ML20094P269
| ML20094P269 | |
| Person / Time | |
|---|---|
| Site: | River Bend  |
| Issue date: | 03/31/1992 |
| From: | Greene C, Leatherwood L, John Miller GULF STATES UTILITIES CO. |
| To: | |
| Shared Package | |
| ML20094P268 | List: |
| References | |
| EA-CA-91-0001-S, EA-CA-91-0001-S2-R00, EA-CA-91-1-S, EA-CA-91-1-S2-R, NUDOCS 9204070317 | |
| Download: ML20094P269 (17) | |
Text
. . _ _ _ _ . _ . -
EA-CA-910001 S2 !
Revision 0 -
STEADY STATE CORE PilYSIC3 METilODS FOR llWR DESIGN AND ANALYSIS r
Appendix G March 1992 i .
Principal lingintCI Craig 11. Greene r
Contributors Lynn A. leatherwood Gary W. Scronce i
Phu Van Vo Review: *se O, IM'='; :( 3[MPIT1 ,
L. A. Leatherwood Date Supervisor - Core Analysis
-, t
'm y ) .)Y3 , h, !; ^ -l2{>/]7-2, b Approve., . .
Date J.S'l h) iller Director - Engineering Analysis Gulf States Utilities Company River Bend Station P.O. Box 220 St Francisville, Louisiana 70775 f
9204070317 920326 PDR ADOCK 05000458 P PDR
-,_a......._...
!!A CA 914XX)l S2 Revision 0 IMPORTANT NOTICE RiiGARDING CONTl!NTS OF Tills DOCUMiiNT r
PL11ASli R11AD CARiiFULLY This document was prepared by Gulf States Utilities Company for the use of the U.S.
Nuclear Regulatory Commission in matters regarding the operating license for the River Bend Station. To the best of the issuer's knowledge, this document contains work performed in accordance with sound engineering practice and is a true and accurate representation of the facts.
The work reported herein is the property of Gulf States Utilities Company, and any usage other than as described above is prohibited. Other than for the intended usage, neither Gulf States Utilities Company, nor any of its employees or officers, nor any other person acting on its behalf:
~
Makes any warranty or re' presentation, express or implied, i'ith respect to the accuracy, completeness, or usefulness of the information contained in this report, or that the use of any information, apparatus, n.ethod, or process disclosed herein would not infringe privately owned rights; or Assumes any liabilities with respect to the use of, or for damages resulting
, from the use of, any information, apparatus, method, or process disclosed in this report, k:
_ _ _ __ _ -- a
4 EA CA 910001-S2 Revision 0 APPENDIX G ADDITIONAL INFORMATION REQUEST NO. 2 1
I!A CA-91-0001 S2 Revision 0 LIST OF TAllLilS Iabic Sublest l' age G- 1.1 Comparison of GSU and Fuel Vendor Loss of Feedwater licating !
Analyses for RilS Cycle 3 ............................ 6 I G-1.2 Summary of Loss of Feedwater licating Analysis with Similar l Assumptions for RBS Cycle 3 . . . . . . . . . . . . . . . . . . . . . . . . . . 7 !
G 2,1 Comparison of GSU and Fuel Vendor Control Rod Withdrawal lirror ,
Analyses for RBS Cycle 3 ............................ 9
-G 2.2 Summary of Control Rod Withdrawal Error Analysis with Similar Assumptions for RilS Cycle 3 . . . . . . . . . . . . . . . . . . . . . . . . . . 10 ,
G-3.1 Summary of Differences Between GSU and Fuel Vendor Standby Liquid Control System Shutdown Margin Analyses for RilS Cycle 3 .. 12 G-3.2 Summary of Standby Liquid Control System Shutdown Margin Analysis with Similar Assumptions for RBS Cycle 3 ............ 13 G-4.1 Quad Cities Gamma Scan 1.ocal Power Benchmark Results, RMS lirror for A ssembly GE11002 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 G-4.2 Quad Cities Gamma Scan Local Power Benchmark Results, Summary of Overall Bench mark . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 2
I
IIA CA-910001-S2 Revision 0 APP 11NDIX G SUhthiARY On January 31,1991, Gulf States Utilities (GSU) issued 1.icensing Topical Report liA-CA 91-0001 ht, which documented GSU's core analysis methods for performing nuclear fuel reload safety analyses. On December 9,1991, the U.S. Nuclear Regulatory Commission (NRC) provided 21 questions concerning GSU's core analysis methods. On January 8,1992, GSU responded to the NRC's questions by providing liA CA-91-0001-S1, Appendix F to the original topical r port. Following the Psuance of Appendix F, the NRC asked three additional questions on February 4,1992, regarding information in GSU topical repor*. IIA-CA 91-0001 ht. The questions asked the technical reasons for the GSU results for the Loss of Feedwater lleating analysis, the Control Rod Withdrawal lirror analysis, and the Standby Liquid Control System cold shutdown margin analysis being nonconservative relative to fuel vendor results. GSU reanalyzed these events for RilS Cycle 3 with similar assumptions to that of the fuel vendor and found the revised results were in close agreement with fuel vendor results. This Appendix describes those assumptions and results of GSU reanalysis.
During the preparation of these responses, an error was discovered :n the cycle depletion analyses which defined the core statepoints at which each of the subject analyses was performed. The error, which involved analytical options within SIMULATii-li, resulted in part of the exposure history being analyzed under the Coarse Mesh Diffusion Theory (CMDT) nuclear option rather than the Modified Coarse Mesh Diffusion Theory (MCMDT) stipulated in the methodology description in Chapter 2 of the base report. -
Since MCMDT and CMDT are permutations of the same nodal coupling theory, the differences were small. This App:ndix describes the differences.
This Appendix also contains a reanalysis of the Quad Cities gamma scan local power benchmarks prompted by the discovery of an error in GSU's original analysis. The error was located in the normalization logic of the single-purpose routine used to interpret local power benchmark data. The correction of the normalization logic improved the statistical results and the overall standard deviation for the local power Quad Cities benchmarks.
See Section G-4 for a discussion of the correction.
3
EA CA-910001 S2 Revision 0 NRC Question: 1 G- 1. Explain the nonconservatism (rt.ative to the fuel vendor) of the GSU Loss of Feedwater lleating (LFWH) analysis for RilS Cycle 3. !
l GSU Response: j The GSU calculations and the fuel vendor calculations were performed with slightly different assumptions and analytical conditions. While the GSU calculations are
,. tulated differently from the fuel vendor calculations, the analytical results remain
, wrvative relative to best estimate predictionr, of plant phenomena. The differences between the two analyses are summarized in Table G-1.1.
Differences in hydraulic modeling and exposure distribution were investigated as potential sources of conservatism in the fuel vendor analysis.- Neither of these differences were foend to have a sign 10 cant effect on the Operating Limit h1CFR (OLhtCPR) calculated for the LFWH transient. Reevaluation of the LFWH transient with htCh1DT also resulted in negligible effect on the OLhtCPR.
The main difference between the GSU and fuel vendor analyses is in the void and Doppler reactivity, which are included implicitly in the cross sections used by the nodal simulator code. While the GSU coefficients are numerically close to the fuel vendor results, the combination of differences drives a substantial difference in Gnal power level for the LFWH event. Overall accuracy of the Doppler and void feedback mechanisms in the GSU modeling is demonstrated in the benchmark sections of the main report.
To quantify the real difference m analytical results, the GSU methodology was exercised using the fuel vendor assumptions. The results of this reanalysis are shown in Tabie G-1.2. The comparative results show close agreement between the GSU analysis and the fuel vendor analysis when similar analytical assumptions are employed.
,Te reanalysis accounted for differences in void and Doppler reactivity feedback (ficients by adjusting the fuel temperature degendence of the fast group absorption cross section in the Sih1ULATE-E analysis until the Onal conditions agreed with those in the fuel vendor analysis. While this measure did not duplicate the fuel vendor coefficients exactly, it provided a reasonable approximation of the Gnal effects with a minimum impact on the nuclear modeling.
4
wumi - c__ - , - ;-
FA vi$
EA-CA-91-(XX)l S2 Revision 0 ag Conservatism ir.troduced by the use of a high initial power assumptien assures final M results which are conserwitive relative ;o expected plam behavior. The use of reactivity adjustments to raise the final power level provides additional consernt4sm in tne result but is not necessary for the establishment of conservative operating lim.ts for the reactor pre.
E ,
The fuel vendor results shown in Table G-1.2 am raw analytical output. In the i determination of core operating limits, the fuel vendor adds additicnal conservatism to the calculated ACPR befo.e dd.ermining the OLMCPR. This adjustment accounts for performing the analysis only at end of cycle conditions. GSU analyses for core operating limits are expected to include iJdetermination of the most conservative exposure point for the LFWil transient during the operating cycle, which would eliminate the need for this allooance.
NI E
e 1
1 5
\
EA-CA-91-0001-S2 Revision 0 Table G-1,1 Comparison of GSU and Fuel Vendor Loss of Feedwater Heating Analyses for RBS Cycle 3 i Parameter GSil' Euel vends Exposure Distribution Actual Haling BOC3 core average exposure, GWd/Te 10.7F; 10.48 EOC3 core average exposure, GWd/Te 20.22 19.98
)
Cycle exposure ia analysis, GWd/Te 20.22 19.98 _
Hydraulic Modeling River Be .o Generic BWR/6 '
.BWR hwuJ 4
Initial Conditions Core Power, %NBR 102 102 ,
Core Pressure, psia 1059.1 10$ii.8 Core Inlet Subcooling, BTU /lb 23.7 23.25 Core Flow /Mlb./hr 84.5 84.5
) ICPR 1.293 1.374
~ Final Conditions Core Power, %WGR 114.9 117.2 Core Pressure, psia 1083.8 1062.6
- Core Inlet Subcooling, BTU /lb, M.9 39.25 Core Flow, Mlb./hr 84.5 85.9
~
3 MCPR 1.217 1.278 s CASMO
,. Reactivity Coefficients Proprietary
, analysis methodology ACPR 0.076 L.096 OLMCPR 1.14 1.15
'Results of the original LFWH analysis reported in Section 7.0 of the base report.
Except for OLMCPR, these values were taken from Table 7.2.
6
EA CA-91-0001-S2 Revision 0 l
Table G-1.2 Summary of Loss of Feedwater lleating Analysis with Similar Assumptions for RBS Cycle 3 E;trameter QSl! Fuel Vendet Initial Conditions Core Power, % NBR 102 102 Core Flow, Mlb,,,/hr 84.5 84.5 __
Core Pressure, psia 1058.8 1058.8 Core inlet Subcooling, BTU /lb. 23.25 23.25 ICPR 1.315 1.374 Final Conditions Core Power, % NBR 117 ' 117.2 Core Flow, Mlbm/hr 85.3 85.9 Core Pressure, psia 1062.6 106?.6 Core inlet Subcooling, BTU /lb, 39.25 39.25 MCPR 1.222 1.278 ACPR8 0.093 0.096 OLMCFR 1.15 1.15
' Calculated value; operating limit is determined by adjusting ACPR to allow for uncertainties in core operation.
7
EA-C A-91-0001-S2 Revision 0 NRC Question:
G-2. Explain the nonconservatism (relative to the fuel vendor) of the GS" Control Rod Withdrawal Error (CRWE) analysis for RBS Cycle 3.
GSU Response:
] The GSU calculations and the fuel vendor calculations were performed with slightly different assumptions and analytical conditions. While the GSU calculations are formulated differently from the fuel vendor calculations, the analytical results remain conservative relative to best-estimate predictions of plant phenomena. Ta differences between the two analyses are summarized in Table G-2.1.
Differences in hydraulic modeling and exposure distribution were investigated as potential sources of conservatism in the fuel vendor analysis. Neither of these differences were found to have a significant effect on the OLh1CPIl calculated for the CRWE transient.
When the CRWE transient was reevaluated with htCh1DT, the OLh1CPR difference between GSU and fuel vendor results was reduced by 0.02, bringing the two methodologies into agreement. The apparent nonconservatism in the CRWE analysis was primarily a result of the selection of nuclear n ' de in the Slh1ULATE-E calculation.
The results of the GSU and fuel vendor analyses of the CRWE transient are compared in Table G-2.2.
8
?
EA-CA-91-0001-S2 Revision 0 Table G-2,1 Comparison of GSU and Fuel Vendor Control Rod Withdrawal Error Analyses for RBS Cycle 3 Parameter OSU Fuel Vendor Exposure Distribution Rodded Haling Depletion
~
BOC3 core average exposure, GWd/Te 10.70 10.43 EOC3 core average exposure, GWd/Tc Not used 19.98 Cycle exposure in analysis, GWd/Te 3.00 2.76 Hydraulic hiodeling River Bend Generic BWR/6 FIBWR hiodel Strongest Rod Location 32-17 32-17 OLh1CPR 1.16 1.18 9
I EA-CA-91-0001-S2 Revision 0 Table G-2,2 Summary of Control Rod Withdrawal Error Analysis with Similar Assumptions for RBS Cycle 3 Earameter 01U Fuel VendfI
~ Exposure Distribution Haling Haling BOC3 core average exposure, GWd/Tc 10.48 10.48 _
EOC3 core average exposure, GWd/Te 19.98 19.98 Cycle exposure in analysis, GWd/Te 2.76 2.76 Hydraulic Modeling River Bend Generic BWR/6 FIBWR Model Strongest Rod Location 32-17 32-17 OLMCPR 1.18 1.18 10 l
l EA-CA-91-0001-S2 l l
Revision 0 I
i NRC Question:
i G-3. Explain the nonconservatism (relative to the fuel vendor) of 9e GSU Standby Liquid _ Control System (SLCS) cold shutdown margin for RlL Cycle 3. !
GSU Response:
The GSU calculations and the fuel vendor calculations were performed with slightly different assumptions and analytical conditions. While the GSU calculations are l formulated differently from the fuel vendor calculations, the analytical results remain i conservative relative to best-estimate predictions of plant phenomena. The differences between the two analyses are summarized in Table G-3.1.
The GSU method calculates the soluble boron worth by adjusting the thermal group absorption cross section consistent with boron worth calculated by CASMO. This calculation is conservatively applied to the SLCS shutdown margin analysis by selecting conservative adjustment factors over the appropriate uposure and void history intervals for each fuel type in the core. Actual boron worth predictions were not made available to GSU by the fuel vendor; however, the fuel vendor method involves approximation of the borated k. for each fuel type bv applying a conservatively low boron reactivity worth to a non-borated k calculation. The fuel vendor c;timates this reactivity worth at 70%
void history, which is deterministically conservative relative to expected void history in the core.
To quantify the diffetence between the fuel vendor models and the GSU models, the SLCS cold shutdown margin was reanalyzed using the GSU modeis and the fuel vendor methods identified above. The results of this reanalysis are shown in Table G-3.2. The l comparative results show close agreement between the GSU analysis and the fuel vendor analysis when the analyses are performed on the same basis. SLCS shutdown margins calculated with CMDT are slightly greater than those calculated with MCMDT under the same conditions.
i 11
! IIA-C A-91-0001-S2 Revision 0 Table G-3.1 Summary of Differences Between GSU and Fuel Vendor Standby Liquid Control System Shutdown Margin Analyses for RBS Cycle 3 Earameter [iS1! Fuel Vendor Soluble boron concentration in detailed analysis, ppm 660 600' Core average exposure at beginning of cycle 3, GWd/Te 10.69 10.482 Boron worth convention Conservative Deterministic
' Fuel vendor analyses contain an arbitrary allowance of 0.01 ok for the difference between 600 ppm and 660 ppm.
2 Fuel vendor analysis is based on projected end of previous cycle conditions, while GSU analysis is based on observed end of previous cycle conditions 12
I l
' EA-C A-91-0001 -S'2 Revision 0 Table G-3.2 Summary of Standby Liquir' Control System Shutdown Margin Analysis with Similar Assumptions for RBS Cycle 3 EaratnelcI DSl! Fuel Vendor Critical Eigenvalue 1.00046 1,0002 Borated Eigenvalue' O.97902 0.97890 _
SLCS Shutdown Margin 0.021 0.021 Calculated at 600 ppm boron concentration 13 l
l
)
EA-C A-91-0001-S2 Revision 0 Benchmark Reanalysis G-4. Correction of Published Benchmark Analysis: Quad Cities Local Power Gamma Scan Benchmarks During an independent technical review of the Quad Cities gamma scan benchmark analysis, a logical error was discovered in the computer program used to perform the benchmarks. This error affected only the extraction of CASMO-generated data for bundles containing nonfueled rods; in the benchmark, only assembly GEH002 was affected by the error.
The logical error concerned normalization of the calculated gamma intensity values. The original coding detected the end of local power distribution information by the presence of a zero power value in the data array. The calculated power level for nonfueled rods is zero; hence, the normalization factor for the calculated local power array was truncated at the water rod in the 8x8 bundle evaluations. Resolution of the coding error improved the accuracy of the GEH002 benchmark.s to a level consistent with benchmarks performed by others .
The corrected results are compared with the original results in Table G-4.1. A revised statistical analysis of the entire benchmark is given in Table G-4.2; the overall standard deviation for the local power benchmarks was calculated to bc 3.1 % rather than the 3.6%
reported in Section 6.3 of the base report.
A. Dyszel, K.C. Knoll, J.H. Emmett, E.R. Jebsen, C.R. Lehmann, A.J. Roscioli, R.M.
Rose, J.P. Spadaro, and W.J. Weadon, " Qualification of Steady State Core Physics Methods for BWR Design and Analysis," PI NF-87-001-A, Pennsylvania Power & Light Company (1987).
14 1
I
EA-CA-91-0001-S2 Revision 0 Table G-4.1 Quad Cities Gamma Scan Local Power Benchmark Results, PMS Error for Assembly GEH002 Elevation Reported Value Revised Value p 15.0 0.03031 0.024 21.0 0.04285 0.025 51.0 0.0552 0.027 56.0 0.04895 0.028 87.0 0.03235 0.030 93.0 0,03287 0.031 123.0 0.02655 0.023 129.0 0.02730 0.025 Table G-4.2 Quad Cities Ga.uma Scan Local Power Benchmark Results, Summary of Overall Benchmark Assembly Mpmber of Poinfs Standard Deviation GEH002 434 2.676 %
CX0672 313 3.295 %
GEB159 300 2.900 %
,-. GEB161 69 3.169 %
l CX0214 310 3.561 %
Sample 1426 3.090 %
15 l
l
,. _