ML17221A588

Rev 0 to Criticality Safety Analysis St Lucie New Fuel Storage Vault W/4.5% Enriched 14x14 Fuel Assemblies

ML17221A588
Person / Time
Site:	Saint Lucie
Issue date:	03/31/1987
From:	SIEMENS POWER CORP. (FORMERLY SIEMENS NUCLEAR POWER
To:
Shared Package
ML17221A586	List: ML17221A585 ML17221A587 ML17221A588
References
XN-NF-87-43, XN-NF-87-43-R, XN-NF-87-43-R00, NUDOCS 8801270017
Download: ML17221A588 (43)

Text

XN-NF-87-43 CRITICALITYSAFETY ANALYSIS ST. LUCIE NEW FUEL STORAGE VAULT WITH 4.5% ENRICHED 14x14 FUEL ASSEMBLIES MARCH 1987 MARCH 1987 RICHLAND, WA 99352 ADVANCEDNUCLEARFUELS CORPORATION ANAfFIUATEOF KRARWfRK IINION Qxwu

I' l

XN-NF-87-43, Rev. 0 CRITICALITY SAFETY ANALYSIS ST. LUCIE NEW FUEL STORAGE VAULT WITH 4.5% ENRICHED l4xl4 FUEL ASSEMBLIES MARCH I 987

CUSTOMER DISCLAIMER IMPORTANT NOTICE REGARDING CONTENTS AND USE OF THIS DOCUMENT PLEASE READ CAREFULLY Advanced Nuclear Fuels Corporation's warranties and representations con-ceming the subject matter of this document are those set forth in the Agreement between Advanced Nuclear Fuels Corporation and the Customer pursuant to which this document Is Issued. Accordingly, except as otherwise expressly pro.

vlded In such Agreement, neither Advanced Nuclear Fuels Corporation nor any person acting on its behalf makes any warranty or representation, expressed or Implied, with respect to the accuracy, oompieteness, or usefulness of the infor-mation contained in this document, or that the use of any Information, apparatus, method or process disclosed in this document will not infringe privately owned rights; or assumes any liabilities with respect to the use of any Information, ap.

paratus. method or process disclosed in this document.

The ktformation contained herein Is for the sole use of Customer.

In order to avokf Impairment of rights of Advanced Nuclear Fuels Corporation 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 gn the patent use ofthe term) ofsuch information until so authorized In writing by Advanced Nuclear Fuels Corporation or until after six (6) months followingtermination or expiration ofthe aforesaid Agreement and any

- extension thereof. unless otherwise expressly provkfed In the Agreement. No rights or licenses In or to any patents are Implied by the furnishing of this docu-ment.

XN NF F00.765 (1/87)

C'N-NF-87-43, Rev. 0 CRITICALITY SAFETY ANALYSIS ST. LUCIE NEW FUEL STORAGE VAULT WITH 4.5% ENRICHED 14xl4 FUEL ASSEMBLIES MARCH 1987 TABLE OF CONTENTS SECTION I.O 2.0

SUMMARY

FUEL PARAMETERS

~Pa e I

2 3.0 4.0 5.0 5.1 5.2 5.3 5.4 6.0 6.I 6.2 6.3 6.4 6.5 7.0 STORAGE RACK GEOMETRY CALCULATION METHODS MODERATION AND SPACING EFFECTS Removal of Fuel Rods (Fully Flooded)

Optimum Interspersed Moderation Within Racks Rack Spacing Effects (Bundle-Bundle Spacing)

Fuel Handling Accidents METHODS VERIFICATION Reference 2 Experiments Reference 3 Data Reference 4 Data Reference 5 Data Acceptability Limit REFERENCES 6

6 7

II l2 l5 I5 l6 l7 IS 20 22

l y,

a s y

XN-NF-87-43, Rev. 0 LIST OF TABLES Table 2.0 S.l Bundle Parameters Single Full Water Reflected Bundle System Full Water Density, Infinite Length Bundle Moderation Effects (Fuel Rod Removal) (XSDRNPM Results)

Pacae 5.2a Infinite System Data (Zero Leakage)

Interspersed Moderation Effects (XSDRNPM Results) 5.2b Finite System Data (New Fuel Racks)

Interspersed Moderation Effects Moderation Between Rack Edge and Walls (Reflector Not Close-Fitted) (KENO-Va Results)

IO 5.3 5.4 6.I 6.2 6.3 6.4a 6.4b Infinite System Data (Zero Leakage) Fully Flooded State Bundle Spacing Effects (XSDRNPM Results)

Fuel Handling Accidents Fully Flooded, Full Water Reflection (KENO-Va Results)

Benchmark Results Data of Reference 2

Benchmark Data From Reference 3'eference 4 Data KENO with Hansen-Roach Cross Sections Fuel Design Parameters Reference 5 Data Low Density Moderation Between Bundles KENO Results l2 13 l6 l7 IS IS l9 LIST OF FIGURES

~Fi ure 2.0 Rod Arrangement

~Pa e

XN-NF-87-43, Rev. 0 Page I

CRITICALITY SAFETY ANALYSIS ST. LUCIE NEW FUEL STORAGE VAULT WITH 4.5% ENRICHED l4xl4 FUEL ASSEMBLIES MARCH I 987 I.O

SUMMARY

The criticality safety of the new fuel storage vault with 4.5% enriched l4xl4 bundles is assessed in accordance with NUREG-0800 and ANSI/ANS-57.3-I983.

The subject system meets the applicable criticality safety criteria subject 'to the limits and controls given below.

I.

Fuel Design - As specified in Section 2.0 2.

Bundle-Bundle Spacing Within Racks - 2I" nominal (center-center)

(l4" minimum bundle pitch is acceptable) 3.

Loading. Arrangement - Rows 5 and 6 of IOxl0 array locked out 4.

Dissolved boron (greater than or equal to l720 ppm) in water during fuel movements ln normally flooded systems unless 4" (minimum)edge-edge bundle spacing (two bundles) is assured at all credible accident conditions.

'J II ti

)II f

XN-NF-87-43, Rev. 0 Page 2

2.0 FUEL'ARAMETERS The key bundle design parameters used in these calculations are listed in Table 2.0.

The bundle is a 14xl4 design with five guide tubes.

Since the guide tubes are much larger than the fuel rods, the bundle may be more easily visualized as a 7x7 array of two cell types.

Type F is a 2x2 fuel rod array, and type G is a single guide tube.

Using this 7x7 description, the bundle is composed as shown in Figure 2.0.

Table 2.0 Bundle Parameters Parameter Enrichment (wt% U-235)

Pellet Diameter (inch)

Pellet Density (%TD)

Pellet Dish Volume (%)

Active Fuel Length (inch)

Clad ID/OD (inch)

Rod Pitch (inch)

Gd/Boron Content Fuel Rods per Bundle Guide Tube ID/OD (inch)

Desi n Value 4.50 (max.)

'.3700 94.0 1.0 136.7 0.378/0.440 0.5800 Variable 176 1.035/ I. 115 Model Value.

4.50 0.3700 95.0 0

136.7 (min.)

0.378/0.440 0.5800 None 176 1.035/1.115

XN-NF-87-43, Rev.

0 Page 3

ROW/COL I

2 3

4 5

6 7

F F

F F

F F

F F

F F

F F

F G

F F

F G

F F

F F

F F

F F

F G

F F

F F

F F

F F

F G

F F

F G

F F

F F

F F

F Key: F

=

2x2 Fuel Rod Array G

=

Guide Tube Figure 2.0 Rod Arrangement

II

~.1

XN-NF-87-43, Rev.

0 Page 4

3.0 STORAGE RACK GEOMETRY The new fuel vault contai'ns a IOxl0 array of storage cells on a 2I inches nominal square pitch.

Rows 5 and 6 have been designated as unavailable for fuel bundle storage and are physically locked out of service.

Therefore, the racks modeled contained only 80 bundles.

The racks were modeled in accordance with Figure 9.I-'I of the FSAR.

The racks were modeled with concrete reflection (30 cm thick) at the four walls, the floor, and at the top of the racks (I4 feet above the floor).

The modeled spacings between the edge of the racks and the walls are 40.5 inches (north), 37.5 inches (south), 3I.5 inches (east),

and 36.0 inches (west).

All.materials of construction were neglected in the model.

All neutron absorptions occur in the fuel, the moderator, or the reflector'.

This is a

- conservative model.

XN-NF-87-43, Rev. 0 Page 5

4.0 CALCULATION METHODS All computer codes and cross sections are part of the SCALE~

~ system.

The neutro'n multiplication factors, kjnf and keff, were calculated using KENO-Va, a three dimensional Monte Carlo code, or using XSDRNPM, a one dimensional discrete ordinates transport code.

The l6 group (Hansen-Roach) cross sections were used with resonance corrections by BONAMI/NtTAWL.

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

Evidence of methods verification is presented later in this document.

J

XN-NF-87-43, Rev. 0 Page 6

5.0 MODERATION AND SPACING EFFECTS To assure safety at all credible accident conditions, the following effects were assessed.

I.

The effect of removing fuel rods from a fully flooded bundle; i.e., increase the water/fuel volume ratio (Vw/Vf) within the bundle.

2.

Achieve optimum interspersed moderation within the racks.

3.

Effect of bundle-bundle spacing within the racks.

4.

Fuel handling accidents resulting in two or more bundles placed closely together at fully flooded and reflected conditions.

5. I Removal of Fuel Rods (Full Flooded)

Generic models with Vw/Vf ratio s in the range of 2.03 to 4.0 were calculated.

The nominal dimensions for the pellet and clad were used with a variable rod pitch to yield the desired Vw/Vf.

The calculation sequence for each Vw/Vf case was:

I.

Prepare -self-shielded cross sections using BONAMI/NITA>NL.

2.

Perform cell weiglfng of the unit rod cell using XSDRNPM.

3.

Using the yell weighted (homogeneous) cross sections from step 2, model an infinite length bundle surrounded by 30 cm of water.

The bundle and the water reflector were modeled as concentric cylindrical regions.

The results are in Table 5.I.

I

'I

• I 1

II I

XN-NF-87-43, Rev. 0 Page 7

Table S.l Single Full Water Reflected Bundle System Full Water Density, Infinite Length Bundle Moderation Effects (Fuel Rod Removal)

(XSDRNPM Results)

Vw/Vf 2.03 (nom.)

2.5 3.0 3.5 4.0 0.888 0.903 0.9 I 0 0.9 I I

, 0.907 The optimum Vw/Vf is near 3.5.

Since the bundle is relatively small (7.98 inch surface-surface or 8.I2 inch across homogeneous cells), the neutron leakage is adequate to maintain the fully flooded keff well below 0.95 with any number of removed fuel rods.

5.2 0 timurn Inters ersed Moderation Within Racks For these

cases, the nominal bundle design and the nominal rack design parameters were modeled.

The water within and between the bundles in the rack was modeled as uniform in the range zero to l00 volume percent.

Infinite and finite systems were modeled.

The infinite system results were used to estimate the optimum interspersed water density.

The finite system was then modeled with water densities near the estimated optimum.

The calculation sequence for each water density was:

XN-NF-87-43, Rev. 0 Page 8

I.

Prepare self-shielded cross sections using BONAMI/NITAWL.

2.

Perform cell weighting of the unit rod cell using XSDRNPM.

3.

Model an infinite array of infinite length bundles on 2I inch centers.

As described before, the fuel and moderation were concentric cylindrical regions.

The k;nf results for the rod lattice and for the bundle lattice (2! inch centers) are listed in Table 5.2a.

Table 5.2a Infinite System Data (Zero Leakage)

Interspersed Moderation Effects (XSDRNPM Results)

Water Density (Vol%)

0 (normal)

I 3

5 8

IO l2 l5 20 40 60 80 IOO kinf (Rod Lattice) 0.7534 0.7784 0.8204 0.86 I 7 0.9240 0.9640 1.0020 I.0546 I. I 300 l.3 I 72 I.4080 I.457 I I.4838 kinf (Bundle Lattice)

(21" Centers) 0.7534 I.0480 l.3344 I.3874 l.3294 I.2563 I. I 749 I.0543 0.8858 0.6548 0.7 I 3 I 0.8063 0.890 I

r 0

4 e

, ~

h

(

XN-NF-87-43, Rev. 0 Page 9

The above data indicate that an infinite array of infinite length bundles is adequately subcritical at all interspersed water densities greater than 20 percent and also with zero interspersed moderation.

Although the optimum water density for an infinite system is near 5 percent, leakage effects in a finite system will typically result in higher optimum water densities and lower keff's.

The actual finite system was explicitly modeled using KENO-Va.

The parameters modeled for the fuel and for the racks were given in Sections 2.0 and 3.0.

The results for low density interspersed moderation conditions are listed in Table 5.2b.

I II

XN-NF-87-43, Rev. 0 Page IO Table 5.2b Finite System Data (New Fuel Racks)

Interspersed Moderation Effects Moderation Between Rack Edge and Walls (Reflector Not Close-Fitted)

(KENO-Va Results)

Water Density (Vol%)

5 8

IO l2 20 0.903

+ 0.004 I 0.963

+ 0.0047 0.96I

+ 0.0049 0.942

+ 0.0044 0.800

+ 0.0045 Interpolated Data Below (Polynomial Regr'ession) 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 I 0.0 I 0.5 I I.O I I.S I 2.0 I 3.0 I 4.0 I 5.0 I 6.0 17.0 I 8.0 I 9.0 20.0 0.903 I 0.9 I 87 0.93 I 8 0.9426 0.95 I I 0.9576 0.9620 0.9646 0.9655 0.9648 0.9626 0.9590 0.9542 0.9483 0.94 l4 0.9253 0.9067 0.8866 0.8662 0.8463 0.8280 0.8 I 22 0.8000

0 l~

XN-NF-87-43, Rev. 0 Page I I The KENO calculations (finite system) were performed using 83 (typical) genera-tions of 300 neutrons.

To establish the 95 percent upper limit on the Monte Carlo calculations, the appropriate one-sided "Student" t (80 degrees of freedom) is I.66.

Using the highest interpolated keff (9 percent water) and the highest KENO standard deviation (0.0049), the 95 percent upper limit is:

keff (95% UL)

Oo9655

+

I o66 0+0049 0+9736 Since the keff is less than 0.98 with 95 percent confidence, the system is demonstrated to be adequately subcritical at optimum moderation.

5.3 Rack S acin Effects (Bundle-Bundle S acin

)

The peak keff with low density interspersed moderation will change little if any with credible changes in bundle-bundle spacings.

However, the optimum inter-spersed water density will tend to increase with decreasing bundle-bundle spacings.

The effect of bundle-bundle spacings on the reactivity of an infinite array of flooded infinite length bundles was assessed as described in Section 5.2 (infinite system) except that the bundle-bundle spacing was decreased and also the water density was fixed at l00 percent.

The results are in Table 5.3.

y p

~

f

XN-NF-87-43, Rev.

0 Page l2 Table 5.3 Infinite System Data (Zero Leakage)

Fully Flooded State Bundle Spacing Effects (XSDRNPM Results)

Bundle Spacing (Center-Center)

(inches) 2 I (nominal)

I8 l4 l2

8. I 2 Bundle Spacing (Edge-Edge)

(inches)

I 2.88 9.88 5.88 3.88 0.00 0.890 0.894 0.930 I.024 I.484 An explicit KENO-Va model of the l4 inch center-center spacing condition resulted in a k;nf of 0.938

+ 0.0060.

Thus, the KENO and XSDRNPM resuJts are statistically identical.

An infinite array of flooded infinite length bundles is adequately subcritical (keff less than 0.95) at all bundle pitches greater than

!4 inches.

No credible combination of dimensional tolerances, eccentric positioning, or accident condition would result in a pitch approaching l4 inches.

Therefore, the flooded state is acceptable.

5.4 Fuel Handl in Accidents As shown in Table 5.2a, an infinite array of edge-edge bundles (infinite rod lattice) is adequately subcritical for all interspersed water densities less than IO percent.

~

IC 0

wJ

~

L

~

XN-NF-87-43, Rev.

0 Page l3 It was also shown that an infinite array of flooded bundles is adequately sub-critical if the bundle pitch is at least l4 inches.

Close placement of two flooded bundles was also explicitly modeled using KENO-Va.

This covers potential accidents at the Fuel Inspection

Elevator, Upender and the Fuel Transfer Tube, as well as generic cases such as dropping a

bundle or moving a bundle next to another flooded bundle.

Finite bundle arrays with various spacings were modeled.

In each

case, the array was reflected by 30 cm of full density water.

The KENO Va results are in Table 5.4.

Table 5.4 Fuel Handling Accidents Fully Flooded, Full Water Reflection (KENO-Va Results)

Bundle Array Bundle-Bundle Spacing (Inches Ed e-Ed e)

Ix I 2x I 2xl 2xl 0

2.0 4.0 0.897

+ 0.0057 I.OI6 + 0.0061 0.992

+ 0.0056 0.920

+ 0.0053 It is seen that two closely placed bundles may become critical if flooded and reflected.

About four inches of water between two adjacent bundles is adequate to assure safety.

XN-NF-87-43, Rev. 0 Page l4 Full flooding and close placement of bundles represent two independent and improbable accident conditions in the new fuel vault.

Since either condition alone has been shown to be acceptable, the new fuel.vault is adequately subcritical.

In other systems which are normally flooded and which may accidentally allow close placement of bundles, additional controls are required.

Assurance of dissolved boron (l720 ppm minimum) in these locations during fuel movement will I

lower the maximum credible keff by about 0.20, resulting in reactivities far below the 0.95 limit.

'I

XN-NF-87-43, Rev. 0 Page l5 6.0 METHODS VERIFICATION Supplemental benchmarking of the methods employed in this analysis were performed.

Critical experiments documented in references 2-5 were modeled using methods identical to those of this report.

The critical experiments include bundle arrays with variable bundle-bundle

spacings, and with and without neutron absorber rods/plates between the bundles.

6.I Reference 2 Ex eriments Reference 2 experiments include a 3x3 array of l4xl4 bundles.

The rods contain 2.46 percent enriched UO2 pellets on a l.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 effects of neutron absorbers.

The other cases, not selected for benchmarking, include'effects such as dissolved boron content and slight temperature changes.

The critical moderator height was determined in these experiments.

The reported keff's were normalized to a 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 6.I.

XN-NF-87-43, Rev. 0 Page l6 Table 6.I Benchmark Results Data of Reference 2

Case Number 232I 23I7 2378 2396 2420 Average keff (Observed)

I.0030

+ 0.0009 I.0083

+ O.OOI2 l.0000

+ 0.0010 I.OOOI

+ O.OOI9 0.9997

+ O.OOI5 I.0022

+ O.OOI 6 keff (Calculated) 0.997

+ 0.005 I.004

+ 0.004 I.009

+ 0.005 I.004

+ 0.004 I.002

+ 0.004 I.0032

+ O.OOI 9 keff (95'L) l.007 I.O I 2 l.019 I.0 I 2 I.O IO I.O I 20 The 95 percent upper limit on the calculated keff, which is the parameter used in judging acceptability, exceeds the observed keff in every case.

The average of the individual biases (calculated minus observed) is 0.00098

+ 0.0028.

6.2 Reference 3 Data Reference 3 includes data on experiments using 2.35 and 4.3I percent enriched UO2 'rods in a Ix3 bundle array.

Only the 4.3! percent enriched cases were selected for this benchmark.

These cases were either Sxl3 bundles (2.54 cm rod pitch) or 16xl2 bundles (I.892 cm pitch).

The critical separation between the bundles was determined 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 l.000.

t a

kit I

XN-NF-87-43, Rev.

0 Page 17 Table 6.2 Benchmark Data From Reference 3

Rod Pitch (cm)

Distance to Steel Wall (cm)

Neutron Absorbers 2.54 2.54 2.54 0

6.6 26.16 0.999

+ 0.006 1.001

+ 0.005 1.012

+ 0.005 1.892 1.892 1.892 1.892 1.892 6.6 13.21 54.05 1.96 1.96 Boroflex Boral 0.999

+ 0.004 0.998

+ 0.004 1.008

+ 0.005 1.003

+ 0.004 0.997

+ 0.005 Average 1.0021 The average bias is 0.0021

+ 0.0019.

The 95 percent upper limit on the KENO keff exceeds the observed value in each case.

6.3 Reference 4 Data A single, undermoderated 22x22 array of 4.742 percent enriched rods with various patterns of 25 "water holes" (removed fuel rods) was tested to determine the critical moderator height.

Since the bundle(modeled contans guide tubes, the n

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 mode I ing.

XN-NF-87-43, Rev. 0 Page IS Table 6.3 Reference 4 Data KENO with Hansen-Roach Cross Sections Case I670 l674 I680 Average keff (Ex licit) 0.995

+ 0.0054 0.996

+ 0.0054 I.OOO

+ 0.005l 0.997 keff (Homo eneous)

I.002

+ 0.0055 I..OOI

+ 0.0054 I.OOS

+ 0.0063 I.0027 The bias is -0.000I7

+ 0.00I5 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 observed keff of unity.

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

Table 6.4a Fuel Design Parameters Rod Diameter (cm)

Enrichment

(% U-235)

UO2 Density (%TD)

Fuel Length (cm)

Clad Clad ID/OD (cm) 0.79 4.742 94.7 I 90 Aluminum 0.82/0.94

I 0

I, U

XN-NF-87-43, Rev. 0 Page l9 Four flooded I8xl8 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 (low density),

and polyethylene balls (higher density).

These experiments were modeled using the SCALE system as documented in reference 6.

Selected cases were modeled here for comparison.

These experiments are useful in validating the optimum moderation calculations.

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

The three cases selected had a IO 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, I.97 and 4I.47 percent, respective'ly.

Table 6.4b Reference 5 Data Low Density Moderation Between Bundles KENO Results H Density (cCm/cc) 0 0.0022 0.0464

" ff (l6 8rou

)

I.OI2 + 0.0046 I.OI2 + 0.0059 I.036

+ 0.0045 "eff (27 Grou

)

0.985

+ 0.0053 I.024

+ 0.0050 For the three l6-group cases, the bias is 0.020

+ 0.008.

The results presented, agree well with the complete result set of reference

6. The l6 group (Hansen-Roach) results appear to be slightly conservative.

XN-NF-87-43, Rev. 0 Page 20 These results indicate that the low density moderation results are accurate or perhaps slightly conservative.

6.5 Acce tability Limit Pooling data from references 2,

3 and 4 (flooded cases),

the average and standard deviation of the systematic bias are 0.00I I and 0.001I, respectively.

Clearly, there is no significant systematic bias.

Based on the limited replication of the low moderation cases (reference 5), and the complete results in reference 6, the bias at this condition will be conservatively set equal to that for the flooded cases.

Using the criteria of ANSI/ANS-8.I7-I984 (and other similar documents),

the

- maximum allowable, calculated keff is established as follows.

LIMIT = A - B - C - D The terms are defined below.

LIMIT:

Maximum accePtable keff.

A:

Mean keff from appropriate benchmarks.

The assigned value is I.OO I I.

B:

An allowance for uncertainties in parameter A.

The assigned value is 0.00I I.

~

C:

An allowance for uncertainty in keff 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.

~,4

~

~

XN-NF-87-43, Rev. 0 Page 2I D:

An arbitrary margin to ensure subcriticality.

This is set to 0.05, except for optimum moderation conditions where the value is 0.02.

Therefore, the acceptability limit is 0.9S, or 0.98 (optimum moderation).

It should be noted that allowances B and C are often pooled before applying a

confidence level multiplier (usually about 2.0).

In this format:

LIMIT = A - D - K + B + C where K is the confidence level multiplier.

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 very close.

The ANS/ANSI format is more conservative by 0-0.0004 for typical KENO standard deviations.

XN-NF-87-43, Rev. 0 Page 22

7.0 REFERENCES

(I)

"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)

S. 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. Manaranche, et. al., "Critical Experiments with Lattices of 4.75% wt%

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

(5)

J. C. Manaranche, et. al., "Dissolution and Storage Experimental Program with U(4.75)02 Rods," ANS Trans, Vol. 33, pp. 362-364.

(6)

A.

M. Hathout, et.

al.,

"SCALE System Cross Section Val idat ion for Criticality Safety Analysis," ANS Trans, Vol. 35, pp. 281-283.

1 A