less than 0.90 with a 95% probability at a 95% confidence level. Reactivity effects of abnormal and accident . conditions have also been evaluated to assure that under credible abnormal conditions, the reactivity will be less than the design basis limit value.

The design basis fuel is the General Electric 8x8R assembly with an infinite multiplication factor of 1.35 in the normal uncontrolled core geometry (6-inch lattice spacing) using nominal dimensions at an ambient temperature of 20 0C and no moderator j voids. This infinite multiplication factor corresponds to a uniform enrichment of 3.05 wt% U-235 without gadolinium burnable poison and without burnup. Once the rack design was completed, calculations were made for higher enrichments to determine the

l. fuel burnup required or the minimum gadolinium loading necessary to assure that the keff will be less than the design basis l reactivity (0.90).

l i

4-2 Applicable codes, standards, and regulations, or pertinent sections thereof, include the following:

O General Design Criterion 62, Prevention of Criticality in Fuel Storage and Handling.

O USNRC Standard Review Plan, llUREG-0800, Section 9.1.1, New Fuel Storage, and Section 9.1.2, Spent Fuel Storage.

O USIIRC letter of April 14, 1978, to all Power Reactor Licensees -

OT Position for Review and Acceptance of Spent Fuel Storage and Handling Applications, including modification letter dated January 18, 1979.

O USNRC Regulatory Guide 1.13, Spent Fnel Storage Facility Design Basis, Rev. 2 (proposed), December, 1981.

O U S il R C Regulatory Guide 3.41, Validation of Calculational Methods for Nuclear Criticality Safety (and related ANSI N16.9-1975). .

O ANSI /ANS-57.2-1983, Design Requirements for Light Water Reector Spent Fuel Storage Facilities at Nuclear Power Plants.

O ANSI N210-19',a, Design Objectives for Light Water Reactor Spent Fuel Storage Facilities at Nuclear Power Plants.

O ANS-8.17-1984, Criticality Safety criteria for the Handling, Storage and Transportation of LWR Fuel Outside Reactors.

To assure the true reactivity will always be less than the calculated reactivity, the following conservative assumptions were made:

.,n- , - + , . , , - , -e s,wg .----.-m-.2,-,cy .,m , ..-------,--.--r,

4-3 O Moderator is pure, unborated water at a temperature within the design basis range corresponding to the highest reactivity.

O Lattice of storage racks is assumed infinite in all directions, i.e., no credit is taken for axial or ranial neutron leakage (except in the assessment of certain abnormal / accident conditions). Criticality safety analyses are based upon km.

O Neutron absorption in minor structural members is neglected, i.e., spa'cer grids are replaced by water.

l l

4-4 4.2

SUMMARY

OF CRITICALITY SAFETY ANALYSES 4.2.1 Normal Operatino Conditions The basic calculations supporting the criticality safety of the Millstone Unit No. 1 fuel storage racks are summarized in Table 4.1. Based upon the design basis b of 1.35*

(b is calculated for an infinite array) in the standard core geometry, the maximum b in the storage rack is 0.895 (95%

probability at the 95% confidence level) including all known uncertainties. Thus, the fuel storage rack satisfies the design basis requirement of a maximum keff less than or equal to 0,90.

Calculations were also made to define the minimum burnup required (ragardless of gadolinium content) and the minimum gadolinium required (regardless of burnup), either of which would permit higher enrichment fuel to be. stored in the racks. These calculations are summarized in Figures 4.1 and 4.2 for initial enrichments up to 3.8 wt% U-235. The data in either of these figures may be used as a criterion for acceptable storage in the fuel racks with assurance that the maximum b will be less than

~0.90. Figure 1 for the minimum Gd 023 loading assumes a minr im of three (3) fuel rods containing the gadolinia.

l The calculated b was 1.3499 for the GE 8x8R fuel assembly with 3.05% uniform enrichment in the standard core geometry at 20 0C, corresponding to 14.78 grams U-235 per axial centimeter in the enriched zone.

t i

l i

4-5 Thus, there are three independent criteria, any one of which may be used to determine the acceptability of a fuel assembly for storage in the Millstone Unit No. 1 spent fuel storage racks (i.e., whether storage is within the design basis limit of a km less than 0.90). These criteria are:

O A uniform (average). enrichment equal to or less than 3.05% (km in the standard core geometry of 1.35),

or O A gadolinia loading (in a minimum of three fuel rods) within the acceptable domain of Figure 4.1, or O An average fuel burnup within the acceptable domain of

, Figure 4.2.

It may be noted that even if fuel of 3.3% enrichment without any gadolinia were to be loaded into the rack, the km would not exceed 0.95 with all known uncertainties included.

l l

l

4-6 Table 4.1 e

SUMMARY

OF CRITICALITY SAFETY ANALYSES Temperature assumed for analysis 20 0 C Reference km (CASMO-2E)(1) 0.8833(1)

Calculational bias 0.0013 Uncertainties Bias 10.0018 B-10 concentration 0.0044 Boraflex thickness i0.0054 Boraflex width i0.0013 Lattice spacing i0.0015 Flow Channel Bulge 0.0049 SS thickness 10.0004 Fuel enrichment 0.0040 1

Fuel density iO.0030

! Statistical combination (2) 10.0105 Eccentric assembly position Negative l

Total 0.8846 1 0.0105 Maximum reactivity 0.895 l

(1) NITAWL-KENO gave a bias-corrected kw of 0.881 1 0.006 (95%/95%).

L l

(2) Square root of sum of squares.

4-7 4.2.2 Abnormal and Accident Conditions

' lione of the credible abnormal or accident conditions that have been identified will result in exceeding'the limiting reactivity (keff of 0.90). The effects on reactivity of credible abnormal and accident conditions are summarized in Table 4.2 below.

Table 4.2 REACTIVITY EFFECTS OF ABllORMAL A11D ACCIDE11T C011DITIO11S Accident / Abnormal Conditions Reactivity Effect Temperature increase llegative Void (boiling) llegative Assembly dropped on top of rack llegligible Lateral rack module movement liegligible Misplacement of a fuel assembly 11egligible or llegative Lost / missing Boraflex sheet + 0.0027 l

4-8 4.3 REFERENCE FUEL STORAGE CELL 4.3.1 Fuel Assembly Desian Snecifications The design-basis fuel assembly is a standard 8x8R array of BWR fuel ~ rods containing UO2 clad in Zircaloy. For the nominal design case, fuel of uniform 3.05 wt% U-235 enrichment was assumed, corresponding to 14.78 grams U-235 per axial centimeter of fuel assembly. Design parameters are summarized in Table 4.3 In the standard core geometry (6.00-in. assembly pitch at- 20 0C), this fuel assembly has a calculated infini.te multiplication factor of 1.35 gadolinium burnable poison. Other fuel assembly configurations analyzed are shown in Table 4.3.

The 8x8EB fuel assembly exhibits the same reactivity as- the reference 8x8R assembly and the remaining assemblies all have lower reactivities.

4.3.2 Storace Rack Cell Specifications The design basis storage rack cell consists of an egg-critu n teture, Lllustrated in Figure 4.3, with fixed neutron abscrb- a. Aterial (Boraflex) of 0.0206 g/cm 2 boron-10 areal density (0.017 g B-10/cm2 minimum) positioned between the fuel assembly storage cells in a 0.076 inch gap. This arrangement provides a nominal center-to-center lattice spacing of 6.32 in.

Manufacturing tolerances, used in evaluating uncertainties in reactivity, are indicated in Figure 4.3. The 0.075-in.

stainless-steel box which defines the fuel assembly storage cell has a nominal inside dimension of 6.095 in. This allows adequate clearance for inserting / removing the fuel assemblies, with or without the Zircaloy flow liner.

4_9 ,

Table 4.3

' FUEL ASSEMBLY DESIGN SPECIFICATIONS Fuel Assembly Designation 7x7/

7x7R 8x8 j FUEL ROD DATA Outside diameter, in. 0.570/0.563 0.493 '

Cladding thickness, in. 0.032/0.037 0.034 Cladding material Zr-2 Zr-2 Pellet density, ga UO2/cc 10.41 10.41 (95% theoretical density)

Pellet diameter, in. 0.488/0.477 0.416 Maximum enrichments, w/o U235 2.60 2.74 (Assembly average)

Active enriched fuel length, in. 144.0 144.0 Distance from s.) of nosepiece to 7.39 7.39 ;

where enriched fuel zone '

begins, in.

WATER ROD DATA Outside diameter, in. -

0.493 Wall thickness, in. - 0.034 Material -

Zr-2 Number por assembly None 1 FUEL ASSEMBLY DATA Number of fuel rods 49 63 Fuel rod pitch, in. 0.738 0.640 Fuel channel outside dimension, in. 5.438 5.438 Fuel channel wall thickness, in. 0.080 0.080 Fuel channel material Zr-4 Zr-4 i

i

4-10 Fuel Assembly Designation 8x8R/ BP8x8R/

PBxBR GE8x8EB FUEL ROD DATA outside diameter, in. 0.483 0.483 Cladding thickness, in. 0.032 0.035 Cladding material Zr-2 Zr-2 Pellet density, ga UO2/cc 10.41 10.41 (95% theoretical density)

Pellet diameter, in. 0.410 0.410/0.411 Maximum enrichments, w/o U235 3.00/3.80 3.00/3.80 (Assembly average)

Active enriched fuel length, in. 133.24 133.24 Distance from end of noseplece to 13.39 13.39 where enriched fuel zone begins, in.

WATER ROD DATA Outside diameter, in. 0.591 0.591 Wall thickness, in. 0.034 0.030 Material Zr-2 Zr-2 Humber per assembly 2 2 FUEL ASSEMBLY DATA Number of fuel rods 62 62 Fuel rod pitch, in. 0.640 0.640 Fuel channel outside 5.438 5.438 dimension, in.

Fuel channel wall thickness, in. 0.080 0.080 Fuel channel material Zr-4 Zr-4 i.

4-11 4.4 AllALYTICAL METHODOLOGY In the fuel rack analyses, criticality analyses of the high density spent fuel storage racks were performed with the AMPX-KEllO computer package (Refs. 4-1 and 4-2), using the 27-group SCALE

cross-section library (Ref. 4-3) with the 11ITAWL subroutine for U-238 resonance shielding effects (11ordheim integral treatment). Benchmark calculations are presented in Appendix A and indicate a bias of 0.0106 1 0.0048 (95%/95%). In the geometric model used in KE110, each fuel rod and its cladding were described explicitly and reflecting boundary conditions (zero neutron current) were used in the axial direction and at the centerline of the water-gap between storage cells. These boundary conditions have the effect of creating an infinite array of storage cells in all directions.

The CASMO-2E computer code (Refs. 4-4, 4-5 and 4-6), a two-dimensional multigroup transport theory code for fuel assemblies, has also been benchmarked (Appendix A) and was used as the primary method of analysis as well as a means of evaluating small reactivity increments associated with manufacturing tolerances. CASMO-2E benchmarking resulted in a SCALE is an acronym for .S_tandardized Computer Analysis for Licensing Evaluation, a standard cross-section set developed by OR11L for the US11RC.

e ,,-- - - , . - - - - - , - .

4-12 calculational bias of 0.0013 ! 0.0018 (95%/95%). Two group diffusion theory constants were edited in the output of CASMO-2E and used in a one dimensional diffusion theory routine to evaluate reactivity effects of the Boraflex axial length.

A third independent method of criticality analysis, utilizing diffusion / blackness theory, was also used for additional confidence in results of the primary calculational methods, although no reliance for criticality safety is placed on the reactivity value from the diffusion / blackness theory technique. This technique, however, is used for auxiliary calculations of the small incremental reactivity effect of eccentric fuel positioning that would otherwise be lost in normal KENO statistical variations, or would be inconsistent with CASMO-2E geometry limitations. Cross sections for the diffusion /

blackness theory calculations were derived from the NULIF '

computer code (Ref. 4-7, supplemented by a blackness theory ,

routine that effectively imposes a transport theory boundary condition at the surface of the Boraflex neutron absorber.

Shielded cross-sections were then used in the spatial diffusion theory code, PDQ07 (Ref. 4-8), in two dimensions, to calculate reactivities.

Comparison of the three independent methods of analysis for the reference design resulted in the following data which confirms the validity of the analytical methodology.

t J

4

. ._-_,-.-____-__.._.m_. . . , . - . - , , ,,..,,_.,_..---,-_._..r._,- ,..,.,. - -.. .. --

'4-13 Maximum km Analvtical Method Bias-corrected km (95%/95%)

AMPX-KENO 0.8809 i 0.0063 0.887 CASMO-2E 0.8846 i 0.0018 0.886 Diffusion-blackness 0.8802 0.880 theory 4.5 CRITICALITY ANALYSES AND TOLERANCE VARIATIONS 4.5.1 Nominal Desian Case For the design basis reactivity of a fuel assembly .

(1.35 - in the standard core geometry corresponding to 3.05% l uniform enrichment), the storage cell infinite multiplication i factor, km, is 0.8846 (bias corrected CASMO). With a Ak of -

0.0105 for all known uncertainties statistically combined, the maximum km is 0.895 which is slightly less than the design basis limit of a 0.90 for km. The GE 8x8R and the GE 8x8EB show ,

essentially the same reactivity in the storage cell while the 7x7 and earlier 8x8 designs are of lower enrichment and hence lower reactivity in the central enriched zone of the fuel assembly.

Reactivity effects of the axial Boraflex length and the natural uranium blankets are considered in Section 4.5.3. below, and confirm that the GE 8x8R (or 8x8EB) fuel assemblies exhibit the L higher reactivity (km) and were therefore used as the design basis fuel assembly.

i I

I

~- _

4-14 4.5.2 Uncertainties Due to Manufacturina Tolerances 4.5.2.1 Boron Loadina Variation The Boraflex absorber sheets used in the storage cells are nominally 0.065-inch thick, with a B-10 areal density of 0.0206 g/cm2 . Independent' manufacturing tolerance limits are i 0.007 inch in thickness and i0.009 g/cm3 in B-10 content. This assures that at any point where the minimum boron concentration -

(0.0191 gram B-10/cm 3) and minimum Boraflex thickness (0.038 inch) may coincide, the boron-10 areal density will not be less than 0.017 gram /cm2. Differential CASMO-2E calculations indicats that these tolerance limits result in an incremental reactivity uncertainty of 10.0044 Ak for boron content and 10.0054 for Boraflex thickness variations.

4.5.2.2 Boraflex Width Tolerance variatie,q The reference storage cell design uses a Boraflex width of 7.75 0.063 inches. A positive increment in reactivity '

occurs for a decrease in Boraflex absorber width. For a reduction in vidth of the maximum tolerance, 0.063 inch, the calculated positive reactivity increment is +0.0013 Ak. !

l l

4.5.2.3 Storace Cell Lattice Pitch Variation

[ The design storage cell lattice spacing between fuel l

assemblies is 6.32 in. Increasing the lattice pitch reduces ,

reactivity. For the manufacturing tolerance of iO.02 in., the ;

corresponding maximum uncertainty in reactivity is i 0.0015 Ak as determined by din.erential CASMO calculations.

l i

l

[

t h

! t L .. _ ~

y 4 -- .

4-15 L 4.5.2.4 Stainless Steel Thickiness Tolerances The nominal stainless steel box thickness is 0.075 i 0.005 inch. The maximum positive renetivity effect of the expected stainless steel thickness tolerance variation was calculated (CASMO-2E) to he i .0004 Ak.

4.5.2.5 Fuel Enrichmqnt and Density Variation The nominal design enrichment is 3.05 wt% U-235. CASMO calculations of the sensitivity to small enrichment variations yielded an average coefficient of 0.0080Ak per 0.1 wt% U-235, in the enrichment range from 3.0 to 3.1%. For an estimated to:erance on percent U-235 enrichment of to.05, the maximum uncertainty is 1 0.0040Ak due to the tolerance on fuel j enrichment. I Calculations were also made to datt rmine the sensitivity to tolerances in 002 fuel density. These calculations indicate that the storage rack km is increased by 0.0030Ak for the expected tolerance (to 97% T.D.) in fuel density. A lowar fuel density results in correspondingly lower values of reactivity. Thus, the maximum uncertainty due to the tolerances et U02 density is 10.0030Ak.

4.5.2.6 Eirconium Flow Channel Elininatien of the zirconium flow channel results in a small decrease in reactivity. More significant is a small ;

positive reactivity effect resulting from potential bulging of [

the zirconium channelc which moves the channel wall outward ,

4-16 toward the Boraflex absorber. For the maximum expected bulging (te 5.93 in, outside dimension) uniformly throughout the assembly, an incremental reactivity of +0.0049Ak could result (differential CASMO calculations). Since actual bulging of the flow channel would not be the maximum everywhere in all assemblies, the positive reactivity effect has been statistically combined with the reactivity effect of other mechanical deviations.

Fuel assembly bowing results in a negative reactivity effect and is treated as an abnormal condition (Section 4.7.4 below).

4.5.3 Reactivity Effects of Boraflex, Axial Lenoth Based upon diffusion theory constants edited in the CASMO-2E output (reference "2esign and a special case with water replacing the Boraflex), one-dimensional axial calculations were made to evaluate the reactivity effect of reduced Boraflex axial lengths and the 6-inch natural U02 blanket at both ends of the reference 8x8R fuel assembly. These calculations used a thick (30 cm.) water reflector and neglected the higher absorption of the stainless steel structural material above and below the active fuel. The cases evaluated included the older zuel with a 144 inch cntiched zone (7x7 fuel of 2.60% uniform enrichment and 8x8 fuel of 2.74% uniform enrichment) and the newer 8x8R or 8x8EB fuel with a 133-inch enriched zone (3.05% enrichment) and 6-inch natural uranium axial blankets top and bottom. With fresh unburned fuel, and without credit for the distribution in enrichments or gadolinium burnable poison actually present, an axial length of 133 inches was found to be acceptable, i.e., the keff with axial leakage included was less than the reference i

4-17 design km. With a 133 inch Boraflex length, the calculated keff values were as follows:

7x7 jyu.8 8x3EB Reference..km 0.978 0.882 0.880 0.683 In practice, the fuel burnup and/or gadolinium burnable poison will provide a large margin in reactivity below the design basis limit. In addition, an additional 4 inches (137 inch axial length) of Boraflex (3%) is provided as an allewance for possible y radiation-induced shrinkage.

4.6 GADOLINIUM AND BURNUP CREDIT Credit for fuel burnup and the presence of gadolinium burnable poison is necessary to justify the criticality safety of fuel with initial enrichments greater than 3.05%. For enrichments up to 3.8% U-235, a parametric fuel-burnup study was made involving three to seven fuel rods containing various concentrations of gadolinia burnable poison. From these calculations, it was possible to define the minimum concentration of gadolinia in the smallest number of rods (three) that would allow the racks to safel.y accommodate the fuel assemblies at their highest burnup-dependent reactivity.

Figure 4.4 illustrates the burnup-dependent reactivities for the gadclinia loadings identified as the minimum necessary for acceptable storage of fuel af the indicated enrichments. For each enrichment, the highest reactivity (a,0.87 km) is at least 0.0 Mk less than the reference design assembly km l

-_ _ J

4-18 (0.883) which provides an adequate margin for uncertainties in the burnup calculations and assures that the maximum b with all uncertainties added remains less than the 0.90 design basis limit. The Gd 023 data in Figure 4.4 is that summarized previously in Figure 4.1 showing the minimum Gd 02 3 concentration in at least three fuel rods. Increasing the number of rods ,

containing gadolinium (at any given percentage) reduces the reactivity at zero burnup but.has only a relatively minor effect on the maximum burnup-dependent reactivity, eg., a small increase in the burnup of highest reactivity and a corresponding small decrease in peak reactivity. Thus, the limiting gadolinium concentrations in Figure 4.1 may be considered to apply for any number of fuel rods containing the burnable poison (minimum of three rods) with the understanding that a larger number of poisoned rods would be slightly more conservative.

The burnup curves without gadolinium (dashed curves) in Figure 4.4 were used to determine the minimum burnup required for ,

storage of the higher enriched fuel regardless of the gadolinium content. In each case, an uncertainty due to burnup was assumed ,

to be 0.0005 times the burnup in MWD /kgU (a method of estimating uncertainty used and accepted in several prior rack licensing applications). Results of the estimat.6.1 burnups and reactivities required are listed below based upon the 0.8833 reference design b:

1 t

t E

4-19 i

i Table 4.4 LIMITIl1G BURNUP VALUES FOR VARIOUS INITIAL ENRICHMEllTS Burnup Enrichment Upcertainty b Burnup, MWD 2py 3.05 0 0.8833 0 Reference 3.2 0.0006 0.8827 1.20 3.4 0.0017 0.8816 3.40 3.6 0.0027 0.8806 5.45 3.8 0.0037 0.8796 7.42 With all uncertainties added, the maximum reactivity at the various burnups indicated above will bo 0.895 b. These burnup data are presented in Figure 4.2. Thus, fuel assemblies which have reached or exceeded the burnups indicated in Figure 4.2 may be stored in the fuel racks regardless of the gadolinium content.

These criteria would normally be expected to encompass most of !

the fuel discharged from the reactor. However, for fuel assemblies of less than the required burnup in Figure 4.2 would meet the design basis criterion only with the gadolinium burnable poison loading in Figure 4.1. !

Figure 4.4 also shows that with no gadolinium present and with 3.8% enriched fuel, the reactivity at zero burnup (fresh fuel) is 0.934. With the uncertainty of 0.0105 added, the maximum b would be 0.945. This confirms that a criticality accident cannot occur in the Millstone Unit No. 1 racks regardless of the fuel loaded, up to an enrichment of at least 3.8%. -

t

w. --

+

b 4-20 4.7 ABNORMAL AND ACCIDENT CONDITIONS 4.7.1 Temeerature ard Water Density Effects The moderator temperature coefficient of reactivity is l negative and a conservative moderator temperature of 20 0 C was assumed for the reference design. This assures that the true ;

reactivity will always be lower.

Temperature effects on reactivity have been calculated and the results are shown in Table 4.5. Introducing voids in the water internal to the storage cell (to simulate - boiling) decreased reactivity, as shown in the table.

Table 4.5 EFFECT OF TEMPERATURE AND VOID ON CALCULATED REACTIVITY OF STORAGE RACK 3 Case Incremental Reactivity Change, Ak 200C Reference 500C -0.007 300C -0.014 1200C -0.026 i 1200 C + 20% void -0.072 i

e r

i i

1

-,--r - ,,

V 4-21 4.7.2 bbnormal Location of Fuel Assembly i

Since the storage rack criticality calculations were made assuming an infinite array of storage cells with no neutron leakage, positioning a fuel assembly outside and adjacent to the actual finite rack cannot add reactivity. Rather, because of the actual neutron leakage, this would result in a lower kerg than the k. calculated for the infinite array.

4.7.3 Eccentric Fuel Assembly Positionino The fuel assembly is normally located in the center of -

the storage rack cell with bottom fittings and spacers that mechanically prevent lateral movement of the fuel assemblies, j

Nevertheless, calculatior.s with the fuel assembly moved into the corner of the storage rack cell (four-assembly closter at closest !

app.oach), resulted in a small negative reactivity e Sect. Thus, i the nominal case, with the fuel assembly positioned in the center of the storage cell, yields the maximum reactivity.

4.7.4 Zirconium fuel channel Distortion Consequences of bulging of the zirconium fuel channel are treated as a mechanical deviation in Section 4.6.2.6 above.

i Bowing of the zirconium channel (includir.g fuel rods) results in a local negative reactivity effect analogous to that of eccentric

! positioning the fuel assembly toward one side of the storage i

fi y - - .. , -..,,w --,---w y-. z_, - - - - , + - - - . . . . , - - - - . . - - - - - - - . ,

l 4-22 t

cell'. Thus, any bowing that might occur will result in a reduction in reactivity.

4.7.5 Dropned Fuel Assembly For a drop on top of the rack, the fuel assembly will cc:ne to rest horizontally on top of the rack with a minimum separation distance from the . fuel of more than the 12 inches sufficient to preclude neutrons coupling (i.e., an effectively infinite separation). Maximum expected deformation under seismic or accident conditions will not reduce the minimum spacing ,

between fuel assemblies to less than 12 inches. Consequently, fuel assembly drop accidents will not result in a significant increaoe in reactivity due to the separation distance.

i 4.7.6 Fuel Rack Lateral Movement ;

Normally, the individual rack modules in the spent fuel pool are separated by a water gap of several inches. For finite fuel racks, this separation would reduce the actual maximum reactivity of the racks. Should lateral motion of a fuel rack occur, closing the gap between racks (for whatever reason), the ,

reactivity would, in the limit, only approach the limiting i reactivity of the reterence infinite array.

4.7.7 Lost or Missina Absorber Sheet l An assumption that a Boraflex absorber plate was lost l

or missing (for whatever reason) from a 9x9 storage cell module

[ (replaced by water and allowing direct neutron communication between two adjacent fuel assemblies separated only by water and

I i

[

4-23 the steel box walls), was estimated to result in an increa<,e in reactivity of 0.0027Ak. Thus, even under this condition, the reactivity remains within the design basis critorion.

4-24 l 8

4.8 REFERENCES

i

-4.1 Green, Lucious, Petrie, Ford, White, Wright, "PSR- t 63/AMPX-1 (code package), AMPX Modular Code System for 4

Generating Coupled Mul*,igroup Neutron-Gamma Libraries from ENDF/B,"_ ORNL '.'M-3 7 0 6, Oak Ridge National Laboratory, March 1976.

4.2 L.M. Petrie and N.F. Cross, "KENO-IV, An Improved Monte Carlo Criticality Prograta," ORNL-4938, Oak Ridge National Laboratory, November 1975. >

4.3 R.M. . Westfall et al., "SCALE: A Modular Code System for Performing Standardized Computer Analyses for

, Licensing Evaluation," !!UREG/CR-0200, 1979.

4.4 A. Ahlin, M. Edenius, H. Haggblom, "CASMO -

A Fuel Assembly Burnup Program," AE-RF-76-4158, Studsvik report (proprietary). t 4.5 A. Ahlin and M. Edenius, "CASMO -

A Fast Transport Theory Depletion Code for LWR Analysis," ANS Transactions, Vol. 26, p. 604, 1977. !

t 4.6 M. Edenius et al., "CASMO Benchmark Report,"

Studsvik/RF-78-6293, Aktiebolaget Atomenergi, March

, 1978. .

4.7 W.A. Wittkopf, "NULIF -

Neutron Spectrum Generator, [

Few-Graup Constant Generator and Fuel Depletion Code," l i' BAW-426, The Cabcock and Wilcox Company, August 1976. i t

! 4.8 W.R. Cadwell, PDQ107 Reference Manual, WAPD-TM-678, !

Bettis Atomic Power Laboratory, January 1967. ,

I l 4.9 M.G. Natrella, Experimental Statistics National Bureau !

l of Standardo, Handbook 91, August 1963.

i h

l l 4

l

)

[

l l >

L

)

MILLSTONE UNIT 1 3.cc .. .. .... .... .... .... ..., ... .... .... .

2.75 .

~

2.50 -

e .

d 2.25 .

~

!!E ' 2.00 [ '

n . ACCEPTABLE DCMAIN :

o : OF Gd203 LOADING :

se 1 .7 .,

e :

m .

~

m, 1.50 .

s : :

/

-o 1.25 .

~

1.ca . ,

c'7 ~* '

Ub ACCEPT'5LE DON AIN '

OF Gd20 LOADING :

0.: :C -

e .

i : -

O 25 .

/ .

4 e e e

e e 9 e e f 1 9 _ *9 9 e e

9 6 e * ? e e e e y 9 e e e ,

5.0 3.1 3.2 3.3 3.4 3.5 3.5 3.7 3.8 3.3 i

INITIAL ENRICHMENT. WTs U-235 i

! FIGURE !. 1 MINIMUM GADOLINIUM LOADING, AS Gd203, l IN MINIMUM OF THREE FUEL RODS PER ASSEMBLY i

d g4 9 .= ,- 6 g

- - - , ,----1"-' "

4-20 MILLSTONE UNIT 1 s .... .... .... .... .... .... .... ... ...,

~

/

7 6

3 : ACC EPTMBIF g -

BURNL PDOMAI N -

s e

y .

s . -

a. .

E 4 x -

2 .

C .

a .

w , .

o u

U NACCEPTABLE -

j BU RNUP DO M AI N .'

2 .

. i .

3

/ -

l

- . /. . . . .,,, .... .... ,,,. .., . . ,,..-

?.C 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 .E9 INmAL ENRICHMENT, Wiz U-235 i

I

{ FIGURE 4.2 MINI!!UM REQUIRED FUEL BURNUP IN l

THE SPENT FUEL STORAGE RACKS i

(

l

. . ~ - '

_ 4-27 90RAR.EX

~

5.28" 0.03" l l'

2 0.'02061 gm B-10/CW i t -......y.......,O.0 85" THK

^

9m m mmm .mmm mmmmm __ """

2r n.ow i ! 0 0 0 0 0 0 0 0) il .075"33 I

F B

fj 000000d0 LINER 3 ca. co s.278 io j j oooooooo j j OO 00000 ;

Sed OOO l

A fl l 0000000

"" " 5 ll00000000 jl

, I j00000000 1 ,

p 6.__ _____J! --

.. . . . . . y c. . . . _ = 1 P 6.32t0.02" 20 MIX N 0.075" GAP l i

concuts N mu er w NCT TO SCALE J

) FIGURE 4.3 CROSS SECTION OF SPENT FUEL STORAGE CE'.L 9

e I

a 4

f

4-28 0.94 m ..o..,oz o.... .u... ..o.. ,<,...

. . . ...oo,io; .....,-.

r . .

0.93 6

.s .

. - ss .

. s '

O.92 - s s ,

wmi cm w* .s s s ,-

m s

." s s

'7 s s wTHcv' ca -

x 0.91 -

e s s g

. s s

-\ s -

N ',

C w

s s .

Q 0.90 6

,s 3 s

s s,

s s

. s s 2 [ 's s 's 's -

E

% O ~8 0 s s ,

s s- g \

u s s s s e

H, 0 **

i

. 5, 's K s i,s ..

3 's [ [s -

A E

d

's ' C D ."

u 0~:7 '

} : l ,

l - - - M "

G 2:

0.w: w

/'fl

-f l

7 A- 3.?qwith 3.4 % W0 23 O ~* ~'

/ --

B - 3.4E vith 1.0 % E0 3

~

c- 3.67q with 1.7 % M;03 -

l -

D- 3.87J vith 2.4 % 0-d;03 .

O.51, l .i -

,i .... .. ... ii.,

! O 1 2 3 4 5 6 7 8 9 FJiEL SURNUP, MWD /KgU I

FIGURE 4.4 1.URNUP DEPENDENT REACTIVITIES IN

'.HE MILLSTONE UNIT 1 FUEL STORAGE RACK l

I e ass == e

i 5-1 5,0 THERMAL-fiYDRAULIC CONSIDERATIONS A primary objective in the design of the high density spent fuel storage racks for the Millstone Unit No. 1 fuel pool is to ensure adequate cooling of the fuel assembly cladding. In the following section, a brief synopsis of the design basis, the method o* analysis, and the numerical results is provided.

Similar methods of thermal-hydraulic analysis have been used in previous licensing efforts on high density spent fuel racks for Feumi 2 (Docket 50-341), Quad Cities 1 and 2 (Dockets 50-254 and 50-265), Rancho Seco (Decket 50-312), Grand Gulf Unit 1 (Docket 50-416), Oyster Creek (Docket 50-219), Virgil C. Su:nmer (Docket 50-395), Diablo Canyon 1 and 2 (Docket Nos. 50-275 and 50-323), Byron Units 1 and 2 (Docket 50-454, 455), and St. Lucio Unit One (Docket 50-335).

The analyses to be carried out for the thermal-hycraulic qualification of the rack array inay be broken dcwn into the following categories:

(i) Pool decay heat evaluation and pool bulk temperature variation with time.

(ii) Determination of the maximum pool local temperature at the instant when the bulk temperature reaches its maximum value, and also at the instant when the heat emission rate frem the discharged fuel assemblies is at its peak value.

5-2 (iii) Evaluation of the maximum fuel cladding temperature to establish that bulk nucleate boiling at any location is not possible.

(iv) Evaluation of the time-to-boil if all heat rejection paths from the cooler are lost.

(v) Evaluation of the pool bulk temperature and local water temperature for the case when one of the fuel pool cooling pumps becomes inoperative.

(vi) Compute the effect of a blocked fuel cell opening on the local water and maximum cladding temperature.

The following sections present a synopsis of the methods employed to perform such analyses.

5.1 DECAY HEAT CALCULATIOllS FOR THE SPENT FUEL This section covers requirement III.1.5(2) of the llRC's "OT Position for Review and Acceptance of Spent Fuel Storage and Handling Applications", issued on April 14, 1978. This llRC requirement states that calculations for the amount of thermal energy removed by the spent fuel cooling system shall be made in accordance with Branch Technical Position APCSB 9-2, "Residual Decay Energy for Light Water Reactors for Long Term Cooling" (Ref. 5-1).

): >

i 5-3 5.1.1 H_apig a

Millstone Unit No. 1 is a GE supplied Boiling Water Reactor - with a rated power of 652 Mwt(e). The nominal rated thermal output of the reactor is ' approximately 2011 Mwt. The

reactor is equipped with - 580 fuel assemblies. Therefore, the average -operating power per assembly, Po, is 2011/580 = 3.467 Mwt, which corresponds to 11.83 million stu/hr.

The spent fuel pool cooling system consists of two fuel pool coolers' deployed during normal discharges, and in addition, one shutdown cooler during a full core discharge (abnormal) condition. The. required thermal performance data 'on these coolers may be found in Table 5.1.

r The object of this analysis is to obtain the decay heat '

load and the bulk pool temperature profile for the spent fuel !

l pool system. Since the heat load and maximum temperature depend on the assumed batch size and in-core exposure periods, the analysis was based on several conservative assumptions, including those discussed below.

The total storage capacity of the pool, after the 1988 .

reracking, will be 3229 locations, of which 1732 are currently l occupied. i I

In order to bound all results, all computations were !

j carried out for a point in time (in the future) when the pool has ;

just enough open locations left to accept a full core offload. !

r I

i 4

I f

r f

s 5-4 h

We assumed that the total-exposure period is 1825 days and that the normal batch size is 1/3rd core (196 assemblies) for

, future fuel cycles.

We analyzed two scenarios of fuel offload to the pool, as follows: [

.i (a) Normal discharge:

To obtain an upper bound on the normal discharge heat generation rate, we assume the following:

4 (i) 17 batches, 145 assemblies each, discharged ,

at 372 days intervals with 1488 days of i exposure in the reactor (to).

(ii) The latest batch of 204 assemblies (an upper bound on the normal batch size), 60 months (1900 days) (to) exposure, discharged after

! 150 hours0.00174 days 0.0417 hours 2.480159e-4 weeks 5.7075e-5 months of de:ay in-the reactor.

~

The above set of assumptions will yield bounding

results for heat generation because the decay heat load monotonically increases with to and with increasing batch size.

We note that after the transfer of the la'.est batch in the pool, the remaining available storage is marginally

, insufficient to accept a full core offload; i.e., we i have assumed maximum possible inventory of fuel in the

! pool.

U i

e 5-5 !

(b) Full core discharge:

In keeping with the above normal scenario, we further assumed that the full core offload takes place 3 months !

after the latest-normal discharge in' step-(ii) above.

The transfer of fuel begins 250 hour0.00289 days 0.0694 hours 4.133598e-4 weeks 9.5125e-5 months s- af ter reactor i shutdown.

5.1.2 Model Descr!otion i

NNECO utilized a methodology consistent with the requirements of NUREG-0800 (Ref. 5-1) to compute the heat ;

dissipation requirements in the spent fuel pool. The total decay heat consists of fission product and heavy element decay heat. f Total decay heat (P)-for a fuel assembly is given as a linear [

funct4.on of Po and as an exponential function-of to and rs!

P = Po f (_to, Is) where: (

P =

total decay heat per fuel assembly, linear !

function of Po i Po =

average operating power per fuel assembly to =

cumulative exposure time of the fuel assembly i in the reactor

(

rs = time elapsed since reactor shutdown Holtec proprietary computer code "DECAY" was utilized in performing this calculation.

I 5-6 The appropriate uncertainty factor, k, was applied in accordance with ,llUREG-0800 (Ref. 5-1). Furthermore, the operating power, Po, was taken equal to the rated power, even though the reactor may be operating at less than its rated power during much of the exposure period for the batch of fuel assemblies. Finally, the computations and results reported here are based on the discharge taking place when the inventory of fuel in the pool will be at its maximum, resulting in an upper bound on the computed decay heat rate.

Having determined the heat generation rate, the next task was to evaluate the time-dependent temperature of the pool water.

The pool bulk temperature was determined using the first law of thermodynamnics (conservation of energy). Holtec proprietary computer code BULKTEM treats the generalized pool cooling problem wherein the number of heat exchangers removing heat from the fuel pool can be suited to the particular plant system ccnfiguration.

A number of simplifying assumptions were made which render the analysis conservative. These include o The heat exchangers were assumed to have maximum fouling. Thus, the temperature effectiveness, P, for the heat exchanger utilized in the analysis is the lowest postulated value calculated from heat exchanger technical data sheets. Table 5.1 contains the heat exchanger thermal data.

O !!o credit was taken for the imprevement in the film coefficients of the heat exchanger as the cperating temperature rices due to monotonic reduction in the water kinematic viscosity with temperature rise. Thus, the film coefficient used in the computatiens are lower bounds.

5-7 i t

O No credit. was taken for heat loss by evaporation of the pool water. , j O No credit was taken for heat loss to pool walls and I pool floor slab. l l

The basic energy conservation relationship for the pool heat I

exchanger system yields:

dt Ct " Q1-Q2 dr r

where i

=

Ct Thermal capacity of stored water in the pool ,

t = Temperature of pool water at time, I f

l Q1

= Heat generation rate due to stored fuel assemblies in the pool; Q1 is a known function of time, t >

from the preceding section.

[

= Heat removed in the fuel pool cooler Q2 ,

I i

This equation is solved as an initial value problem by noting :

that the cooler heat removal rate must equal the heat generation (

rate from previously discharged assemblies. Hence f Weool P (Tin - tcool) = PCONS :

where the parameters are as follows:

PCONS: Heat generation rate from provicualy stored l assemblies ;

t i

s h

I i ,

t

F 5-8 i

i Weool: Coolant thermal flow rate P Temperature effectiveness of the fuel pool- !

cooler.

Tin: Coincident pool water temperature (initial value before beginning of discharge) ;

.teools Coolant inlet temperature The above equation yields: ,

PCONS >

Tin =

+ tcool .

Ucool P The value of Tin computed from the above formula is the initial (

value of the pool water temperature (at the start of fuel j discharge). !

5.1.3 Bulk Pool Temoerature Resules The following values of maximum pool bulk temperature

[

were calculated:

Normal discharge: 1370F (two fuel pool coolers) f Full Core Offload: 121.2 0 F (two fuel coolers plus I one shutdown cooler acting in parallel) ;

i l

I I

l 5 t

5-9 5.2- LCCAL PCOL WATER TEMPERATURE

., 5.2.1 Basia In. order to determine an upper bound on the maximum fuel cladding temperature, a series of conservative assumptions were made. The most important assumptions are listed below:

O In general, a discharge batch contains fuel with different levels of exposure and burn-up. In other

, words, the reactor operating time, to (in ASB-9.2 heat generation formula) may be different for different fuel i

assemblies. It can be seen from ASB-9.2 that the residual heat generation rate is a monotonically increasing function of to. Therefore, assuming an upper bound value on to will ensure that in-pool heat emission rate is conservatively calculated. It is 4 further assumed that every fuel assembly in the pool 4

has this upper bound heat emission rate.

O As shown in Figure 2.1 in Section 2, the modules occupy an irregular floor space in the pool. For the hydrothermal analysis, a circle circumscribing the

, actual rack floor space is drawn (Figure 5.2). It is

, further assumed that the cylinder with this circle as its base is packed with fuel assemblies at the nominal laycut pitch.

O The actual downcemer space aLound the rack module group varies, as shown in Figure 2.1. The nominal downcemer gap available in the pool is assumed to be the total gap avai'.able around the idealized cylindrical rack; thus, the maximum resistance to downward flew is incorporated into the analysis (Figures 5.3 and 5.4).

O Mo downe:mer ficw is assumed to exist between the rack me du '. e s .

i i

i-t 5-10 l i

h 5.2.2 Model-Descriotion

Based on the above assumptions, a conservative l idealized model for the rack assemblage was obtained. In this model, the watar flow is axisymetric about the vertical axis of -

the circular rack assemblage, and thus, the flow is two-dimensional (axisymmetric three-dimensional). Figure 5.3 shows a typical "flow chimney" rendering of the thermal hydraulies model.

4

] The governing integral equation to characterize the flow field in the pool can be solved for the lower plenum velocity field (in the radial direction) and axial velocity (in-cell velocity i field), by using the method of collocation. It should be added !

. t that the hydrodynamic loss coefficients which enter into the !

formulation of the integral equation were also taken from well- i t

recognized sources (Ref. 5-3) and wherever discrepancies in !

l. reported values exist, the conservative values were consistently l used. Reference 5-I gives the details of mathematical analysis j

used in this solution precess. ;

i i

i After the axial velocity field is evaluated, it is a j straight-forward matter to ec=pute the fuel assembly cladding j temperature. The knowledge of the overall flow field enables l pinpointing of the storage location with the minimum axial flow !

l (i.e., maximum water outlet te=peratures). This is called the ;

j rest "choked" locatica. In order to find an upper bound on the l temperature in a typical cell, it is assumed that it is located f at the most choked location. Knowing'the global plenum velocity field, the revised axial flow through this choked cell can be I calculated by solving the Bernoulli's equation for the flow i s

I I

i k i

i

-_., , _ __._ ____ -, _ _._ -,_ _ .._ - _ ,.- - __.._.~ _ - ,.. _ .._,, _ _ _ _ _ _ , , _ _ . _ .

1 5-11 circuit through this cell.- Thus, an absolute upper bound on the water exit temperature and maximum fuel cladding temperature was obtained. It is believed that, in view of~the aforementioned assumptions, the temperatures calculated in this manner overestimate the temperature rise that will actually occur in the pool. Holtec's computer code THERPOOL, based on the theory of of Ref. 5-4, automates this calculation. The analysis embodied in THERPOOL has been accepted by the 11RC on several dockets.

Table 5.3 gives the output for THERPOOL for both.the normal and full core off-load scenarios of discharge.

5.3 CLADDI!1G TEMPERATURE r

The maximum specific power of a fuel array qA can be given by:

9A = q Fxy where:

Fxy adial peaking factor q - average fuel assembly specific power

. The data on radial and axial peaking factors may be found in

! Table 5.2. Based on this data, tillECO performed calculations for a total peaking factor of 3.23 with two bounding values of the radial peaking factor (1.5 and 1.9).

The maximum temperature rise of pool water in the most disadvantageously placed fuel assembly was computed for all loading cases. Having determined the maximum local water temperature in the poet, it is then possible to determine the l

5-12 maximum fuel cladding temperature. A fuel red can produce F Tot times the aver, age heat emission rate over a small length, where FT ot is the total peaking factor. The axial heat dissipation in a rod is known to reach a maximum in the central region, and taper off at its two extremities. For the sake of added conservatism it was assumed that the peak heat emission occurs at the top where the local water temperature also reaches its f maximum. Furthermore, no credit was taken for axial conduction of heat along the red. The highly conservative model leads to simple algebraic equations which directly give the maximum local

! cladding temperature, te.

Table 5.3 summari::es the key output data. It was found that the maximax (maximum in time and space) value of the local water

, temperature is well below the nucleate boiling condition values of l 235 0F. The incremental cladding temperature is too small to produce significant thermal stresses.

Assuming a fouling factor of 0.001 hr-sq.ft OF/ Btu on the fuel cladding surface produces an incremental temperature rise of less than 10 F. It is to be noted that the assumed fouling resistance is substantially in excess of the values cu.atemarily used for fuel pool cooler tuce surface.

5.4 TIME-TO-30IL Calculations were also performad to determine the time elapsed before bulk boiling of the pool begins if all engineered heat removal paths are lest. Table 5.4 given the time-to-boil

l l 5-13 l

for both normal and full core off-load conditions, assuming that the loss of cooling occurs at the instant when the pool bulk temperature has reached its maximum value. The ensuing rate water level drop was also computed.

5.5 SINGLE PUMP FAILURE SCENARIO Even though the two pumps circulating the fuel pool water through the two fuel pool coolers are qualified as Seismic Category I, analyses were performed to determine the maximum pool bulk temperature under the scenario when one pump becomes unavailable immediately af ter the reactor shutdown, and remains unavailable for a long time. The maximum pool water temperature for the normal disci.arge case under this scenario reaches 140.4 0F. Thus, the temperature of supporting concrete structures remains safely below the ACI recommended limit of 150 0 p, The offect of single pump failure during or after a full core discharge case is even less pronounced because of the dominating effect of the shutdown cooler which is aligned in parallel with the fuel pool coolers after a full core discharge.

The maximum bulk temperature in this case is less than 125 0 7, Reanalyzing for the local pool water temperature for the most limiting of the cases in Table 5.3 (case 2) yields the marimum temperature as follows.

Maximum local pool water temperature = 196.1 0 7 Maximum fuel cladding temperature - 248.3 0 F

_ _ ~ , . .

1 5-14 It is clear from the above that there "is _ a . wide margin against localized boiling even with only one pump operating and thi fuel with the maximum radial peaking factor located at the most unfavorable location in the pool.

Finally, if we postulate that the second pump also fails at the instant that-the pool bulk temperature has reached the peak value, then the pool water temperature will rise steadily. For normal batch discharge, the time-to-boil is reduced to 13.6 hrs (from 14.3 hrs in Table 5.4).

If the remaining fuel pool cooler pump and the shutdown cooler pump both fail at the instant that the bulk pool temperature reaches its maximum value in the aftermath of a full core discharge, then the time-to-boil is 8.5 hrs (reduced from 8.87 hours0.00101 days 0.0242 hours 1.438492e-4 weeks 3.31035e-5 months in the no-pump failure case).

These time periods are sufficient for plant personnel to j provide alternate emergency cooling measures, i

l 5.6 BLOCKED CELL SCEllARIO t

i j litiECO also performed thermal-hydraulic analyses of the case l

. where a channelled fuel assembly is accidentally dropped on top of a rack containing freshly discharged fuel. The channel on the 4

fuel assembly maximizes the blockage of gravity driven convection through the most disadvantageously placed storage cell. We assume l

that 80% of the cross sectional opening in a storage cell is l completely blocked under this condition. Il!1ECO's analyses show that the local water temperature sustains an incremental temperature rise of a maximum of 11 0F for the worst orientation i

I

5-15 of the assembly on the storags cell. Thus for the single pump operating case (one pump disabled), the maximum local pool water temperature is less than 208 0F. The corresponding local boiling temperature of water (33' high column of water) is 235 07, substantial margin against localized boiling therefore exist, and the fuel cladding would not be subject to bigh thermal stresses.

5-16 Table 5.1 HEAT EXCHANGER DATA (1) Fuel Pool Cooler The following data characterize the fuel pool coolers with the two pumps working.

Tubeside Flow Rate: 314000 lb/hr The temperature effectiveness is defined as:

tubeside outlet - tubeside inlet P"

shellside inlet - tubeside outlet substituting the appropriate numbers from the heat exchanger datasheet, we have 112.5 - 125 Temperature effectiveness p=

85 - 125 or p = 0.3125 (ii) Shutdown Cooler Coolant Flow Rate: 900,000 lb/hr Temperature effectiveness = 0.39 (iii) Fuel Pool Coolers with only One Pumn Working The flow rate through each cooler drops down to 214462 lbs per cooler. The corresponding temperature effectiveness p is 0.428.

l r

5-17 Table 5.2

. AXIAL AND RADIAL PEAKING FACTOR DATA AXI?L RADIAL PEAKING PEAKING CASE I.D. CASE IDENTIFICATION FACTOR FACTOR A Normal discharge and 2.1533 1.5 Bulk Pool Temperature is maximum.

B llormal discharge and 1.7 1.9 Bulk Pool Temperature is maximum.

l

5-18 Table 5.3 LOCAL MAXIMAX POOL WATER AND FUEL CLADDING TEMPERATURE LOCAL MAXIMAX BULK POOL FUEL CASE CASE POOL WATER CLADDING MO. IDENTIFICATION TEMP. TEMP. TEMP.

1 Normal discharge and 137.0 181.0 233.2 Bulk Pool Temperature is maximum.

Case A of Table 5.2 2 Normal discharge and 137.0 192.7 244.9 Bulk Pool Temperature is maximum.

Case B of Table 5.2 3 Normal discharge and 119.0 166.0 223.2 the Heat Generation Rate is maximum.

Case A of Table 5.2 4 Normal discharge and 119.0 178.5 235.8 the Heat Generation Rate is maximum.

Case B of Table 5.2 5 Abnormal discharge 121.0 160.8 206.2 and Bulk Pocl Temperature is maximum.

Case A of Table 5.2 6 Abnormal discharge 121.0 171.6 217.0 and Bulk Pool Temperature is Maximum.

Case B of Table 5.2

k 5-19 Table 5.3 (continued) 7 Abnormal Discharge 121.0 161.0 206.4 and the Heat Generation Rate is maximum.

Case A of Table 5.2 8 Abnormal discharge 121.0 171.6 217.0 and the Heat Generation Rate is maximum.

case B of Tabic 5.2

a 5-20 l

Table 5.4 TIllE-TO-BOIL DATA RATE OF WATER LEVEL TI!4E-TO-BOIL DECREASE C011DITIOli (hours) (inch /hr)

!!ormal ,14.3 1.70 Full Core Offload 8.87 3.30 6

h P

I t

b I

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

5-21 5.7 REFEREllCES 5.1 IlUREG-0800, U.S. !!uclear Regulatory Commission, Standard Review Plan, Branch Technical Position ASB 9-2, Rev. 2, July 1981.

5.2 Singh, K.P., Journal of Heat Transfer, Transactions of the ASME, August, 1981, Vol. 103, "Some Fundamental Relationships for Tubular Heat Exchanger Thermal Performance".

5.3 General Electric Corporation, R&D Data Books, "Heat Transfer and Fluid Flow", 1974 and updates.

5.4 Singh, K.P. et al., "Method for Computing the Maximum Water Temperature in a Fuel Pool Containing Spent lluclear Fuel", Heat Transfer Engineering, Vol. 7, !!o .

1-2, pp. 72-82 (1986).

'$y 5-22 -( ,

88 8g

3 nn n w wa si

<c a

e g,g

~

5

'N % :

Ni n E

7 i .

. I A E .

i 28 8 '

5o a

<gn 5 IE l$ E. -

.. t l

IDEALIZATION CF F ACX ASSIMILY FIGUR1' Sal i

. . . . ~

. _ _ . - , - - - - - - - - - - - - - ~ ' ~ ' ~ ' ' ' ~ ' ' ' " ~ ~ ~ ~ ~

.> .' .i

~

' ^

/

I _

J (

/ /

O

' , i T '

our

/ -

/ .

0 T

lN g g 91 s I w *

.J w

H E E 3 J C U

> c '

$ l 1 - O HEAT ADOITION 1 i

=-

I' C v JL I

I

./ .

V -

y ! /

/2

/1 -

7 I 'i j ,

l/ l t/ L_ /

i

'TNERM At. CHIMNEY PLQW MCDEL PlGURE 5.2

1 4i e

4 5-24 A 'j I 'f 4

.A- 44 -

3

h:i .

,l s'

e

. 's.

3

4. ,, .

1

.A .

T .I.]

2 t k' . '

( '

i;t

? 4k

\. .

. d.

t.1 * .

.* 6. '.

1

.A

.A -

aAcx ,%*

%{. s

.I p;. .

. sk

': 9

P $ r* P P P

. Y i 4.J t i i i

s. . i i 9-

.*.i; 4.,

i. . -

"7,. .

w. '/

(9.*.

.s

r. ..

Nvann

! COMER' j.1 . .

' - ,1 .

.s -

h. . ,

,y 9 '.,, ,

c r

, '.t 9 ;. > . ; > '. - ,. >

s. ,.

d. ' . s. M, UM. c. 6* .

8 @$ ,-1 De 'Jwi' .'\

t . , a

.y ,

\ PLENUM CONVE@CN CUM MENT 3

!!$ WEPCOL FIGURE 5.3

6-1 6.0 STRUCTURAL ANALYSIS, The purpose of this section is to demonstrate the structural adequacy of the Millstone Unit No. I spent fuel rack design under normal and accident loading conditions. The method of analysis presented herein uses a time-history integration method similar to that previously used in the licensing reports on high density spent fuel racks for Fermi 2 (Docket lio. 50-341), Quad Cities 1 and 2 (Docket 11os. 50-254 and 5 0-2 t,5 ) , Rancho Seco (Docket !!o.

50-312), Grand Gulf Unit 1 (Docket flo . 50-416), Oyster Creek (Docket !!o. 50-219), V.C. Summer (Docket !!o. 50-395), and Diablo Canyon Units 1 and 2 (Docket 11os. 50-275 and 50-323). The results show that the high density spent fuel racks are structurally adequate to resist the postulated stress combinations associated

, with level A, B, C, and D conditions as defined in References 6-1 and 6-2.

6.1 ANALYSIS OUTLIllE (FOR 11EW PROPOSED RACK MODULES)

The spent fuel storage racks are Seismic category I equipment. Thus, they are required to remain functional during and after a Safe Shutdown Earthquake (Raf. 6-3). As noted previously, these racks are neither anchored to the pool floor nor attached to the sidewalls. The individual rack modules are not interconnected. Furthermore, a particular rack may be completely loaded with fuel assemblies (which corresponds to greatest rack inertia), or it may be completely empty. The coefficient of friction, p, between the supports and pool floor i ,

6-2 is another indeterminate factor. According to Rabinowic: (Ref.

6-4), the recults of 199 tests performed on austenitic stainless steel plates submerged in water show a mean value of p to be 0.503 with a standard deviation of 0.125. The upper and lower bounds (based on twice the standard deviation) are thus 0.753 and 0.253, respectively. Two separate analyses were performed for the rack assemblies with values of the coefficient of friction equal to 0.2 (lower limit) and 0.8 (upper limit), respectively.

Analyses performed for the geometrically limiting rack modules focus on limiting values of the coefficient of friction, and the number of fuel assemblies stored. Cases studied are for the largest rack module with maximum aspect ratio (Medule D-2), and the smallest module (Module A). Typical simulations are:

O Fully loaded rack (all storage locations occupied),

p = 0.8; 0,2 (p = coefficient of friction)

O Rack half full, p = 0.8, .2 The simulations were performed using unchanneled fuel (weight of 680# per cell). The case of nearly empty racks is found not to be critical.

The seismic analyses were performed utilizing the time-history methed. Pcci slah acceleration data in three orthegenal directions was developed and verified to be statistically independent. The acceleration data for the center of the pool ficor was censervatively applied to all the racks regardless of 1ccation. This is a censervative approach since the accelerations were higher at the center of the floor than at the sides.

I

6-3 The objective of the seismic analysis is to determine the structural response (stresses, deformation, rigid body motion, etc.) due to simultaneous application of the three statistically independent, orthogonal excitations. Thus, recourse to approximate statistical sununation techniques such as the "Square-Root-of-the-Sum-of-the-Squares" method (Ref. 6-5) was avoided.

For nonlinear analysis, the only practical method is simultaneous application.

Pool slab acceleration data are provided for two earthquakes: Operating Basis Earthquake (OBE) and Safe Shutdown Earthquake (SSE). Figures 6.1 - 6.3 show the time-histories corresponding to the SSE condition.

The seismic analysis was performed in three steps, namely:

1. Development of a nonlinear ctnamic model consisting of inertial mass elements and gap and friction elements.

2. Generation of the equations of motion and inertial coupling and solution of the equations using the i

"component element time integration scheme" (References 6-6 and 6-7) to determine nodal forces and displacements.

3. Computation of the detailed stress field in the rack (at the critical location) and in the support legs i using the nodal forces calculated in the previous step. These stresses are checked against the design limits given in Section 6.5.

i A brief description of the dynamic model follows.

4 i

i

6-4 6.2 FUEL RACK - DYNMtIC MODEL Since the racks are not anchored to the pool slab or attached to the pool walls or to each other, they can execute a wide variety of rigid body motions. For example, the rack may slide on the pool floor (so-called "sliding condition"); one or more legs may momentarily lose contact with the liner ("tipping condition"); or the rack may experience a combination of sliding and tipping conditions. The structural model should permit simulation of these kinematic events with inherent built-in conservatisms. Since the modules are designed to preclude the incidence of inter-rack impact, it is also necessary to include the potential inter-rack impact phenomena in the analysis to demonstrate that such impacts do not occur. Lift off of the support legs and subsequent liner impacts must be modelled using appropriate impact elements, and coulomb friction between the rack and the pool liner must be simulated by appropriate piecewise linear springs. These special attributes of the rack dynamics require a strong emphasis on the modeling of the linear and nonlinear springs, dampors, and stop elements. The model outline in the remainder of this section, and the model description in the following section, describe the detailed modeling technique to simulate these effects, with emphasis placed on the nonlinearity of the rack seismic response.

6-5 6.2.1 Outline of Model for Computer Code DYNARACK s a. The fuel rack structure is a folded metal plate assemblage welded to a baseplate and supported on four legs. The rack structure itself is a very rigid structure. Dynamic analysis of typical multicell racks has shown that the motion of the structure is captured almost completely by the behavior of a six degree-of-freedom structure, where the movement of the rack cross-section at any height is described ,in terms of the six degrees-of-freedom of the rack base. The rattling fuel is modelled by five lumped masses located at H, .75H, .5H, .25H, and at the rack base, where H is the rack height as measured from the base.

b. The seismic motion of a fuel rack is characteri::ed by random rattling of fuel assemblies in their individual storage locations. Assuming a certain statistical coherence in the vibration of the fuel assemblies exaggerates the ,

computed dynamic loading on the rack structure. This assumption, however, greatly reduces the required degrees-of-freedom needed to model the fuel assemblies which are represented by five lumped masses located at different levels of the rack. The centroid of each fuel assembly mass can be located, relative to the rack structure centroid at that level, so as to simulate a partially loaded rack.

c. The local flexibility of the rack-support interface is a modeled conservatively in the analysis,

d. The rack base support may slide or lift off the pool floor.

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

f. Fluid coupling between rack and assemblies, and between rack and adjacent racks, is simulated by introducing appropriate '

inertial coupling into the system kinetic energy. Inclusion of these effects uses the methols of References 6-4 and 6-6 t for rack / assembly coupling and for rack / rack coupling (see Section 6.2.3 of this report).

a

6-6 >

g. Potential impacts between rack and assemblies are accounted ;

for by appropriato "compression only" gap elements between masses involved. .

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

i. The supports are modeled as "compression only" elements for the vertical direction and as "rigid links" for dynamic anclysis. The bottom of a support leg is attached to a frictional spring as described in Section 6.3. The cross- section inertial properties of the support legs are computed and used in the final computations to determine support leg stresses.

j. The effect of sloshing was shown to be negligible at the .

bottom of the pool and is hence neglected, ;

k. The possible incidence of inter-rack impact is simulated by a series of gap elements at the top and bottom of the rack in the two horizontal directions. The most conservative case of adjacent rack movement is assumed; each adjacent rack is assumed to move completely out of phase with the .

rack being analyzed.

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

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

n. The rattling of the fuel assemblius inside the storage locations causes the "gap" between the fuel assemblies and the cell wall to change from a maximum of twice the neminal gap to a theoretical =ero gap. Therefore, the fluid coupling !

coefficients (Ref. 5-8) utilized are based on non-linear vibration theory (Re[. 6-9). Studies in the literature show that inclusion of tne nonlinear effect (i.e., vibration amplitude of the same order of magnitude as the gap) provides a more accurate characterization of the equipment response (Ref. 6-10).

k

l L

I i-l 6-7 r

o. The cross coupling effects due to the movement of fluid from one interstitial (inter-raok) space to the adjacent one is ,

modelled using potential flow and Kelvin's circulation '

theorem. This formulation has been reviewed and approved by the Nuclear Regulatory Commission, during the post-licensing multi-rack analysis for Diablo Canyon Unit I and II reracking project.

Figure 6.4 shows a schematic of the model. Six degrees-of-freedom are used to track the motion of the rack structure. i Figures 6.5 and 6.6, respectively, show the inter-rack impact l

springs and fuel assembly / storage cell impact springs at a !

particular level.

i As thown in Figure 6.4, the model for simulating fuel ,

assembly motion incorporates five lumped masses. The five f

rattling masses are located at the baseplate, at quarter height, 7 at half height, at three quarter height, and at the top of the rack. Two degrees-of-freedcm are used to track the motion of each h rattling mass in the hori:: ental plane. The vertical motion of !

cach rattling maso is assumed to be the same as the rack base.

t 6.2.2 Model Description The absolute degeees-of-freedom associated with each of ;

the mass locations are identified in Figure 6.4 and Table 6.1.

The rattling masses (nodes 1*, 2*, 3*, 4*, 5*) are described by translational degrees-of-freedom q7-(116 t

Ui(t) is the pool floor slab displacement seismic l time-history. Thus, there are sixteen degrees-of-freedom in the !

system. Not shown in Fig. 6.4 are :.he gap elements used to model i

the support legs and the impacts with adjacent racks. i l

1 i

1 I

l r

6-8 '

Table 6.1 DEGREES OF FREEDOM Displacement Rotation Location Ux Uy U Ox Oy Oz

(!! ode) 1 PJ P2 P3 94 95 96 i

2 Point 2 is assumed attached to rigid rack at the top most point.

2* P7 P8 3* pg pio ;

4* Pil P12 5* pl3 P14 1* P15 P16 where:

pi = qi(t) + U l(t) i = 1,7,9,11,13,15

= qi(t) +U2(t) i = 2,8,10,12,14,16

= qitt) +U3(t) i=3 Ui(t) are the 3 known earthquake displacements.

i t

i

6-9 6.2.3 Fluid Counling An effect of some significance requiring careful modeling is the so-called "fluid coupling effect". If one body of mass (ml) vibrates adjacent to another body (mass m ), and both bodies are submerged in a frictionless fluid medium, then

!!ewton's equations of motion for the two bodies have the form:

.. oo (mi + tt11) X1 - 1112 X2 = applied forces on mass mi oo ,

-!121 N1 + (m2 + l122) X2 = applied forces on mass m2 N1 , 'X2 denote absolute accelerations of mass mi and m2, '

respectively.

!!11, 1112, !!21, and !!22 are fluid coupling coefficients which !

depend on the shape of the two bodies, their relativo disposition, etc. Frit: (Ref. 6-9) gives data for tilj for various body shapes and arrangements. It is to be noted that the above equation indicates that the effect of the fluid is to add a certain amount of mass to the body (!!11 to body 1), and an ;

external force which is proportional to the acceleration of the adjacent body (mass m2). Thus, che acceleration of one body affects the force field on another. This force is a strong function of the interbody gap, reaching large values for very small gaps. This inertial coupling is called fluid coupling. It has an important offect in rack dynamics. The lateral motion of a fuel assembly inside the storage location will encounter this effect. So will the motion of a rack adjacent to another rack.

These effects are included in the equations of motion. For 6 example, the fluid coupling is between nodes 2 and 2* in Elgure l

6-10 6.4. Furthermore, the rack equations contain coupling terms which model the effect of fluid in the gaps between adjacent racks. The -

coupling terms modeling the effects of fluid flowing between adjacent racks are computed assuming that all adjacent racks are vibrating 1800 out of phase from the rack being analyzed.

Therefore, only one rack is considered surrounded by a hydrodynamic mass computed as if there were a plane of sy: metry I

located in the middle of the gap region.

Finally, fluid virtual mass is included in the vertical direction vibration equations of the rack; virtual inertia is also added to the governing equation corresponding to the rotational degree-of-freedom, g6(t).

6,2.4 Damping In reality, damping of the rack motion arises from material hysteresis (material damping), relative intercomponent motion in structures (structural damping), and fluid drag effects (fluid damping). In the analysis, a maximum of 4%

structural damping is imposed on elements of the rack structure during SSE seismic simulations. This is in accordance with the FSAR and !!RC guidelines (Ref. 6-11). Material and fluid damping are conservatively neglected. The dynamic model has the j provision to incorporate fluid damping effects; however, no fluid i damping has been used for this analysis.

)

i

6-11 6.2.5 Impact Any fuel assembly node (e.g., 2*) may impact the corresponding structural mass node 2. To simulate this impact, ,

four compression-only gap elements around each rattling fuel assembly node are provided (see Figure 6.6). The compressive loads developed in these springs provide the necessary data to evaluate the integrity of the cell wall structure and stored array during the seismic event. Figure 6.5 shows the location of the impact springs used to simulate any potential for inter-rack impacto. Section 6.4.2 gives more details on these additional impact springs.

t 6.3 ASSEMBLY OF THE DYNAMIC MODEL The cartesian coordinate system associated with the rack has the following nomenclature O x= Horizontal coordinate along the short direction of ;

rack rectangular platform ,

O y = Horizontal coordinate along the long direction of the rack rectangular platform

, o z = Vertically upward As described in the preceding section, the rack, along with the base, supports, and stored fuel assemblies, is modeled for the general three-dimensional (3-D) motion simulation by a mixteen degree-of-freedom model. To simulate the impact and !

sliding phenomena expected, 64 nonlinear gap elements and 16 ,

nonlinear friction elements are used. Gap and friction elements, !

with their connectivity and purpose, are presented 'n Tcble 6.2.

t i

l l

6-12 Table 6.2 11U!!BERIt!G SYSTE!! FOR GAP ELE!!EllTS AllD FRICTIOli ELEMEllTS I. ?!cnlinear Springs (Gap Elements) (64 Total) s fiumber flode Location Descriptien 1 Support S1 3 compression only element 2 Support S2 Z compression only element 3 Support S3 Z compression only element 4 Support S4 Z compression only element 5 2,2* X rack / fuel assembly impact element 6 2,2* X rack /f'uol assembly impact element 7 2,2* Y rack / fuel assembly impact element 8 2,2* Y rack / fuel assembly impact element 9-24 Other rattling masses for nodes 1*, 3*, 4* and 5*

25 Bottom cross- Inter-rack impact elements section of rack (around edge)

Inter-rack impact elements

. Inter-rack impact elements

. Inter-rack impact elements 44 Inter-rack impact elements 45 Top cross-section Inter-rack impact elements

. of rack Inter-rack impact elements

. (around edge) Inter-rack impact elements

. Inter-rack impact elements

. Inter-rack impact elements 64 Inter-rack impact elements

6-13 l

Table 6.2 (continued)

NUMBERI!;G SYSTE!1 FOR GAP ELE!4E!1TS AND FRICTIO!1 ELE!4E!1TS II. Friction Elements (16 total)

Number Node Location Description 1 Support S1 X direction friction 2 Support S1 Y direction friction 3 Support S2 X direction friction 4 Support S2 Y direction friction ,

5 Support S3 X direction friction i 6 Support S3 Y direction friction ;

7 Support S4 X direction friction l 8 Support S4 Y direction friction i 9 S1 X Slab moment 10 S1 Y Slab moment 11 S2 X Slab moment I 12 S2 Y Slab moment i 13 S3 X Slab moment 14 S3 Y Slab moment 15 S4 X Slab moment 16 S4 Y Slab moment +

1 J

l l i

r I

i ,

t-i f t

i t

i 6-14 i i

If the simulation model is restricted to two dimensions (one ;

hori: ental riotion plus vertical motion, for example) for the !

purposes of nodel clarification only then a descriptive model of l l.

the simulated structure which includes gap and friction elements ,

is shown in Figure 6.7. :

e impacts between fuel essemblies and rack show up in the gar .< ments, having local stif fr ess K I , in Fj gure 6.7. In Table ,

6.2, gap elements 5 through 8 are for the vibrating mass at the .

top of the rack. The support leg spring rates Ks are modeled by elements 1 through 4 in Table 6.2. !!ote that the local compliance {

of the concrate floor :. 77 eluded in Ks. To simulate sliding ;

potential, friction eleme 2 plur 8 and 4 plus 6 (Table 6.2) are shown in Figure 6.7. The friction of tne support / liner j interface is modeled by a piecewise linear spring with a suitably ,

large stiffness Kf up to the limiting lateral load, p!!, where 11 is the current compression load at the interface between support and liner. At every time step during the transient analysis, the ,

current value of !! (either zero for liftoff condition, or a !

.i compressive finite value) is computed. Finally, the support [

rotational friction springs Kg reflect any rotational restradnt that may be offered by the foundation. This spring rate is j calculated using a modified Bousinesq equation (Ref. 6-4) and is !

included to simulate the resistive moment of the support to i counteract rotation of the rack leg in a vertical plane. This rotation spring is also nonlinear, with a :ero spring constant value assigned after a certain limiting condition of slab moment !

loading is reached. !

i r

l' j

l

(>. ^

f, :. '

6-15

-The nonlinearity of these springs (friction elements 9,-11, 13, and 15-in Table 6.2) reflects the edging; limitation imposed on the' base of the rack support legs. .If this .effect is neglected; any support leg bending, induced by liner / baseplate friction forces, is resisted by the leg acting as a beam cantilevered from the rack baseplate. This leads to higher predicted loads at the support leg - baseplate junction.

The spring rate Ks modeling the effective compression stiffness of the structure in the vicinity of the support, is computed from the equation:

1 1 1 1

_ = _+ _ +

KS K1 K2 K3 where:

Ki= spring rate of the support leg treated as a tension-compression member = ESUPPORT

ASUPPORT/h (h = length of support leg)

K2 = 1.05EcB/(1-7/2) = local spring rate of pool slab (Ee =

Young's modulus of concrete, and B = length of bearing surface)

K3 = spring rate of folded plate cell structure above support leg (same form as K2 with E chosen to reflect the local stiffness of the honeycomb structure above the leg)

For the 3-D simulation, all support elements (listed in Table 6.2) are included in the model. Coupling between the two horizontal seismic motions is provided both by the offset of the

6-16 l

fuel ~ assembly group centroid which causes the rotation of the 1 entire rack and by the possibility of liftoff of one or more support legs. The potential exists for the rack to be supported on one or more support legs during any instant of a complex 3-D seismic - event. All of these potential events may be simulated during a 3-D motion and have been observed in the results.

6.4 TIME IUTEGRATION OF THE EQUATIONS OF MOTION 6.4.1 Time-History Analysis Usino Multi-Decree of Freedom

, Rack Model Having assembled the structural model, the dynamic equations of motion corresponding to each degree-of-freedom are written by using Lagrange's Formulation. The system kinetic energy can be constructed including contributions from the sel'.d structures and frc ' the trapped and surrounding fluid. A s' gle

. rack is modelled in detail. The system of equations can be represented in matrix notation as:

_ {M] {q) = {Q} + {G}

where the vector {Q} is a function of nodal displacements and velocities, and {G) depends on the coupling iner'ia t and the ground acceleration. Premultiplying the above equations by [M]-1 renders the resulting equation uncoupled in mass.

We have: {q'} = [M]-1 {Q} + (M]-1 {G)

As noted earlier, in the numerical simulations run to verify structural integrity during a seismic event, all elements of the fuel assemblies are assumed to move in phase. This will

6-17 provide- maximum impact force level, and induce additional conservatism in the time-history analysis.

This equation set is mass uncoupled, displacement coupled, and is ideally suited for numerical solution using a central difference scheme. The computer program "DYllARACK"* is utilized for this purpose.

Stresses in various portions of the structure are computed from known element forces at each instant of time.

Dynamic analysis of typical multicell racks has shown that the motion of the structure is captured almost completely by the behavior of a six-degree-of-freedom structure; therefore, in this analysis model, the movement of the rack cross-section at any height is described in terms of the rack base degrees-of-freedom (q1(t),...q6(t). The remaining degrees-of-freedom are associated with horizontal movements of the fuel assembly masses.

In this dynamic model, five rattling masses are used to represent fuel assembly movement in the horizontal plane. Therefore, the This code has been previously utilized in licensing of similar racks for Fermi 2 (Docket 11o. 50-341), Quad Cities 1 and 2 (Docket llo s . 50-254 and 265), Rancho Seco (Docket flo . 50-312), Oyster Creek (Docket 11 o . 50-219), V.C. Summer (Docket flo. 50-395)e and Diablo Canyon 1 and 2 (Docket 11os .

50-275 and 50-323), St. Lucie Unit I (Docket lio. 50-335) and Byron Units I and II (Docket !!os. 50-454, 50-455).

'it 6-18 final -. dynamic model consists - of six ' degrees-of-f reedom for the rack plus ten additional mass degrees-of-freedom for the five ' ~ .

rattling masses. The totality of fuel' mass is included in the simulation and is distributed among the five rattling masces.

6.4.2 Evaluation of Potential for Inter-Rack Impact Since the racks are closely spaced, the simulation includes impact springs to model the potential for inter-rack impact, especially for -low values of the friction coefficient between the support and the pool liner. To account for this potential, yet still retain the simplicity of simulating only a single rack, gap elements were located on the rack at the top and at the baseplate level. Figure 6.5 shows the location of these gap ' elements. Loads in these elements, computed during the dynamic analysis, are used to assess rack integrity if inter-rack impact were shown to occur.

6.5 STRUCTURAL ACCEPTANCE CRITERIA There are two sets of criteria to be satisfied by the rack modules:

a. Kinematic Criterion This criterion seeks to ensure that the rack is a physically stable structure. The Millstone Unit No. 1 racks are designed to preclude inter-rack impacts.

Therefore, physical stability of the rack is considered along with the criterion that inter-rack impact or rack-to-wall impacts do not occur.

6-19

b. Stress Limits ,

The stress limits of the ASME Code, .Section III, Subsection NF, 1983 Edition are used since this code provides .the most - appropriate and consistent set of limits for various stress types and various loading conditions. The following loading combinations are applicable (Ref. 6-1).

Loadino Combination Stress Limit D+L Level A service limits D + L + To D + L + To + E D + L + Ta + E Level B service limits D + L + To + Pf D + L + Ta +.E' Level D service limits D+L+Fd The functional capability of the fuel racks should be demonstrated.

where:

D = Dead weight-induced stresses (including fuel i assembly weight)

L = Live Load (0 for the structure, since there l are no moving objects in the rack load path).

l = Force caused by the accidental drop of the Fd heaviest load from the maximum possible height.

Pf = Upward force on the racks caused by i

postulated stuck fuel ascembly E = Operating Basis Earthquake l

t

6-20 E' = Safe Shutdown Earthquake To = Differential temperature induced loads (normal or upset condition)

Ta = Differential temperature induced loads (abnormal design conditions)

The conditions Ta and To cause local thermal stresses to be produced. The worst situation will be obtained when an isolated storage location has a fuel assembly which is generating heat at the maximum postulated rate. The surrounding storage locations are assumed to contain no fuel. The heated water makes unobstructed contact with the inside of the storage walls, thereby producing the maximum possible temperature difference between the adjacent cells. The secondary stresses thus produced are limited to the body of the rack; that is, the support legs do not experience the secondary (thermal) stresses.

6.6 MATERIAL PROPERTIES The data on the physical properties of the rack and support materials, obtained from the ASME Boiler & Pressure Vessel Code, Section III, appendices, and supplier's catalog, are listed in Tables 6.3 and 6.4. Since the maximum pool bulk temperature (except for the full core discharge case) is less than 1500 , this is used as the reference design temperature for evaluation of material properties.

6-21 Table 6.3 RACK MATERIAL DATA Young's Yield Ultimate Modulus Strength Strength Material E (psi) Sy (psi) Su (Psi) 304 S.S. 27.9 x 106 23150 68100 Section III Table Table Table Reference I-6.0 I-2.2 I-3.2 Table 6.4 SUPPORT MATERIAL DATA Material 1 ASTM-240, Type 304 27.9 x 106 23,150 68,100 (upper part of support psi psi psi feet) 2 ASTM 564-630 27.9 x 106 120,600 145,000 psi psi psi

6-22 6.7 STRESS LIMITS FOR VARIOUS CONDITIONS

~

-The~following stress limits are derived from the guidelines of the ASME Code,Section III, Subsection NF, in conjunction with the_ material properties data of the-preceding section.

6.7.1 Normal and Ueset Conditions (Level A or Level B)

a. Allowable stress in tension on a net section

=Ft = 0.6 Sy or Ft= (0.6) (23,150) = 13,890 psi (rack material)

Ft = is equivalent to primary membrane stresses Ft= (.6) (23,150) = 13,890 psi (upper part of support feet)

= (.6) (120,600) = 72,360 psi (lower part of support feet)

b. On the gross section, allowable stress in shear is:

Fy = .4 S

(.4)y(23,150) = 9,260 psi (main rack body)

Ft= (.4) (23,150) = 9,260 psi (upper part of support feet)

= (.4) (120,600) = 48,240 psi (lower part of support feet)

6-23

c. Allowable stress in compression, Fa:

2

[1 - (k1 2Ce ]Sy r

Fa "

5 kl kl 3 3

( ) + [3 ( ) 8Cc] -

[( ) 8Ce ]

3 r r 6-20 where:

(2n 2 E) 12

/

Cc

l 3 Sy kl/r for the main rack body is based on the full height and cross section of the honeycomb region.

Substituting numbers, we obtain, for both support leg and honeycomb region:

Fc = 13,890 psi (main rack body)

Fa = 13,890 psi (upper part of support feet)

= 72,360 psi (lower part of support feet)

d. Maximum allowable bending stress at the outermost fiber due tc flexure about one plane of symmetry:

Fb = 0.60 Sy = 13,890 psi (rack body)

Fb = 13,890 psi (upper part of support feet)

= 72,360 psi (lower part of support feet)

e. Combined flexure and compression:

fa Cmx fbx Cmyf by

+ + <1 Fa DxFbx DF y by

Ti t:-

6-24 where:

fa = Direct compressive stress in - the section fxb = Maximum flexural stress along x-axis fby = Maximum flexural stress along y-axis Cmx

=

Cmy = 0.85 6-21 fa Dx = 1 -

F'ex fa Dy=1-F'ey where:

1; 2g F'ex ey = 2 23 ( )

rbx,y and the subscripts x,y reflect the particular bending plane of interest.

I

_,.------+,=,.w--,y--

,---,,e- 7,,we,y,wy, ,,7.,y9,v--w,ww,, w--1, e - w w e v v.-e-- s- w -em+ve--+e

6-25

f. . Combined flexure and' compression (or tension):

fa f bx' fby *

< 1.0 0.6S y Fbx Fby The above requirement should be met for both the direct tension or compression case.

6.7.2 Level D Service Limits F-1370 .(Section III, Appendix F), states that the limits for the Level D condition are the minimum-of.1.2 (Sy /Ft) or: (0.7Su/Ft) times the corresponding limits. for Level A condition. SinceLl.2 S y is less than 0.7 Su for,the lower part of the support feet, 'the factor is 1.40 for the lower section under SSE conditions. The factor for the upper portion of the support foot is.2.0.

Instead ~ of tabulating the results of these six different stresses as dimensioned values, they are presented in a dimensionless form. These so-called stress factors are defined as the ratio of the actual developed stress to its specified limiting value. With.this definition, the limiting value of each stress factor is 1.0 for the OBE and 2.0 or 1.40 for the SSE l condition.

16 . 8 RESUITS FOR THE NEW SPEUT FUEL RACKS Figures 6.1, 6.2, and 6.3 show the pool slab motion in horizontal x, horizontal y, and vertical directions. This motion is for the SSE earthquake.

8 1

6-26 Results are ' abstracted here for the A module (the smaller module) and for'the D-2 module (largest aspect ratio).

1 A complete synopsis of the analysis of the modules A and D-2, subject to the SSE earthquake motions, is presented in a summary Table 6.5 which gives the bounding values of stress factors R1 (i = 1,2,3,4,5,6). The stress factors are defined as:

R1 = Ratio of direct tensile or compressive stress on a ,

net section to its allowable value (note support feet only support compression)

R2 = Ratio of gross shear on a net section to its allowable value R3 = Ratio of maximum bending stress due to bending about the x-axis to its allowable value for the section R4 = Ratio of maximum bending stress due to bending about the y-axis to its allowable value -

R5

= Combined flexure and compressive factor (as t defined in 6.7.le above)

R6 = Combined flexure and tension (or compression) factor (as defined in 6.7.lf above)

As stated before, the allowable value of R1 (i =1,2,3,4,5,6) is I

l 1_for the OBE condition and 2 for the SSE (exceot for the lower i section of the succort where the factor is 1.40) i The dynamic analysis gives the maximax (maximum in time and in space) values of the stress factors at critical locations .

in the rack module. Values are also obtained for maximum rack i displacements and for critical impact loads. Table 6.5 presents r

e l f l

l _ _ _ _ _ _ _ _ _ _ _ - _ - _ _ _ _ _ _ - -

6-27 Table 6.5

, VALUES OF DIMENSIONLESS STRESS FACTORS Stress Factors

R1 R2 R3 R4 R5 R6 (Upper values for rack base, lower values for the most Analysis stressed location in the Number AnalysisHI.D. support foot) -

1.a SSE loads, y = .8 a a a a a a Module A, fully loaded -- -- -- -- -- --

unchannelled fuel b b b b c c 1.b Same as 1. a , - a a a a a a channelled fuel .

a a a b c c 2.a Same as 1.a, p = 0.2 a a a a a a a a a a e c 2.b Same as 1.b, p = 0.2 a a a a a a a a a a e c 3.a SSE loads, Module A, a a a a a a y = .8, half full, -- -- -- -- -- --

unchannelled fuel b b b b c c 3.b Same as 3.a, channelled a a a a a a i fuel, SSE loads, Module A -- -- -- -- -- --

b b b b c c See legend definition on the next page, please.

i; 6-28 Table 6.5 (continued) 4.a Module D2,.all cells a a a a a a occupied, unchannelled -- -- -- -- -- --

fuel, p = 0.8 b b b b c c

'4.b Module D2, all cells a a a a a a occupied, channelled -- -- -- -- -- --

fuel, p = .8 b b b b c c 5.a Module 02, half full a a a a a a unchannelled fuel, - - - -' - -

p = .8. b b b b c c 5.b .Same as 5.a, with a a a a a a unchannelled fuel -- -- -- -- -- --

b b b b c c 6.a Module D2, half full, a a a a a a unchannelled fuel, - - - - - -

SSE loads, p = 0.2 b b b b' c. c 6.b Same as 6.a, except a a a a a a channelled fuel, p = 0.,2 -- -- -- -- -- --

b b b b c c Legend: a: Stress factor less than 50% of the respective allowable b: Stress factor less than 75% of the respective allowable c: Stress factor less than 100% of the respective allowable, i

4 6-29 It is found that the results. corresponding to SSE are most critical in comparison to the corresponding allowable )

limits. The results given herein are for the SSE. The maximum stress factors (R1) are below the limiting value for the SSE condition for all sections. It is noted that the critical load factors' reported 'for the support feet are all for the upper l segment of the foot and are to be compared with the limiting value of 2.0.

Analyses have been carried out to show that significant margins of safety exist against local deformation of the fuel storage cell due to rattling impact of fuel assemblies and against local overstress of impact bars due to inter-rack impact.

Analyses have also been carried out for the OBE condition to demonstrate that the stress factors are below 1.0.

Results obtained for all rack sizes and shapes are enveloped by the data presented herein. Overturning has also been considered for the cases where racks are adjacent to open areas.

6.9 IMPACT ANALYSES 6.9.1 Imcact Loadina Between Fuel Assembly and Cell Wall The local stress in a cell wall is estimated from peak impact loads obtained from the dynamic simulations. Plastic analysis is used to obtain the limiting impact load that can be tolerated. The limit load is more than twice the maximum rattling load.

, - =-~ . .. . . ___ . . .~ ._ . -

?

e v

6-30 5

6.9.2 Imoacts Between Adiac'ent Racks I

I All 'of the dynamic analyses assume, conservatively, that adjacent racks move completely out of phase. Thus, the f' highest- potential for inter-rack impact is achieved. The displacements obtained from the dynamic analyses are less than

50% of the rack-to-rack spacing or rack-to-wall spacing.

Therefore, we conclude that no impacts between racks or between {

racks and walls occur during the SSE event.

i. 6.10 WELD STRESSES critical weld locations under seismic loading are at [

, the bottom of the rack at the baseplate connection and at the !

welds on the support legs. Results from the dynamic analysis f using the simulation codes are surveyed and the maximum loading i

is used to qualify the welds on these locations.

i 6.10.1 Basenlate to Rack Welds and cell-to-cell Welds l f

Section NF permits, for the SSE condition, an aliewable !

weld stress r = .42 Su = 28,600 psi. The calculated weld stress, ,

i based on the highest load factor, is below this value for the j 4

baseplate to rack welds. !

t t

L

, The critical area that must be considered for cell-to-

, cell welds is the weld between the boxes. Where stitch welding is [

i I

i:

L,1:

6-31 used, this weld As continuous near the baseplate but is a stitch weld (2" on 9.0" spacing) as we move up tho' box. The critical shear stress in this weld region for the SSE condition is less than 15000 psi where account is taken ut the stitch welds or spot welds used in this region. Ilear the bottom of the rack the shear stress dominates, and no other stresses on the weld need be considered.

Stresses in the box welds may also develop due to fuel assembly impact with the cell wall near the top of the rack.

This will occur if fuel assemblies in adjacent tubes are moving out of phase with one another so that impact loads in two adjacent cells are in opposite directions which would tend to separate the channel from the tube at the weld. Our analysis demonstrated that the maximum weld shear stress in this area is less than 10000 psi.

f 6.10.2 Jientina of an Isolated Cell Weld stresses due to heating of an isolated hot cell are also computed. The assumption used is that a single cell is heated, over its entire length, to a temperature above the value associated with all surrounding cells. lio thermal gradient ,

in the vertical direction is assumed so that the results are conservative. Using the temperatures associated with this unit, the analysis demonstrated that the stitch welds along the entire cell length do not exceed the allowable value for a thermal loading condition.

-v

r-6-32 a

lfl 4 6.11 RESULTS OF'EXISTIl1G RACK AllALYSIS i

The existing spent fuel racks are : assemblea into six

s. supermodules and were originally qualified using -the response ,

spectrum method. The original qualification.was for supermodules !

braced .against the wall and against each other so that deformation need not be considered. In the requalification, we V examined the supermodule in the southwest corner of the pool which is made up of six modules (2 type A and 4 type B). These modules were assumed to behave as a single rigid rack since they ,

are each. connected together by plate connectors over nearly the full height. DYNARACK was used to requalify the rack module as a rigid rack supported on 24 supports. The compressive spring rates for each of -the support feet locations were adjusted to -

account for the actual number of supports being simulated at that l location.

The rack module is assumed to be free standing and is subject to the 3-D time history. The dimensions and material properties were abstracted from li!!ECO Report "Millstone Unit 1, l Spent Fuel Pool Modifications", July 1976, llRC Docket 11o. 50-245.

i The rack structural properties were calculated based on a cross section of 143" x 146" x 164" (height) gridwork. The supermodule was assumed to be fully loaded with 440 fuel assemblies having weight = 643#. Hydrodynamic forces were incorporated into the analysis using the appropriate rack-rack or rack-wall spacing. Gap elements to simulate inter-rack impact j were included at the baseplate level since the design of the !

i L

- - , . s w-..w4 , - . . v,----~,m,e.,._w r,--v.,--, ,.,,_.,_-,_.,_w_.--y,,,,,,g,,y.-9 ,, _. -,m_ .,,.,,-.--y, - . . , _ _ , - - - - , - - - - - --

6-33 existing modules is such as to permit contact at the baseplate level. Simulations were carried out for coefficients of .8 and

.2 with the rack beidg' full'for each case.

The dynamic analysis showed that maximum deflections in either direction were below .45" at the top of the supermodule.

Since this is less than 50% of the rack-to-rack spacing, no impacts will occur when the racks are considered free-standing.

Fuel assembly to cell impact loads are less than 1600# per cell which is well below the allowable load for the existing ra,ck structure (cell wall thicknesses = 0.125"). Impact loads between racks, at the permitted hard points at the rack base plate level, were below 20,000#.

The baseplate is considered as a rib reinforced structure. The critically loaded beams were re-checked for structural integrity, and found to be stressed below the material allowable. The requalification of the existing racks as free standing supermodules is based on the premise that the connector plates between individual modules in a supermodule can effect the appropriate shear transfer between the modules. These .125" connector plates were subjected to the maximum in-plane shear stresses implied by the DYNARACK seismic results and were found to neither buckle nor induce an overstress in the attachment welds. Structural checks, similar to those mandated for the new racks, were carried out on the existing racks and the appropriate Ri factors are within SSE allowables.

t

6-34 6.12 FLOOR SLAB AND BUILDING ANALYSIS

- later -

6.13 DEFINITION OF TERMS USED IN SECTION 6.0 S1, S2, S3, S4 Support designations pi Absolute degree-of-freedom nurler i qi Relative degree-of-freedom number i y Coefficient of friction Ui Pool floor slab displacement time history in the i-th direction x,y coordinates horizontal direction z coordinate vertical direction K

I Impact spring between fuel assemblies and cell K Linear component of friction spring f

Ks Axial spring of support leg locations N Compression load in a support foot K

R Rotational spring provided by the pool slab Subscript i When used with U or X indicates direction (i = 1 x-direction, i=2 y-direction, i = 3 z-direction)

6-35 6.14 REFERE!1CES 6.1 US11RC Standard Review Plan, 11UREG-0800 (1981).

6.2 ASME Boiler & Pressure Vessel Code,Section III, Subsection 11F (1983).

6.3 U S !! R C Regulatory Guide 1.29, "Seismic Design Classification," Rev. 3, 1978.

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

6.5 USilRC Regulatory Guide 1.92, "Combining Modal Responses and Spatial Components in Seismic Response Analysis,"

Rev. 1, February, 1976.

6.6 "The Component Element Method in Dynamics with Application to Earthquake and Vehicle Engineering," S.

Levy and J.P.D. Wilkinson, McGraw Hill, 1976.

6.7 "Dynamics of Structures," R.W. Clough and J. Penzien, McGraw Hill (1975).

6.8 "Mechanical Design of Heat Exchangers and Pressure Vessel Components," Chapter 16, K.P. Singh and A.I.

Soler, Arcturus Publishers, Inc., 1984.

6.9 R.J. Fritz, "The Effects of Liquids on the Dynamic Motions of Immersed Solids," Journal of Engineering for Industry, Trans. of the ASME, February 1972, pp 167-172.

6.10 "Dynamic Coupling in a Closely Spaced Two-Body System Vibrating in Liquid Medium: The Case of Fuel Racks,"

K.P. Singh and A.I. Soler, 3rd International Conference on lluclear Power Safety, Keswick, England, May 1982.

6.11 US!!RC Regulatory Guide 1.61, "Damping Values for l Seismic Design of !!uclear Power Plants," 1973.

l r

6-36 e

O O

_ _ _ _ -_ g I

I I l l :o I I i i 1 -O N

I I I I .I I I I I I O

-O l_____J_____! ' _ _ _ _

J_______g i i

__-.- i i :O I Ld 1 1 1 -O

~

(A i - N l [

O ~_- :OA i ; i ;

-O l _ _ _ _ g J _ _ _ _ _ w _ _ _ J _ _ _ _ ._ J _ _ _ _ _ . g Q l 8 g I l- - l l @M-I I I : v e 1 -'

I O w

-O x l _ _ . _C J -: --_ '

5 I 9i

~

J______-6 I :o 0 E I

r.

t_

1 =

~

I I -O O 1 I l : '

l

- - i l :o

-O l _ _ _ ___J _

- _ _ -- _ J. y

-O 1 i 37 ! 1

O

_o g I I -

1 I i i _ D__ I I :

1 I -D l 1 E

i_____J___ -

_____J_____J-----:OOF-- -

1 I I I l

-O 1 1 I I i :

I I l 1 I .

O I I I I i :O i i i

i..... ie 3.ei...e iigiie iei iiij iei..eie i iiiiiiieiig o o O O O 00 v N O, N t ql o

6 6 o o o l l l l l s,3 303y

6-37 O

O i I l-

______d 1 I I :O o' I I I I I -O I I I I I b I I I I I O

-O l_____J__

i l_____J_____J_______g I

1 I :O j j -o I I

==E__

l 1

N i wg -~_---- g ; .O A m -O '

l____Wg J _ _ _ _g ___J_____J_____._ g (_)

1 1 - 1 I :Oq)

I hl W l 1

- O (/)

1 Wl _

l I : v I 3i i I O ~

j _ _ _ _ =l -i v. _ a. J

_ _ _ _ j _ _ _ _ _ j _ _ _ _ _ _.- 0 0 Wl

=-

l m

1 I :O E i yI __ _

l _ _z____ =

i C #U

! ! I ~ -% i : 2 I._ ,

_ _ _ J _ _ _ _ n- _ J ._ _ _ _ J _ _ _ _ . OL O -

I I I I I I -n i I i i 1 1 :~

l i I I I I :g i I I I I I :O t i

_I

.........g.i.....s.isiiesee.s;iisissoieg..iiiiiiigiiiiiiisegO i !

-d o o o O O O N

00 '

i n N " 9 "

91 6 6 6 o o o o .

I l l I (s,e) 3 333y

6-39 83

7

/ /
^
/ /2*
4. - --- - ypg P7 g6 HI#
j o 3*
= P 10 a
RACK GEOMETRIC CENTER LINE N '9 H/4 H ,4
P12 H/4 P11 o5'
'P14 ja
. 4 /l H/4 A /c S4 P13 #5 /f f
.O
/$
~
Y I [
/g y(C "=
/, X 1
' Y --> ag
.. A of 3
< Yg :e g[ 1-p LCNG OlRECTION M swe. .
7q/ A y /
s
-gg . s g :.
ca TYP. FRICTION 7 ELEMENT 3'
't i
i
! SOHEMATIC MCCEL FC A OYNR AOK FIGURE 6.4 i
l
6-40 TYPIC AL TCP IMP ACT ELEMENT ,
s
/
r' 5
f
^
M &$ '
a v i
' m h7 3 r .
~
[(M R ACK STRUCTURE TYP. ECTTCM IMP ACT ELEMENT f
~
^&
/
A
~ m
'/\
v- /
y s'
v
/
v
'v /=
R ACK.TC A ACX IMP AC~ SP RINO3 FIGURE 6.5
6-41 L
Y 4 .
CELL WALL 7
I f
M ASS i Xg [.
y 4l
. / ,! FUEL ASSEMBLY! CELL g IMP ACT SPRING / / / /
Yg
/ / / / / >X ,
IMP ACT SP AING A R R AN0!YENT AT NC::E i FIGURE 6.6
6-42 FL'EL ASSY/ CELL IMPACT ~ SPRING , X I
%M A A l
0.25H F/2 %
~ VIE.W I A
0.25H RACK N y s qh[w U h n g
-N1 fv- f l A TY?ICA* ~
FJi n :NG MASS 0.25H s; I .. ra:c::N 1
-M D tv- T c, $f $ $ N-INT m C: S S??:N3, K i ? 2 ;-/ / VA- ,- - 6 C w ,, . s. -
FCL'NATICN EGTAT:CNAL COMPLIANCI SPR:NG, K ^ R FIGL?.E 6.7 Ta'0 DIMENSIONAL VII'a' CF TdE SPRING-MASS S:ML'LAT CN
i i n 7-1 . 7.0- OTHER MECHAllICAL LOADS 7.1 MECHAllICAL LOADIllGS 7 .1.1 - Fuel Handling In addition to the seismic analysis presented in Section.6.0, the racks were also analyzed for several mechanical loading and accident conditions. The fuel handling bridge crane is capable of exerting a 900 lb. downward thrust and a 1700 lb. upward thrust on a fuel
~
assembly during fuel manipulation. The 900 lb. downwards thrust,. together with 800 lb. fuel assembly weight, produces a net downwarda load of 1700 lbs. on the rack. Calculations show that a load of 1700 lb. (in upwards or downwards direction) applied on a 1" characteristic dimension of the rack will produce a local stress of approximately 14000 psi. Since the yield stress of the ASTM 240-304 material is 25000 psi, even local plasticity is precluded. 7.1.2 Dropped Fuel Accident I A fuel assembly (weight - 800 pounds) is dropped from 36 inches above a storage location and impacts the base. Local failure of the baseplate is acceptable; however, the rack design should preclude impact with the pool liner. The suberiticality of the adjacent fuel assemblies is not to be violated. Calculated results show that the baseplate is not pierced and the rack feet loading on the liner is well be. low that caused by
7-2 seismic loado. The maximum depth of basaplate penetration is conservatively shown to be less than the baseplate nominal thickness. 7.1.3 Dropped ,Eugl Accident II One fuel assembly dropping from 36 inches above the rack and hitting the top of the rack. Permanent deformation of the rack is acceptable, but is required to be limited to the top region such that the rack cross-sectional geometry at the level of the top of the active fuel (and below) is not altered. Analysis dictates that the maximum local stress at the top of the rack is limited to 21000 psi which is less than material yield point. Thus, the functionality of the rack is not affected 7.2 LOCAL BUCKLING OF FUEL CELL E1LLS The allowable local buckling stresses in the fuel cell walls are obtained by using classical plate buckling analysis. The following formula for the critical stress has besn used (1). rt 2 Et2 acr " (1) 12 b2 (1 _ 2) where E = 27 x 106 psi, t = .075, b=6.0". The factor p is suggested in Reference 7-1 to be 4.0 for a long panel loaded as shown in Fig. 7.1.
7-3 For the given data Ucr < 10000 psi It shculd be noted that this calculation is based on the applied , stress being uniform along the entite length of the cell wall. In the actual fuel rack, the compressive stress comes from consideration of overall bending of the rack structure during a saismic event and as such is negligible at the rack top and maximum at the rack bottom. It is conservative to apply Eq. (1) to the rack cell wall if we compare a cr with the maximum compressive stress anywhere in the cell wall. 7.3 AllALYSIS OF WELDED JOIliTS Ill RACK Welded joints are examined under the loading conditions arising frem thermal effects due to an isolated hot shell, and due to seismic loadings. Under both sets of load conditions, the weld stresses are found to be below the allowable value of 24000 psi in shear that is given in Table 11F3 2 9.1-1 of AS!4E Section III, Division 1, Subsection 11F, 1980. A. A thermal gradient between cells will develop when an isolated storage location contains a fuel assembly emitting maximum postulated heat, while the surrounding locations are empty. We can obtain a conservative estimate of weld stresses along the length of an isolated hot cell by considering a beam strip uniformly heated by 20 F,0 and restrained frcm growth along one long edge. the configuration is shown in Figure 7.2. L l
T i [ 7-4 , i Using a shear beam-theory, an'd . subjecting the strip to a j uniform temperature rise AT = 20 0F, we can calculate an ;
estimate of the maximum value of the average shear stress in I the strip. The strip is subjected to the following boundary t conditions. ,
a e
'a. Displacement- Ux (x,y) = 0 at x = 0, all yt and at y =
w/2, all x. *
b. Average force Mx, acting on the cross section Hxt = 0 ,
at x = L, all y. ; The final result for wall' shear stress, maximum at x = L, is l found to be given as: -I EaAT
.931 where E = 28 x 106 psi, a = 9.5 x 10-6 in/in 0F and AT = 20 0p, ;
Therefore, we obtain an estimate of maximum weld shear stress in I an isolated hot cell, due to thermal gradient, as , IMAX = 5775 psi 8 B. The critical weld locations, when the loading is seismic, ' are at the bottom of the rack (at the connection to the l a baseplate) and in the welds on the support legs. The results ! from the dynamic analyses using DYNARACK are surveyed and l the maximum loading used to qualify the welds in these ; locations. The ASME Code allowable value of 27300 psi is l used on allowable weld stress. All welds are qualified using SSE seismic results and are based on limit strength of ; the support plate to baseplate weld configuration.
,.._ ,.---,-- _.-- - _ ,. ._~. _ ..._- _ - _----, , _ --
7-5 7.4 Cask Drop Accident The postulated cask drop accident is discussed in Section 8.0.
. _ _ - + - -
7-6
7.5 REFERENCES
7.1 Strength of Materials, S.P. Timoshenko, 3rd Edition, 1956, Part II, pp. 194-197.
7-7 t ' a 1
. -: a ~ - 4 + a * + ->4
> - b + c b m
+ -~i v ~ -
FIG. 7.1 Lb ADING ON R ACK W ALL 4 t .
-. + t A
Heated C e ll W a ll
= =x . H I
r s s s s s s s s s s s s s s' - i y g 3 W eld t i Line T Y FIG. 7.2 . . ' ~ WELDED JOINT IN R'ACK l l L.
8-1 8.0 RADIOLOGICAL CONSIDERATI0!1S This section addresses the radiological consequences of increasing the spent fuel storage capacity from 2184 locations to 3229. Both the direct radiational dose to operating personnel under normal operating conditions and the pc,tential dose consequences expected during accident conditions were considered. 8.1 OPERATIllG PERSOllllEL EXPOSURE The increase in the spent fuel storage capacity from 2184 locations to 3229 should not increase the direct radiational dose to operating personnel because of the extensive shielding r provided by the water depth above the top of the fuel and the i arrangement of the surrounding building structures as delineated below: O Water depth above the fuel is 24 feet. O The east and west walls are six feet thick. O The floor is five feet, four inches thick. ! O The south wall is of varying thickness and faces the reactor cavity. O The north wall is six feet thick for the lower 23.1 feet and four feet, six inches thick for the next fifteen feet, seven and three quarter inches above that. This shielding is sufficient to ensure that dose rates from the spent fuel are insignificant and, in most cases, undetectable. l r I
8-2 8.2 ACCIDENT CONDITIONS 8.2.1 Shinpino Cask Drop .'.ccident The drop of a shipping cask on the spent fuel storage racks is precluded since the crane and rigging used to lift the cask are single failure proof (Ref. 8-1). In addition, crane interlocks and limit switches prevent crane travel over irradiated fuel assemblies when handling shipping casks. 8.2.2 Dropped Fuel Accidents Fuel assembly drop accidents from 17 feet above the racks impacting both the rack tops and baseplate were considered. These accidents are bounded by the previously analyzed FSAR (Ref. 8-2) irradiated fuel drop accident onto the reactor core following a 24 hour decay. The dose consequences for this accident are well within 10CFR100 limits as shown below. Dose Consequence (Rem) Thyroid Whole Body Exclusion Area Boundary 1.54 0.110 Low Population Zone 0.05 0.036 10CFR100 limit 300 25 The fuel drop accidents onto the spent fuel racks are bounded by this accident since the maximum drop height is 17 feet as compared with thirty for the reactor core drop accident, thereby bounding the potential energy and the resultant ability to cause cladding damage. In addition, the irradiated spent fuel would have decayed for a longer period of time.
4 i 8-3
8.3 REFERENCES
8.1 B.K. Grimes, "OT Position for Review and Acceptance of Spent Fuel Storage and Handling Applications," April 14, 1978, and as amended January 18, 1979. 8.2 Millstone Unit One Final Safety Analysis Report. 1 4
T l 9-1 9.0 IN-SERVICE SURVEILLAllCE PROGRAM FOR BORAFLEX 9.1 OVERVIEW A long term surveillance f.regram will be implemented to ensure continued acceptable performance of the Boraflex neutron poison material used in the spent feel racks. This program will be based on current performance information for the Boraflex neutron poison material. Proper documentation will be obtained from the manufacturers of Boraflex and the racks to assure the quality of the neutron poison material and its proper loading in the racks. Visual inspection of the racks will be performed during rack construction to verify that the Boraflex is installed in each of the specified locations in each rack. Uniquely identified samples, taken from material representative of that used as a neutron absorber, will be placed within surveillance capsules and placed within the racks. At specified intervals the samples will be removed and tested for boron content and physical integrity. Irradiation tests have been previously performed to determine the stability of Boraflex in boric acid solution. The results of these tests are documented in test reports of the Bisco Corporation (Refs. 9-1, 9-2 and 9-3). From these tests, there is no evidence indicating any deterioration of the Boraflex naturial through a cumulative irradiation in excess of 1x 10 11 rads gamma affecting the suitability of Boraflex as a neutron
9-2 poison material. Under the proposed surveillance program, calculations have shown that the specimens would require at least five years in the pool environment to approach the level of , cumulative exposure. Direct dosimetry will be utilized, however, to establish an accurate record of cumulative exposure. Periodic testing and examination of poison specimens will take place beyond the point at which cumulative exposures exceed those of documented tests. , The surveillance specimens will initially be examined after i approximately five years of exposure in the pool environment. Several specimens will be checked for physical integrity and other general performance characteristics. This examination will include visual inspection as well as other tests to verify the material stability. This initial surveillance will be used to verify that the performance of the Boraflex is consistent with . current Bisco Products (a division of Brand, Inc.) tast results. Based on the results of the initial surveillance, and results from existing fuel Jack surveillance programs at the 14111 stone
!!uclear Station Units 2 and 3, additional testing will be scheduled to assure acceptable material performance throughout the life of the plant.
i i
i 9-3 9.2 REFERE!!CES , 9.1 J.S. Anderson, "Boraflex lieutron Shielding Material-- Product Performance Data," Brand Industries, Inc., Report 748-30-2 (August 1981). 9.2 J.S. Anderson, "Irradiation Study of Boraflex lieutron Shielding Materials," Brand Industries, Inc., Report 748-10-1 (August 1981). ; 9.3 J.S. Anderson, "A Final Report on the Effects of High Temperature Borated Water Exposure on BISCO Boraflex lieu' tron Absorbing Materials," Brand Industries, Inc., Report 748-21-1 (August 1978). r I I 1 h l l
10-1 10.0 COST ASSESSMEllT AtID Et1VIRCtlMEt1TAL IMPACT ll!!ECO has prepared a cost / benefit assessment of the spent fuel rack modifications in accordance with the requirement i of Section V, Part 1 of Ref. 10-1. The assessment demonstrated that the installation of high density spent fuel storage racks is the most advantageous means of providing additional, necessary spent fuel storage considering public safety and projected costs. 10.1 COST ASSESS!tEllT : 10.1.1 tieed for Increased Storage Capacity i A. Il!!ECO currently has no contractual arrangements with any fuel reprocessing facilities. B. Adoption of this proposed spent fuel storage , expansion would not necessarily extend the time ! period that spent fuel assemblies would be stored j on site. Spent fuel could be sent off site for ' final disposition under existing legislation, but i the govern: cent facility is not expected to be l available before 1998. As matters now stand and until alternate storage facilities are available, spent fuel assemblies on site will remain there. C. It is estimated that the spent fuel pool will be filled, with the proposed increase in storage capacity, in 1999. 10.1.2 fanstruction Cests Tetal construction cost associated with the proposed i modification is approximately 5 million dollars. This figure ; includes the cost of designing and fabricating the spent fuel ! k
10-2 racks, engineering costs, and installation and support costs at the site. 10.2 ENVIROUl1 ENTAL IMPACT g This section discusses the environmental impact due to the spent fuel pool rack modifications. Environmental effects considered include radiological, chemical discharge, and heat dissipation effects. O Padiolocical Effects As discussed in Section 8.0, the increase in spent fuel pool storage capacity will not increase the direct radiational dose to operating personnel, nor will it increase the potential dose consequences expected during accident conditions. O Chanical Dischargon The only chemical discharge that could be affected by the proposed expansion af the spent fuel pools is the powdered ion exchange resins used in the two filter demineralizers. The frequency of resin replacement is determined primarily by the need for water clarity. The particulate material that must be removed to maintain water clarity enters the water during refuelings and is removed well before the next refueling. Therefore, the frequency of resin replacement is independent of the number of spent fuel assemblies stored in the pools and there should be no change in the amount of spent fuel pool purification system filter recin discharged from the plant as a result of the increase in storage capacity.
r 10-3 0 llent Dissipation The two Hillstone Unit tio . I spent fuel pool cooling system heat exchangers are designed to each transfer a total of 7.3 x 106 Btu /hr. It is estimated that the increased storage-will result in about a 10 't increase in heat release when the pool is filled, or an increase of approximately 7.3 x 105 Btu /hr based on the heat exchanger design. When compared to the over 5 x 109 Btu /hr discharged into the environment by each unit, this increase is seen to have a negligible offect on the environment. P F F F 2 P I
10-4
10.3 REFERENCES
10.1 B.K. Grimes, "OT Position for Review and Acceptance of Spont Fuel Storago and llandling Applications," April 14, 1978. I i f k 1 i I f
APPEtiDIX A ~ BE!1CHMARK CALCULATIOtiS f A-1
0 i
1. INTRODUCTION AND
SUMMARY
i The objective of this benchmarking study is to verify both the AMPX (NITAWL)-KENO (Refs. 1 and 2) methodology with the 27-group SCALE cross-section library (Ref s. 3 and 4) and the CASMO- f l 2E code (Refs. 5, S, 7, and 8) for use in criticality calcula- l tions of high density spent fuel storage racks. Both calcu-lational methods are based on transport theory and have been ;
benchmarked against critical experiments that simulate typical f spent fuel storage rack designs as realistically as possible.
] Results of these benchmark calculations with both - methodologies l are consistent with corresponding calculations reported in the [ literature and with the requirements of Regulatory Guide 3.41,* Rev. 1, May 1977. Results of these benchmark calculations show that the l l 27-group (SCALE) AMPX-KENO calculations consistently underpredict ! . the critical eigenvalue by 0.0106
0.0048 ak (with a 95% proba-bility at a 95% confidence level) for critical experimentr selected to be representative of realistic spent fuel storage f
rack configurations and poisen worths. Similar calculations by l Westinghouse suggest a bias of 0.012
0.0023, and the results of 1
! ORNL analyses of 54 relatively "clean" critical experiments show [ l a bias of 0.0100 6 0.0013. Similar calculations with CASMO-2E for clean critical i experiments resulted in a bias of 0.0013 0.0019 (951/951). I ! CASMO-2E and AMPX-KENO intercomoarisen calculations of infinite { l ' arrays of poisoned cell configurations shew very god agreement I and sugges: that a bias of 0.0013 0.0018 is tl.e reasonably expected bias and uncertainty for CASMO-2E esiculations. Validation of Calculational Methods for Nuclear Criticality l Safety. (See also ANSI N16.9-1975.) A-2
The benchmark calculations reported here indicate that either the 27-group (SCALE) AMPX-KENO or CASMO-2E calculations are acceptable for criticality analysis of high density spent fuel storage racks. The preferred methodology, however, is to perform independent calculations with both code packages and to utilize the higher, more conservative value for the reference design infinite multiplication factor.
2. AMPX (NITAWL)-KENO BENCHMARK CALCULATIONS Analyr.s of a series of Babcock & Wilcox (B&W) critical experiments (Ref. 9), which include some with absorber sheets typical of a poisoned spent fuel rack, is surcarized in Table 1, as calculated with AMPX-KENO using the 27-group SCALE cross-section library and the Nordheim resonance integral treatment in NITAWL. The mean for these calculations is 0.9894 0.0019, conservatively assuming the larger standard deviation calculated from the kegg values. With a one-sided tolerance factor (K = 2.502), corresponding to 951 p r o b a b i l i t',' at a 95% confidence level (Ref. 10), the calculational bias is +0.0106 with an uncer-tainty of 0.0048.
Similar calculational deviations reported by Westinghouse (Ref, 11) are also shown in Table 1 and suggest a bias of 0.012 0.0023 (95%/95%). In addition, ORNL (Ref. 12) has analyzed seme 54 critical experiments using the same methodology, obtaining a mean bias of 0.0100 0.0013 (951/95%). These published results are in goed agreement with the results obtained in the present analysis and lend further credence to the validity of the 27-group AMPX-KENO calculational model for use in criticality analy-sis of high density spent fuel storage racks. Variance analysis of the data in Table 1 suggests the possibility that an unknown factor may be causing a slightly larger variance than might be expected frem the Monte Carlo statistics alene. However, such a A-3
i I Table 1 : RESULTS OF 27-GROUP (SCALE) AMPX-KESO CALCULATIONS r
~
OF B&W CRITICAL EXPERIMENTS ' ( Westinghouse ! Experiment Calculated Calculated-meas. l Number k,gg e k,gg i I 0.9839 *0.0049 -0.008 l II 1.0040 0.0037 -0.012
III 0.9985 *0.0046 -0.008 IX(1) 0.9924 0.0046 -0.016 ;
X 0.9907 *0.0039 -0.008 j XI 0.9989 *0.0044 +0.002 ! l XII 0.9932 *0.0046 -0.013 j XIII 0.9890 *0.0054 -0.007 g XIV 0.9830 *0.0038 -0.013 [ XV 0.9852 *0.0044 -0.016 i XVI 0.9875 to.0042 -0.015 . a XVII 0.9811 to 0041 -0.015 , XVIII 0.9784 0.0050 -0.015 t XIX 0.9883 0.0033 -0.016 j XX 0.9922 *0.0048 -0.011 l
XXI 0.9783 *0.0039 -0.017 Mean 0.9894 *0.0011(2) -0.0120 t 0.0010 !
I Bias 0.0106 0.0019 I3} 0.0120
0.0010 Bias (95%/95%) 0.0106 to.0048 0.0120 0.0023 l Maximum Bias 0.0154 0.0143 I
(1) Experiments IV through VIII used B 4C pin absorbers and were ! considered representative of poisoned storage racks. (2)not (3)Calculated Calculated fr:m fr:m k,gg individual values and standard used asdeviations. reference. I l I
t b
A-4 l l _ _ _ . .--_-_______ _ _-__ - ._- i
factor, if one truly exists , is too s=all t. be resolved on the basis of critical-experiment data presently available. No trends in kegg with intra-assembly water gap, with absorber sheet reactivity worth, or with soluble poisen concentration were identified.
3. CASMO-25 BENCHMARK CALCULATIONS 3.1 GENERA:,
The CASMO-O! code is a multigroup transport theory code utilicing transmission probabilities to accomplish two-di=en-sional calculations of reactivity and depletion for BWR and PWR fuel asse : lies. As such, CASMO-2E is veil-suited to the criti-cality analysis of spent fuel storage racks, since general practice is to treat the racks as an infinite medium of storage cells, neglecting leakage effects. CASMC-2E is closely analegous to the EPR!-CPM code (Ref. 13) and has been extensively benchmarked against het and cold crit-ical ex:eriments by Studsvik Energiteknik (Refs. 5, 6, 7, and 8). Reperted analyses of 26 critical ex:eriments indicate a mean kegg of 1.000 0.0037 ( le t . Yankee Atemic (Ref. 14) has also reported results of extensive benchmark calculations with CA5MO-2E. Their analysis of 54 Strawbridge and Sarry critical experi-ments (Ref. 15) using the reported buckling indicates a rean cf 0.99B7 0.0009 (le), or a bias of 0.0013 0.0018 (with 951 pr:bability at a 95% confidence level). Calculations were repeated f:r seve; of the Strawbridge and Sarry experiments K Significantly large trends in k , with water gap and with ab-sorber sheet reactivity worth have,,'5een re;crt1d ( Re f . 16 ) for AMPX-KENO calculations with the 123-group GAM-THERMCS library. A-5 e
selected at randem, yielding a mean ke gg of 0.9987
0.0021 (le),
thereby confirming that the cross-section library and analytical methodology being used for the present calculations are the same as those used in the Yankee analyses. Thus, the expected bias for CASMO-2E in the analysis of "clean" critical experiments is 0.0013 t 0.0013 (951/95%). 3.2 BENCHMARK CALCULATIONS CASMO-2E benchmark calculations have also been made for the B&W series of critical experiments with absorbe sheets, simu-lating high density spent fuel storage racks. Hewever, CASMo-2E, as an assembly code, cannot directly represent an entire cere configuratien. without introducing uncertainty due to reflector constants and the appropriateness of their spectral weighting. For this reason, the poisoned cell configurations of the central assembly, as calculated by CASMO-2E, were benchmarked against L
' corresponding calculations with the 27-group (SCALE) AMPX-KENO code package. Results of this ccmparisen are shewn in Table 2.
Since the dif ferences are well within the n:rmal KENO statistical variation, these calculations confirm the validity of CASMO-2E calculations f:r the typical high density poisoned spent fuel ! rack confi;urati:ns. The differences sh wn in Table 2 are also consistent with a bias of 0.0013 0.0013, de te r-ined in Ee: tion 3.1 as the expe::ed bias and uncertainty of CASMO-2E calcula-tiens. Yankee has atte p ed such calculations (Ref. 14) using CASMO-2E-generated constants in a two-dimensional, four-group POO =cdel, obtaining a : ean k,gg of 1.005 for 11 p isoned cases and 1.009 for 5 unpeisened cases. Thus, Yankee benchmark calculations suggest tnat CA5Mo-2E tends to slightly everpredict reactivity. A-6
Table 2 RESULTS OF CASMO-2E BENCHMARK (INTERCOMPARISCN) CALCULATIONS t k- (1) B&W Experiment No.(1) ,AMPX-XEN0(2) CASMO-2E Ak XIX 1.1203
0.0032 1.1193 0.0010 !
XVII 1.1149
0.0039 1.1129 0.0020 >
XV 1.1059
0.0038 1.1052 0.0007 Interpelated(3) 1.1024

0.0042 1.1011 0.4013 XIV 1.0983

0.0041 1.0979 0.0004 1 XIII 1.0992

0.0034 1.0979 0.0013 Mean i 0.0038 0.0011 Uncertainty 0.0006 SWR f'Je1 ra:k 0.9212

0.0027 0.9218 -0.006 (1) Infinite array of central assemblies of 9-assembly B&W criti- !
(2) cal fconfigura:icn t:m AMP.t-KENO (Ref. 9). c rrected for bias of 0.0106 ok. (3)kI5:e:pc'ated item Fig. 25 of Ref. 9 to soluble boren concen-
ation a: critical conditien. ,
t t I I a l e i A-7
REFERENCES TO APPENDIX A
1. Green, Lucious, Petrie, Ford, White, Wright, "PSR-63/AMPX-1 (code package), AMPX Modular Code System for Generating Coupled Multigroup Neutron-Gamma Libraries from ENCF/B,"
ORNL-TM-3706, Oak Ridge National Laboratory, March 1976.
2. L. M. Petrie and N. F. Cross, "KENO-IV, An Improved Monte Carlo Criticality Program," ORNL-4933, Oak Ridge National Laboratory, November 1975.

3. R. M. Westfall et al., "SCALE: A Modular Cods System for Perf:rming Standardized Computer Analyses for Licensing Evaluation," NUREG/CR-0200, 1979.

4. W. E. Ford, III et al., "A 213-Neutron Group Master Cross-section Library for Criticality Saf ety Studies," ORNL/TM-4, 1976.

5. A. Ahlin, M. Edenius, H. Haggblem, "CASMO - A Fuel Assembly Burnup Program," AE-RF-76-4153, Studsvik report ,
(pr:prietary). l
6. A. Ahlin and M. Edenius, "CASMO - A Fast Transport Theory Depletion Code f er LWR Analysis," ANS Transactions, Vol. 26, ,

p. 604, 1977. ,

7. M. Edenius et al., "CASMO Senchmark Report," Studsvik/RF-l 7S/6193, Aktietolage: At:mener;i, March 1973. ,
f
8. "CASMO-2E Nuclear Fuel Assembly Ana'ysis, A plication Users l
Manual," Rev. A, Control Data Corporatien, 1932.
9. M. N. Baldwin et al., "Critical Experiments Sup crting Close Pr:ximity Water Storage of P wer Reactor Fuel," S AW-14 3 4-7, The Bab:::k & Wilecx C:mpa".y, July 1979.

10. M. G. Natrella. Exreri ental Statistics, National Sureau of ;
Standards, Handt::s 91, August 1963. al., "Cc paris:ns of Experi-ents and
11. B. F. C::ney et Cal:ulati:ns f:: LWR Storage Geome: ries," Westinghouse NES ,
ANs Trans actions , vol. 39, p. 5 31, N:vember 19 81. ' Knight, "S: ale System Cross-se: tion } 12. R. M. Westfall and J. R. Validation with Shipping-cask Critical Experiments," ANS Transactions , Vol. 33, p. 368, Novem:er 1979, i
13. "The EPRI-CPM Data Library," ARMP Computer Code Manuals _,
' Part I!, Chapter 4, COM3, Electric F wer Researen Inst.tute, November 1975. i i A-8 :
REFERENCES TO g?ENCTX A (Continued)
14. E. E. Pilat, "Metheds for the Analysis of Boiling Water Reacters (Lattice Physics)," YAEC-1232, Yankee Atemic Electric Co., Cecember 1980.

15. L. E. Strawbridge and R. F. Barry, "Criticality Csiculatiens for Unifer., Water-medersted Lattices," Nuclear Science and Engineerine, Vol. 23, p. 53, September 1965.

16. S. E. Turner and M. K. Gurley, "Evaluation of AMPX-KINO Benchtsrk Calculations for High Censity Spent Fuel St: rage Racks," Nu: lear Science and Engineerine, 80(2): 230-237, February 1952.
i i l l A-9 ' _____.____1}}

B12844, Rev 0 to Spent Fuel Pool Capacity Expansion Project Description & Sar

Contents

Text

1.0 INTRODUCTION

SUMMARY

4.8 REFERENCES

5.7 REFERENCES

7.5 REFERENCES

8.3 REFERENCES

9.2 REFERENCES

10.1 INTRODUCTION

10.4 REFERENCES

SUMMARY

SUMMARY

4.8 REFERENCES

7.5 REFERENCES

8.3 REFERENCES

10.3 REFERENCES

SUMMARY

Navigation menu

B12844, Rev 0 to Spent Fuel Pool Capacity Expansion Project Description & Sar

Text

1.0 INTRODUCTION

SUMMARY

4.8 REFERENCES

5.7 REFERENCES

7.5 REFERENCES

8.3 REFERENCES

9.2 REFERENCES

10.1 INTRODUCTION

10.4 REFERENCES

SUMMARY

SUMMARY

4.8 REFERENCES

7.5 REFERENCES

8.3 REFERENCES

10.3 REFERENCES

SUMMARY

Navigation menu

Search