ML20078G616
| ML20078G616 | |
| Person / Time | |
|---|---|
| Site: | Crystal River  |
| Issue date: | 12/31/1993 |
| From: | Sarah Turner HOLTEC INTERNATIONAL |
| To: | |
| Shared Package | |
| ML19311B716 | List: |
| References | |
| HI-931111, NUDOCS 9502030185 | |
| Download: ML20078G616 (33) | |
Text
,.
l CRITICALITY SAFETY EVALUATION OF THE POOL A j
SPENT FUEL STORAGE RACKS IN CRYSTAL RIVER UNIT 3 WITH FUEL OF 5.0% ENRICHMENT T
r Prepared for the 1
FLORIDA POWER CORPORATION i
by i
Stanley E. Turner, PhD, PE December 1993 Holtec Project 3119530514 Holtec Report HI-931111 230 Normandy Circle 2060 Fairfax Ave.
Palm Harbor, FL 34683 Cherry Hill, NJ 08003 9502030185 950126 PDR ADDCK 05000302 P
r._
-0, TABLE OF CONTENTS
1.0 INTRODUCTION
1 2.0
SUMMARY
AND CONCLUSIONS 2
2.1-Normal Storage Conditions 2
2.2 Abnormal / Accident Conditions 3
3.0 CRITICALITY SAFETY ANALYSES.
4 3.1 Puel Assembly Specifications 4
3.2 Storage Rack Specifications 4
3.3 Manufacturing Tolerances and Uncertainties 4
3.4 Calculational Nhthodology 5
3.4.1 Computer Codes 5
3.4.2 Axial Distribution in Burnup 6
3.4.3 Checkerboard Configuration.
7 i
4.0 REFERENCES
8 l
.)
l l
l l
1 t
L 4
-. b -
I List of Tables Table 1
SUMMARY
OF CRITICALITY SAFETY ANALYSES 9
POOL A STORAGE RACKS Table 2 FUEL ASSEMBLY SPECIFICATIONS 10 i
List of Figures Fig. 1 ACCEPTABLE BURNUP DOMAIN IN POOL A 11 Fig. 2 Pool A FUEL STORAGE CELL 12 l
l i
4 11 l
i
1.0 INTRODUCTION
O The present study is part of an evaluation of the fuel storage facilities at Crystal-River Unit 3 in order to qualify the i
facilities for fuel of 5.0% average initial enrichment.
This report addresses the Pool A spent fuel pool while a companion report evaluates the new-fuel vault (HI-931110).
The CR3 Pool' A storage racks are designed to accomodate spent fuel which has attained a minimum average burnup that is dependent on the average initial enrichment of the fuel assembly.
These racks W for fuel use a B C matrix absorber and were previously qualified 4
of 4.5% enrichment burned to 7.0 MWD /KgU and employ a water flux-l trap between storage cells as a means of augmenting reactivity control.
In the present study, the previous curve of limiting i
burnups is extended to encompass 5.0% enriched fuel.
The effect of the axial distribution in burnup has also been considered as specified by Regulatory Guide 1.13 (draft, Rev.2).
Calculations were made with both: the CASMD-3 program and the N7TAWL-KENO-Sa code package.
CASMD-3 was used for burnup and restart calculations and to define an equivalent enrichment for use in the KENO-Sa calculations.
Both normal and accident conditions j
are assessed.
Credit Lor the soluble poison normally present in the pool water is allowable under accident conditions (double contingency principle).
To assure the criticality safety under all conditions and to conform to the requirements of General Design Criterion ' 62,
" Prevention of Criticality in Fuel Storage and Handling",
the definitive criteria contained in the April 14, 1978 USNRC letter and draft Regulatory Guide 1.13 (Rev. 2) are applicable.
i Calculations were also made to demonstrate the acceptability of a checkerboard loading pattern of fresh un-burned fuel of 5.0%
enrichment.
1.
2.0 SUP99LRY AND CONCLUSIONS 2.1 Nomal Storace Conditions The spent fuel storage racks in Pool A use stainless steel boxes to define the storage cells with a B,C matrix neutron absorber of l
0.015 gms B-10/cm' areal density. A water-gap between the absorber panels affords a
flux-trap to augment reactivity control.
Calculations were made for fuel burned to 10.5 MWD /KgU for fuel of 5.0% initial enrichment.
Table i summarizes the criticality safety analysis for 5.0%
enriched fuel at a burnup of 10.5 MWD /XgU.
The maximum k,,,
is i
0.9435, including uncertainties, which is within the Regulatory guideline (k,,, of 0.95) and is therefore acceptable.
The previous W
burnup limit curve was extended to include 5.0% enriched fuel
[
and the updated curve is shown in Figure 1.
The limiting burnup curve in Figure 1 is well described by a linear fit as shown below,
('
and may be used to calculate the minimum burnup for any initial enrichment, E,
up to 5.0%.
l Acceptable Burnup_in MWD /KgU
= 7.0
- E - 24.5 This fit is the same as previously determined, extended to include 5.0% enriched fuel.
Based upon the calculations reported here (see Table 1 and Figure 1), it is concluded that fuel of 5.0% initial enrichment is acceptable for storage in Pool A of the Crystal River Unit 3 spent fuel storage facilities, provided the fuel has attained a minimum burnup of 10.5 MWD /KgU. Minimum burnup specifications are shown in,
I Figure 1 for other enrichments (from the previous evaluation) with -
~
assurance that the maximum reactivity is within the regulatory limit.
calculations were also made for checkerboard arrangements of fresh 5.0% enriched fuel.
These calculations show that a checkerboard arrangement with empty cells (i.e. filled only with water or non-j fissile bearing material) is acceptable with a maximum k,,,
of 0.833.
2.2 Abnormal / Accident Conditions The reactivity consequences of abnormal / accident conditions were considered in the previous analysis") and found to be within acceptable bounds.
However, with the higher enrichment fuel (5.0%), the consequence of a mis-placed fuel assembly could differ from that previously evaluated.
Calculations with a mis-placed fuel assembly (fresh assembly of 5.0% enrichment accidentally loaded into a Pool A cell) resulted in a maximum k,,,
of 0.946 (including uncertainties) with all other cells filled with fuel of the maximum permissible reactivity.
This is within the Regulatory guideline even without the allowable credit for soluble boron.
1 3-
(-
a.O CRITICALITY SAFETY ANALYSES 3.1 Fuel Assembly Specifications The fuel assemblies used in the analyses is the Babcock & Wilcox 15 x 15 fuel assembly, the same as that used in the previous analyses").
Table 2 attached lists the design specifications for the fuel used in the analyses.
3.2 Storage Rack _Soecifications The storage rack cell design, illustrated in Figure 2, is composed of B,0 absorber material sandwiched between two 0.060 inch thick stainless steel boxes of 8.9375 inch inside dimension.
The cells are arranged on a 10.50 inch lattice spacing with a 1.173 inch water gap between the storage cells.
The stainless steel tabs connecting the storage boxes have a slightly negative reactivity effect and were neglected in the calculations.
The B C absorber 4
has a thickness of 0.075 inches and a B-10 loading of 0.015 i 0.003
+
gms B-10/cm'.
3.3 Manufacturina Tolerances and Uncertainties The small reactivity increments associated with manufacturing tolerances obtained in the previous evaluation") (i 0.0097) were assumed to remain applicable.
Combined with the uncertainty in bias (i 0.0024, Appendix A) results in a total uncertainty of i 0.0100.
Fuel of 5.0% enrichment also requires an increase in the allowance for uncertainty in the depletion calculations. As in the original evaluation for 4.5% fuel, the depletion uncertainty was assumed to be 5.0% of the reactivity decrement from beginning-of-life to the burnup of 10.5 MWD /KgU.
This allowance amount,s to 0.0035 Ak which is conservatively treated as an additiive term rather than being statistically combined with the other uncertainties..
Previous calculations have demonstrated a continuous reduction in
- ~
f.
reactivity with storage time (after Xe decay) primarily due to Pu-241 decay and An-241 growth. No credit is taken for this reduction in reactivity except to acknowledge an additional level of conservatism in the calculations.
t 3.4 Calculational Methodoloov 3.4.1 Computer Codes The principal methods of analysis were CASMO-3(2), a two-dimensional multigroup transport theory code for fuel assemblies and NITAWL -
KENO-Sa(3), a three dimensional Monte Carlo code package, using the 27-group SCALE
- cross-section library.
The calculational methods used for the present evaluation are comparable to those used in the original calculations, differing only in that updated versions of the codes were used, ie, CASMO-3 rather than CASMO-2E, and KENO-Sa rather than KENO-4.
Results of these codes are not significantly different from those of the earlier versions, and benchmarking of the updated codes resulted in a bias of 0.0000 i 0.0024 for CASMO-3 and 0.0103 i 0.0018 for NITAWL - KENO-Sa (95% probability, 95%
confidence level")).
A summary of the detailed bench-marking j
unalyses is included in Appendix A.
CASMO-3 was also used both for burnup calculations and for restart calculations in the rack geometry.
Since KENO-Sa cannot perform burnup
- analyses, CASMO-3 is used to define an equivalent enrichment, ie, the U-235 enrichment that yields the same reactivity in the racks as the burned fuel.
It was found that an enrichment of 3.4% yields the same reactivity in the storage racks as 5.0% fuel burned to 10.5 MWD /KgU.
Independent check calculation SCALE is an acronym for Etandardized Computer Analyses for Licensing Evaluation, developed for the USNRC by the Oak Ridge National Laboratory. d
l for the reference case with NITAWL-KENO-Sa (3.4%
equivalent t'
enrichment) gave a bias corrected k,,,
of 0.9280 1 0.0010 (without uncertainties) which confirms the CASFO-3 calculation (k,,, of 0.9300).
In the geometric model used in the calculations, each fuel rod and its cladding were described explicitly in both the CASE-3 and KENO-Sa models.
Reflecting boundary conditions (zero neutron current) were used in the radial direction which has the effect of creating an infinite array of storage cells in X-Y directions.
In the KENO-Sa
- model, the actual fuel assembly length was used in the axial direction, assuming thick (30 cm) water reflectors top and bottom.
Monte Carlo (KENO-Sa) calculations inherently include a statistical uncertainty due to the random nature of neutron tracking. To minimize the statistica.1 uncertainty of the KENO-calculated reactivity, a minimum of 500,000 neutron histories in 1000 generations of 500 neutrons each, were accumulated in each calculation.
3.4.2 Axial Distribution in Burnup i
Initially, fuel loaded into the reactor will burn with a slightly skewed cosine power distribution.
As burnup progresses, the burnup distribution will tend to flatten, becoming more highly burned in the central regions than in the upper and lower ends. At high burnup, the more reactive fuel near the ends of the fuel assembly (less than average burnup) occurs in regions of lower reactivity worth due to j
neutron leakage.
Consequently, it would be expected that over most of the burnup history, distributed burnup fuel assemblies would exhibit a slightly lower reactivity than that calculated for the average burnup.
As burnup progresses, the distribution, to some extent, tends to be self-regulating as controlled by the axial power distribution, precluding the existence of large regions of significantly reduced burnup.
, 1
Among others, Turnerm has provided generic analyses of the-axial
(
burnup effect based upon calculated and measured axial burnup
' distributions.
These analyses confirm the minor and generally negative reactivity effect of the axially distributed burnups at values less than about 30,000 Mwd /MtU.
Because a burnup of only 10.5 MWD /KgU is necessary for 5.0% fuel, the reactivity consequence of the axial distribution in burnup will be slightly negative.
3.'
checkerboard Configuration Fuel up to 5.0%
initial enrichment may also be stored in a checkerboard pattern, alternating with cells filled with only water or non-fissile material.
For this case, the maximum reactivity, including uncertainties, was calculated to be 0.8331, which is well below the USNRC guideline.
O
_7_
8
(
4.0 REFERENCES
1.
S.E Turner, " Criticality Safety Analysis of the Crystal River Unit 3 Pool A Spent Fuel Storage Rack", Southern Science Office of Black & Veatch, SS-162 (Undated) 2.
A.
- Ahlin, M.
- Edenius, H.
Haggblom, "CASMO
-A Fuel Assembly Burnup Program," AE-RF-76-4158, Studsvik report (proprietary),
A. Ahlin and M. Edenius, "CASMO - A Fast Transport Thaory Depletion Code for LWR Analysis,"
ANS Transactions, Vol.
j 26, p. 604, 1977.
N. Edenius et al., "CASMO Benchmark Report,"
Studsvik/ RF-78-6293, Aktiebolaget Atomenergi, March 1978.
3.
R.M.
- Westfall, et.
al.,
"NITAWL-S: Scale System Module for Performing Resonance Shielding and Working Library Production" in SCALE:
A Modular Code System for nerformina Standardized Computer Analyses for Licensina Evaluation NUREG/CR-0200,1979.
L.M. Petrie and N.F.
Landers," KENO Sa. An Improved Monte Carlo criticality Program with Supergrouping" in Scale: A Modular Code System for nerformina Standardized Cocouter Analyses for Licensino Evaluation, NUREG/V-0200, 1979.
R.M.
Westfall et al.,
" SCALE:
A Modular Code System for nerformina Standardized Computer Analyses for Licensina Evaluation," NUREG/V-0200, 1979.
4.
M.G.
- Natrella, Exoerimental Statistics, National Bureau of Standards, Handbook 91, August 1963.
5.
S. E. Turner, " Uncertainty Analysis - Burnup Distributions", in Proceedinas of a Workshop on the use of Burnuo Credit in Soent Fuel Transport Casks, Sandia Report SAND-89-0018, October 1989.
,,:1-Table 1 SUl@mRY OF CRITICALITY SAFETY ANALYSES POOL A S'IORAGE RACKS Fuel Enrichment, wtt U-235 5.00 Design Burnup, MWD /KgU 10.50 l
Reference Temperature,
- F 68 l
1 Reference km (CASPD-3) 0.9300 Calculational Bias, Ak")
0.0000 Axial Burnup Distribution Negative Ak Uncertainties (2) to.0100 Total 0.9300
- 0.0100 Ak allowance for depletion to.00035 calculations (2)
Maximum Reactivity (k=)
0.9435 l
Regulatory Limit ( k,,,
0.95 U) Appendix A (2) Section 3.3 i '
4
~
TABLE 2 FUEL' ASSEMBLY SPECIFICATIONS Fuel Rod Data 1
f Outside dimension, in.
0.430 Cladding ID, in.
0.377 Cladding thickness, in.
0.0265 Cladding material Zr-4 Pellet diameter, in.
0.369 2
UO density, g/cm 10.420 i 0.166 2
Enrichment, wt.% U-235 5.0 i 0.02 Fuel Assembly Data Number of fuel rods 208 (15x15 array) i Fuel rod pitch, in.
0.568 Control rod guide tube Number 16 O.D.,
in.
0.530 Thickness, in.
0.016 Material Zr-4 Instrument thimble Number 1
0.D.,
in.
0.493 Thickness, in.
0.02' Material Zr-4 p a
i.
11 10
)
I 9
/
B ACC EPT AEL E B JRNJP
/
3 D 3 MAIN
/
q 7
g
/
6
/(
5
/
i f
4
[
l 3
g I
/
LRJMIIP" AIL: IUNLP l
/
D0ftIN l
2 1
f 6
1
/
0 i
3.0 3.5 4.0 4.5 5.0 1
INITIAL FLEl. DRIOfENT. Wts U-236 l
l F ig. 1 Accept.oble Bur n up Dome tn in Pool A
.. (
i e
1 l
l
.-.l.-.-.
6.687
- 0.063" B4C WATRIX.-.-------l.-.
l 2;
l 4
l 0.075" THK.015 i 0.003 gmB-10/cm e
a B&W 15x15 FUEL l
.l O
38' '"-
i l
DENSITY 10.42 g/cc e
00 e
NUMBER 208 (15 X 15) i OOO cuo io 0.377 iN.
i OOOO ROD OD 0.430 IN.
l i
OO900 PITCH 0.568 IN.
i i
O00000 THlWBLES 0.060 SS i
O000000 OD 0.530IN. ~
- 1NNER & OUTER BOX i
O0000000
$0. 77 8 IN.
"[737:
i 00900900 j
0000000000
{
000e0000000 i
000000000900 1
00000m^^^^^O0 8.937" BOX l.D.
ggt 10.50" PITCH d
l
. -.l-8 i
NOT TO SCAL 2 FIG. 2 CR3 P00L A FUEL STORAGE CELL l l (
l
'I
---_- a
i r
l t
APPENDIX A
]
1 i
BENCHMARK CAIEUIATIONS 1
I s
by Stanley E. Turner, PhD, PE HOLTEC INTERNATIONAL November, 1993
{
t
'l I
e l.
TABLE OF CONTENTS
1.0 INTRODUCTION
AND
SUMMARY
A-1 2.0 NITAWL-KENO 5a BENCHMARK CAIEUIATIONS A-2 3.0 CASMO3 BENCHMARK CAIEUIATIONS A-4 1
4.0 WORKER ROUTINE A=5 5.0 CI4SE-PACKED ARRAYS A-6
6.0 REFERENCES
A-7 l
b I
i i
h a
O
. ~..,
r_
(
List of Tables Table 1 RESULTS OF 27-GROUP (SCALE) NITAWL-KENO 5a.
A-9 CAICUIATIONS OF B&W CRITICAL EXPERIMENTS Table 2 RESULTS OF 27-GROUP (SCALE) NITAWL-KEN 05a
. A-10 CAIEUIATIONS OF FRENCH and BNWL CRITICAL EXPERIMENTS i
Table 3 RESULTS OF CASNO3 AND NITAWL-KENO 5a..... A-11 BENCHMARK (INTERCOMPARISON) CAIEUIATIONS Table 4 Intercomparison of WORKER-NITAWL-KEN 05a.
. A-12 and CASMO3 Calculations at Various Temperatures Table 5 Reactivity Calculations for Close-Packed
. A-13
[
Critical Experiments l
l List of Floures Fig. 1 COMPARISON OF CASNO AND KEN 05A CALCUIATIONS.. A-14 AT VARIOUS ENRICHMENTS IN REPRESENTATIVE FUEL STORAGE RACK Fig. 2 COMPARISON OF CASNO3 AND KEN 05A A=15 TEMPERATURE DEPENDENCE 11
(
a
r.
1.0 INTRODUCTION
AND
SUMMARY
g l
The objective of this bes.chmarking study is to verify both the NITAWL-KENO 5a ta) methodology with the 27-group SCALE c
cross-section library and the CASMO3 code (3) for use in criticality sa"ety calculations of high density spent fuel storage racks. Both calculational methods are based upon transport theory and have been benchmarked against critical experiments that simulate typical I
spent fuel storage rack designs as realistically as possible.
Results of these benchmark calculations with both methodologies are consistent' with corresponding calculations reported in the literature.
l Results of the benchmark calculations show that the 27-group (SCALE) NITAWL-KEN 05a calculations consistently under-predict the critical eigenvalue by 0.0103 i 0.0018 ok (with a 95%
i probability at a 95% confidence level) for critical experiments (')
)
that are as representative as possible of realistic spent fuel i
storage rack conff crations and poison worths.
j Extensive benchmarking calculations of. critical experi-ments with CASMO3 have also been reported (5), giving a mean k,,, of 1.0004 i 0.0011 for 37 cases.
With a K-factor of 2.14(') for 95%
l probability at a 95% confidence level, and conservatively neglect-I ing the small overprediction, the CASMO3 bias then becomes 0.0000 j
i 0.0024. CASMO3 and NITAWL-KEN 05a intercomparison (
,Jiations of infinite arrays of poisoned cell configurations (representative of c
in that 0.0 0 i 0.0024 is a asonable b s and cran ty for CASMO3 calculations.
Reference 5 also documents good agreement of heavy nuclide concentrations for the Yankee core isotopics, agreeing with the measured values within experimental
)
error.
A-1 j
(
i
The benchmark calculations reported here confirm that
($
either the 27-group (SCALE) NITAWL-KEN 05a or CASMO3 calculations are acceptable for criticality analysis of high-density spent fuel storage racks.
Where possible, reference calculations for storage rack designs should be performed with both code packages to provide independent verification.
CASMO3, however, is not reliable when large water gaps ( > 2 or 3 inches) are present.
2.0 NITAWL-KEN 05a BENCHMARK CALCULATIONS Analysis of a series of Babcock & Wilcox critical experiments"), including some with absorber panels typical of a poisoned spent fuel rack, is summarized in Table 1, as calculated with NITAWL-KEN 05a using the 27-group SCALE cross-section library and the Nordheim resonance integral treatment in NITAWL.
Dancoff factors for input to NITAWL were calculated with the Oak Ridge SUPERDAN routine (from the SCALE (2) system of codes). The mean for these calculations is 0.9899 i 0.0028 (1 o standard deviation of the population).
With a one-sided tolerance factor corresponding to 95% probability at a 95% confidence level"), the calculational bias is + 0.0103 with an uncertainty of the mean of i 0.0018 for the sixteen critical experiments analyzed.
Similar calculational deviations have been reported by ORNd7) for some 54 critical experiments (mostly clean criticals without strong absorbers), obtaining a mean bias of 0.0100 i 0.0013 (95%/95%).
These published results are in good agreement with the results obtained in the present analysis and lend further credence to the validity of the 27-group NITAWL-KEN 05a calculational model for use in criticality analysis of high density spent fuel storage racks.
No abnormal deviations in k,,, with intra-assembly water gap, with absorber panel reactivity worth, with enrichment or with 1
poison concentration were identified with the 27 group SCALE library, comparable to those previously observeds) with the 123-t group GAM-THERMOS cross-section library.
~
A-2 s
9
1 Additional benchmarking calculations were also made for I
a series of French critical experiments") at 4.754 enrichment and for several of the BNWL criticals with 4.26% enriched fuel.
Anallysis of the French criticals (Table 2) showed a tendency to overpredict the reactivity, a result also obtained by ORNd10)
The calculated k,,,
values showed a trend toward higher values with decreasing core size.
In the absence of a significant enrichment effect (see section 3 below), this trend and the overprediction is attributed to a small inadequacy in NITAWL-KEN 05a in calculating neutron leakage from very small assemblies.
Similar results were observed for the BNWL series of critical experiments ("), which are also small assemblies (although significantly larger than the French criticals).
In this case (Table 2),
the calculated mean k,,,
was 0.9959 1 0.0013 (1 e population standard deviation).
Because of the small size of the BNWL critical experiments (compared to the B&W criticals used to determine the KEN 05a bias) and the absence of any significant g
enrichment effect, the results also suggest a small inadequacy of NITAWL-KEN 05a in treating large neutron leakage from very small assemblies.
Since the analysis of high-density spent fuel storage racks generally does not entail neutron leakage, the observed inadequacy of NITAWL-KEN 05a is not significant.
Furthermore, omitting results of the French and BNWL critical experiment analyses from the determination of bias is conservative since any leakage that might enter into the analysis would tend. to result in overprediction of the reactivity.
e
(
A-3 4
t I
3.0 CASMO3 BENCHMARK CALCULATIONS The CASMO3 code is a multigroup transport theory code utilizing transmission probabilities to accomplish two-dimensional calculations of reactivity and depletion for BWR and PWR fuel assemblies.
As such, CASMO3 is well-suited to the criticality analysis of spent fuel storage racks, since general practice is to treat the racks as an infinite medium of storage cells, neglecting leakage effects.
1 CASMO3 is a modification of the CASMO-2E code and has been extensively benchmarked against both mixed oxide and hot and cold critical experiments by Studsvik Energiteknik s)
Reported t
analyses (5) of 37 critical experiments indicate a mean k,,, of 1.2004 1 0.0011 (10).
To independently confirm the validity of CASMO3 (and to investigate any effect of enrichment),
a series of i
calt.ulations vore made with CASMO3 and with NITAWL-KEN 05a on identical poisoned storage cells representative of high-density j
spent fuel storage racks.
Results of these. intercomparison calculations * (shown in Table 3 and in Figure 1) are within the normal statistical variation of KEN 05a calculations and confirm the bias of 0.0000 i 0.0024 (95%/95%) for CASMO3.
Since two independent methods of analysis would not be expected to have the same error function with enrichment, results j
of the intercomparison analyses (Table 3) indicate that there is no significant effect of fuel enrichment over the range of enrich-ments involved in power reactor fuel.
Intercomparison between analytical methods is a techni endorsed by Reg. Guide 3.41, " Validation of Calculatio,que nal Methods for Nuclear Criticality Safety".
A-4
~(7 A second series of CASMO3 and KEN 05a intercomparison calculations consisting of five cases from the BAW critical experiments were analyzed for the central cell only. The calculat-ed results, also shown in Table 3, indicate a, mean difference within the 95% confidence limit of the KENO 5a calculations.
This lends further credence to the recommended bias for CASMO3.
4.0 WORKER ROUTINE The WORKER routine was obtained from ORNL and is intended to interpolate the hydrogen scattering matrices for temperature in order to correct for the deficiency noted in NRC Information Notice 91-66 (October 18, 1991). Benchmark calculations were made against CASMO3, based on the assumption that two independent methods of analysis would not exhibit the same error.
Results of these i
calculations, shown in Table 4,
confirm that the trend with temperature obtained by both codes are comparable.
This agreement i
I establishes the validity of the WORKER routine, in conjunction with f
NITAWL-KEN 05a, in calculating reactivities at temperatures between i
20'C and J.20*C.
The deficiency in the NITAWL hydrogen scattering matrix at temperatures above 20 *C does not appear except in the presence of a large water gap where the scattering matrix is important.
i However, the absolute value of the km from CASMO3 is not reliable in the presence of a large water gap, although the relative values should be accurate.
In the calculations shown in Table 4 and in l
Figure 2, the absolute reactivity values differ somewhat but the 4
trends with temperature are sufficiently in agreement to land credibility to the WORKER routine over the temperature range from i
20*C to 120*C.
I L
A-5 l
4 L..
-=
5.0 CLOSE-PACKED ARRAYS l
(~
The BAW close-packed series of critical experiments"2) intended to simulate consolidated fuel, were anslyzed with NITAWL-KENO 5a.
Results of these analyses, shown in Table 5, suggest a slightly higher bias than that for fuel with normal lattice i
spacings.
Similar results were obtained by ORNLU33 Becauae there are so few cases available for analysis, the maximum bias for close-packed lattices may be taken as 0.0155, including uncertain-ty.
This would conservatively encompass all but one of the cases measured.
r K
i i
l i
1 I
A-6
(
t
'y 6.O REFERENCES TO APPENDIX A r~
1.
Green, Lucious, Petrie, Ford, White, and Wright, "PSR /NITAWL-1 (code package) NITAWL - Modular Code System For Generating Coupled Multigroup Neutron-Gamma Libraries from ENDF/B", ORNL-TM-3706, Oak Ridge National Laboratory, November 1975.
2.
R.M. Westf all et. al., " SCALE: A Modular System for Performing Standardized Computer Analysis for Licensing Evaluation",
NUREG/CR-0200, 1979.
3.
A.
- Ahlin, M.
Edenius, and H.
- Haggblom, "CASNO A Fuel Assembly Burnup Program", AE-RF-76-4158, Studsvik report.
A. Ahlin and M. Edenius, "CASNO - A Fast Transport Theory Depletion Code for LWR Analysis", ANS Transactions, Vol. 26,
- p. 604, 1977.
"CASNO3 A Fuel Assembly Burnup Program, Users Manual",
Studsvik/NFA-87/7, Studsvik Energitechnik AB, November 1986
(
4.
M.N.
Baldwin et al., " Critical Experiments Supporting close Proximity Water Storage of Power Reactor Fuel", BAW-1484-7, The Babcock & Wilcox Co., July 1979.
5.
M. Edenius and A. Ahlin, "CASMO3: New Features, Benchmarking, and Advanced Applications", Buclear Science and Enaineering, 100, 342-351, (1988) 6.
M.G.
Natrella, Exnerimental Statistics, National Bureau of Standards, Handbook 91, August 1963.
7.
R.W.
Westfall and J. H. Knight, " SCALE System Cross-section Validation with Shipping-cask Critical Experiments",
- AHS, Transactions, Vol. 33, p. 368, November 1979 8.
S.E.
Turner and M.K.
- Gurley,
" Evaluation of NITAWL-KENO Benchmark Calculations for High Density Spent Fuel Storage Racks",
Nuclear Science and Engineerina, 80(2):230-237, February 1982.
A-7 1
t
9.
J.C. Manaranche, et. al., " Dissolution and Storage Experiment with 4.75% U-235 Enriched U0 Rods", Hustlear Technoloav, Vol.
50, pp 148, September 1980. 2 10.
A.M.
Hathout, et.
al.,
" Validation of Three Cross-section Libraries Used with the SCALE System for Criticality Analy-sis", Oak Ridge National Laboratory, NUREG/CR-1917, 1981.
11.
S.R.
Bierman, et.
al.,
" Critical Separation between Sub-23% Enriched UO,fi critical Clusters of 4.29 Wt. 4 Rods in Water with Fixed Neutron Poisons",
Battelle Paci c Northwest Laboratories, NUREG/CR/0073, May 1978 (with August 1979 errata).
12.
G.S. Hoovler, et al., " Critical Experiments Supporting Underwater Storage of Tightly Packed Configurations of Spent Fuel Pins", BAW-1645-4, Babcock & Wilcox Company (1981).
13.
R.M. Westfall, et al., " Assessment of Criticality Computation-al Software for the U.S.
Department of Energy Office of Civilian Radioactive Waste Management Apslications", Section 6, Fuel Consolidation Applications, ORNL/CSD/TM-247 (undated),
I
Table 1 RESULTS OF 27-GROUP (SCALE) NITAWL-KEN 05a CAICUIATIONS OF B&W CRITICAL EXPERIMENTS Experiment Calculated a
Number k,,,
1 I
0.9922 i 0.0006 II 0.9917 i 0.0005 III 0.9931 i 0.0005 IX 0.9915 i 0.0006 X
0.9903 i 0.0006 XI 0.9919 i 0.0005 XII 0.9915 i 0.0006 s
XIII 0.9945
, i 0.0006 XIV 0.9902 i 0.0006 I
XV 0.9836 i 0.0006 XVI 0.9863 i 0.0006
)
XVII 0.9875 i 0.0006 XVIII 0.9880 i 0.0006 XIX 0.9882 1 0.0005 XX 0.9885 1 0.0006 XXI 0.9862 i 0.0006 Mean 0.9897 i 0.00070)
Bias (95%/95%)
0.0103 i 0.0018 03 Standard Deviation of the Mean, calculated from the k,,, values.
A-9
L.
a
- e -
Table 2 RESULTS OF 27-GROUP (SCALE) NITAWL-KEN 05a CALCULATIONS OF FRENCH and BNWL CRITICAL EXPERIMENTS French Experiments Separation Critical Calculated Distance, cm Height, cm k,,,
0 23.8 1.0302 i O.0008 2.5 24.48 1.0278 i 0.0007 5.0 31.47 1.0168 i 0.0007 10.0 64.34 0.9998 i 0.0007 BNWL Experiments Calculated i
Case Expt. No.
k,,,
No Absorber 004/032 0.9942 1 0.0007 SS Plates (1.05 B) 009 0.9946 i 0.0007 SS Plates (1.62 B) 011 0.9979 i 0.0007 SS Plates (1.62 B) 012 0.9968 i 0.0007 SS Plates 013 0.9956 i 0.0007 SS Plates 014 0.9967 i 0.0007 Zr Plates 030 0.9955 1 0.0007 Mean 0.9959 i 0.0013 A - 10 i
a
.O L
Table 3 RESULTS OF CASNO3 AND NITAWL-KEN 05a BENCHMARK (INTERCOMPARISON) CALCULATIONS Enrichment")
k*
Nt. % U-235 NITAWL-KEN 05a(8)
CASMO3 l8kl 2.5 0.8376 i 0.0010 0.8386 0.0010 3.0 0.8773 1 0.0010 0.8783 0.0010 3.5 0.9106 i 0.0010 0.9097 0.0009 4.0 0.9367 i 0.0011 0.9352 0.0015 4.5 0.9563 i 0.0011 0.9565 0.0002 5.0 0.9744 i 0.0011 0.9746 0.0002 Mean 0.0008 l
Expt. No.(3)
XIII 1.1021 i 0.0009 1.1008 0.0013 XIV 1.0997 i 0.0008 1.1011 0.0014 XV 1.1086 i 0.0008 1.1087 0.0001 XVII 1.1158 i 0.0007 1.1168 0.0010 XIX 1.1215 i 0.0007 1.1237 0.0022 Mean 0.0012 08 Infinite array of assemblies typical of high-density spent fuel storage racks.
(23 k, from NITAWL-KEN 05a corrected for bias.
(3)
Central Cell from BAW Critical Experiments A - 11 e
t a b
-e
.o Table 4 I
Intercomparison of WORKER-NITAWL-KEN 05a and CASM03 Calculations at Various Temperatures Temperature CASMO3 W-N-KEN 05a(*8 4*C 1.2276 1.2345 i 0.0014 17.5'C 1.2322 1.2328 i 0.0015 l
25'C 1.2347 1.2360 1 0.0013 f
50*C 1.2432 1.2475 i 0.0014 75'c 1.2519 1.2569 i 0.0015 120*C 1.2701 1.2746 i 0.0014
- Corrected for bias
(
A - 12
('
4
l*
i-Table 5 I
Reactivity Calculations for close-Packed critical Experiments l
Calc.
BAW Pin Module Boron Calculated No.
Expt.
Pitch Spacing Conc.
k,,,
No.
cm en ppa KS01 2500 Square 1.792 1156 0.9891 i 0.0005 1.4097 KSO2 2505 Square 1.792 1068 0.9910 i 0.0005 1.4097 KS1 2485 Square 1.778 886 0.9845 i 0.0005 Touching KS2 2491 Square 1.778 746 0.9849 i 0.0005 Touching KT1 2452 Triang.
1.86 435 0.9845 i 0.0006 i
Touching KT1A 2457 Triang.
1.86 335 0.9865 i 0.0006 i
Touching KT2 2464 Triang.
2.62 361 0.9827 1 0.0006 Touching KT3 2472 Triang.
3.39 121 1.0034 i 0.0006 Touching A - 13 4
9
/
1.00
[
0.95 N
~
sz mg e.9e s
CASW KENO-Ee 0.85 0.80 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 i
COrPARISON OF CA KENOGe CALCLLATIONS AT VARIOUS ENRICMS IN REPRESENTATIVE FLE1. STORAGE RACK l
l t
l i
1 t,
A - 14 I
[
l O
i i
I
- o-l 1.28 l
~
1.27_
e f) 1.26,
--/
e g
.s l
~
5 f**h
--[/
E
/
(
1.24 5 ).
1.23 5!
1.22" O
20 40 60 80 100 120 140 Te mpera tur e, Degrees C r
Fta. 2 COtFARISON OF CAST 10-3 en d KEN 05e TEtPERATURE DEPENDENCE L
e f
A - 15 i
s.
p i
A
1 ATTACHMENT 3 PROPRIETARY INFORMATION
_ _ _ _ _ _ _ _ _ _. _ _ _ _ _ _.