ML20211C049

Rev 0 to Final Rept Criticality Safety Analysis Hb Robinson New Fuel Storage Vault W/4.2% Enriched 15x15 Fuel Assemblies

ML20211C049
Person / Time
Site:	Robinson
Issue date:	09/02/1986
From:	Gerrald L, Pieper J, Skogen F SIEMENS POWER CORP. (FORMERLY SIEMENS NUCLEAR POWER
To:
Shared Package
ML14190A959	List: ML20211C049 ML20211C064
References
XN-NF-86-100, XN-NF-86-100-R, XN-NF-86-100-R00, NUDOCS 8610210259
Download: ML20211C049 (54)

Text

XN-\ =-86 00 l

=

\ A _ R E 309-~

CR- CA_ - Y SA;=: Y A\ A_YS S

. 3. 903 \SO\ \ EW =UE_ S- ORAGE VAU_--

l W- -i 4.2% E\ 9 C-EJ 15 X15 =UE_ ASSEV 3_lES g AUGUS- 1986 I SE3- EMBER 1986 l

-,2 I

l- EXXON NUCLEAR COMPANY, INC.

puo2;gggggaggg1 P

I I XN-NF-86-100, Rev. O Issue Date 9/2/86 I

l FINAL REPORT l

CRITICALITY SAFETY ANALYSIS H. B. ROBINSON NEW FUEL STORAGE VAULT WITH 4.2% ENRICHED 15xl5 FUEL ASSEMBLIES AUGUST 1986 1

I I

Prepared by L. D. Gerrald I

I I

E I

i 1

ERON NUC_ EAR COV DANY NC.

I

.I

l 1 i

XN-NF-86-100, Rev. 0 l

!ssue Date: 9/2/85 W I I

l FINAL REPORT l l

l CRITICALITY SAFETY ANALYSIS H. B. ROBINSON NEW FUEL STORAGE VAULT l

l WITH 4.2% ENRICHED 15xl5 FUEL ASSEMBLIES AUGUST 1986 l

l 5

Prepared by:

L. D. Gerrald, Senior Engineer Date Corporate Licensing i

Reviewed by: --

I/24/94 J. E./Pieper, Engineer ' Date Corporate Licensing J.32. - WrIs1s F.'3. Skogen, Managery Date PWh Neutronics N

I Approved by:

C. W. Malody, Manager ,/

Corporate Licensing V

_._ f 26

/ Date

/

W T. W. Potten, Manager

$fAS//b Date Neutronics and Fuel Management

<g 3,s I

I _

I I

CUSTOMER DISCLAIMER I

IMPORTANT NOTICE REGARDING CONTENTS AND USE OF THIS DOCUMENT PLEASE READ CAREFULLY Exxon Nucteer Company's warranties and representations concoming the subject matter of this document are those set forth in the Agreement between Exxon Nucieer Company, Inc. and the Customer pursuant to which

' this document is issued. Accordingly, except as otherwise expressly provedad in such Agreement, netther Exxon Nuclear Company, Inc. nor any person acting on its behalf makes any worrenty or representation, expressed or implied, with roepect to the accuracy, comodeteness, or usefulness of the information contained in this document, or that the use of any information, apparatus, method or process disclosed in this document will not infringe privetsey owned rights; or assumes any liabilities with respect to the use \

of any information, apparatus, method or process discioned in this document.

The information contained herein is for the solo use of Customer, in order to avoid imperment of rights of Exxon Nuclear Company, Inc.

in patents or inventions which may be included in the information contained in this document, the recipient, by its acceptance of this document agrees not to publish or make public use (in the patent use of the term) of such information until so authorized in writing by Exxon Nucteer Company, Inc. E or until after six (6) months following termination or expiration of the aforesaid Agreement and any extension thereof, unless otherwise expreesty provided in the Agreement. No rights or licenses in or to any patents are implied by the fumashing of this document.

I i

i MPHep-F00,76e i

I I  ;

l ii XN-NF-86-100, Rev. O i

FINAL REPORT I

l CRITICALITY SAFETN ANALYSIS H. B. ROBINSON NEW FUEL STORACE VAULT WITH 4.2% ENRICHED 15xl5 FUEL ASSEMBLIES AUGUST 1986 TABLE OF CONTENTS l

1 SECTION Page l.

I l.0 l.1 INTRODUCTION Summary I

I 2.0 FUEL PARAMETERS 2 3.0 STORAGE RACK CEOMETRY 4

{

4.0 CALCULATION METHODS 5 4.1 Normal and Credible Abnormal Conditions 5 4.l.1 Maximum Allowable Multiplication Factor 5 4.1.2 Fuel Assembly Effects 6 1 4.1.3 Fuel Assembly Arrangement Effects 7 4.1.4 Other Assembly Spacing Effects 7 4.1.5 Fixed Absorbers 8 5.0 MODERATION EFFECTS 9 I 5.1 Fully Flooded Conditions 9 I 5.2 5.2.1 5.2.2 Low Density Moderator Effects K-infinite Calculations K-eff Calculatioris for Finite System 11 12 14 6.0 ALTERNATE BUNDLE LOADING ARRANGEMENTS 17 6.1 Option A (73 Bundles) 17 l

6.2 Option B (69 Bundles) 19 6.3 Option C (73 Bundles) 22 6.4 Option D (72 Bundles) 23 l

7.0 SENSITIVITY STUDIES 26 A 7.1 Bundle Offset Effects 27 8.0 FUEL HANDLING ACCIDENTS 23 8.1 Bundle Drop Accidents 28 8.2 Misplaced Bundles Within Rocks 30 8.3 Bundle-Bundle Interactions During Handling 33 I _

iii XN-NF-86-100, Rev. O l

~

TABLE OF CONTENTS (Cont.d)

Page I

SECTION 9.0 METHODS VERIFICATION 35 9.1 Reference 2 Experiments 35 9.2 Reference 3 Data 36 9.3 Reference 4 Data 37 9.4 Reference 5 Data 38 9.5 Acceptability Limit 40 9.6 Cross Section Comparisons 42 10.0 COMPUTER INPUT LISTINGS 43 10.1 KENO Input: K-Infinite Case at Full Flooding 43 10.2 KENO Input: New Fuel Vault, Option D Model, 5%

Water 44 ,

I l.0 REFERENCES 45 I

I I

I I

i 8

I I

I

iv XN-NF-86-100, Rev. O LIST OF TABLES l Table Page 2.0 Bundle Parameters 2 l 5. l a Fuel Assembly Composition 9 5.lb Rod Pitch (Moderation) Effects Full Density Water l Moderation XSDRNPM Results 10 5.lc Generic Bundle Parameters !I

5. 2.1 a KENO Rcsults for K-Infinite Calculations Knight-I l

Modified Hansen-Roach Cross Sections Interspersed Moderation Ef fects Bundles on 20.875 Inch Centers 13 5.2. l b XSDRNPM Results infinite Array of Edge-Edge Bundles 14 l

5.2.2 New Fuel Storage Rock Arrangement 16 6.1 Option A Arrangement Moderation Ef fects 18 f 6.2 Option B Arrangement Moderation Effects 20 6.4 Option D (72 Bundles) Moderation Effects 24 l

7.0 XSDRNPM Results Bundle Spacing Effects 100%

Interspersed Water Density 1

26 l 8.2 Fuel Handling Interactions Array / Spacing Effects 34 9.1 Benchmark Results Data of Reference 2 36 l

9.2 Benchmark Data From Reference 3 37 9.3 Reference 4 Dato KENO-IV With Hansen-Roach Cross Sections 38 9.4a Fuel Design Parameters 39 9.4b Reference 5 Data Low Density Moderation Between Bundles KENO Results 40 9.6 Cross Section Comparisons 42

g l

I v XN-NF-86-100, Rev. O L LIST OF FIGURES I Figure Page 2.0 Rod Arrangement Within Bundle 3 5.2.2 New Fuel Storage Rack Arrangement 15 6.la Option A (73 Bundles) 17 6.lb Option A - Estimated Normalized Fission Densities 19 l 6.2a Option B (69 Bundles) 20 6.2b Option B Estimated Normalized Fission Densities 21 1 1 l

6.3 Option C (73 Bundles) 23 i

i 6.4 Option D (72 Bundles) 24 8.2a Option D.! (74 Bundles) 31

~

8.2b Option D.2 (76 Bundles) 32 8.2c Option D.3 (76 Bundles) 33 1

i l

l I

I I

L

I I XN-NF-86-100, Rev. O Page 1 FINAL REPORT I-CRITICALITY SAFETY ANALYSIS g' H. B. ROBINSON NEW FUEL STORAGE VAULT s WITH 4.2% ENRICHED 15xl5 FUEL ASSEMBLIES AUGUST 1986 I

I I

I l.0 INTRODUCTION 8 The criticality safety of the new fuel storage vault with 4.2 6 enriched 15x15 l

bundles is demonstrated in accordance with NUREG-0800 and ANSI /ANS-57.3-1983.

.. l.1 Summary The new fuel storage vault meets the applicable criticality safety criteria subject to the limits and controls given below. )

1. Fuel design - os specified in Table 2.0.

l

2. Vault loading arrangement - any of the options in Section 6. '

I I

i

l I

I XN-NF-86-100, Rev. O Page 2 2.0 FUEL PARAMETERS Key bundle design parameters used in these calculations are listed in Table 2.0.

I All fuel rods in the model were 4.20% enriched. The arrangement of the fuel rods and the instrument / guide tubes is shown in Figure 2.0.

Table 2.0 Bundle Parameters Parameter Design Value Model Value Enrichment (wt% U-235) 4.20 (max.) 4.20 Pellet Diameter (inch) 0.3565 0.3565 Pellet Density (%TD) 94.0 + 1.5 95.0 Pellet Dish -Volume (%) 1.0 0 I. Stock Length (inch) 132 (enr) + 12 (nat)* 144 (min.)

Clad ID/OD (inch) 0.364/0.424 0.364/0.424 g Rod Pitch (inch) 0.563 0.563 3 Gd 023 Content .

Variable None Fuel Rods per Bundle 204 (max.) 204 Guide / Inst. Tube ID/OD (inch) 0.511/0.544 0.511/0.544

• NOTE: The design length for some UO 2-Gd 02 3 stacks may be different but the enriched zone will be 132 inches maximum.

I I

I I

I I 1

XN-NF-86-100, Rev. O Il l Page 3 l 1

ROW / COL I 2 3 4 5 6 7 8 9 10 11 12 13 14 15 I F F F F F F F F F F F F F F F 2 F F F F F F F F F F F F F F F 3 F F G F F G F F F G F F G F F 4 F F F F F F F G F F F F F F F 5 F F F F G F F F F F G F F F F [l W

6 F F G F F F F F F F F F G F F )

7 F F F F F F F F F F F F F F F  !

8 F F F G F F F I F F F G F F F 4 I

9 F F F F F F F F F F F F F F F 10 F F G F F F F F F F F F G F F 11 F F F F G F F F F F G F F F F 12 F F F F F F F G F F F F F F F 11 F F G F F G F F F G F F G F F 14 F F F F F F F F F F F F F F F g 15 F F F F F F F F F F F F F F F g Key: F = Fuel Rod G = Guide Tube I = Instrument Tube Figure 2.0 Rod Arrangement Within Bundle l

I I

I I

I I

I I J

I XN-NF-86-100, Rev. O L Page 4

~

l 3.0 STORAGE RACK GEOMETRY r

L The new fuel storage vault contains 105 cel!s. The racks are in a 10x10 array with half of an additional row.

The bundles are on a nominal 21-inch square pitch within the racks. The modeled pitch was 20.857 inches for all locations.

The total height of the rocks is about 14 feet. The new fuel vault was reflected

- within 30 cm of concrete at the four walls, the floor, and at 14 feet above the floor (ceiling).

All rack materials of construction were neglected in the models.

L__

Thus, the models are conservative in geometry, reflection, and neutron absorption E eifects.

l

=

m W

l I

I I XN-NF-86-100, Rev. O Page 5 4.0 CALCULATION METHODS I

All computer codes and cross sections used are part of the SCALE (I) system.

The neutron multiplication factors, kinf and keff, were calculated using KENO-IV or XSDRNPM.

I The 16 group Knight-modified Hansen-Roach cross section library was used for most calculations. Reference cases were also calculated using the 27 group library (subset of 218 group library based on ENDF/B-IV) and with the 123 group GAM-THERMOS library. For the reference calculations, the resonance self-shielding calculations were performed using NITAWL.

I All codes and cross sections have been extensively benchmarked against critical experiment data.

Evidence of methods verification is ' presented later in this document.

I 4.1 Normal and Credible Abnormal Conditions I The conditions specified in Section 6.2.4 of ANSI /ANS-57.3-1983 have been evaluated.

Maximum Allowable Multiplication Factor I 4.1.1 Sections 6.2.4.1.1 through 6.2.4.1.3 of ANSI /ANS-57.3 relate to methods validation and establishing the maximum allowable multiplication factor. As detailed in Section 9.0 of this report, the maximum allowable factors are 0.95 at full flooding, and 0.98 at optimum moderation.

I I

I

I XN-NF-86-100, Rev. O Page 6 4.1.2 Fuel Assembly Effects Sections 6.2.4.1.4 and 6.2.4.1.5 of ANSI /ANS-57.3 specify that the most reactive fuel assembly be evaluated. The parameters considered include:

1. Maximum fissile loading. This analysis covers a fuel design described in Section 2.0 of this report. Based on a 95.5 percent theoretical density pellet density and a 1.0 volume percent dish, the average pellet stack density is 94.55 percent theoretical density moxinium while the modeled value is 95 percent theoretical density. The modeled pellet stock length is 12 inches I

greater than the design value. The model contains the maximum credible E number of fuel rods. The effect of pellet diameter tolerance is . negligible.

Therefore, the maximum fissile loading is conservatively modeled.

2. Fuel rod diameter. The effect of tFe 0.0015 inch diameter tolerance is negligible.

3. Cladding effects. The clad and the pellet-clad gap were explicitly modeled at the nominal dimensions. Tolerance ef fects on the order of 0.002 inch are negligible. The clad war, modeled as pure zirconium at 100 percent theoretical density.

4. Pellet density . As stated earlier, the modeled pellet density conservatively I

simulates the maximum credible pellet density.

5. Fuel rod pitch and number of rods in assembly. The modeled pitch is the nominal value. The modeled number of fuel rods is the maximum credible value. The ef fects of removing fuel rods from o flooded assembly are detailed in Section 5.1.

6. Absence of fuel rods in certain incotions. The instrument / guide tubes were explicitly modeled. Other effects are given in Section 5.1.

I Il

I XN-NF-86-100, Rev. O Page 7 I 7. Burnable poison. None.

8. Distribution of fissile content. All rods were 4.2% enriched. The locations are shown in Figure 2.0.

I 4.l.3 Fuel Assembly Arrangement Effects I Sections 6.2.4.l.6 specifies that the most reac tive credible arrangement be evaluated.

l. Spacing between assemblies. All storage cells are on a 21-inch nominal square pitch. The cells were conservatively modeled on 20.875 inch centers.

Sensitivity study results for bundle spacing effects are in Section 7.0 of this report.

2. Moderation between assemblies. Moderation ef fects were evaluated by I modeling the vault with uniformly interspersed water of various densities in the range 0-l.0 gm/cc. Results are detailed in Sections S.2 and 6.0 of this report.

3. Fixed neutron absorbers. None.

4.l.4 Other Assembly Spacing Effects I

Section 6.2.4.1.7 o f ANSI /ANS-57.3 specifies that the following ef fects be evaluated:

1. Eccentricity of assembly locations and other spacing effects. Evcluation in Section 7.0 of this report.

I I

z I

XN-NF-86-100, Rev. O Page 8

.I

2. Dimensional tolerances. The bundle pitch was conservatively modeled as 20.875 inches for all locations. In addition, the results in Section 7.0 indicate that closer spacings would be acceptable.

3. Construction materials. The effects of neutron absorptions in the materials of I

construction were conservatively neglected.

4. Moderator / fuel density effects. Already covered.

5. Burnable poison. None.

6. Cell wall materials. None.

4.1.5 Fixed Absorbers I

Section 6.2.4.2 of ANSI /ANS-57.3 allows credit for fixed absorbers. None were modeled in this analysis.

All criticality safety aspects of ANSI /ANS-57.3 have been covered.

I I

I I

I I

I IJ

I I XN-NF-86-100, Rev. O Page 9 I

5.0 MODERATION EFFECTS 5.1 Fully Flooded Conditions ,

I If fuel rods are removed from an assembly, the keff for flooded assemblies may be effected.

The modeled fuel design (i.e., 204 fuel rods and 21 instrument / guide tubes) is composed as noted in Table 5.la.

I Table 5.la I Fuel Assembly Composition Material Volume Percent Fuel 28.55 Pellet-Clad Gap I Clad (Zr)

Water l.21 I1.43 58.86 Total 100.0 The average water / fuel volume ratio (Vw/Vf) is 2.06. If the entire 15xl5 array was fuel rods, the Vw/Vf would be 1.76.

The ef fect of changes in Vw/Vf on kef f was evaluated for generic assemblies using XSDRNPM. The generic assemblies were modeled with the nominal pellet and clad dimensions but with rod pitches to yield Vw/Vf ratios in the range 1.0 to 4.0. The calculations were made as follows:

1. Self-shielded 16 group cross sections were generated using BONAMI/NITAWL.

I I

I

I XN-NF-86-100, Rev. O Page 10

.I

2. The unit rod cell was cell-weighted using XSDRNPM. The king for the infinite rod lattice is listed in Table 5.lb.

3. These cell-weighted (homogeneous) cross sections were then used in a XSDRNPM model of 8.445" x 8.445" x infinite bundles on 20.875 inch centers.

Full density water was within and between bundles. The assembly and the water between assemblies were modeled as concentric cylindrical regions. The kinf for the infinite bundle lattice is listed in Table 5.lb.

Table 5.lb I

Rod Pitch (Moderation) Effects Full Density Water Moderation XSDRNPM Results Rod k ing kinf Pitch (Rod (Bundle Vw/Vf (cm)_ Lattice) Lattice) 1.0 1.2470 1.3765 0.8333 1.5 1.3700 1.4458 0.8896 2.0 1.4829 1.4791 0.9209 2.5 1.5877 1.4939 0.9378 3.0 1.686I l.4977 0.9454 3.5 1.7790 1.4933 0.9465 5 4.0 1.8673 1.4838 0.9432 5 The optimum Vw/Vf is near 3.5. An infinite array of flooded bundles on 20.875 inch centers is acceptable of any rod pitch. The octual assembly (l.43 cm pitch) is obviously undermoderated even with 21 instrument / guide tubes.

Due to the fixed 8.445 inch square bundle modeled with various rod pitches, the bundles are not 15xl5 rod arrays. The generic bundle parameters are listed in Table 5.lc.

Ii IJ

I I XN-NF-86-100, Rev. O Page lI I Table 5.lc i I Gendric Bundle Parameters Array Vw/Vf (Rods / Edge) Total Rods I.0 17.2 295.9 I 1.5 2.0 2.5 15.7 14.5 I3.5' 245.1 209.2 I82.5 3.0 12.7 16I.8 3.5 12.1 145.4 4.0 I l .5 I32.0 I

The peak reactivity condition (Vw/Vf = 3.5) corresponds to about 80 removed fuel rods; i.e., about 59 in addition to the 21 instrument / guide tubes. The reactivity of the generic bundle filled with rods at optimum pitch is bounding for the actual bundle with removed rods.

5.2 Low Density Moderator Effects Flooding with full density water is the most reactive state for a single bundle, or for an array of bundles placed closely (one rod pitch or less) together.

As the edge-edge spacing between bundles in a flooded array is increased above zero, the array keff declines to the asymptotic value for a single bundle with full water reflection; i.e., bundle-bundle interactions eventually become negligible. This asymptotic keff is 0.917 3 0.005. It will be shown that the 21 inch (20.875 inches) center-center spacing between bundles in the rocks is more than adequate to result lI in negligible bundle-bundle interaction at full flooding.

!I I

I

I XN-NF-86-100, Rev. O Page 12 I

However, if the water density within the rocks is lowered below 100 percent (full g density water flooding), the bundle-bundle interactions will increase. As the 5 decrease in water density continues, the competing effects of decreased moderation and increased interactions typically result in a peak keff value. This is the optimum moderation or peak reactivity condition. This peak reactivity condition was determined for the new fuel vault. The calculations were made using 16 group cross sections (Hansen-Roach) and with the KENO-IV code'(3-D Monte Carlo).

The appropriate U-235/U-238 nuclides in the Knight-modified Hansen-Roach library were selected based on the effective moderator cross section (sig-m (ef f)) at each of the interspersed moderation conditions.

The interspersed moderator was water at various percentages of full density.

5.2.1 K-infinite Calculat;oas A single bundle was explicitly modeled using KENO-IV. The bundle was centered in E

l a 20.875" x 20.875" x infinite cuboid of between-bundle water of the same density 5 as the within-bundle water. The fuel zone length was infinite.

l With specular reflection at the six faces of the cuboid, the model simulates an infinite x infinite array of infinite length bundles on 20.875 inch centers.

The KENO results at several moderator densities are listed in Table 5.2.la.

e I

I

! I I

l IJ L

E XN-NF-86-100, Rev. O Page 13 l

I l Table 5.2.la KENO Results for kinf Calculations Knight-Modified Hansen-Roach Cross Sections interspersed Moderation Effects l Bundles on 20.875 Inch Centers l

Water Density Sig-m (eff) DANCOFF

(%) (U-235/U-238) (Bundle-Average) k i nf I

l I 640/28.4 0.810 1.070 + 0.0029 2.5 717/31.8 0.758 1.303 7 0.0037 5 776/34.5 0.712 1.399 7 0.0035 7.5 825/36.6 0.684 1.366 7 0.0030 j 10 880/39.1 0.650 1.292 T 0.0040 1.192 ! 0.0037 I 895/39.7 12.5 0.631 15 931/41.4 0.607 1.101 ! 0.0035 20 1007/44.7 0.564 0.944 ! 0.0039 l 100 1593/70.7 0.243 0.917 i 0.005 1

These results indicate that an infinite system would be adequately subcritical at all interspersed water densities greater than about 20 percent.

For an infinite system, the optimum interspersed water density is about five l percent.

B l Cell weighted macroscopic cross sections were prepared for each of the moderation levels in Table 5.2.la. The cell weighting was based on the KENO fluxes from the ki nf calculations. These macroscopic cross sections were used in subsequent calculations which employed a homogeneous bundle model.

For reference, the k nf i values for these homogeneous bundles are listed in Table 5.2.lb. These data support the earlier statement that the most reactive state for edge-edge bundles is at full flooding.

I XN-NF-86-100, Rev. O Page 14 1

Table 5.2.Ib I

XSDRNPM Results E Infinite Array of Edge-Edge Bundles g Water Density

(%) k i nf I 0.800i E 2.5 0.8405 5 5 0.8969 7.5 0.9528 10 1.0039

12.5 1.0515 t 15 1.0989 l 20 1.1748 g i 100 1.5069 m l

l i lt is noted that the 1.5069 value above is slightly higher than the peak value in I

Table 5.lb. This is attributed to:

1

1. Slight dif ferences in XSDRNPM/ KENO models, i.e., Cylindrical / Cuboidal models.

2. The weighting environment in XSDRNPM was an infinite rod lattice, while that in KENO is ef fectively one bundle in a sea of water. The overage i ef fective moderation level in the KENO case is higher than that for the XSDRNPM case.

Therefore, the homogeneous bundle model used is appropriate and conservative.

5.2.2 K-Effective Calculations for Finite System l

The vault contains 105 bundle storage locations on 21-inch nominal centers. The

! modeled center-center spacing was 20.875 inches (conservative).

I

w <

l

[

L 1-XN_NF_86_100, Rev. O Page 15 i The storage rocks are a 10x10 array with half of an eleventh row as shown in Figure 5.2.2. The homogeneous bundle model was used in these calculations.

[ _ . ___

ROW / COL I 2 3 4 5 6 7 8 9 10

[ 11 M M M M M F F F F F 10 F F F F F F F F F F 9 F F F F F F F F F F

[ 8 F F F F F F F F F F 7 F F F F F F F F F F 6 F F F F F F F F F F

( 5 4

.F F

F F

F F

F F

F F

F F

F F

F F

F F

F F

L 3 F F F F FL 8- F F F F

[ 2 I

F F

F F

F F

F F

F F

F F

F F

F F

F F

F F

KEY:F = Bundles on 20.875 inch centers M = Moderation-filled (no fuel)

Figure 5.2.2 New Fuel Storage Rock Arrangement The vault was reflected with 30 cm of close_ fitted concrete at all six faces. The kef t results for the rocks full of 14 foot long bundles (rock length filled) are given in Table 5.2.2. It should be noted that the actual fuel length is 12 feet, including six inches of natural uranium at each end. Unless specifically stated otherwise, all results, except those of Table 5.2.2, are for a 12 foot long enriched zone (no

[ natural).

[

[

C. - - - - - - - - - --- -- -- -

I XN-NF-86-100, Rev. O Page 16 I

Table 5.2.2 I

New Fuel Storage Rocks E l05 Bundles of 14 Foot Length g Moderation Effects Water Density

(%) kerf I 0.754 + 0.0049 2.5 0.928 + 0.0051

'5 1.109 ~ 0.0048 7.5 1.154[0.0044 10 1.144 + 0.0039 12.5 1.082 + 0.0045 Table 5.2.2 results indicate:

1. The peak keff for the finite system with concrete reflection occurs with an interspersed water density near 7.5 percent.

2. The kef f for water densities in the approximate range of 2.5 percent to 20 percent is unacceptable if all locations are filled with bundles.

I I

I I

I I

I 1

XN NF-86-100, Rev. O Page 17 6.0 ALTERNATE BUNDLE LOADING ARRANGEMENTS Since the kef f of the new fuel vault may exceed 0.98 if fully fooded, and with optimum interspersed moderation, alternate bundle loading patterns were investi-gated to determine which patterns would have a peak k eff ess l than 0.98.

l l

6.1 Option A (73 Bundles)

The vault was loaded as shown in Figure 6.la. The basic pattern in the 10x10 part of the array is full loading of one peripheral row / column, and a checkerboard pattern of the internal 8x8 subarray. All of the five " extra" locations are filled I with bundles.

I j The model contains 73 bundles, and 32 empty locations.

I l

ROW / COL l 2 3 4 5 6 7 8 9 10 l 11 M M M M M F F F F F 10 F F F F F F F F F F 9 F F M F M F M F M F l 8 F M F M F M F M F F g 7 F F M F M F M F M F g 6 F M F M F M F M F F i 5 F F M F M F M F M F l 4 F M F M F M F M F F I

l 3

2 l

F F

F F

M F

M F

F F

M F

M F

F F

M F

M F

F F

M F

M F

F F

F F

\

KEY: F = Fuel Bundle (moderated)

M = Empty (moderation only)

Figure 6.la Option A (73 Dundles)

I I

I XN-NF-86-100, Rev. O Page 18 I

The keff of the option A arrangement is shown in Table 6.1 at various interspersed moderator densities.

I 1

Table 6.1 Ill Option A Arrangement l Moderation Effects l Water Density keff keff g

(%) (14' Bundle) (12' Bundle) g 1 0.729 + 0.0055 0.697 + 0.0044 2.5 0.909 ! 0.0051 0.859 + 0.0049 5 1.005 7 0.0048 0.961 ! 0.0051 {l m

7.5 0.96I i 0.0041 0.949 i 0.0045 1

1 The Table 6.1 results indicate a peak keff (12 foot bundle length) less than 0.972 1 l (95% confidence) at five percent interspersed water. The axial leakage ef f ect is seen to decrease with increasing moderation, l

I.

The estimated normalized fission densities for the bundles at five percent water is l shown in Figure 6.lb. These values have an average standard error of about six percent relative.

l I' l

l I

l l

1 I

l I I

I I

I l

XN-NF-86-100, Rev. 0 Page 19 I

j. ROW / COL i 2 ~3 4 5 6 7 8 9 10 1I 0.78 0.69 0.69 0.60 0.62 l 10 0.39 0.74 0.80 0.99 1.06 0.92 1.04 0.98 0.79 0.60 I

9 0.60 1.03 1.30 1.34 1.20 0.88 8 0.71 1.1 I l.47 1.56 1.51 0.98 7 0.97 l.37 l.48 1.50 1.45 0.96 l 6 0.82 1.12 1.54 1.49 l.49 0.88 I 5 4

3 0.86 0.74 1.10 0.73 1.43 1.20 1.33 1.65 1.33 1.62 1.54 1.40 1.51 1.32 0.74 0.77 0.88 0.62 2 0.65 0.70 1.15 1.26 1.00 0.58 1 0.48 0.56 0.67 0.63 0.81 0.79 0.89 0.68 0.58 0.36 I

i Figure 6.lb Option A - Estimated Normalized Fission Densities 1

3 The interior bundles have the highest fission densities even with a checkered arrangement.

6.2 Option B (69 Bundles)

I This arrangement has two peripheral rows / columns of the 10x10 array filled. The I interior 6x6 subarray contains moderation only.

locations are filled with bundles. The arrangement is shown in Figure 6.2a.

As before, the five "e x t ra" The model contains 69 bundles, and 36 empty locations.

I I

I I

I

I XN.NF.86-100, Rev. O Page 20 I

ROW / COL I 2 3 4 5 6 7 8 9 10 I

11 M M M M M F F F F F 10 F F F F F F F F F F 9 F F F F F F F F F F 8 F F M M M M M M F F 7 F F M M M M M M F F 6 F F M M M M M M F F 5 F F M M M M M M F F 4 F F M M M M M M F F 3 F F M M M M M M F F 2 F F F F F F F F F F I F F F F F F F F F F KEY: F = Fuel bundle (moderated)

M = Empty (moderation only)

Figure 6.2a ,

Option B (69 Bundles)

The kef f of the option B arrangement is shown in Table 6.2 of various interspersed moderator densities.

.... .. .= . ..

Table 6.2 Option B Arrangement Moderation Effects Water Density

(%) keff 2.5 0 767 + 0.0054 l 5 0.852 + 0.0055 7.5 0.885 I 0.0057 10 0.914 7 0.0047 l l 2.5 0.884 + 0.0057 5 IJ

I XN-NF-86-100, Rev. O Page 21 The peak keff is less than 0.924 (95 % confidence). The optimum water density is about 10 percent.

When compared to the option. A arrangement with only four more bundles, it is apparent that interior bundles tend to raise the keff more than peripheral bundles.

I The estimated normalized fission densities for the option B arrangement with 10 percent interspersed water are listed in Figure 6.2b. These values have on average standard error of about six percent relative.

ROW / COL I 2 3 4 5 6 7 8 9 10 I lI 10 9

l.90 2.62 2.84 2.90 1.88 0.36 0.52 0.74 1.02 1.60 2. I I 3.01 3.80 3. I 7 2.5 I 0.54 0.57 0.68 1.07 1.17 1.77 2.49 2.95 3.09 2.66 8 0.38 0.47 2.36 2.00 7 0.38 0.43 1.57 f.30 6 0.29 0.29 1.02 0.97 5 0.19 0.17 0.72 0.68 I 4 3

2 0.14 0.11 0.17 0.13 0.13 0.23 0.20 0.24 0.15 0.24 0.30 0.40 0.52 0.45 0.65 0.47 0.62 0.37 0.19 0.20 0.21 0.17 0.08 0.17 0.27 0.34 0.34 0.31 I

l Figure 6.2b Option B - Estimated Normalized Fission Densities The fission density data indicate that the bundles nearest the five " extra" locations have much higher reactivities than the other bundles.

I I

I I

I XN-NF-86-100, Rev. O Page 22 To assure adequate Monte Carlo sampling of the most reactive region, the option B arrangement case with 10 percent water was repeated with most of the first generation neutrons started in the bundle at location row 10, column 8 as depicted in Figure 6.2b. The resulting keff was 0.911 3 0.0049 which is statistically identi-col to the result using the cosine start typically employed in KENO calculations.

For reference, the keff for the option B arrangement, except with the five " extra" locations containing moderation-only (10 percent water), is 0.870 1 0.0043. With g the five " extra" locations without fuel, the model includes 64 bundles. W 6.3 Option C (73 Bundles)

This is a minor variation of the option B arrangement in that four additional bundles (2x2 array) were added to the center of the 10x10 part of the array. The five " extra" bundles were in this model for a total of 73 bundles as shown in Figure 6.3 I

I I

I I

I I

I 9 I

XN-NF-86-I00, Rev. 0 l Page 23

. .== ....

ROW / COL I 2 3 4 5 6 7 8 9 10 I ll 10 9

M F

F M

F F

M F

F M

F F

M F

F F

F F

F F

F F

F F

F F

F F

F F

I 8 7

6 F

F F

F F

F M

M M

M M

M M

M F

M M

F M

M M

M M

M F

F F

F F

F 5 F F M M F F M M F F 4 F F M M M M M M F F I 3 F F M M M M M M F F 2 F F F F F F F F F F 1 F F F F F F F F F F KEY: F = Fuel bundle (moderated)

M = Empty (moderation only)

I Figure 6.3 Option C (73 Bundles)

~ ~

E The kef f values at 5, 7.5 and 10 percent water are 0.873 1 0.0052, 0.890 1 0.0048, and 0.913 1 0.0052, respectively. Addition of these bundles does not increase the peak keff (reference option B) because they are in a small array decoupled from the main array, and because the kef f is dominated by the bundles near the five

" extra" bundles.

6.4 Option D (72 Bundles)

Another variation of the option B arrangement is shown in Figure 6.4. The five

" extra" bundles are removed and a 4x2 array is placed in the center of the otherwise moderation-only 6x6 subarray.

I I

I

Il I

XN-NF-86-100, Rev. 0 1 Page 24 l ROW / COL i 2 3 4 5 6 7 8 9 10 ll M M M M M M M M M M 10 F F F F F F F F F F 9 F F F F F F F F F F 8 F F M M M M M M F F 7 F F M M M M M M F F 6 F F M F F F F M F F 5 F F M F F F F M F F 4 F F M M M M M M F F 3 F F M M M M M M F F 2 F F F F F F F F F F

! F F F F F F F F F F KEY: F = Fuel bundle (moderated)

M = Empty (moderation only)

Figure 6.4 Option D (72 Bundles) l l The keff results are listed in Table 6.4.

Table 6.4 Option D (72 Bundles)

Moderation Effects Water Density

(%) keff l 2.5 0.835 + 0.0043 l

5 0.897 7 0.0048 7.5 0.895 i 0.0046

~

10 0.878 + 0.0048 I

IJ

I I XN-NF-86-100, Rev. O Page 25 This pattern has the highest margin of safety and the flattest response to water fI i

density. The normalized fission densities are much more uniform for this option as compared to the previously tabulated data.

I I

I I

I I

I I

I I

I I

I I

l u

-- XN-NF-86-100, Rev. O Page 26 H 7.0 SENSITIVITY STUDIES The assumed temperature for these models is 200C or less. Since'keff will tend to l decrease with increasing temperature, the model is conservative.

~

l No credit is taken for neutron absorption in the steel / aluminum materials of the i rocks. All neutron absorptions occur in the fuel, the moderator, or the reflector.

1 This is a conservative model.

m L The key parameter for sensitivity evaluation is the bundle spacing within the rocks.

The modeled pitch between bundles is conservative. For reference, the ef fect of

[

l pitch on k ngi was evaluated for a flooded system. For lower moderator densities, the peak kef f will change little, if any, but the optimum water density will j increase with decreasing bundle-bundle spacing.

These calculations were performed using XSDRNPM. The bundle and water between l bundles were modeled as concentric cylindrical regions of infinite length. Results are listed in Table 7.0.

l t

l I

Table 7.0 XSDRNPM Results

. Bundle Spacing Effects 100% interspersed Water Density Center-Center Spacing (inch) k i nf

,I 20.875 0.9I55 20.0 0.9 i 6 I j l 9.0 0.9I72 5 17.0 0.9227 15.0 0.9403 13.0 1.0000 I

I

I-XN-NF-86-100, Rev. O  !

Page 27  !

The Table 7.0 results indicate:

1. The XSDRNPM result at 20.875 inch pitch is statistically identical to the KENO value in Table 5.2.la (0.917 + 0.005).

I

~

2. The ki nf would be less than 0.95 for all pitches greater than about 15 inches.

7.1 Bundle Offset Effects The bundles are constrained on 21-inch nominal centers in the subject rocks by channels or guides slightly larger than the bundle dimensions.

The nominal guide dimensions are 9.4375 inch square. Therefore, assuming a 20.875 I

inch pitch for the guides, if a- 2x2 array of bundles were brought together as closely as possible at the near edges of the contiguous guides, the bundles would be on 19.8825 inch centers. Based on the results in Table 7.0 this spacing would be acceptable even for an infinite system. This could not be achieved actually since the spacing is increased on two sides of the guides as the bundles are moved together.

I I

I I

I I

I1

I XN-NF-86-100, Rev. O Page 28

^

b.0 FUEL HANDLING ACCIDENTS The accidents considered include:

r-

1. A single bundle placed edge-edge to the side or top of the new fuel rocks

(" bundle drop").

2. Accidental placement of a bundle into an unauthorized rock location.

F

^

3. Two bundles not in the rock placed closely together.

~

8.1 " Bundle Drop" Accidents The low density moderation results in Section 6 assumed close-fitted concrete reflection at the sides, bottom, and at the top of the 14 foot tall rocks. A a

j close-fitted reflector actually exists only at the floor, and at the south (-Y in I KENO) wall. Therefore, the neutron leakage from the actual system would be l

r greater than that modeled. Placement of a bundle next to the rack would require c- movement of the reflector away from the rock. The close-fitted reflector model l conservatively bounds any credible effects of external vertical bundles as shown y

l below.

I l A quantitative comparison may be deduced from the peak keff data for the option B case:

1 I The keff declined f rom 0.914 to 0.870 when the five " extra" bundles were 1.

removed from the model.

2. The ef fect of one bundle would be less than 20 percent of the apparent 0.044 delta k because the five " extra" bundles also interact among themselves, while a single bundle interacts only with the main system.

I

I XN-NF-86-100, Rev. O Page 29

3. Note that this delta k also results from decreased reflector effectiveness.

The option D orrangement (5 percent water) was modeled with more realistic reflector spacings from the rock edge:

West (-X): 40 inches East (+ X): 14 inches South (-Y): 0 inch North (+Y): 24 inches moderation only (no concrete. open area)

The KENO result is 0.872 + 0.0043. This is 0.025 lower than the 0.897 + 0.0048 result with the close-fitted reflector.

If a bundle is placed horizontally on top of the rocks, the nearest fuel would be I

about 20 inches below. Inieracting neutrons would have to pass through steel, concrete and moderation. As described earlier, the reflection conditions modeled conservatively bound the effect of this interacting b'undle. It is also noted that the E

assumed 12 foot fuel length ef fectively models several horizontal bundles placed 5 directly on the top of the bundle array, rather than 20 inches away at the top of the rocks.

Therefore, the horizontal bundle drop case is conservatively demonstrated as a safe condition.

Note that the above evaluation assumes two independent and improbable accident conditions (optimum moderation and optimally placed bundle) which is not credible.

I I

I I

I

L E XN-NF-86-100, Rev. O Page 30 1

H 8.2 Misplaced Bundles Within Rocks m

' The four optional loading patterns described in Section 6 require administrative controls to assure that bundles are placed only in designated rack locations.

Accidental misplacement of one or more bundles was evoluoted for effect on criticality safety. The reference loading pattern for this accident analysis is option D at five percent interspersed water.

p The misplaced bundles used were either 24 feet long or 14 feet long, while the l

^

remaining bundles were 12 feet long. The " ceiling" (concrete) was at the top of

~ the tallest bundle (14 or 24 feet above floor). These models simulate multiple simultaneous accidents:

L 1. Placing two bundles (end-end) in an unauthorized rock location (two accidents).

f 2. Achieving optimum interspersed moderation throughout the system.

Thus, these cases also conservatively demonstrate that safety is retained during 1

rack loading and unloading operations; i.e., when the bundles in the rocks may "see" g a bundle above the 12 foot elevation.

I The rock model was slightly more conservative thon that described earlier for l

option D in that the array was 10x10 with close-fitted concrete reflection. The previous model had the eleventh row filled with moderation between the row 10 bundles and the reflector. This reflection change increased the option D keff from I

0.897 1 0.0048 to 0.908 1 0.0048.

f in the first case, the 24 or 14 foot bundle was placed in location (5,8) as shown in Figure 8.2a.

t 1

I XN-NF-86-100, Rev. O Page 31 ROW / COL I 2- 3 4 5 6 7 8 9 10 10 F F F F F F F F F F 9 F F F F F F F F F F 8 F F M M M M M M F F 7 F F M M M M M M F F a 6 F F M F F F F M F F 5 F F M F F F F F F F I 4 F F M M M M M M F F 3 F F M M M M M M F F E 2 F F F F F F F F F F 5 i F F F F F F F F F F I

KEY: F = Fuel bundle (moderated)

M = Empty (moderation only)

Figure 8.2a Option D.! (74 Bundles)

The keff for the above arrangement is 0.907 1 0.0040 (24 feet), or 0.919 1 0.0039 (14 feet). The closer reflector is worth more than the longer bundle.

in a second case, the bundles were placed into cells (7,5 and 8,5). Thus, the I

moderation-only zone was bridged in the north-south direction. The arrangement is shown in Figure 8.2b.

I I

I I

I I

I

'I XN-NF-86-100, Rev. O Page 32 l

lI ROW / COL I 2 3 4 5 6 7 8 9 10 10 F F F F F F F F F F 9 F F F F F F F F F F

8 F F M M F M M M F F 7 F F M M F M M M F F l 6 F F M F F F F M F F 5 F F M F F F F M F F

!g 4 F F M M M M M M F F lg 3 F F M M M M M M F

F F

F F

2 F F F F F F F 1 F F F F F F F F F F

I l

I KEY:F = Fuel bundle (moderated) l M = Empty (moderation only)

Figure 8.2b Option D.2 (76 Bundles)

'I The keff for the above arrangement is 0.918 3 0.0040 (24 feet) or 0.931 1 0.0046 I (14 feet).

In the third case, the bundles were placed into each of the locations (7,8) and (8,8) as shown in Figure 8.2c.

I I

I I

I I -

I.

XN-NF-86-100, Rev. O Page 33 ROW / COL I 2 3 4 5 6 7 8 9 10 10 F F F F F F F F F F g 9 F F F F F F F F F F 5 8 F F M M M M M F F F 7 F F M M M M M F F F 6 F F M F F F F M F F 5 F F M F F F F M F F 4 F F M M M M M M F F 3 F F M M M M M M F F E 2 F F F F F F F F F F 5 I F F F F F F F F F F I

KEY:F = Fuel bundle (moderated)

M = Empty (moderation only)

Figure 8.2c Option D.3 (76 Bundles)

The keff for the above arrangement is 0.918 1 0.0040 (24 feet) or 0.927 1 0.0044.

A fourth case had a 14 foot bundle in on authorized location in the middle of the array (5,5). The keff is 0.907 1 0.0044.

Therefore, criticality safety is assured during loading / unloading and in the event of credible bundle misplacement within the rock.

8.3 Bundle-Bundle Interactions During Handling Placing two or more bundles closely together was considered as a hypothetical I

accident condition.

I I

Ij

I

'I XN-NF-86-100, Rev. O I Page 34 lI I These few bund!e arrays would require flooding and reflection by nearly full density water to approach criticality. Although flooding and handling accidents are independent and unlikely, they were considered simultaneously for information I purposes. Either event alone would not lead to unsafe conditions.

All cases assumed fully flooded bundles (12 foot length) with full water reflection (30 cm water) at all six faces of the smallest cuboid enclosing the system model.

I Table 8.2

'I Fuel Handling Interactions Array / Spacing Effects I Bundle Array Center-Center Spacing (inch) keff Ix1 0.921 + 0.0056 2xl 8.445 1.048 ! 0.0053 1.012 ~ 0.0058 I 2xl 2xl 2xI 10.445 12.445 i3.445 0.943 - 0.0056 0.940 I 0.0060 2xl I4.445 0.923 I 0.0049 2x2 8.445 1.214 + 0.0050 I These results indicate:

1. The kef f for a single, fully reflected bundle (12 foot length) is statistically identical to that for an infinite x infinite array of infinite length bundles on 20.875 inch centers. As expected, the 12.43 inches of water between adjacent bundle envelopes is adequate to isolate fully flooded bundles.

I 2. Two closely placed bundles may become critical if flooded and fully reflected.

'I I

I

I I XN-NF-86-!00, Rev. O Page 35 9.0 METHODS VERIFICATION Supplemental benchmarking of the methods employed in this analysis were 3 performed. Critical experiments documented in references 2-5 were modeled using methods identical to those of this report. The critical experiments include bundle crrays with variable bundle-bundle spacings, and with and without neutron absorber rods / plates between the bundles.

il 9.1 Reference 2 Experiments j

(E Reference 2 experiments include a 3x3 array of 14xl4 bundles. The rods contain i

2.46 percent enriched UO2 Pellets on a 1.636 cm square pitch. Five of the experiments were selected for this benchmark. These cases contain little, if any, dissolved boron in the moderator (water), and include the desired effects of neutron absorbers. The other cases. not selected for benchmarking, include effects such as dissolved boron content and slight temperature changes.

l l

l The critical moderator height was determined in these experiments. The reported kef f's were normalized to o constant moderator height for each of the two classes of experiments. Therefore, the observed keff's are not all unity. The data are in Table 9.l.

I I

E

,I I

'I I

I

~ XN-NF-86-100, Rev. O Page 36 I

Table 9.1 Benchmark Results Data of Reference 2 kerf keff kerf Case Nurnber (Observed) (Calculated) (95% UL) 232I l.0030 + 0.0009 0.997 + 0.005 1.007 2317 1.0083 I 0.0012 1.004 7 0.004 1.012 2378 1.0000 7 0.0010 1.009 I 0.005 1.019 2396 1.0001 + 0.0010 1.004 + 0.004 1.012 2420 0.9997 [ 0.0015 1.002 7 0.004 1.010 Average 1.0022 1 0.0316 1.0032 3 0.0019 1.0120 The 95 percent upper limit on the calculated keff, which is the parameter used in I

judging acceptability, exceeds the observed keff in every case. The average of the individual biases (calculated minus observed) is 0.00098 + 0.0028.

9.2 Reference 3 Data Reference 3 includes data on experiments using 2.35 and 4.31 percent enriched UO2 rods in a lx3 bundle array. Only the 4.31 percent enriched cases were selected for this benchmark. These cases were either 8x13 bundles (2.54 cm rod pitch) or 16xl2 bundles (1.892 cm pitch). The critical separation between the bundles was deter-mined with various neutron absorbers between the bundles and with various spacings to a thick steel wall.

In these cases, the observed keff's are all I.000. El' I

I I

I ;

]

XN-NF-86-100, Rev. O Page 37 l

- Table 9.2 Benchmark Data From Reference 3 I

I Distance to Rod Pitch Steel Wall Neutron (cm) (cm) Absorbers keff I

l 2.54 0 0.999 + 0.006 l

I 2.54 2.54 6.6 26.16 1.001 + 0.005 1.012 + 0.005 1.892 6.6 0.999 + 0.004 l

i I 1.892 1.892 1.892 13.21 54.05 1.96 Boroflex 0.998 + 0.004 1.008 + 0.005 1.003 + 0.004 l.892 1.96 Boral 0.997 + 0.005 Average 1.0021 I

The overage bias is 0.0021 + 0.0019. The 95 percent upper limit on the KENO keff exceeds the observed value in each case.

l 9.3 Reference 4 Data A single, undermoderated 22x22 array of 4.752 percent enriched rods with various patterns of 25 " water holes" (removed fuel rods) was tested to determine the critical moderator height. Since the 15xl5 bundle assumed in the storage rocks contain 21 guide / instrument tubes, the reference 4 data are useful to verify the methods, particularly the homogeneous representation of the bundle.

Three cases were calculated using KENO with explicit modeling and homogeneous modeling.

'I I

I

I XN-NF-86-100, Rev. O Page 38 Table 9.3 Reference 4 Dato KENO-IV with Hansen-Roach Cross Sections keff keff Case (Explicit) (Homogeneous) 1670 0.995 + 0.0054 1.002 + 0.0055 1674 0.996 ~ 0.0054 1.00I i 0.0054 l 1680 I.000{0.005I l.005 i 0.0063

, Average 0.997 1.0027 l

l The bias is -0.00017 + 0.0015 for all six cases.

The homogeneous model results appear to be about 0.005 higher than the explicit model results, but this bias is not significant. All results agree very well with the l observed kef f of unity.

9.4 Reference 5 Data The rod design here is identical to that of the reference 4 data:

I I

I I

I l

I l

l l

IJ

I l

e XN-NF-86-100, Rev. O Page 39 I

Table 9.4a Fuel Design Parameters I Rod Diameter (cm) 0.79 Enrichment (% U-235) 4.742 I UO2 Density (% TD)

Fuel Length (cm) 94.71 90 Clod Aluminum Clad ID/OD (cm) 0.82/0.94 I

i Four flooded 18xl8 bundles were placed in a 2x2 array spaced by various thick-nesses of various between-bundle moderators. These moderators included air, water, expanded polystyrene, polyethylene powder (Iow density), and polyethylene balls (higher density).

I These experiments were modeled using the SCALE system os documented in reference 6. Selected cases were modeled here for comparison.

These experiments are useful in validating the optimum moderation calculations.

,I Experimental data with low density moderation within and between bundles are not available.

I The three cases selected had a 10 cm spacing between bundles. This spacing was filled with either air, polystyrene, or polyethylene powder. The corresponding hydrogen densities were 0, 0.0022, and 0.0464 gm/cc, respectively. The water densities to yield these H densities are 0, l.97 and 41.47 percent, respectively.

I 1

I -

I XN-NF-86-100, Rev. 0 ,

Page 40 7

Table 9.4b Reference 5 Data Low Density Moderation Between Bundles KENO Results H Density keff keff (gm/cc) (16 Group) (27 Group) 0 1.012 + 0.0046 0.985 + 0.0053 3

0.0022 1.012 - 0.0059 1.024 + 0.0050 3

0.0464 1.036 + 0.0045 I

For the three 16-group cases, the bias is 0.020 + 0.008.

I The results presented agree well with the complete result set of reference 6. The 16 group (Hansen-Roach) results appear to be slightly conservative.

These results indicate that the low density moderation results are accurate or pechaps slightly conservative.

9.5 Acceptability Limit Pooling data from references 2, 3 and 4 -(flooded cases), the average and standard deviation of the systematic bias are 0.00ll and 0.00..! i, r'Espectively. Clearly, there is no significant systematic bias. Ba'sp! on the limited replication of the low moderation cases (refe. enc,e 5) and the complete results in reference 6, the bias at this condition v.dl be conservatively set equal to that for the flooded cases.

Using the criteria of ANSI /ANS-8.17-1984 (and other similar documents), the maximum allowable calculated keff is established as follows.

Limit = A - B - C - D El Il

I I XN-NF-86-100, Rev. O Page 41 I

The terms are defined below.

A: Mean keff from appropriate benchmarks. The assigned value is 1.0011.

I B: An allowance for uncertainties in parameter A. The assigned value is 0.0011.

C: An allowance for uncertainty in kef f calculations. This allowance is variable and is included in the final keff result; i.e., the 95 percent upper limit statement. Therefore, the acceptability limit is not adjusted and C is set to zero.

D: An arbitrary mygin to ensure subcriticality. This is set to 0.05 except for optimum moderation conditions where the value is 0.02.

i e

5 Therefore, the acceptability limit is 0.95 or 0.98 (optimum moderation).

l.t should be noted that allowances B and C are of ten pooled before applying a confidence level multiplier (usually about 2.0).

I In this format:

I Limit = A - D - K * /B2+C2 I where K is the confidence level multipliar.

Since the KENO standard deviation is typically 0.003-0.006, the sum of squares is dominated by the KENO variance. The limits calculated by the two methods are I very close. The ANS/ ANSI format is more conservative by 0-0.0004 for typical KENO standard deviations.

I I

I XN-NF-86-100, Rev. O Page 42 9.6 Cross Section Comparisons Selected models were calculated using the 27 group ENDF/B-IV and 123 group I

GAM-THERMOS cross section libraries with NITAWL processing.

For the infinite array of flooded bundles on 20.875 inch centers, the kinf results are 0.906 1 0.0048 (27 group) and 0.913 3 0.0057 (123 group). These results agree well with the 0.917 1 0.005 result using the 16 group Hansen-Roach library.

The option D model results with 5 and 7.5 percent interspersed water are tabulated below.

Table 9.6 I

Cross Section Comparisons Water Density keff kerf ke rf

(%) (16 Group) __ (27 Group) (123 Group) 5 0.897 + 0.0048 0.888 + 0.0047 0.873 + 0.0051 7.5 0.895 1 0.0046 0.867 i 0.004 0.869 i 0.0054 All replicates with other cross sections confirm the adequacy of the 16 group results.

I I

I I

I II

I I XN-NF-86-100, Rev. O Page 43 I

10.0 COMPUTER INPUT LISTINGS Typical KENO input listings are provided for reference.

10.1 KENO Input: K-Infinite Case at Full Flooding This is the explicit model for on infinite array of 15xl5 bundles on 20.875 inch centers.

HB ROBINSON, 4.2% ENR, K-INF, BUNDLES ON 20.875" CENTERS, FLOODED 300.0 I 83 300 3 16 6 7 3 7 9 2 15 15 I 7 1 0 2000 00 1 0 0 0 0 0 00 0 0 6*-1 1 -92508 6.2909E-04 I I 92509 3.5867E-04 I

1 9281I 92812 l.0803E-02 1.1443E-02 I I 2 502 1.00 8100 4.6466E-02 3 40100 4.2535-2 BOX I

l FUEL ROD CYCL 1 0.452755 500.0 -500.0 CYLI O 0.46228 500.0 -500.0 I CYLI 3 0.53848 500.0 -500.0 CUBO 2 0.71501 -0.7I501 0.71501 -0.71501 500.0 -500.0 BOX 2 I INST / GUIDE TUBE CYLI 2 0.64897 500.0 -500.0 CYLI 3 0.69088 500.0 -500.0 CUBO 2 0.71501 -0.71501 0.7I501 -0.7I501 500.0 -500.0 I '

15XI5 BUNDLE: 21.4503 CM SQUARE (8.445" SO)

CORE 0 10.72515 -10.72515 10.72515 -10.72515 500.0 -500.0 CUBO 2 26.5I I25 -26.5I I25 26.5I I25 -26.5I I 25 500.0 -500.0 I II O I I i 15 I I 15 I 2 3 13 10 363 III O 2 3 13 10 10 13 3 II I O 2 4128 88I III O I 2 5116 5 ll 6 III O 2 6 10 4 3 13 10 1II 0 2 881 4 12 4 I Il 9 END KENO I

I I

I XN-NF-86-100, Rev. O Page 44 10.2 KENO Input: New Fuel Vault, Option D Model, 5% Water HB ROBINSON, 4.2% ENR, KEFF, BUNDLES ON 20.875" CENTERS, NEW 300.0 83 300 3 16 6 3 3 3 10 2 10 11 I g

-3 0 0 1000 00 1 0 0 0 0 0 00 0 0 3 MACRO MIXTURE FOR 5% WATER-BUNDLE I -92005 1.0 2 502 0.05 33011.0

' 2 PERIMETER ROW FILLED, INTERNAL 6 ROWS EMPTY EXCEPT FOR 4X2 ARRAY OF BUNDLES 5 EXTRA LOCATIONS EMPTY 144" FUEL,14' VAULT BOX l a BUNDLE ON 20.875" CENTERS CUBO I 10.72515 -10.72515 10.72515 -10.72515 152.4 -213.36 E CUBO 2 26.5I I25 -26.5I I25 26.5i I25 -26.5I I25 213.36 -213.36 BOX 2 EMPTY LOCATION (MODERATION-FILLED)

CUBO 2 26.5I I25 -26.5I I25 26.5I I25 -26.5I I25 213.36 -213.36 REFL 3 6*30.0 301 g I i 10 1 I ll 1 1I I 0 g 2 381 38I II I O 1 471 56I III 0 2 1 10 I ll 11 I I I I 9 END KENO I

I I

I I

I t

I I

I I XN-NF-86-100, Rev. O Page 45 I

Il.0 REFERENCES I (1) " SCALE: A Modular Code System for Performing Standardized Computer Analyses for Licensing Evaluation," NUREG/CR-0200.

(2) M. N. Baldwin, et. al., " Critical Experiments Supporting Close Proximity Water Storage of Power Reactor Fuel," BAW-1484-7, July 1979.

(3) 5. R. Bierman and E. D. Clayton, " Criticality Experiments with Subcritical Clusters of 2.35 wt% and 4.31 wt% U-235 Enriched UO2 Rods in Water with Steel Reflecting Walls," NUREG/CR-1784 (PNL-3602), January 1981.

(4) J. C. Monaranche, et. al., " Critical Experiments with Lattices of 4.75% wt%

U-235 Enriched UO2 Rods in Water," ANS Trans, Vol. 28, pp. 302-303.

(S) J. C. Manaranche, et. al., " Dissolution and Storage Experimental Program with U(4.75)O2 Rods," ANS Trans, Vol. 33, pp. 362-364.

I (6) A. M. Hathout, et. al., " SCALE System Cross Section Validation for Criticality Safety Analysis," ANS Trans, Vol. 35, pp. 281-283.

I I

I I

I I

I . __

I l

XN-NF-86-100, Rev. 0 Issue Date: 9/2/86 FINAL REPORT CRITICALITY SAFETY ANALYSIS H. B. ROBINSON NEW FUEL STORAGE VAULT WITH 4.2% ENRICHED 15.0S FUEL ASSEMBLIES AUGUST 1986 I DISTRIBUTION L. D. Gerrald C. W. Malody J. E. Pieper F. B. Skogen I I. Z. Stone J. W. Hulsman T. W. Potten L. J. Federico CP&L/H. G. Shaw (SI)

Document Control (S)

I I

I I

I