ML20029C190

Criticality Analysis of Seabrook Station New & Spent Fuel Storage Racks,Feb 1991.

ML20029C190
Person / Time
Site:	Seabrook
Issue date:	02/14/1991
From:	Adli D, Caccipouti R, Napolitano D YANKEE ATOMIC ELECTRIC CO.
To:
Shared Package
ML20029C186	List: ML20029C185 ML20029C187 ML20029C190
References
YAEC-1778, NUDOCS 9103260401
Download: ML20029C190 (63)

Text

9 h

w. ...

. [ ,

J7--- YANKEE ATOMIC ELECTRIC COMPANY i  :

ii 4'

I ,~ .~ . . - -

) I, f l 3 .

I i

k h

" [

u j 1 .

. YAuxei:

8 1 g i

5- ,

I o ,

w L

i]'c f'

-t een -

g.,

i

-u L.

k.

\. 9103260401 s t o 3 t cs I P DFi (4 DOC F 05,0004t.3 P PDR

__ ~ .- ___. _.-. _ . - . . .. .._. . - . . . _ . . . - _ - - . . . - _ . . . _ . - - _ _

I l

, I I: C a

W

! Criticality Analysis of Seabrook Stations's

,g

=

New and Spent Fuel Storage Racks February 1991 by D. G. Napolitano D. G. Adli l

i i

I I

I l I '

I I

,I Yankee Atomic Electric Company i Nuclear Services Division 1 580 Main Street i g Bolton, Massachusetts 01740 l

l.l

'I I ,

1

---_.-.. - . . . - _ . . - - - - . _ _ , _ - . - - . .-.-1

I I

Prepared by: hhetu 2/TN!dto d//Y Y/

D. G. Napolitano Senior Engineer /(Dat'e )

heactor Physics roup i

Nuclear Engineering Department hamilak Of 6(A/C D. G . Adli, Nuclear Engineer A /kl Ai I (Date)

I Reactor Physics Group Nuclear Engineering Department Approved by: A ddLA 9/

R //J . Cacci ti, Manager / (Dat'e )

Rtractor Phy a Group I- Nuclear Eng nemiing Department Ys&$) f l. k l kl li .~ C. Slifez, Director (Datfe)

Nuclear Engineet $g Department i

I I

I I

t I

t I j I 11 l

l DISCLATMER OF RESPONSIBILTTY l

l This document was prepared by Yankee Atomic Electric Company

(" Yankee"). The use of information contained in this document by anyone other than Yankee, or the Organization for which this document was prepared a under contract, is not authorized and, with respect to any unauthorhed use, neither Yankee nor its officers, directors, agents, or employees assume any obligation, responsibility, or liability or make any warranty

~

or representation as to the accuracy or completeness of the material contained in this document.

l t

I I

i I

i iii

I~

APETRACT I This report presents a revi6 ion to the criticality analysis for i Seabrook Station's new and spent fuel storage racks. The original analysis justified the placement of fuel with enrichment up to 3.5 w/o '85U in the new fuel vault and spent fuel racks. This analysis jubtifies the placement of fuel with enrichment up to 5.0 w/o '8 'U . The present analysis is performed with: KENO-Va Monte Carlo, CASMO-3 integral transport theory and SIMULATE-3 nodal diffusion theory. This report coverli the criticality I analysis of both the spent fuel racks and the new fuel vault. Criticality of the spent fuel racks is studied as a function of enrichment, burnup and fuel placement. A proposed Technical Specification for the placement of I' fresh fuel with enrichment up to 5.0 w/o 885 U in the spent fuel racks is presented. Criticality of new fuel vault is studied as a function of moderator density and fresh fuel enrichment. A proposed Technical Specification for the placement of fresh fuel with enrichment up to 5.0 w/o

'85 0 in the new fuel vault is presented.

I I

I I

I iv l

TABLE OF CONTENTS Pace DISCLAIMER OF RESPONSIBILITY , . . . . . . . . . . . . . . . . iii ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . iv TABLE OF CONTENTS . . . . . . . . . . . . . . . . . . . . . . V LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . . Vi LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . vii

1.0 INTRODUCTION

. . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1 Regulations and Design Basis . . . . . . . . . . . . . . 1 1.2 YAEC Criticality Safety Methods . . . . . . . . . . . . . 2 2.0 SPENT FUEL POOL CRITICALITY ANALYSIS . . . . . . . . . . . . . 4 2.1 Spent Fuel Rack Mechanical Design . . . . . . . . . . . 4 2.2 KENO-Va Modelling . . . . . . . . . . . . . . . . . . . 4 2.3 CASMO-3 Modelling . . . . . . . . . . . . . . . . . . . 5 1 2.4 2.5 K,gg vs. Enr$ chment . . . . . . . . . . . . . . . . . . .

Sensitivity Analysis . . . . . . . . . . . . . . . . . .

5 5

2.6 Infinite Array Maximum Fresh Fuel Enrichment 6 I

2.7 Infinite Array Burnup Credit Analysis . . . . . . . . . 7 2.0 CASMO-3/ TABLES-3/ SIMULATE-3 Modelling . . . . . . . . . 8 2.9 Fresh Fuel Checkerboard Analysis . . . . . . . . . . . . 9 2.10 Fresh and Burnt Fuel Checkerboard Analysis . . . . . . . 9 2.11 Axial Effects 10 1 2.12 Accident Situations

. . . . . . . . . . . . . . . . . . 11 3.0 NEW FUEL VAULT CRITICALITY ANALYSIS . . . . . . . . . . . . . 37 3.1 New Fuel Vault Mechanical Design . . . . . . . . . . . . 37 3.2 KENO-Va Modelling . . . . . . . . . . . . . . . . . . . 37 E 3.3 K.rt vs. Moderator Density . . . . . . . . . . . . . . . . 30 g 3.4 K e, vs. Enrichment et Optimum Density . . . . . . . . . . 38

4.0 CONCLUSION

S . . . . . . . . . . . . . . . . . . . . . . . . . 47 1

5.0 REFERENCES

. . . . . . . . . . . . . . . . . . . . . . . . . . 48 APPENDIX A - Validation of CASMO-3/ SIMULATE-3 for Fuel Storage Burnup Credit Criticality Analysis . . . 49 9

I I

I ~

v I

l L

I LTST OF TABLES r'

k Number Title Ppge 2.1 Nominal Spent Fuel Rack Design Speed . . . . . . . . . . 12 I 2.2 Nominal Fuel Asssmbly Design Specs . . . . . . . . . . . 13 5

2.3 Criticality Analysis Tolerances . . . . . . . . . . . . 14

~

2.4 Sensitivity Analysic Results . . . . . . . . . . . . . . 15 2.5 Boraflex Shrinkage Bias . . . . . . . . . . . . . . . . 16 2.6 Spent Fuel Rack K,un vs. Enrichment . . . . . . . . . . 17 2.7 Axial Boraflex Effects . . . . . . . . . . . . . . . . . 18 2.8 Axial Burnup/ Moderator History Effects . . . . . . . . . 19 3.1 New Fuel Vault Kerr vs. Void, Fully Loaded with 3.5 w/o "5U Fuel . . . . . . . . . . . . . . . . . . 39 l 3.2 New Fuel Vault Ker, vs. Loading and Assembly Enrichment 3 at " Optimum Moderation" . . . . . . . . . . . . . . . . 40 A.1 Surry Unit 1, Cycles 1 and 2, SIMULATE-3 Eigenvalues . . 52 A.2 Surry B Fuel Assembly Discharge Exposures From Cycle 2 . . . . . . . . . . . . . . . . . . . . . . 53 A.3 CASMO-3/ SIMULATE-3 B&W Fuel Storage Criticals Results . 54 A.4 CASMO-3/ SIMULATE-3 PNL Flux Trap Criticale Results . . . 55 I

I I

vi

LIST OF FIGURES

! Number Title Pace 2.1 Seabrook Station Spent Fuel Pool Arrangement . . . . . . 20 j 2.2 Storage Rack Module, Radial . . . . . . . . . . . . . . 21 i

2.3 Storage Rack Module, Axial . . . . . . . . . . . . . . . 22 2.4 Storage hack Unit Cell for Criticality Analysis . . . . 23 2.5 Storage Rack KENO-Va Model . . . . . . . . . . . . . . . 24 2.6 Storage Rack CASMO-3 Model . . . . . . . . . . . . . . . 25 2.7 Storage Rack CASMO-3 Model, Nominal Dimensions . . . . . 26 2.8 Storage Rack Kerr vs. Enrichment, CASMO-3 and KENO-VA Comparison ....... . . . . . . . . . . . . . . . . 27 B 2.9 Storage Rack K,3f,3 vs. Enrichment . . . . . . . . . . . . 28 2.10 Storage Rack K,gg vs. Burnup and Initial Enrichment . . . 29 2.11 Storage Rack Single Unit Burnup Credit Tech Spec . . . . 30 2.12 CASMO-3 Fuel Seament Calculation . . . . . . . . . . . . 31 2.13 SIMULATE-3 Two Dimensional Fuel Rack Checkerboard . . . 32 2.14 SIMULATE-3 Three Dimensional Fuel Rack Canister . . . . 33 2.15 Storage Rack Fresh Fuel Checkerboard, SIMULATE-3 and KENO-Va Comparison . . . . . . . . . . . 34 2.16 Storage Rack Checkerboard K.tr vs. Burnup and Initial Enrichments . . . . . . . . . . . . . . . . . . 35 2.17 Storage Rack Burnup Credit and Checkerboard Tech Spec . 36 3.1 Seabrook Station New Fuel Vault . . . . . . . . . . . . 41 3.2 Fully Loaded New Fuel Vault KENO-Va Model . . . . . . . 42 3.3 Partially Loaded, 72 Assembly Capacity, New Fuel Vault KENO-Va Model . . . . . . . . . . . . . . . . . . . . . 43 3.4 Partially Loaded, 81 Assembly Capacity, New Fuel Vault KENO-Va Model . . . . . . . . . . . . . . . . . . . . . 44 3.5 New Fuel Vault835K,3f,3 vs. Void, Fully Loaded with 3.5 w/o U Fuel . . . . . . . . . . . . . . . . . . 45 3.6 New Fuel Vault K,3f,3 vs. Enrichment and Vault Loading at " Optimum Moderation" . . . . . . . . . . . . . . . . 46 vii

I I

1.0 JNTRODUCTION I 1.1 Reculations and Desion Basta, i The applicable codes, standards and regulations of criticality safety for spent fuel and new fuel storage include the following:

• General Design Criterion 62 - Prevention of Criticality in Fuel Storage and Handitng

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

• ANSI /ANS-57.2-1983, Design Requirements for Spent Fuel Storage Facilities At Nuclear Power Plants, Section 6.4.2.

t

• ANSI /ANS-57.3-1983, Design Requirements for New Fuel Storage Facilities at LWR Plants, Section 6.2.4.

I These regulations and guides require that for spent fuel racks the maximum calculated K.gc, including margin for uncertainty in calculational method and mechanical tolerances, be less than or equal to 0.95 with a 95%

probability at a 95% confidence level.

For new fuel vaults, a dual criteria applies in which the maximum cal culated K.rg, including uncertainties, is less than or equal to 0.95 when flooded, and less than or equal to 0.98 under conditions of " optimum moderation."

In order to assure the true reactivity will always be less than the i calculated reactivity, the following conservative assumptions are made in calculating the criticality safety limits for the spent fuel racks:

• pure, unborated water at 68 *F is used in all calculations, e a 2D infinite array with no radial or axial leakage is modelled, and

• neutron absorption from spacer grids is neglected, i.e. replaced by water.

Because the new fuel vault is normally dry, and low density moderation or " optimum moderation" produces strong coupling between assemblies, the following conservctive assumptions are made:

• the vault is water tight, e unborated water is introduced uniformly throughout the vault and in the space between fuel pins, I 1 I

I

• water density is varied uniformly f rom flooded to dry,

• ne'atron absorption f rom spacer grids is neglected, and

.E

• a 3-D semi + infinite array is modelled with reflection from the 5 floor and walls.

g Unlike the spent fuel rack analysis, the assumption of an infinite E array with no radial or axial leakage is totally unrealistic at low water

densities typical of " optimum moderation." Siirple 2D modelling of the new fuel vault array produces erroneously high values of K. under conditions f

of " optimum moderation."

Thus, the axial leakage is modelled along with

^

explicit reflection from the walls and floor, 1.2 The YAEC Criticality Safety Methods

>l Yankee Atomic Electric Company (YAEC) has developed and validated a combination of criticality safety methods based ont KENO-Va Monte Carlo",

I CASMO-3 LWR lattice integral transport"I, PDQ-7 fine mesh diffusion theory"' and SIMULATE-3 nodal burnup credit analysis"'U. This permits f criticality analysis by several independent methods and allows the flexibility to handle various LWR fuel types, fuel storage errays and

, criticality safety assumptions. These methods and their applications are

, described in more detail below.

In the NITAWL-S/ KENO-Va methodology, the NITAWL-S code prepares a working nuclide library and performs resonance self-shielding for 8"U. In this analysis, the 123 group data is used in all KENO-Va calculations. The I working nuelide library along with case specific compositions and rack geometry data are input to KENO-Va. KENO-Va performs a multi-group, Monte Carlo eigenvalue calculation. The results from KENO-Va analysis are Ktr vs. generation, fluxes and reaction rates. Since Monte Carlo is stochastic in nature, results will always have some uncertainty. In this analysis, KENO-Va is used to verify the CASMO-3 spent fuel rack criticality results and to perform the new fuel vault criticality analysis I CASMO-3 is an integral transport lattice code with a hierarchy of energy condensation and spatial detail leading to a seven-group,

'I transmission probability model of the fuel rack unit cell, COXY. The 40 micro-group nuclear data is used in all CASMO-3 calculations. CASMO-3 is flexible enough to handle up to a 19x19 fuel assembly array with storage f

canister regions, poison sheets, and water gaps. CASMO-3 can perform transport theory burnup credit analysis. Hot full power lattice depletions 2

I

can be executed, and cold zero power restarts in rack geometry can be performed. CASMO-3 can produce few-group cross sections for PDQ fine mesh cif fusion theory analysis. Also, CASMO-3 can produce two-group homogenized cross section for nodal burnup credit criticality analysis on fuel storage arrays.

In this analysis, CASMO-3 is used to study: rack Nrr vs. fresh fuel enrichment, unit cell sensitivity to mechanical perturbatione, and rack Ntr vs. burnup. Also, CASMO-3 is used to generated homogenized two group cross sections for nodal burnap credit analysis using SIMULATE-3. Since the results of CASMO-3 calculations are deterministic, Nic vs. enrichment is I monotonic and smooth. Also, a reactivity change, AK, f rom mechanical perturbation is not overwhelmed by stochastic uncertainty like the AK from Monte Carlo would be.

The use of KENO-Va and CASMO-3 for fuel storage criticality analysis has been validated by comparison to 21 B&h fuel storage critical experiments."'" The methodology bias and uncertainty determined from this validation will be used in the calculation of Ntr at a 95/95 probability /

I confidence level.

l .

In order to address the isotopic and axial burnup distribution issues associated with burnup credit, Yankee Atomic has developed and extended lW its incore methodology to excore fuel storage criticality analysis in fuel storage racks or casks.'" This methodology is based on the advanced nodal method CASMO-3 / TABLE S- 3 / S IMULATE-3 0" and permits direct coupling of incore reactivity characteristics with excore storage array criticality analysis- Two dimensional checkerboard and three dimensional axial fuel -

rack criticality analysis is performed with SIMULATE-3 nodal diffusion I

theory.

The une of SIMULATE-3 nodal diffusion theory in fuel storage burnup credit criticality analysis has been validated by comparison to reactor criticals'O, measured assembly burnups i ",10 B&W"' fuel storage criticals end 11 PNL flux trap criticals."4 These results are summarized in Appendix A, and they show that the SIMULATE-3 advance nodal methodology is valid and accurate for fuel storage burnup credit criticality analysis when the arrays are nodalizable.

3

I 2.0 SPENT FUEL RACK CRITICALITY ANAtYSIS 2.1 Spent Fuel Rack Mechanical Desian Presently, the Seabrook Station spent fuel pool contains six free-I standing and self-supporting modules allowing spece for 660 fuel assemblies (see Figure 2.1). Each rack module comprises an array of Boraflex poison fuel storage cells with a center-to-center spacing of 10.35" (see Figure l 2.2). Each storage cell is welded to a grid base and welded together at the top through an upper grid to form a integral structure 17 3.75" in g height (see Figure 2.3). Criticality control is by the flux trap 5 principlet fast neutrons leaking from stored assemblies are thermalized in the 1.086" water gap between cells and are then absorbed in the Boraflex sheets. The unit cell for spent fuel rack criticality analysis is shown in Figure 2.4. The nominal rack dimensions and mechanical tolerances that

***""*" *" '"' "*'* * *'Y """ Y'** "'" S $"*" *" Teb es 2.1 and 2.2 iI

W

The analysis assumes that standard Westinghouse 17x17 assemblies are used in present and future cycles. Table 2.3 shows the nominal fuel assembly design specifications. It is assumed that inconel grids are used I

in assembly burnup calculations, and that no grids are used in fuel storage calculations. This is a conservative epproach because inconel grids cause a hard neutron spectrum in burnup calculations resulting in more reactive fuel with burnup, and no grids in the fuel storage racks creates a soft spectrum resulting in more reactive fvel in storage. No removable burnable absorber pins are included in the fuel storage rack calculations. However, II a burnable absorber history penalty is included for conservatism in defining the burnup credit criticality safety limits. This is described in Section 2.7.

' 2.2 KENO-Va Modelling The KENO-Va model of the spent fuel rack unit cell is an explicit pin by pin model (see Figure 2.5) . Reflecting boundary conditions are applied i

at the sides, top and bottom simulating a two dimensional infinite array, g A 123 group working library is created by NITAWL for criticality analysis 3 of the racks versus fresh fuel enrichment and for fresh fuel checkerboard analysis. Separate resonance calculations are performed for 2"U at each enrichment with Dancoff factors calculated by Sauer's method."" The Ystr calculation for each enrichment is an average of three independent calculations with: different starting seeds for the random number

I 4

I

generator, a cosine starting distribution and 300 neutrons per generation for 103 generation, skipping the first three. Thus, each case is a result of-90,000 histories.

2.3 CASMO-3 Modellina

{

j The two dimensional CASMO-3 model of the Seabrook spent fuel racks is

( shown in Figure 2.6. This model is based on the seven group transmission probability routine, COXY. Reflecting boundary conditions simulating an infinite two dimensional array are implicit in COXY. In this fuel storage rack model, the pin cells representing the fuel assembly are appropriately homogenized square cells surrounded by an explicit inner water gap, steel canister wall, Boraflex sheets, steel outer wrapper, and flux trap water gap. The model makes use of half diagonal fuel storage symmetry. The nominal dimensions are shown in Figure 2.7.

2.4 L , vs. Enrichment CASMO-3 and KENO-Va fuel storage rack K.rt vs. enrichment is plotted on Figure 2.8. Both sets of calculations are at nominal mechanical dimensions and 6B 'F system temperature. The agreement between CASMO-3 and KENO-Va is excellent over the range of enrichment from 1.6 to 5.0 w/o "5U.

This agreement establishes the validity of the CASMO-3 fuel storage rack model and reactivity at high enrichments. Also, based on these results, an enrichment between 3.75 and 4.00 w/o "SU will be the maximum fresh fuel enrichment when all uncertainties are included, 2.5 Sensitivity Analysis h

Calculation of YS at a 95/95 probability / confidence level requires an evaluation of reactivity effects of the mechanical uncertainties associated with a particular rack and fuel assembly design. CASMO-3 is i used . to' determine the sensitivity of the racks to these mechattical uncertainties. The reactivity effects of the mechanical tolerances listed in Table 2.3 are shown in Table 2.4. The final root-sum-of-squares mechanical uncertainty is 0.01086 AK.

When Boraflex poison is subject to gamma irradiation from spent fuel, it undergoes shrinkage in width and length. A maximum shrinkage factor of 4% in width and length can be expected during life. " CASMO-3 is used to determine '.he AK penalty due to a 4% shrinkage in width. This penalty is 5

F applied as a bias in the determination of maximum fresh fuel. The results are shown in Table 2.5. An average bias of +0.00609 AK is used in the calculation of Knin. Further analysis of the shrinkage in length will be analytod with SIMULATE-3. It will be shown that there is no axial penalty

{ due to the compensating effect of leakage.

. 2.6 Infinite Arrav Maximum Presh Fuel Enrichment i

A determination of maximum fresh fuel enrichment without administrative controls is made by adding all uncertainties to the nominal 4 K.:: values vs. enrichment and then solving for the enrichment at which K,e I = 0.95, the NRC limit.

K.r t is calculated at 95/95 probability / confidence level by the following equation:

Knin*K,,,,+ AKg AKgl(6K,) ' + (6K,V (1) wheret K. = K,:: of the nominal configuration, AKo = calculational bias, AK. = boraflex shrinkage bias, AK, = 95/95 calculational uncertainty, and AK, = 95/95 mechanical uncertainty, For CASMO-3 based fuel storage criticality calculations, AKa = -0.00251 lP and 6K, = 0. 00 8 53. "' From Tabl e 2. 4, AK. - 0. 0108 6 and f rom Table 2. 5 AKa

= 0.00609. Thus, the total uncertainty, AK , applied to the nominal K rr i e values ist i AK, = 00251 + . 00 609 +/ ( 008 53 ) 2 + ( . 010 8 6 ) * = , 017 3 9 (2)

Knfn vs. enrichment is given in Table 2.6 and is plotted in Figure 2.9. Interpolating between the Knin values f sr 3.75 and 4.0 w/o "SU gives "5

a maximum fresh fuel enrichment of 3.7' w/o U for K = 0.95 with uncertainties.

6

2.7 Infinite Atrav Furnur, credit Analysis This section addresses the amount of burnup necessary to get burnt fuel with enrichment higher than 3.75 w/o 8"U in the spent fuel racks.

Single assembly CASMO-3 hot full power depletion calculations and insertion into rack geometry are performed f or fuel enrichments up to 5.00 w/o U 8 ".

- These calculations serve two purposes to study the burnup credit sensitivity of the fuel storage racks and to generate fuel-rack segment cross sections for SIMULATE-3 checkerboard calculations.

I All single assembly burnup calculations are performed at hot full power average conditions given in Table 2.2 with restarts to rack geometry at 6B 'r, no xenon, conditions. The rack K,,, V a. . burnup for various initial enrichments are plotted in Figure 2.10.

I Included in rigure 2.10 ic a maximum reactivity acceptance line based on the 0.95 NRC limit minus uncertainties as a function of burnup.

uncertainties as a function of burnup include the total uncertainty defined The by Equation 3 ( AK.= 0. 017 3 9 ) and two burnup depended components. The two burnup dependent components account for the effect of 2D vs. 3D modelling I of burnup/ moderator history and the eff ect of burnup absorber (BA) history.

Both of these effects are positive contributions to reactivity as a function of burnup. The 2D to 3D component accounts for the positive effects of less exposure and high '"Pu concentration at the top of a fuel assembly. The 2D to 3D penalty is assumed to be a linear increase in reactivity of .02 AK from 0 to 50 GWd/MtU. The BA history component I accounts for the ef fect of depleting with pyrex bas and then removing them.

This is assumed to be a linear increase in reactivity of .01 AK from 0 to 12 Gwd/MtU and then no increase thereafter. Thus, the maximum reactivity acceptance line is given by

?L,= . 9 5 . 017 3 9 . 0 2

• B . 01

• B g3) for 0 $ B $ 12 Gwd/MtU, and 16,= . 9 5 . 017 3 9 . 0 2

• B 01 (4)

> u for 12 < B $ 50 Gwd/MtU.

This approach is equivalent to applying all uncertainties to the CASMO-3 calculated fuel rack K.rt 's. burnup. The follow maximum reactivity 1

7

values are used for the range from 0 to 10 Gwd/MtU 4 ,00GWd/NtU=. 95 . 01739 .02 0 _. 0l e 0 = . 93261 (5) bu u

( ,05GNd/HtU=. 95 .01739 .02 e5. 01 5 =. 92644 (6)

I rw,010GNd/NtU=. 9 5 . 017 3 9 . 02 10. 01 10 = . 9202 8 du u (7) l The intersection of rack K.fr vs. burnup for each enrichment with the g maximum reactivity acceptance line defines the minimum assembly burnups W necessary to meet 0.95 with all uncertainties. These enrichment /burnup combinations are plotted to define the single unit (infinite array) burnup credit acceptance criteria (see Figure 2.11). This criteria defines two regions: a region of acceptable burnup and enrichment for placement in the racks and a region of unacceptable burnup and enrichment. Since this criteria is based on an infinite array assumption, it is conservative. The fuel assemblies with characteristics in the region of unacceptability can be made ecceptable by checkerboarding this fuel with fuel of lower enrichment and/or higher burnup. Such fuel is available from Cycle 1 and g subsequent cycles. Further development of this burnup credit acceptance y criteria including the option for fuel assembly checkerboard placement is presented in Section 2.10.

I 2.8 CASMO-3/ TABLES-3/ SIMULATE-3 Modellina The CASMO-3 cases used to study rack K.tr vs. burnup and enrichment also provide homogenized two group cross sections and assembly discontinuity factors (ADFs) necessary for nodal burnup credit analysis (see Figure 2.12). In this methodology, each fuel-enrichment-rack combination becomes a fuel segment, a unique set of two group cross sections and ADFs. These segments are collected by TABLES and processed into a binary cross section library for use by SIMULATE. This library covers pool conditions from 68 'r to 150 'r and soluble boron concentration from 0 to 2000 ppm. However, the criticality safety limits are defined by calculations at 68 'T and 0 ppm soluble boron. SIMULATE-3 assembles the fuel-rack segments into a nodal array representing the fuel storage racks.

Four nodes radially and twelve n. des axial'.y are model'.ed as in incore 8

analysis. SIMULATE-3 is used in two dimensional (2D) checkerboard (see Figure 2.13) and three dimensional (3D) axial (see Figure 2.14) criticality analysis of the spent fuel racks.

2.9 Fresh Fuel Checkerboard Analysis In this analysis, a series of 2D fresh fuel checkerboard cases are executed with SIMULATE-3 and are also cross validated with an explicit pin by pin KENO-Va model. These fresh fuel checkerboards study the criticality of Cycle 1 fuel, assuming no burnup, placed nex+ to fresh fuel of enrichments from 3.5 to 5.0 w/o "SU. The Cycle 1 fuel comprised 65 r

assemblies of 1.6 w/o 8"U, 64 assemblies of 2.4 w/o "SU and 64 assemblies of 3.1 w/o '"U. These fuel assemblies are of low enough enrichment and sufficient number the*. they will allow the placement of a large batch of 5.0 w/o 8"U fuel in a checkerboard configuration. '

Figure 2.15 thows a SIMULATE-3 and KENO-Va reactivity comparison for the fresh fuel checkerboards. Except for the 1.6 checkerboards agreement is within 1%. The 1.6 checkerboards are an extreme enrichment split.

These cases are within 2% with KENO-Va predicting higher. Overall, agreement gets better with the more limiting cases, i.e. the 3.1 checkerboards. The results also show that checkerboarding with the 1.6 and 2.4 w/o 8"U fuel from Cycle 1 will allow the placement of fresh fuel with enrichment up to 5.0 w/o 8"U. If uncertainties are included, then 3.1 w/o 2"U fuel can be checkerboarded with up to 4.5 w/o '"U fuel.

2.10 Fresh and Burnt Fuel Checkerboard Analysis In this analysis, a series of fresh fuel and burnt fuel checkerboard cases are executed with SIMULATE-3. The fresh fuel is set at a maximum of 5.0 w/o 8"U. The burnt fuel is varied in initial enrichment from 3.5 to 5.0 w/o and assembly burnup from 0 to 30.0 Gwd/MtU. Based on the checkerboard unit reactivity, a second line of demarcation will be defined allowing the maximum permissible enrichment /burnup combinations to be checkerboarded with up to 5.0 w/o "SU.

The SIMULATE-3 checkerboard K.tr vs. burnup at various initial enrichments are plotted in Figure 2.16. The intersection of the checkerboard K.,, vs. burnup for each initial enrichment with the maximum reactivity acceptance line, defined in Section 2.7 aTd with the total CASMO-3/ SIMULATE-3 uncertainty given in Appendix A, yields the minimum 9

1 m... . . . . . . . . _ _ _ _ _ _ _ . _ _ _

B l

alternate assembly burnup necessary to meet .95 with uncertainties. These B alternate enrichment /burnup combinations are plotted in Figure 2.:' which defines the final proposed Burnup Credit and Checkerboarding Technical Specificatior.. This proposed Technical Specification defines three regions: 1, 2 'nd 3 and the following conditions hold for the Seabrook Spent Fuel Racks:

1 may be stored anywhere, 2 must not be stored next to 3, and 3 must be stored next to 1 or empty locations.

2.11 Agial rffects In the previous analysis, criticality was calculated in 2D with both CASMO-3 and SIMULATE-3. Usually, 2D analysis is conservative because axial leakage has a negative effect on reactivity. However, there are two issues leading to potential nonconservatism in the 2D analysis: Boraflex shrinkage and axial burnup/ moderator history. SIMULATE-3 is used to analyze these axial effects. Comparison to 2D calculations is made to determine if previous axial penalties are conservative.

The Boraflex poison sheets have a nominal length of 141.25" (358.78 cm) and do not entirely cove. the active fuel length 144" (365.76 cm). In addition, gamma irradiation from spent fuel will shrink the Boraflex by 4%

to 135. 6" (344. 42 cm) . SIMULATE-3 is used to study the foreshortening of I

the Boraflex. Since this requires discrete axial detail of fuel segments without Boraflex, 24 nodes axially are modelled. These calculations are performed at 0 Gwd/MtU and over an enrichment range of 3.75 to 4.25 w/o 2"U. The results are shown in Table 2.7. The results show that 2D to 3D (358.78 cm of Boraflex) modelling of the fuel storage racks results in a

.002 AK reduction in reactivity.

I If an additio- ihrinkage in Boraflex is modelled, the results are still less than . analysis. Therefore, additional penalties for axial boraflex shrinkage are not necessery.

Axial burnup/ moderator density history effects have a complicated I

nonlinear ef fect on storage array reactivity. At 0 Gwd/MtU the effects o:'

axial leakage are negative. As burnup proceeds incore, the central region s become more burnt than the top and bottom of the fuel assembly owing to the buckled shape of the axial power distribution. This leads to more reactive fuel at the bottom and top. Usually, the top is less exposed than the bottom because of slight bottom peaking of the power distribution due to 10 I  !

l g moderator density ef fects. In addition, the top becomes even more reactive

@ than the bottom with burnup due to the greater buildup of 8"Pu from moderator density history ef fects. Generally, wher' the reactor is at power these effects are not pronounced, but when core power is reduce to zero and/or coolant temperature is reduce, there can be a strong positive effect . ,

in reactivity f rom the redistribution of flux to the top of the fuel. This

_E effect is especially pronounced in the criticality analysis of burnt fuel stored racks. Previous analysis (see Section 2.7) has chosen to bound this effect by a linear penalty with burnup. This allows simpler 2D analysis )

to provide the crSticality safety limits.

SIMULATE-3 in used to study the effects of axial burnup/ moderator L history relative to 2D analysis. These calculations were performed with g

3.1 w/o "SU fuel, assembly average exposure and moderator history characteristics of twice burnt fuel irom Seabrcok Station, Cycle 1 and 2, design calculcticns. The results are given in Table 2.0, and they show that the linear penalty given is section 2.7 is bounding over the range from 0 to 25 GWd/HtU. This is especially true whcn ' n penalty for burnable absorber history is also applied.

2,12 Accident situations Accident situations include: a complete misloading of the fuel storage

'g racks with fresh fuel of 5.0 w/o 8"U enrichment, an assembly on ty f the T racks, E d an assembly next to the s;. des of the racks. Credit is . . awed for the presence of soluble boron (2000 ppm) in accident situations. This refueling concent2= tion of soluble boron provides a 30% reduction in reactivity over the ua. borated situation and more than adequately suppresses reactivity effects from the above accident situations.

CASMO-3 storage rack calculations with 5.0 w/o '"U fuel give a K g: of 0.80605 with 2000 ppm in the pool water versus 0.98405 with no boron in the pool water. This 2D infinite array calculation bounds the complete misloading of the fuel sto-- acks with 5.0 w/o 8"U f uel . CASMO-3 single assembly aalculations witi. 1 w/o "SU fuel assembly outside the racks in the pool water give K.,, of 0.72844 at 2000 ppm versus 0.97695 et 0 ppm.

J This ca*Julation bounds the ef fect of an assembly on top of and next to thc sides of the racks.

R

,g 11 I I

I

, Table 2.1 Nominal Spent Fuel Rack Design Specifications I Rack Meche_. eal Design Canister Center-to-Center Spacing A_

10.35 ctn 26.2890 I Inner Canister ~ welop Canister Wall Thickness 8.90 0.09 22.606 0.2286 Boraflex Width 7.46 18.9484 Boraflex Thickness 0.071 0.1803 Rapper Thickness 0.02 0.0500 Flux Trap Gap Thickness 1.086 2.7584 i

I Rack Compositions

. tat . "s Steel-304 pensity 7.90 Constituents N/o)

Si=0.51, Cr=17.4, Mn=1.99, Fe=68.35, Ni=11.7 I Boraflex Water 1.83 0.9982 Si=18.9, O=17.0, C=21.0, H-2.1, B-41.0 H-11.19, 0-88.1 I

'I og

>I I

I I

I I

I 12 I

E G Table 2.2 g Nominal ruel Assembly Design Specifications I

Assembly Mechanical Design in, em Assembly Pitch, in core 8.466 21.5036 Rod Pitch, 0.496 1.2598 Number of Grids, in core 7 Grid Material Inc718 & SS304 Grid + Sleeve Weight, lb/kg 1.611 0.7308 I- Active Core Height 144.0 365.76 ruel Rod Mechanical Design outside Diameter 0.374 0.9500 Diametral Gap 0.0065 0.0165 Pellet Diameter 0.3225 0.8192 Pellet Compositions UO, Clad Thickness 0.0225 0.0572 I Clad Haterial Zirc-4 I Guide Tube Mechanical Design Outside Diameter Inside Diameter 0.484 0.448 1.2294 1.1380 Tube Material Zirc-4 Average Operating Conditions (For Burnup Credit. Analysis)

Pressure, psia 2250 Moderator Temperature, 'K 583 I Moderator Temperature, 'M Fuel Temperature, 'K 566 900 (For Moderator History)

Soluble Boron, ppm 500 Power Density, W/gU 38.13 I ruel Assembly Compositions Density Grids + Sleeves Pellet Stack Density 3.99*

10.36 genetituent (w/ot 84.6 w/o Inc-718, 15.4 w/o SS-304 CASMO-3 Default Clad Density 6.55 CASMO-3 Default I . ,cm axial, no grids in the fuel rack calculations I

I 13 I

E Table 2.3 Criticality Analysis Tolerances I

l Center-to-Center Spacing 1 0.060" ( .152 4 cm)

Canister Envelope 1 0.050" t.1270cm)

Boraflex Width 1 0.075" (.1905cm) j Boraflex Thickness 1 0.010" ( 0254cm) i Boraflex Boron Loading 1 5 w/o B Fuel Stack Density i 0.06 g/cc

Fuel Enrichment 1 0.05 w/o "50 lI.

I 4

i

I

) -

il.

.I jI II I

{ 14

!I ,

I Table 2.4 Sensitivity Analysis Results Tolerance -

/ + AK/2 >

Oenter-to-Center Spacing 0.93895/0.92659 *0.00618 0.92619/0.93999 10.00690 I Canister Envelope Boraflex Width Boraflex Thickness 0.93397/0.93201 0.93629/0.92989

• 0.00098

• 0.00320 Boraflex Boron Loading 0.93665/0.92920 *0.00373 Fuel Stack Density 0.93198/0.93328 10.00065 Fuel Enrichment 0.93006/0.93518 0.00256 I Root S-tm of Squares 0.01086 I

I I

I I

g I-I-

LI I

I E 15

i g

Table 2.5 f

I. Boraflex Shrinkage Bias l

'I w/o U235 3.50 3.50 Boraflex Condition Nominal 4% Shrinkage dra 0.91927 0.92528 AK

+0.00601 1

I l

3.75 Nominal 0.93263 3.75 4% Shrinkage 0.93874 +0.00611  ;

4.00 Nominal 0.94480 i

4.00 4% Shrinkage 0.95096 +0.00616 Average Bias +0.00609 I

^

3 I

I  ;

I I l I

I I

I l l l 16 I

L Table 2.6 Spent Fuel Rack K,sm vs. Enrichament t w/o 835 U 1m J ,3,, u 1.60 0.74021 0.75760 2.40 0.83892 0.85631 3.10 0.89477 0.91216 3.50 0.91927 0.93666 3.75 0.93263 0.95002 1 4.00 0.94480 0.96219

~

4.25 0.95584 0.97323 4.50 0.96603 0.98342 J 4.75 0.97538 0.99277 l 5.00 0.98405 1.00144 I

I I

I I

I >

I 17

l . = ,- - - -

m,

-Table 2.7 e Amirl Boraflex Bffects-

[

Case ,,_2, e ._. AK 2D Base Case - 3.-75 w/o Fuel 0.93273

h. - 3D 358.78 cm Boraflex, 24 Nodes 0.93076 -0.00197 3D 344.42 cm Boraflex, 24 Nodes 0.93246 -0.00027 .

~

f -_

i 2D-Base Case - 4.00 w/o Fuol 0.94486

-3D-358.78 cm Boraflex, 24 Nodes 0. v 290 -0.00196 1

3D'344.42 cm Boraflex,.24 Nodes 0.94462 -0.00024 i

2D'. Base Case - 4.25 w/o Fuel 0.95593 3D 358.-78 cm Boraflex,-24 Nodes 0.95392 -0.00201 3D 344.42.cm Boraflex, 24 Nodes 0.95565 -0.00028

{. '

D p

C C

x  ;

(i

(;

[ .

[ .

(

7

L-I L

Table 2.8 Axial Burnup/ Moderator Bistory Effects f'

Burnup (GWd /Mt U) 2D K.,g_, 3D K-r: AK Penaltv*

0 0.69486 0.89280 -0.00206 -

0.84678 0.84401 -0,00277 +0.00240

( 6 0.00276 +0.00480 0.80433 +0.00157 12 19.83 0.74P73 0.75659 +0.00786 +0.00793 25.50 0.71327 0.72487 +0.01160 +0.01020

• 2D to 3D component of Equation 3 l

I I

I I

I I

I I

I i

19

!I4 J

!I i h il Area Prerently Not Used t

I i

I 37.5* Storage Rack Module

/

t 10 x 1l 10x11 10 x 11 J

! I 10 x 11 10 x 11 10 x 11 1

1 I __ 27  :

jI i

{

~

l 1

Figure 2.1 i

, The Seabrook Station Spent Fuel Pool Arrangement 1 1 l 20

I l

i

L 7"

i h

E Storage Canister _10.35"_., Boraflex Sheets Flu Tr i .. _ \/

I I m  :

10.35' -

o ,

l I

I I .

Figure 2.2 Storage Rack Module, Radial 21

I I

I ,, ~

I

~

I

~ ,, - - "

Top Grid l -- 1

Storage Canister I h Bornfsx Sheet I Flux Trap Gap I

173.75*

I I

I /

Bottom Grid I

r Leveling Pad l I /

r- rn "

I I Figure 2.3 Storage Rack Module, 4xial g 22 I

I-L

[-

1**

( =  :

Rack Canister 17x17 Fuel Assemb;y ( - =' /[ HaN Flux Trap Gap standard Design 'l.................b'

• ..Y.......  :.. +c L ~ a- . ~~

s. .a..

m. e

.. ..a.

. .... .:a/ . ,. .. ,

[

.. .. .. .. .. .u

{  :..

.. a ni

( g...................

.c

{

8.90* .  ;

[

[

[

Storage Rack Unit Cel For Criticality Analysis 1

23

---

_--_.____.___-____m._ _ _ _ _ _ _ _

+

r i

m.

a um nuium a un I g: disiliiisMdiisWsiissi,P".

g;: >r v0:oc_+4oooo4.00o,.oo

v. y ar. v. v 'sissississiik 'i3 Par.v. 'y4vd v v ,P""vn6 " siiisings W,gg s

g3 g i g l >^ O o O k> # > % f >. 4 #

E:>>00<b.44WCit###',b;ttcoe9<,,>0<,,>0<,,t9oO'I i

ROlO t.A E

B : r, ,c,0 , c 3 4 ,93 ,. c 3 4v pv 3, ,' ,v,0cv3v ,4Ijli. B I  ;

i v v v v v v v v,4v v,s,0.

> % O'O $ O.O'O $ 0 4 O C O O. 4 0 (g g 2 ;E

>0t I E.

s, * ,, A' ,4

v. A,,,,.,. M f A,'>v 0*,,*tvA,, i 4vvd', $~ 0 it i O, sS,c d,,,g: E J(g,* A,,

! > 4 o O k 4 4 0-b O 4 O O.D M O O 4 I E s: e,,) > Q26

r v fv,i e v.,,0 frvv 4 o 4r6 & y4v4A29v fv 9v.y'Mav vfAi 4.4Ei g

4 E >;: r.0;vr,a Cv f vy 4. ,5c, O.$.Ow. $ 4. 'y4, 0 .,t' yA 'fv Cv. .,e 4 421i I E : >> 0 e o e <,g, e ? ,> c ?g> c o o o ,>

r, g

i OCO#40c l 1WWWWWWW'y,4.4004404--WWWWWWWW'em EBummassamma' mask I

Figura 2.5 Storage M ck KENO-Va Model 24

4 F .

4 L-F.

L-N Stestar#apper I ICFD*N#1

{ __- ,

'minismiisiir

'~-$ 1M/ FuelLattbe

...,' - jI j  ;

{  : - . ww om i

~

B 'E- B l1 .g.. .

- g!

carewwe ecm

.'7.. -

I  ; . E c 0. 4-5 ~ B -B . -

c. ;

5.

naaragww w

.c .. . .

. L ri iv .

.i.

l j .: f' E . E, 10 -

B ; B- -. 'i -

!.,..-  ;. ~-

.j' .

g;g,g l . g - *g f . .:

..g.- '

!. . L.. ... g3:, ,. ;3' i  :

--g..- ' ;. . l 1 .i i o 75

..:. 1 -

t:

. . ; y M !lpk g g

.3 Q / ..; f . , *i ' '.

I --

- 6 r -( (CFO,EE)

~

c I

I  ;

Figure 2.6 Storage Rack CASMO-3 Model ,

25 ,

1

I I

I GAW=.2312 BXW 2286 CRS-2.2886 I

I .

ABL-18.9484

. a;.6909 I -

.--- b g 6909 i= 1.. 4 n f'_.

I -

_ _T GAN=.2313 CRD 'SE' CRT=.2311 T=.0508,T = 1803  :

ABL-18.9484 5 2 CRS-2.2886 I

I I

I I a-e - WZ- 1 =- -

I 2e I-

I I.

I "

m I _

l 0.9 y

I o

g -

I o CASMO3 a KENO Va I 0.7 1 2 3 4 5 Enrichment (w/o)

I I '

I

I Figure 2.8 Storage Rack K.,, vs. Enrichment, CASM0-3 and EENO-Va Comparison 3 2, I

I l]

1 m

I 1.0 l .95 NRC Umit ,/

0.95 -

0.9 E-0.85-l #

0.8 I

g 0.75 --

g 3.75 w/o U235 Maximum Fresh Fuel I

0.7 . . . - -

-1 2 3 4 5 Enrichment (w/o U235) t Figure 2.9 Storage Rack K,3f,5 vs. Enrichment 28 I

I - 95, Uncertainty 4.00 WO U235 .

I

• 4.25 WO U235 o 4.50 WO U235 O 4.75 WO U235 A 5.00 WO U235 1.0 x

l NN NNN 0.95 N j

5

$N x s N N l ] N x N_^x N N s N N N x%.l N I

0.9 m N.

N N N '

I N I 0.85 0

1

,P, 2 3 4

5 6

7 8

9 e

10 Bumup (GW6MtU) i I

I 4

Figure 2,10 Storage Rack K.,, vs. Burnup and Initial Enrichment I 2e I 1

1 I

10 I I /

}

, y=

6 -

/

Acceptable for Placement 8

i-to

(

/

3

. / Not Acceptable for Placement Without Checkerboarding a

g 2- 7 f

E /

I 3.25 3.5 3.75 4.0 4.25 4.5 4.75 5.0 3.0

=

Initial Enrichment (w/o U235)

I I

g.

,, ,_... ..., m 1. -1:e: cm -- -i-i- 1 I

U '

99999999999999999 h 9999 9 9 9999 99 OG 99 99 80 99 9 9 9 J. 9 99 $9 99 99 99 99:

9 Fuel Storage Rack i '9 9 9' Unit cell je 9 I 99 99 99 99 L lg Og l l9.. .

.Y.Y.9099

.S .Y9 .9 OS 999 999 99999 99 99 99999 l g.S9999999999999999,i I

l s '

i

..s CASMO-3 COXY y ,R Calculation I l

' f Ei . 9 Sk i N!, .

, \, Y ' ' ' 3

.o. '

.A ts l4 . - ;' 6.. -

m.-E ' :6;E ' iM 'A..

. G , ~ *: i ;'. . i . Al iM -

l , Wl',c5 ? s si'?!.%.ff,: .7

';~ t ..AE i t1B :4 ";

, .l ..

g

f.  :,,

g% ru D,E ,

3 E.3, VE g, D, g 2 E.2, VEra Fuel-Rack-ADFui, ADFv2, ADFit, ADF 2 Segment Data Figure 2,12 CASMO-3 Fuel Segment Calculation 31

W I

Fresh Fuel  :

Bumt Fuel ,- '

I Unitof Anabis .. .

l .

~

-j i

l i

I i

Figure 2.13 SIMUIATE-3 Two Dimensional Fuel Rack Checkerboard 32

ll i

I / /// / /l ..-

d Top Water Reflector Nodes

. n ___. _ _ .$f JL

/

/

/

/

/

.E /

i y

<g u

• 365.76 cm 358.78 cm / 4N kadi I

/ 12 or 24 Nodes Axla!

Fuel Boraflex j

/

/

'E /

!g' /

i /

/

g ,

/

'.W //

! y y --- --

// = Bottom Water Reflector

;/ noggg .

/

/

.l '

'l

!I I

iIi Figure 2,14 SIMUIATE-3 Three Dimensional ma al Rack Canister

33

1.0 I

I 0.95 i * '

J q f ,

-~-

o .

] 0.9 --

l 1; o O O

O.85 =t  ; O SIMULATE-31.6 W/O 0 KENO-Va 1.6 W/O

-(

A SIMULATE-3 2.' 'N/O o KENO-Va 2.4 W/o y SIMULATE 3 3.1 W/O

• KENO Va 3.1 W/O 0.8 .

3.5 4.0 4.5 5.0 -

Altemate Assembly Enrk:tunent (w/o U235)

Figure 2.15 Storage Ract Fresh Fuel Checkerboard, SIMULATE-3 vs. ITNO-Va Comparison 34 l

I

I j

j i

I 1.0

- 95 Uncertainty 5.0/3.5 Dumt

.5 Bu I { "' _- N N N I l

' A NNxx D x i i" x NN I NN I

OM 0 10 15 20 25 30 0

Altemate Assently Exposure (GWdWtV)

R Figure 2.16 Storage Rack Checkerboard K.,, vs. Burnup and Initial Enrichment 35 l

l'.

' 25

!  ! l l l ,

1 can be stcred anywhere 2 rnust not be stored next to 3 3 rnust be stored next to 1 or empty locatlons

(

20 --

Region 1

i. 15 - j E

I I 10

/ 7' '

I -

/

Region 2 -/

/

/ -

Region 3

!\

I o 2.5 2.75 3.0 3.25 3.5 3.h5 4.0 4.25 4.5 4.75 5.0 Initial Enrichtnent (w/o U235) g; Figure 2.17 Storage Rack Burnup Credit and Checkerboard Tech Spec I-l8 36 I

I 3.0 NEW PUEL VAULT CRTTICALITY ANALYSIS I 3.1 New Fuel Vault Mechanical Desion The new fuel storage vault is a temporary storage area for fresh, unirradiated fuel . Assemblies can be arranged in a 5 x 18 array with a 21" minimum center-to-center spacing (see Figure 3.1) . Assemblies are held in I place at top and bottom by grids which provide the necessary center-to-center spacing. The vault is surrounded by one foot thick concrete walls with the outer row of assemblies one foot away f rom the walls. Criticality control in the vault is essentially by wide separation between assemblies.

The space between and within assemblies is normally air (void) . Moderator

>g is only introduced by abnormal situations, like fires, which require fire W fighting foam or water mist. Since the intrusion of water by foam or mist cannot be totally precluded, the criticality of the vault is studied as a function of moderator density with particular emphasis on conditions of low density, 0.1 to 0.05 g/cc, or " optimum moderation". The criticality safet3 limits on maximum fresh fuel enrichment and maximum number of assemblies I are defined at the " optimum moderation" condition with K,u,3 5,0.98.

3.2 KENO-Va Mode 11ino The KENO-Va model of the fully loaded new fuel vault involves a basic I unit of analysis which includes the concrete wall, floor and three partial assemblies (see Figure 3.2). This model allows axial leakage with reflection from the floor and radial leakage with reflection from the walls. However, the model is infinite in the 18 canister direction. Since results show that fuel only up to 3.675 w/o 2"U can be allowed in the fully loaded vault at " optimum moderation", two other KENO-Va models are developed to' study partial loading of the vault with higher fresh fuel enrichments.

The K2NO-Va models of the partial loaded new fuel vault are shown in I Figures 3.3 and 3.4. In the first rodel, the central column of the 5 canister direction is lef t empty. Even though the model is infinite in the I 18- canister direction, the model impiles a maximum capacity of 72 assemblies in new fuel vault. If the second model, the central column has alternating empty and loaded locations. Again, even though the model is I- infinite in the 18 canister direction, the model implies a maxi. mum capacity l of 81 assemblaes in the new fuel vault. Other partial loadings are possible, but these are the simplest to model and to specify.

I

I 3.3 L , vs Moderator Density Criticality as a function of void for the fully loaded new fuel vault is given in Table 3.1 and is plotted in Figure 3. 5. The asseA11es are 3.5 w/o 8"U enrichment. Moderator is introduced uniformly throughout all pin I cells, guide tube cells, assembly upper and lower reflector regions and the inter-assembly gap regions. The flooded condition or 0% void correst . ids to water at 68 "F or 0.9982 g/ce, and 100% void is the dry condition.

Figure 3.5 shows that vauAt criticality at 0% void is at about 0.89. Vault criticality decreases steadily with void until a minimum is reached at 65%,

I 0.35 g/cc. After which there is a sharp increase in criticality with a peak at 95% void, 0.05 g/cc, and then a rapid drop in K,c, at 100% void, dry.

I The behavior of K.,, can be understood if one considers that there are two types of moderation occurring in the vault: moderation between assembly I pin cells and moderation in the space between assemblies. The former type of moderation dominates the criticality of the array in high density situations, and the latter dominates in low density situations. This second type of moderation can produce large increases in reactivity. The I moderator density at which the peak occurs is called " optimum moderation".

In the Seabrook new tuel vault, " optimum moderation" occurs at 95% void or about 0.05 g/cc of water. For 3.5 w/o 2"U fuel assemblies K,rr is still below the limit of 0.98. Also, in a fully flooded condition, the K.,, is considerably below 0.95. The optimum moderator density of 0.05 g/cc of water appears to be the most limiting condition (see Table 3.1).

3.4 Lt, vs. Enrichment at Optimum Density I Criticality of the vault vs. loadir.g and enrichment is given in Table 3.2 and plotted with uncertainties in Figure 3.6. From Figure 3.6, it can I be seen that the fully loaded vault, 90 assemblies, has an enrichment limit of . 3. 675 w/o '"U. Also, F'gure 3.6 shows that either partial loading arrangements, 72 or 81, will allow fuel with enrichment up to 5.0 w/o 8" under conditions of " optimum moderation." Thus, the proposed Technical Specification for tr.e new fuel vault is:

Full loading (90 assentlies) is allowed with fuel of enrichment 5 3.675 w/o

"'U, but the loading must be reduced to every other central column location empty (81 assemblies) fur enrichments from > 3.675 and 5 5.0 w/o "SU.

I 38 I

1 l

E Table I 3.1 How Fuel Vault N,, vs. Void, Fully Loaded with 3.5 w/o 8850 Fuel I y1QQ1, KrNo-Va F-t- *O .leues 0 0.07009 1 0.00432 0.88902 20 0.*19973 1 0.00397 0.81429 0 0.10462 1 0.00431 0.72354 I 60 75 80 0.62137 1 0.00346 s.64601 1 0.00448 0.77985 t. 0.( J309 0.63944 0.66591 0.79814 85 0.78012 1 0.00355 0.79828 90 0.89641 1 0.00390 0.91490 92.5 0.94339 0.00398 0.96196 8 95 0.9!)30 1 0.00241 0.97254 97.5 0.81271 1 0.00445 0.83178 100 0.45630 1 0.00300 0.47398 I

I I

I g.

I I

I

- I I 39 I

I I "able 3.2 New ruel Vault K.et vs. Vault Loading and Assembly Enrichment at " Optimum Moderation" 90 Assembly Capacity w/o 825U ,KTNo-Va K.,,t 0 ,1, g, , y 3.50 0.9653020.00241 0.972b4 3.75 0.9651610.00238 0.98238 4.00 0.9762720.00235 0.99347 I 4.25 4.50 0.9844710.00247 0.9925710.00234 1.00175 1.00976 4.75 1.0025520.00249 1.01984 5.00 1.0103210.00259 1.02768 81 Assembly Capacity w/o 8350 VENO-Va K-rt 1 2 LiseL 0.9004310.00150 0.91720 I

3.50 3.75 0.9121710.00171 0.92900 4.00 0.9211710.00166 0.93797 4.25 0.9336310.00167 0.95044 4.50 0.9390810.00166 0.95588 4.75 0.9440710.00170 0.96089 I 5.00 0.9561410.00170 0.97296 I 72 Aseeably Capacity w/o 88!U KENO-Va K-r* i 0 .lesiew I 3.50 3.75 4.00 0.8104010.0022b 0.9105210.90247 0.8297910.00247 0.02753 0.92780 0.84707 4.25 0.8378810.00251 0.85519 4.50 0.8475210.00253 0.86484 I 4.75 5.00 0.8539010.00230 0.8573510.00256 0.87112 0.07469 I

I I

I 40

P' l

L F

I L

CONCRETE WALL s

. . 'p

+ * . * . * *. *

,j

,2 g at,  :

, ,. 3,3- , , :, j :, .,, 33- , / , :j , 2,1* M12$,12" 7 ,

.12 ..

• } ..

i .. . *.'

~

l ,

= - ...*

• + ..

g .

g

. . . 'h .

. ... .e I . . * . . &

O m

• ,, .

• f 4

g

• s. BASIC UNIT OF ANALYSIS

,/.

me, .ms es. me eu. es. een ies aus .e .a' .u. .m. esu. emum '* g , ,

I . .

a .

' - * . y l

STORAGE CELL . .

l

• * ** DOOR +.. .

. . * . . . '. DOOR * **

Figure 3.1 Seabrook Station New Fuel Vault 41

k a

s CONCRETE WALL STORAGE CELLS v ,, ,,

10.6* . .-

o

,_ 12" , 12"  :- ; 71" 33" . y

{

l

{

Figure 3,2 f Fully Loaded New Fuel Vault Kr.NO-Va Model

L l

l CONCRETE WALL STORAGE CELLS EMPTY LOCATION i

T-- .

l 3o.g.

J 1 . .. .

3, 12" 12" _ 71"  :

33"  :

I I

I I

I I

I Figure 3.3 Partially Loaded, 72 Assembly Capacity, New Fuel Vault IG:NO-Va Wodal 43

I I EMPTY LOCATON o

Y c', . ... . .

:e.

I

. e** ,' , ..,.

I 31" ". ,

• * $,4 STORAGE CELLS W ;gi:

I -

. 12" _.

12" _

_ 21" _ _

33 .

I I

I -C I

I I

I I

I rigure 3.4 Partially Loaded, 81 Assembly Capacity, New Fuel Vault KI:NO-Va Model I 44 I

H

< OSB' Umt l n,9 --

Ns 1 0.e l _

3 0.,

\s ,

l

\ "

0.6-1 I

05 o

5 0 10 20 30 40 50 60 70 80 90 100

% Void New Fuel Vault Kuf,, vs. Void, lly aded with 3.5 w/o "50 ruel 45

I I

I 1.05 I  :

/

I "

/

1.0 I j 0.98 limit I i

/

/

i 0.95 --

7 I $ -

I

/

0.9 -

I /

0.85 -

- /

g o 90 unit capacty

'I O 81 unit capacty a 72 urdt capacty 0.8 3.5 3.75 4.0 4 25 4.5 4.75 5.0 2/ 4/91 Enrichment w/o U-235 Figure 3.6 New Fuel Vault Kenes vs. Enrichment and Vault Loading at " Optimum Moderation" 46 I

I 4.0 I CONCtV870tifS.

A Technical Specification for the placement of fresh fuel with enrichment up to 5.0 w/o 8"U in the Seabrook Station spent fuel racks has been developed and is presented in Figure 2.17. This Technical I

Specification is based on burnup credit and checkerboarding of fresh and burnt fuel. Adherence to thic Technical Specification guarantees that the K.tr of the spent racks will be less than or equal to the 0.95 NRC limit.

I A Technical Specification for the placement of fresh fuel with I enrichment up to 5.0 w/o 8"U in the Seabrook Station new fuel vault has been developed and is presented in Section 3.4. This Technical Specification is based on the " optimum moderation" condition and on reduced fuel assembly loading of the vaVit. Adherence to thin Technical Specification guarantees that the Twer of the vault will be below the 0.95 NRC limit in the flooded condition and below the 0.98 NRC limit under conditions of " optimum moderation" at low moderator density.

I I

I I

I I

I I

I I 47 I

L l

5.0 RrrrurNCrs F 1. ORNL/NUREG/CSD-2/V2, "NITAWL-S, SCALE System Module for Performing L Resonance Shielding and Working Library Production", R. M. Westf all, L. M. Petrie, N. M. Greene and J. L. Lucius, October 1981.

~

2. ORNL/NUREG/CSD-2/V1/R2, " KENO-Va, An Improved Monte Carlo Criticality

- Prograin with Supergrouping", L. M. Petrie and N. F. Landers, DNember 19B4

3. STUDSVIK/NTA-80/7, "CASMD-3, A Tuel Assembly Burnup Program", User' Manual, M. Edenius, A. Ahlin and B. Forssen, November 1986.

4. EPRI/ARMP Documentation , "PDQ-7/RARMONY User's Manual", B. M. t Rothleder, March 31, 1983. 1

5. STUDSVIK/SOA-88/02, " TABLES-3P, Library Preparation Code for

~

SIMULATE-3P," D. M. Ver Planck, K. S. Smith and J. A. Umbarger, February 1988.

6. STUDSVIK/SOA-88/01, " SIMULATE-3P Advanced Three Dimeny lonal Two Group Reactor Analysis Code," D. M. Ver Planck, K. S. Smith and J. A.

Umbarger, February 1988.

7. B&W-1484-7, " Critical Experiments Supporting Close Proximity Water Storage of Power Reactor Fuel", N. M. Baldwin, G. S. Hoovler, R. L.

Eng and T. G. Welfare, July 1979.

I 6. YAEC-1622, " Validation of the YAFC Criticality Safety Methodology",

D. G. Napolitano and F. L. Carpenito, January 1988.

9. "Burnup Credit Using Advanced Nodal Techniques," D. G. Napolitano and D. G. Adli, Sandia/EPRI Report, May 1990.

l 10. PNL-6205, " criticality Experiments to Provide Benchmark Dat a on y Neutron Flux Traps," S. R. Bierman, June 1988.

11. "EPRI-CELL Code Description," ARKP Package, Part II, Chapter 5, Pages 5-14, 5-15, 5-7C and 5-77, Octeber 1975.

12. EPRI-NP-6159, "An Assessment of Boraflex Perf ormance in Spent Naclear Fuel Storage Racks," December 1988.

i 13. B. Hubbard, et. al., "MICBURN-3/CASMO-3/ TABLES-3/ SIMULATE-3 Benchmarking of Vermont Yankee Cycles 9 Through 13," YAEC-1683, March 1989.

14. A. DiGiovine, et al., "McGuire Unit 2 SIMULATE-3 Benchmark Analysis Cycles 1 Through 3," YAEC-1608, October 1987,

15. K. Smith, et al., " SIMULATE-3 PWR Benchmark Report, Farley Unit II,"

STUDSVIK/SOA-87/10, April 1987

16. H. Hakansson, et al., " SIMULATE-3 BWR Benchmark, Forsmark Unit 1 Cycles 1 through 5," STUDSVIK/NFA-88/44, July 1988.

17. " Reactor Core Physics Design and Operating Data f or Cycles 1, 2, and 3 of the Surry Unit 1 PWR Power Plant," EPRI NP-79-2-LD, March 1979.

48

i L

APPFNDIX A

^

Validation of CASMD-3/ SIMULATE-3 for Fuel Storace r Burnup Credit Criticality Analysis l

The SIMULATE-3 advanced nodal method is firmly validated for LWR power distribution and reactivity cal culat ions . '"*"'"* "' Here, the methodology is extended to criticality analysis of LWR fuel in racks or casks. Two Cycles of the Surry nuclear power plant"" are analyzed for criticality and burnup performance. Validation of the methodology in a cold fuel storage rack environment is done by comparison to B&W fuel storage criticals"' and PNL flux trap critical s . '"' A composite I

uncertainty and bias based on the weighted average of reactor criticals and fuel storage criticals defines the calculational uncertainty of CASMO-3/ SIMULATE-3 in fuel storage burnup credit applications.

The difficulty of performing critical experiments with burned LWR fuel in racks or in casks necessitates burnup credit methods validation by comparison to core criticals. Nuclear reactor core criticals are relevant benchmarks to burnup credit criticality analysis because these types of criticals tect a methodology's ability to perform depletion and to handle the re0ctivity effects of heteroge wities and strong absorbers. Typical PWR validation studies compare: boron letdown, incore instrument reaction rate integrals, power distribution and control rod worths. In this validation study, the accuracy of the methodology in predicting criticality' and assembly burnup is at issue. Thus, this study will focus on CASMO-3/ SIMULATE-3 criticality prediction by precise core state point calculations and by comparison of measure and calculated assembly burnups.

A series of core follow calculations were performed with the SIMULATE-3 models of Surry Cycles 1 and 2. Reactor operation was explicitly modelled and included: power level, soluble boron, system I pressure, average coolant temperature and control rod bank D height.

model was allowed to converge on an eigenvalue at the specific reactor The conditions. Twenty six exposure state points for Cycle 1, and eleven exposure state pcints for Cycle 2 were modelled. The results are presented in Table A.l. The average eigenvalues and standard deviations are 0.99965 I 1 0.00198 and 0.99982 1 0.00106, for Cycles 1 and 2, respectively.

average eigenvalue from Surry Unit 1, Cycles 1 and 2, reactor criticals is The 0.99970 1 0.00176 for thirty se /en state points calculated.

, 49

An assembly burnup comparison between plant measured and SIMULATE-3 calculated is shown in Table A.2 for a subset of the region B fuel assemblies discharged from cycle 2; i.e., B-01 through B-31. The average difference (bias) and standard deviation (spread of bias) between plant measured and SIMULATE-3 calculated assembly burnup is 478 1 201 mwd /MtU.

The low positive bias indicates accurate power distribution model31ng, but the high spread is more indicative of measurement uncertainty. The uncertainty of the plant incore measurement system is 1 500 Mwd /Mtv.

Both the reactol critical statistics and fuel assembly burnup comparisons are in excellent agreement with the plant and establish a firm basis of confidence in CASMO-3/ SIMULATE-3's prediction of criticality and assembly burnup.

I In order to address the performance of CASMO-3/ SIMULATE-3 in fuel storage applications, a validation of the method was performed on a set of cores f rom the B&W fuel storage criticals and the PNL flux tr&p criticals.

Both set of criticals chosen are nodalizable by the CASMO-3/ SIMULATE-3 I nethodology.

The B&W criticals comprise 3x3 arrays of 14x14 fuel pins. The fuel pins were aluminum *.ubes with pellets of 2.46 % enriched UO 2. Two set of cores were studied: cores II, III, IX and X, and cores XI, XIII, XIV, XV, I XVIII and XIX. The first set varied water gap spacing between fuel clustera without poison sheets and the second set varied poison sheet absorber loading holding the gap spacing between clusters constant. These sets can be used to detect trends with storage array water gap spacing or poison loading, respectively.

The PNL flux trap criticals comprise assemblies of 2x2 arrays with UO, fuel enriched to 4.31 w/o *U. Neutron flux traps were created between fuel arrays using Boral sheets. Experimental criticality was studied as I

a f unction of Boral boron loading, and the space between the arrays and the flux trap sheets.

The results for the 10 B&W criticals are shown in Table A.3. The K.t statistics for CASMO-3/ SIMULATE-3 are 0.9992110.00367 with no significant trends in water gap spacing or poison loading. The results for 11 PNL

'I criticals are given in Table A.4. The Krr statistics for these 11 cores is 1.00293 1 0.00624, again, with no signi'icant trends in gap spacing or poison loading.

50 I

It is worth noting that these results are comparable in performance to KENO-Va Monte Carlo using the 123 group library and CASMO-3/PDQ-7 fine mesh diff usion theory in four groups. Thus, CASMO-3/ SIMULATE-3 nodal I diffusion theory is valid and accurate for LWR fuel storage criticality analysis when the storage array is nodalizable. l I

The CASMO-3/ SIMULATE-3 calculational bias and 95/95 uncertainty can be calculated from a weighted average of the statistics from the core I criticals (37 data points), the B&W fuel storage criticals (10 data points) and the PNL flux trap criticals (11 data points) . For the method bias this is, M=a [37*.00030+10*.00079+11 -0.00293)= .00023 (A-1)

I and the 95/95 calculational uncertainty is, l

I M,=2.030* (37 * . 001768+10 * . 003672+11 006242) = . 00694 (A~2) j I where 2.030 is the 95/95 one sided tolerance factor for 58 data points.

Now, the total uncertainty, AKg for Seabrook spent fuel rack criticality In calculations with CASMO-3/ SIMULATE-3, is from Equation 1, AK, = . 0 0 0 2 3 + . 0 0 6 0 9 +/ ( . 0 0 6 9 4 ) 2 + ( . 010 8 6 ) 2 = 018 7 5 (t.- 3 )

with the CASMO-3/ SIMULATE-3 calculational bias and 95/95 calculational I uncertainty. This total uncertainty is slightly higher than the pure CASMO-3 based number due to the lower negative bias. Also, this total uncertainty is used in defining the maximum reactivity acceptance line for CASMO-3/ SIMULATE-3 checkerboard calculations performed in Section 2.10.

I I

I I 51

I Table A.1 3 Burry Unit 1, Cycles 1 and 2, SI.WUI. ATE-3 Eigenvalues Reactor Conditions Burnup Power Boron Pres T.,. D-Bank Model (Gwd/Mtv) (4) (com) (esia) ('r 1 (inches) Eioenvalue I Cycle 1 0.270 0.288 0.0 41.6 1044 873

'250 2250 551.4 555.8 62.5 87.8 1.00124 1.00030 0.365 75.3 808 2250 565.0 101.1 1.00075 I 0.790 1.009 1.405 74.2 74.6 75.3 808 805 779 2250 2250 2250 564.4 566.0 566.0 117.5 116.8 125.1 1.00143 1.00142 1.00429 I 1.571 2.190 2.990 4.043 75.7 90.8 88.4 89.1 807 787 746 676 2250 2250 2250 2250 561.9 566.4 565.1 565.4 125.1 123.2 129.5 126.9 1.00007 0.99821 0.99969 1.00113

' I 4.990 5.915 6.968 90.9 88.3 94.0 649 582 548 2000 2000 2000 562.0 563.3 563.6 123.8 133.9 132.6 0.99796 1.00088 0.99694 8.045 95.9 458 2000 564.0 133.3 0.99913 I 9.130 9.470 9.995 97.8 98.6 98.7 394 372 364 2000 2000 2000 563.0 563.0 562.0 135.8 136.4 130.7 0.99829 0.99013 0.99404 10.405 99.4 302 2000 563.0 130.3 0.99884 I 11.050 11.630 12.000 96.3 98.6 99.5 265 233 204 2000 2000 2000 562.7 562.0 563.0 137.1 136.4 142.1 0.99857 0.99710 0.99770 12.530 99.8 149 2000 564.0 143.4 0.99968 I 12.960 13.120 13.330 90.6 60.6 59.8 110 145 156 2000 2000 2000 563.0 563.0 560.0 143.4 128.2 135.2 1.00083 1.00215 1.00092 13.560 61.0 148 2000 561.0 132.6 0 , 9 9 9 f,0, Average N,,i o = 0.999651 00198 Bias = 0.00035 Cycle 2 0.115 98.4 624 2250 564.0 137.7 1.00189 0.530 99.6 600 2250 564.0 135.2 0.99948 1.150 98.9 567 2250 565.0 137.1 0.99888 I 1.685 2.020 2.570 99.2 99.1 100.0 523 487 474 2250 2250 2250 566.0 566.0 562.0 142.1 141.5 144,0 1.00002 1.00134 0.99851 3.010 100.0 439 2250 562.0 144.0 0.99871 3.552 100.0 381 2250 566.0 140.8 1.00025 4.025 99.9 351 2250 565.0 143.4 0.99987 4.400 99.1 327 2250 566.0 138.9 0.99920 5.080 99.5 272 2250 565.6 143.4 0.99988 Average A,,+ 0 = 0.99982+.00106 Bias = 0.00018 I

I 52 I

1 I l 1

Table A.2 surry a ruel Assembly Discharge Exposures From cycle 2 serial Measured SIMULATE-3 Difference EHhkfl. (Wwd/NtU) _(Wwd/WtU) _(Wwd/MtU)

B-01 22989 2??51 -238 i B-02 22472 23034 +562 l B-03 22941 23034 +93 l B-04 21474 21342 -132 I

B-05 22724 23034 +310 B-06 22614 22657 -43 B-07 23111 23134 -23 B-00 22974 23134 +160 B-09 22604 22657 +53 B-10 21981 22053 +72 B-11 22932 23034 +102 B-12 21435 21342 -93 B-13 22846 23034 +188 B-14 23426 23592 +166 B-15 22841 22657 -184 B-16 21697 22053 +356 B-17 24286 24177 -109 B-18 21946 22053 +107 B-19 22468 22657 +189 B-20 22914 23134 +220 B-21 24344 24177 -167 B-22 22245 22657 +412 B-23 21066 21342 +276 B-24 21956 22053 +97 B-25 22635 22657 +22 B-26 23550 23592 +42 I B-27 B-28 23032 21570 22751 21342

-281

-228

+286 B-29 22848 23134 I B-30 B-31 22521 21412 22657 21342 Average Difference i 10 = +78 1 201

+136

-70 I

I 53 I

(

w l

Table A.3 CASuo-3/SIMUIATE-3 B&W Fuel Storage Critical Results f

L Soluble Array Gap Fixed Poison Core Boron (pre). Spacina tos) JAggi __E. , u II 1037 0 -

1.00284

{

III 764 1.636 -

1.00335 IX 0 6.544 -

0.99813 X 143 4.908 -

0.99823 XI $14 1.636 SS-304 1.00245 XIII 15 1.636 1.614* 1.00313

{ XIV 92 1.636 1.257 0.99B92 XV 395 1.636 0.401 0.99250

( XVil 487 1.636 0.242 0.99569 XIX 634 1.636 0.100 0.99689 Average K.,, i c = 0.99921 1 0.00367 g 31as = 0.00079 j *w/o B in borated aluminum sheet 1

I I

I I

I I

I I 54 1

L I

Table A.4 CABMo-3/sINULATr-3 INL 71ux Trap critichis Amaults Lattios_ Gap (cus)

• Borel (as/cre')_ K , ,_,_,

,7,,0309

{ 231 214 0.295 0.295 0.45 0.36 1

1.00521 228 0.295 0.13 1.00106 230 0.295 0.05 0.99762 229 0.295 0.0 (A1) 1.01382

~

224 2.186 0.36 1.00686 223 4.077 0.36 0.99880 1 221 7.859 0.36 1.00530 220 9.750 0.36 1.01035 226 5.967 0.45 0.99578 227 5.967 0.13 0.99354 Average K.,, + 0 = 1.00293 1 0.00624 1 Blas = 0.00293

• Flux Trap to ruel Array Separation i

I I

I I

55

_ _ _ _ _ _ _ _ - _ - _ _ _ _ - _ _ _ -