ML14330A510
| ML14330A510 | |
| Person / Time | |
|---|---|
| Site: | Palo Verde  |
| Issue date: | 12/04/2014 |
| From: | Balwant Singal Plant Licensing Branch IV |
| To: | Edington R Arizona Public Service Co |
| Singal B | |
| References | |
| TAC MF4169 | |
| Download: ML14330A510 (11) | |
Text
UNITED STATES NUCLEAR REGULATORY COMMISSION WASHINGTON, D.C. 20555-0001 December 4, 2014 Mr. Randall K. Edington Executive Vice President Nuclear/
Chief Nuclear Officer Mail Station 7602 Arizona Public Service Company P.O. Box 52034 Phoenix, AZ 85072-2034
SUBJECT:
PALO VERDE NUCLEAR GENERATING STATION, UNIT 3- REQUEST FOR ADDITIONAL INFORMATION RE: RELIEF REQUEST 52, ALTERNATIVE TO ASME CODE, SECTION XI REQUIREMENTS FOR FLAW EVALUATION, FLAW CHARACTERIZATION, AND SUCCESSIVE EXAMINATIONS (TAC NO. MF4169)
Dear Mr. Edington:
By letter dated May 16, 2014 (Agencywide Documents Access and Management System (ADAMS) Accession No. ML14142A029), Arizona Public Service Company (APS) submitted a request to the U.S. Nuclear Regulatory Commission (NRC) for the use of alternatives to the American Society of Mechanical Engineers Boiler and Pressure Vessel Code, Section XI requirements related to axial flaw indications identified in a Palo Verde Nuclear Generating Station, Unit 3, reactor vessel bottom mounted instrument nozzle. The NRC staff has reviewed the information provided in your application and determined that additional information, as identified in the enclosure to this letter, is required to complete review of your application.
A draft copy of the enclosed request for additional information (RAI) was provided to Mr. Thomas N. Weber of your staff via e-mail on November 14, 2014. A clarification call was held on November 20, 2014, to discuss the RAI request. Mr. Thomas N. Weber agreed that APS will provide response to the requested information by January 16, 2015.
R. Edington If you have any questions, please contact me at (301) 415-3016 or via e-mail at Balwant. Singal@nrc.gov.
Sincerely, b~~):._S6,~~
Balwant K. Singal, Senior Project Manager Plant Licensing Branch IV-1 Division of Operating Reactor Licensing Office of Nuclear Reactor Regulation Docket No. STN 50-530
Enclosure:
Request for Additional Information cc w/encl: Distribution via Listserv
REQUEST FOR ADDITIONAL INFORMATION RELIEF REQUEST 52 ARIZONA PUBLIC SERVICE COMPANY PALO VERDE NUCLEAR GENERATING STATION, UNIT 3 DOCKET NO. STN 50-530 By letter dated May 16, 2014 (Agencywide Documents Access and Management System (ADAMS) Accession No. ML14142A029), Arizona Public Service Company (APS, the licensee) submitted Relief Request 52 to the U.S. Nuclear Regulatory Commission (NRC) for the use of alternatives to the American Society of Mechanical Engineers Boiler and Pressure Vessel Code, (ASME Code), Section XI requirements related to axial flaw indications identified in a Palo Verde Nuclear Generating Station (PVNGS), Unit 3, reactor vessel bottom-mounted instrument (BMI) nozzle. The NRC staff has reviewed the information provided in the licensee's application and determined that the following additional information is required to complete its review of the application.
Relief Request 52. Item 6, "Duration of Proposed Alternative
RAI-1
As indicated in Relief Request 52, Item 6, the requested duration of the proposed alternative is for the remainder of the Unit 3 licensed operating life, which expires on November 25, 2047.
The NRC approves relief requests on an interval-by-interval basis (1 0-year in service inspection (lSI) interval). Therefore, please modify the requested duration of the proposed alternative of Relief Request 52 to indicate "the remaining life of the third inservice inspection interval" or justify the current description.
Attachments 1 to 6. Supporting Analyses
RAI-2
For each attachment supporting Relief Request 52, please summarize the major areas that were different from the previous submittals supporting relief for one operating cycle. , "ASME Section XI End of Life Analysis of PVNGS Unit 3 RV BMI Nozzle Repair"
RAI-3
Section 2.1, "Stress Intensity Factor Solution," of Attachment 1 of Relief Request 52 proposed an extrapolation rule for estimation of the stress intensity factor (SIF or K1) associated with a flaw size larger than the largest flaw in the finite element method (FEM) model of the PVNGS, Unit 3 bottom head and BMI nozzle. Please discuss the extent of using this extrapolation rule (e.g., 30 percent of the crack front profiles used the extrapolation rule to obtain their K1 values) to demonstrate that the calculated crack growth based on the flK1 values are still valid.
Enclosure
RAI-4
Section 2.4.1, "Screening Criteria," of Attachment 1 of Relief Request 52 mentioned use of the failure mode criteria (linear elastic fracture mechanics (LEFM), elastic plastic fracture mechanics (EPFM), and limit load analysis) of the ASME Code, Section XI, Appendix C, "Evaluation of Flaws in Piping," in the current reactor pressure vessel (RPV) application. Applying the failure mode criteria for piping to RPV does not appear to be justified. The NRC staff determined the criteria separating the LEFM and EPFM regions in ASME Code Case N-749, "Alternative Acceptance Criteria for Flaws in Ferritic Steel Components Operating in Upper Shelf Temperature Range," with the NRC staff's modification acceptable. The NRC staff's position on use of ASME Code Case N-749 is provided in the Attachment for your information. Please address the impact of using the proposed Appendix C criteria instead of ASME Code Case N-749 with the NRC staff's modification to the results of this ASME Code, Section XI end-of-life evaluation.
RAI-5
Section 4.4.2, "Applied Stresses," of Attachment 1 of Relief Request 52 states, in part, that "Residual stresses are obtained from the 3-D weld residual stress calculation documented in Reference [16]." This is different from the approach that was used to support the one-operating-cycle relief, where the residual stresses are inseparable from the combined stresses based on Calculation No. C-7789-00-2, Revision 1, "Palo Verde Bottom Head Instrumentation Nozzle Stress Analysis," by Dominion Engineering (Reference [7] of Calculation 32-9212942-001, "Palo Verde Unit 3 BMI Nozzle Repair- Section XI Analysis for Restart"; ADAMS Accession No. ML13317A072). Please address the following:
- Since operating stresses for the prior and the current analyses are likely to stay the same, the difference in the combined stresses of the two cases reflects the difference in their residual stresses. In light of this observation, please discuss the difference in residual stresses between the prior and current analyses and demonstrate why the current analyses is better representation of the "actual" residual stresses.
- The applied stress intensity factors (SIF) due to residual stresses for four crack fronts are shown in Figure A-1, "SIFs for Uphill Side- Welding Residual Stress."
Please provide the corresponding residual stress distributions for these four crack fronts and demonstrate, based on engineering fundamentals, that the residual stress distribution could vary its shape from one crack front to another as suggested by Figure A-1. . "Corrosion Evaluation for Palo Verde Unit 3 Reactor Vessel Bottom Mounted Instrument Nozzle Modification"
RAI-6
Section 2.0, "Assumptions," of Attachment 2 of Relief Request 52 lists four assumptions used in the corrosion evaluation, which require plant-specific verification. Please provide your verification based on PVNGS plant-specific information.
Attachment 3, "Weld Residual Stress Analysis for PVNGS3 RV BMI Nozzle Repair"
RAI-7
Section 3.3, "Modeling Simplifications," of Attachment 3, item 3 of Relief Request 52 states that surrounding penetrations are far enough away from the modeled BMI nozzle to preclude interaction. Please demonstrate, based on testing/experimental or analytical information, that the spacing distance from the next nearest penetration is enough to preclude interaction.
RAI-8
Section 4.2, "Materials," of Attachment 3 of Relief Request 52 states that ASME SA-508 material is used in lieu of ASME SA-533 Grade B Class 1. Based on the value of the modulus of elasticity at room temperature given in Table 4-3, "Room Temperature Thermal and Mechanical Properties," the ASME SA-508 class appears to be Class 2. There have been revisions to ASME SA-508 Class 2 properties (ASME SA-508 Class 2 is now currently ASME SA-508 Grade 2 Class 1) after the construction code of record for the subject component. This revision lists different values of elastic modulus (E) and coefficient of thermal expansion (a).
Please assess the impact of these differences on the resulting stresses.
RAI-9
Section 4.5.2, "Structural Boundary Conditions," of Attachment 3 of Relief Request 52 mentions the application of post-weld heat treatment (PWHT) of the RVBH (reactor vessel bottom head).
Please provide a description of modelling the PWHT process to demonstrate that the modelling reflects what the BMI penetration nozzle #3 actually experienced. . "ASME Section Ill End of Life Analysis of PVNGS Unit 3 RV BMI Nozzle Repair"
RAI-10
Section 4.3.1, "Design Condition," of Attachment 4 of Relief Request 52 states the design conditions were simulated on the model by applying a uniform temperature with no differential thermal growth throughout the model. The term "no differential thermal growth" needs clarification. In the actual situation, if metals of different coefficients of thermal expansion are brought up to the same temperature, there will be differential thermal growth and thermal stresses.
RAI-11
Section 4.3.3.1, "Repair Nozzle", of Attachment 4 of Relief Request 52, states, in part, that "External Loads are defined at the juncture between the nozzle and the outside surface of the RVBH ... " Since the FEM model includes a portion of the repair nozzle that extends out some distance from the outside surface of the RVBH, please address whether the external loads have been adjusted by the appropriate moment arm, which is the distance the repair nozzle extends out from the outside surface of the RVBH.
RAI-12
Section 4.4.1.2, "Structural Boundary Conditions," of Attachment 4 of Relief Request 52, states, in part, that "Symmetric boundary conditions are applied to the 3 faces corresponding to the cut-off sections of the model." The NRC staff is of the opinion that the surface of geometric symmetry is a preferred choice for the FEM model symmetry. However, symmetric boundary conditions may not necessarily be appropriate to the other two faces, especially the curved surface of the FEM model. Please justify your assumption. This justification should be expanded to include discussion of the assumed symmetric boundary conditions of the FEM model for residual stresses discussed in Section 4.5.2, "Structural Boundary Conditions," of of Relief Request 52.
RAI-13
Section 6.2, "Thermal Analysis," of Att~chment 4 of Relief Request 52, states that bulk fluid temperatures correspond to RPV inlet temperatures. Please explain why the temperatures at the RPV inlet (near top of reactor vessel) are applicable at the bottom of the RPV.
RAI-14
Section 6.2, "Thermal Analysis," of Attachment 4 of Relief Request 52, states that the selection process for time points of maximum and minimum temperature gradients is based on temperature gradients for certain paths as shown in Figure 6-3. In the opinion of NRC staff, an ANSYS macro may be used to consider all the time points in the thermal analysis and use those same time points to be read in the subsequent stress analysis. Selecting time points based on the critical paths in Figure 6-3 involves some judgment, and, therefore, may not capture time points of maximum/minimum thermal stresses. Please justify that selecting the time points based on temperature gradients in Figure 6-3 is at least adequate enough to capture the times of maximum/minimum thermal stresses as compared to a macro-based procedure for the time-selection process.
RAI-15
Section 7.2, "Primary plus Secondary Stress Intensity Range," of Attachment 4 of Relief Request 52, states that a post-processing routine is used to convert component stresses into stress intensities (SI) and Sl ranges. Is this post-processing part of the ANSYS fatigue module mentioned in Section 7.3, "Fatigue Usage Factor Criteria," of Attachment 4? If so, please provide details of internal verification performed for this ANSYS fatigue module to ensure compliance with the requirements of Title 10 of the Code of Federal Regulations (1 0 CFR),
Part 50, Appendix B, "Quality Assurance Criteria for Nuclear Power Plants and Fuel Reprocessing Plants."
RAI-16
Table 7-2, "P+Q Membrane plus Bending Sl Ranges," of Attachment 4 of Relief Request 52, Section 7.2, "Primary plus Secondary Stress Intensity Range," presents the membrane plus bending (P+Q) Stress Intensity (SI) range for HDPath5 shown in Figure 7-1, "Path Lines on Reactor Vessel Bottom Head." The NRC staff notes that this path does not have the highest
P+Q Sl range, but has the highest cumulative fatigue usage factor (CFUF) as shown in Table 7-4, "Cumulative Fatigue Usage Factors." This observation is contrary to those indicated by WDPath1 and NZPath3, where both have the highest values of the P+Q ranges and CFUF in their respective path groupings. Please justify your approach.
RAI-17 of Relief Request 52 does not mention thermal stress ratchet. Did the analyses supporting nozzle repair met the requirement of NB-3222.5, "Thermal Stress Ratchet," of ASME Code, Section Ill? , "Palo Verde Unit 3 - BMI Nozzle Crack Growth Analysis"
RAI-18
Section 5.4, "Growth of Single Edge Notched Plate Crack," of Attachment 5 of Relief Request 52, states that "Review of the results indicates that by path P23 the residual stresses have decayed and thus the stresses will be constant and equal to those at path P23 above that point." However, stresses in Table 5-3 decrease constantly along crack growth beyond P23.
Please explain the inconsistency. , "Natural Frequency and Structural Integrity Analysis for PVNGS3 RV BMI Nozzle Repair"
RAI-19
Figure 6-4, "Remnant Nozzle First Mode," of Attachment 6, Section 6.2, "Natural Frequency Analysis," of Relief Request 52 shows the mode shape of the fundamental mode of the remnant BMI nozzle. The dynamic model has a cut-out (removal of FEM elements) to account for the actual cracks on the nozzle and weld. Similar to a cantilever beam, a cut-out (see Figure 6-4) close to the root of the nozzle will lower its first natural frequency significantly. The NRC staff considers that detaching the nodes without removing elements (i.e., the cut-out) to simulate the crack may be more realistic. Therefore, please explain that an FEM model with detached nodes without cut-out will not increase the first natural frequency of the remnant BMI nozzle to a value close to the reported high pump exciting frequency (the higher of the reported two frequencies).
Attachment:
January 23, 2013 EVIB Determination of Code Case N749.
Vessels and Internals Integrity Branch's Determination on ASME Code Case N-749 "Alternative Acceptance Criteria for Flaws in Ferritic Steel Components Operating in Upper Shelf Temperature Range" January 23, 2013 ASME Code Case N-749 proposed to use an elastic plastic fracture mechanics (EPFM) methodology similar to that of the ASME Code, Section XI, Appendix K, "Assessment of Reactor Vessels with Low Upper Shelf Charpy Impact Energy Levels," to evaluate detected flaws in ferrictic steel components operating in the upper shelf temperature range. Vessels and Internals Integrity Branch (EVIB) of Division of Engineering of the Office of Nuclear Reactor Regulation reviewed the code case and proposed to place a condition on use of Code Case N-749 regarding the applicable temperature range of the ferritic components:
Condition Instead of the upper shelf transition temperature Tc defined in the code case, the following shall be used:
Tc =170.4 oF+ 0.814 x RTNoT (in U.S Customary Units), and Tc =73.6 oc + 0.814 x RTNoT (in Sl Units).
Alternatively, the licensee may use a different Tc value if it can be justified by plant-specific Charpy Curves.
With this condition, EVIB accepts Code Case N-749.
The code case states that the proposed methodology is applicable if the metal temperature of the component exceeds the upper shelf transition temperature, Tc, which is defined as nil-ductility reference temperature (RT NoT) plus 105 °F. The justification for this, as documented in the underlying white paper (PVP2012-78190, "Alternative Acceptance Criteria for Flaws in Ferritic Steel Components Operating in the Upper Shelf Temperature Range") is that the ASME Code, Section XI, K1c curve will give a (T- RTNoT) value of 105 oF at K1c of 200 ksi~inch. Defining an upper shelf transition temperature purely based on linear elastic fracture mechanics (LEFM) data is not convincing because it ignored EPFM data and Charpy data and their relationship to the LEFM data. EVIB performed calculations on several randomly selected reactor pressure vessel surveillance materials with high upper shelf energy values and low RT NoT values from three plants and found out that using Tc. as defined in the code case, is non-conservative because at the temperature of RT NoT + 105 oF their Charpy curves show that most of the materials will not reach their respective upper shelf levels.
EVIB's condition is based on a 2006 paper by Marjorie Ericson Kirk and Mark Ericson Kirk, "The Relationship between the Transition and Upper-Shelf Fracture Toughness of Ferritic Steels,"
Fatigue and Fracture of Engineering Materials and Structures, Volume 29, where the upper shelf transition temperature is defined as the intersection of the Wallin Master Curve (LEFM data) and the Kirk's upper shelf master curve (EPFM data). Using Kirk's model is justified because, in addition to its theoretically motivated approach to adopt the temperature-dependent flow behavior of body-centered cubic materials, this model is also supported by numerous LEFM data and 809 EPFM data in the upper shelf region. EVIB found out that the Tc proposed Attachment
in Code Case N-749 is conservative based on the intersection of mean curves of the two sets of data. However, for evaluating detected flaws in a ferritic component using the EPFM approach, actual or bounding properties (on the conservative side) should be used instead of mean material properties. Therefore, EVIB, with the assistance from Component Integrity Branch of Division of Engineering of Office of Regulatory Research, used the 5th percentile/95th percentile curves of the two sets of data to define the intersections, or the upper shelf transition temperatures. This resulted in a Tc as a linear function of RT Nor:
Tc =170.4 oF+ 0.814 x RTNoT (in U.S Customary Units), and Tc =73.6 oc + 0.814 x RTNDT (in Sl Units).
In conclusion, the condition that the NRC proposes to impose on Code Case N-749 is to use the above two equations, instead of the T c proposed by the code case:
Tc = RTNor + 105 oF (in U.S Customary Units), and Tc =RTNDT + 58.3 oc (in Sl Units).
Alternatively, the licensee may use a different Tc value if it can be justified by plant-specific Charpy Curves.
For comparison, the proposed Tc, expressed in (Tc- RT Nor), is plotted in the attached figure, along with the code case (Tc - RT NDT) and the one based also on the Mark's model but used the mean LEFM and EPFM data. It showed that the code case (Tc- RTNor) bounds the Mark's curve based on mean LEFM and EPFM data, but not the proposed curve which is based on the 5th percentile/95th percentile LEFM and EPFM data.
100 ~----...,.------....,.., - Mean Data
- Proposed 80
-u
- .- 6o 1-;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;~;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;::;;:s.;;:;;~;;;;;;;;;;;;;;;==---r
....c z
1-
~I 40 +-----~--~------~--------~
1-u 20 0
-100 0 100 200 RTNor [oc]
ML14330A510 *via email OFFICE NRR/DORLILPL4-1/PM NRR/DORLILPL4-1/PM NRR/DORLILPL4-1/LA NAME MWatford BSingal JBurkhardt DATE 12/03/2014 12/04/2014 12/03/2014 OFFICE NRR/DE/CVI 8/BC NRR/DORLILPL4-1/BC(A) NRR/DORLILPL4-1/PM NAME SRosenberg* EOesterle BSingal DATE 10/27/2014 12/04/2014 12/04/2014