ML100541556
| ML100541556 | |
| Person / Time | |
|---|---|
| Site: | Peach Bottom  |
| Issue date: | 02/23/2010 |
| From: | Plant Licensing Branch 1 |
| To: | Exelon Corp |
| Hughey J, NRR/DORL, 301-415-3204 | |
| References | |
| Download: ML100541556 (6) | |
Text
Enclosure DRAFT REQUEST FOR ADDITIONAL INFORMATION REGARDING LICENSE AMENDMENT REQUEST TO REVISE TECHNICAL SPECIFICATION 4.3.1.1.A CONCERNING K-INFINITY PEACH BOTTOM ATOMIC POWER STATION, UNITS 2 AND 3 DOCKET NOS. 50-277 AND 50-278 By letter to the Nuclear Regulatory Commission (NRC) dated June 25, 20081, as supplemented by letters dated November 6, 20082, March 9, 20093, June 12, 20094, and December 18, 20095, Exelon Generation Company, LLC, (Exelon) submitted a license amendment request (LAR) for Peach Bottom Atomic Power Station, (PBAPS) Units 2 and 3. The LAR seeks to revise Technical Specification (TS) 4.3.1.1.a concerning the spent fuel pool (SFP) k value. The Nuclear Regulatory Commission (NRC) staff has reviewed the LAR and determined that in order for the NRC staff to complete its evaluation, response to the following request for additional information (RAI) questions is required.
Steam Generator Tube Integrity and Chemical Engineering Branch RAI Questions:
In the December 18, 2009, supplement, the licensee is proposing to lower the k-infinity value from 1.362 to 1.270. As part of the analysis to support this change, the licensee performed an analysis to predict the degradation of Boraflex in the spent fuel pool. In order for the NRC staff to have reasonable assurance that this analysis will conservatively predict the degradation of Boraflex, the NRC staff requires additional information.
RAI 26
NET-264-02, Revision 45, describes an algorithm to predict Boraflex degradation. The NRC staff has the following questions regarding the algorithm:
RAI 26.1: Please discuss if the NET-264-03, Revision 1, Appendix A6, algorithm or a similar algorithm, was used in the current, NET-264-02, Revision 4, analysis.
RAI 26.2: Discuss whether NET-264-03, Revision 1, Appendix A, is still applicable to the Licensing Amendment Request.
RAI 26.3: Describe the methodology, conservatisms, and assumptions of the current, NET-264-02, Revision 4, algorithm to predict Boraflex degradation.
1 Agencywide Documents Access and Management System (ADAMS) Accession No. ML081820302.
2 ADAMS Accession No. ML083190840.
3 ADAMS Accession No. ML090690804.
4 ADAMS Accession No. ML091740446.
5 ADAMS Accession No. ML093521435.
6 ADAMS Accession No. ML091740445.
RAI 26.4: Describe the limitations, such as upper limits on % Boraflex loss, of the current, NET-264-02, Revision 4, algorithm.
RAI 26.5: Describe the differences, such as geometric considerations, of the current, NET-264-02, Revision 4, algorithm to the one described in NET-264-03 Revision 1 Appendix A.
RAI 26.6: Discuss how the current, NET-264-02, Revision 4, algorithm was validated.
RAI 27
NET-264-02, Revision 4, mentions that the Peach Bottom Unit 2 RACKLIFE model was verified by BADGER campaigns; however, the verification of the RACKLIFE model for Unit 3 is not mentioned.
RAI 27.1: Please provide all of the BADGER results and RACKLIFE predictions for Unit 2 and Unit 3 racks.
RAI 27.2: Discuss the validation and verification of the RACKLIFE predictions by the BADGER results for the Unit 2 and Unit 3 racks.
RAI 28
The NET-264-02, Revision 4, analyses is based on Unit 2 with the assertion that Unit 2 bounds Unit 3. The NRC staff is unaware of the bases that justify that Unit 2 bounds Unit 3. Please discuss the similarities and differences of the Unit 2 and 3 spent fuel pools, spent fuel pool racks, and Boraflex material.
RAI 29
NET-264-02, Revision 4, correlates the peak panel loss to an average panel boron carbide loss. The NRC staff is uncertain how this correlation was obtained. Please discuss the correlation methodology.
RAI 30
NET-264-02, Revision 4, states, The original (RACKLIFE) model was updated by Exelon every 6 months to reflect actual fuel discharges into the spent fuel racks through 2008. Discuss whether the RACKLIFE model was updated in 2009 and if there are future plans to update the RACKLIFE model at a 6 month frequency. In addition, please describe plans for future BADGER testing for Units 2 and 3 to verify the RACKLIFE predictions.
Reactor Systems Branch Questions:
RAI questions 31 through 35 pertain to the area of the applicability of the validation as discussed in NET-901-02-05, Revision 4.
RAI 31
Is Table 4-1 intended to state the ranges of parameters that the safety analysis fits within? If so, what is the basis for applying the bias and bias uncertainty up to 5 w/o U-235 when the CASMO validation only goes up to 4.31 w/o U-235? Similarly, H/U and EALF ranges in Table 4-1 are not substantiated by the CASMO validation ranges. Provide the basis for the extrapolation.
RAI 32
Similar to RAI 31 above, H/U and EALF ranges in Table 4-1 are not substantiated by the SCALE validation ranges. Provide the basis for the extrapolation.
RAI 33
In Table 4-1, there seems to be an omission for the absorber plate poison loading.
Please provide the missing data.
RAI 34
Why were experiments with soluble boron used when the analyzed system does not contain any soluble boron? What is the effect of soluble boron on the code bias and bias uncertainty? Justify your approach.
RAI 35
Discuss validation gaps (e.g., fission product validation) and, if appropriate, additional margin adopted to cover validation gaps.
RAI questions 36 through 37 pertain to the normality test performed for the CASMO validation as discussed in NET-901-02-05, Revision 4.
RAI 36
Section 5.2 states that the normality tests were performed for the k values. This approach seems to deviate from NUREG-6698 which uses the calculated keff values.
RAI 36.1: Justify your approach.
RAI 37
Section 5.2 states that k values were tested for normality and the data passed all tests. The staff could not confirm the normality of the k values based on the Shapiro-Wilk test described in NUREG-6698. In addition, the normality of the calculated keff values could not be confirmed.
RAI 37.1: Re-evaluate the CASMO bias and bias uncertainty using the appropriate methods to handle non-normal data or justify not doing so.
RAI questions 38 through 39 pertain to the normality test performed for the CASMO validation as discussed in NET-901-02-05, Revision 4.
RAI 38
The 3-diemnsional SCALE bias was determined by comparing SCALE to critical experiments.
The 2-dimensional SCALE models simulate a fictitious set of experiments used to validate CASMO.
RAI 38.1: Justify using the SCALE bias determined in the 3D model to correct 2D SCALE models.
RAI 39
The CASMO validation was performed in two steps. SCALE was validated against 103 critical experiments resulting in a SCALE bias and bias uncertainty. Then 24 of the 103 critical experiments were represented using the 2-dimentional SCALE models which were used determine the CASMO bias and bias uncertainty. The 2-dimentional SCALE representation of the experiments corrects for the SCALE bias, but it does not seem to account for the SCALE bias uncertainty.
RAI 39.1: Please explain how the SCALE bias uncertainty is accounted for in the maximum keff determination. Presently, it is not clear how the analysis links to the critical experiments.
RAI questions 40 through 43 pertain to the normality test performed for the CASMO validation as discussed in NET-901-02-05, Revision 4.
RAI 40
Please discuss what is to be concluded from the low p values of <1E-4 in Tables 5-2 and 5-5. Does this indicate that the association between the response and predictor is statistically significant?
RAI 41
Please explain what is meant by maybe in Table 5-5. Does this indicate that a statistically significant trend exists? If so, justify its impact on the bias.
RAI 42
The NRC staff could not confirm the r2 value for enrichment in Table 5-5. Please confirm that the correct value was determined in the submittal.
RAI 43
The trending analysis data provided in the submittal is not sufficient for the staff to independently verify that no statistically significant trends exist for CASMO. This is especially the case for enrichment and EALF.
RAI 43.1: Please provide additional information to substantiate your claim that no trends exist for CASMO.
RAI questions 44 through 59 pertain to NET-264-02, Revision 4.
RAI 44
Provide 2-dimensional plots of the CASMO-4 and KENO V.a models used in the analysis.
RAI 45
Describe how the asymmetries of the Peach Bottom spent fuel pool (SFP) racks are modeled in CASMO-4.
RAI 46
Describe how the gamma dose to an individual Boraflex panel is determined.
RAI 46.1: Explain the use of the end of cycle relative power in calculating the gamma dose.
RAI 46.2: Explain why the use of a weighted average end of cycle relative power is appropriate versus a bounding end of cycle relative power.
RAI 46.3: What is the basis for the gamma source term?
RAI 46.4: The analysis assumes a 0.8 weighted average end of cycle relative power; describe the effect a higher weighted average end of cycle relative power would have on the analysis.
RAI 47
How are the KENO bias and bias uncertainty applied in the methodology for assessing the reactivity effects of Boraflex degradation? Justify the approach used and explain how it ensures the Boraflex degradation prediction meets a at a 95 percent probability, 95 percent confidence level.
RAI 48
Provide the data that is plotted in Figure 4-2.
RAI 49
What trend analyses were performed on the data plotted in Figure 4-2 to reach the conclusion that no non-normal behavior was observed?
RAI 50
Provide plots of the Boraflex degradation being modeled in the data in Figure 4-2 so that it is clear how the Boraflex degradation changes from one case to the next.
RAI 51
What are the Boraflex loading reference points for Table 4-3?
RAI 52
Provide the depletion parameters used in determining the limiting peak reactivity of the lattices evaluated. Justify those depletion parameters and Identify how deviations from those parameters would affect the peak reactivity.
RAI 53
With respect to the determination of the reactivity equivalent fresh fuel enrichment (REFFE) fuel assembly, the cross-section bias and CASMO/KENO geometric bias, please provide the following information:
RAI 53.1: How were the CASMO-4 and KENO biases and bias uncertainties applied?
RAI 53.2: How was xenon treated?
RAI 53.3: How were the lumped fission products in CASMO-4 treated? Please provide the following information concerning those lumped fission products:
i)
What actual fission products are represented in each lumped fission product?
ii) What are the cross sections for each lumped fission product? What are the decay constants for each lumped fission product?
iii) Are there any neutron absorbers represented in the lumped fission products?
What are the cross sections for those neutron absorbers? What are the decay constants for those neutron absorbers?
iv) Are there any neutron sources represented in the lumped fission products?
What are the source terms? What are the decay constants for those neutron sources?
v) What is the basis for the cross section for each lumped fission product?
vi) How do the cross sections of the lumped fission products respond to changes in temperature and spectral hardening?
RAI 53.4: Provide 2-dimensional plots of the CASMO-4 and KENO V.a models used in the analysis.
RAI 53.4: Explain how normalizing KENO to CASMO-4 in determining the REFFE affects the determination of the cross-section bias and CASMO/KENO geometric bias.
RAI 54
Table 5-2 compares CASMO-4 and KENO V.a in a standard cold core geometry (SCCG) configuration which both codes are capable of modeling.
RAI 54.1: Explain the differences in the information contained in the third and fourth columns and the differences in the information contained in the fifth and sixth columns of the table.
RAI 54.2: The number of examples provided is insufficient to draw conclusions; provide additional examples and include depleted fuel.
RAI 55
Table 5-3 compares CASMO-4 and KENO V.a in cold SFP rack geometry. Since CASMO-4 cant model the asymmetries associated with the Peach Bottom SFP rack design, explain what is being done in CASMO-4 to model the SFP rack geometry.
RAI 55.1: Explain the differences in the information contained in the third and fourth columns and the differences in the information contained in the fifth and sixth columns of the table.
RAI 55.2: The number of examples provided is insufficient to draw conclusions, provide additional examples and include depleted fuel.
RAI 56
Explain why the information provided in Table 5-3 doesnt match the information provided in Figure 5-1.
RAI 57
Justify not using the KENO V.a uncertainty in the calculation of Keff as stated on page
- 39.
RAI 58
Please provide the applicable sections of Reference 25, Doub, W.B., "Particle Self-Shielding in Plates Loaded with Spherical Poison Particles," Part B of Section 4.2, Naval Reactors Physics Handbook, Volume 1: Selected Basic Techniques, Naval Reactors, Division of Reactor Development, United States Atomic Energy Commission: Washington, D.C.; 1964. Justify why this reference is applicable to the Peach Bottom SFP.
RAI 59
Please explain the large change in the fuel assembly miss placement accident from the previously supplied information.
RAI questions 60 through 62 pertain to GNF-0000-0110-5796, Revision 0.
RAI 60
GNF-0000-0110-5796, Revision 0, states, For fuel pool storage evaluations, the 95/95 bias uncertainty is used as a bias applied to the TGBLA in-core peak cold reactivity. This is the effect of an uncertainty when it is the only uncertainty considered. Please provide the justification for not considering the other uncertainties.
RAI 61
GNF-0000-0110-5796 indicates that the TGBLA06 bias and the 95/95 bias uncertainty are extracted from NEDE-240ll-P-A-16, "General Electric Standard Application for Reactor Fuel (GESTAR II)", October 2007 (ADAMS Accession No. ML091340075). However, the NRC staff could not locate the TGBLA06 bias and the 95/95 bias uncertainty in NEDE-240ll-P-A-16. Please provide clarification regarding the location of the TGBLA06 bias and the 95/95 bias uncertainty in a document currently on the NRC docket, or provide the information anew for the NRC staffs review.
RAI 62
GNF-0000-0110-5796-P Figure 3 contains the description of the recommended reference bundle. So that the staff may compare the those bundles with bundles actually in use at Peach Bottom, please provide similar figures for recent Peach Bottom fuel assemblies.