ML101380215

License Renewal - April 1 Webinar on Metal Fatigue Arizona Public Service Company Presentation Slides

ML101380215
Person / Time
Site:	Palo Verde
Issue date:	04/01/2010
From:	Arizona Public Service Co
To:	Division of License Renewal
Regner L M, NRR/DLR, 415-1906
References
TAC ME0254, TAC ME0255, TAC ME0256
Download: ML101380215 (30)

Text

PVNGS License Renewal Application Metal Fatigue Topics April 1, 2010

Discussion Topics

• Historical and general background

• Metal Fatigue Monitoring Program

• Cycle Counting Questions

• Class 1 Fatigue TLAA Evaluation Process

• Future actions

• NRC staff questions

Historical and general background

Initial development of PVNGS metal fatigue monitoring program

• Program required by PVNGS Technical Specification 5.5.5

• SYS80-PE-DE Compilation of NSSS Responses to Design Bases Dynamic Events for the System 80 Standard Design established applicable design cycles and their 40-year expected accumulation

• Class I metal fatigue analyses incorporated the SYS80-PE-DE limits and in some cases identified additional items of concern

Initial development of PVNGS metal fatigue monitoring program

• The initial Technical Specification program did not require all UFSAR transients to be monitored

• The UFSAR does not incorporate all of the additional limits (e.g. RCP stud tensioning/de-tensioning) - to be discussed later in this presentation

Program Review - 1995

• PV staff identified that not all UFSAR transients were being counted

• Implemented expanded list of transients in January 1996

• ITS upgrade in 1998 established current requirements

• Not counted data between 85 and 95 was assumed to be an accumulation of 25% of the design allowed transients in all but a few cases

Performance of cycle counting since January 1996

• Actual event counting has been in place since January 1996

• Unit 1, 2 and 3 records have now been reviewed for LRA support

Metal Fatigue Monitoring Program

Current PVNGS Metal Fatigue Monitoring Program

• Current methodology:

- Cycle counting to ensure design assumptions are not exceeded

- Specific component locations are not specified

- Exception: Partial cycle (CBF-PC) methodology is used to track spray nozzle usage

Addressing Cycle Count Assumptions

• LRA development prompted PVNGS staff to revisit the assumptions for 1985 - 1995

• Recounts of all three units were reported in the response to RAI B3.1-4 (see RAI response Table 3.1-4)

- Recount Methodology

• Best source selected (logs, MORs, LERs, WOs, interviews)

• Validated assumptions

Cycle Recount

• Recount Methodology

- Best source selected (logs, MORs, LERs, WOs, interviews)

- Reviewed assumption of 25% accumulated cycles

• Actual data

• Comparison to 95 - 05 operating history

Enhanced Metal Fatigue Monitoring Program

• Required by 10 CFR54.21 (c)(1)(iii)

• Scope - All Class I components

• Methodology:

- Based on component locations

- Adds SBF for high usage locations (not discussed in this presentation)

- Continues cycle counting for Global monitoring

• Used for low 40-year design CUF

• No new industry issues since original design

• Industry experience indicates not a problem location

- Expands CUF monitoring

• CBF-C (CUF based on design cycle)

• CBF-PC (CUF based on partial cycle)

• CBF-EP (CUF based on event pairing)

- Establishes appropriate action limits and corrective actions

Enhanced Program Status

• SBF methodology (to be determined)

• FatiguePro is under evaluation for cycle counting and CBF monitoring

• Current manual cycle counting will continue until a suitable software program has been validated

• Enhanced program will be implemented no later than two years prior to the PEO (LRA commitment #39 as revised in Amendment 9)

Cycle Counting NRC Questions Cycle Counting RAIs and Amendments

• Amendment 3 (annual update)

- Added Unit 3 record review results and revised one transient total to replace an estimate with data

- Revised current program action limit discussion to avoid confusion with the enhanced program limits

- Minor clarifications were included in the AMP OE discussion

Cycle Counting RAIs and Amendments (continued)

• Amendment 9 (result of RAI B3.1-1 thru 8)

- RAI B3.1-4 response provided the individual unit cycle totals from best available sources

- Corrected typographical errors

- Included enhanced monitoring implementation schedule

- Committed to selection of a suitable SBF methodology

- Clarified cycle projection methodology

Cycle Counting RAIs and Amendments (continued)

Most recent cycle count related DRAI set (DRAI 4.3-1 thru 9

- DRAI 4.3-1 Transients projected to exceed 40-year limits

• Projections are conservative and may not be reached

• 10CFR54.21(c)(1)(iii) permits aging management for TLAAs that are not validated or projected for the PEO

• NUREG 1800 paragraph 4.3.3.1.1.3 permits the use of GALL programs for aging management

• NUREG 1801 Vol 2 AMP X.M1 allows use of this program to manage aging and recommends no further evaluation if this program is selected to satisfy 10CFR54.21(c)(1)(iii)

• PVNGS took no exceptions to X.M1 and concludes fatigue reanalyses are not required based on projections . Corrective action will be initiated if needed based on program action limits

Cycle Counting RAIs and Amendments (continued)

• DRAI 4.3-1 LRA Transient projection methodology

- Presented for information only

- Concept of a worst case bounding unit

- Conservative Assumptions

• Highest unit total was used for 85-95 and for 95-05

• Lowest operating years was used for accumulation rate

• 42 years were used for projecting and added to above

• Zero accumulation assumed to be linear times 22/40

- If not expected still assumed 1 event

Cycle Counting RAIs and Amendments (continued)

• DRAI 4.3-2 Global monitoring criteria does not seem consistent

- Exceptions are explained in LRA table notes

• DRAI 4.3-3 Clarify the CBF-PC method

• DRAI 4.3-4 Program action limits (see commitment #39)

• DRAI 4.3-5 Some cycles do not agree with the UFSAR

- Some are derived from UFSAR sections other than 3.9.1

- Some come from specific analyses (e.g. RCP studs)

- Addition of cycles to the UFSAR is under consideration

Cycle Counting RAIs and Amendments (continued)

• DRAI 4.3-6 25% cycle assumption

- Review of best source data was presented in the response to RAI B3.1-4

- Compared assumption to actual data from recount efforts or data from 95-05

- Result:

• Four assumptions of 25% cycle accumulation are being reviewed: Rx Trip, Load Reject, Turbine Trip w/o Rx Trip and Depressurization by MSSV at 100% power

• Remaining 25 % assumptions are conservative

Cycle Counting RAIs and Amendments (continued)

• DRAI 4.3-7 Not all cycles were recounted

- Some had accurate data and did not require recount

- Some were not significant (e.g. plant loading at 5%/min)

- Some could not be recovered so the 25%

assumption was used and validated from 95-05 data

Cycle Counting RAIs and Amendments (continued)

• DRAI 4.3-8 Questioned an incorrect table note (had been corrected in Amendment 9)

• DRAI 4.3-9 Some cycles are not included in the UFSAR (similar to DRAI 4.3-5)

- Some come from specific analyses (e.g. RCP studs)

- Addition of cycles to the UFSAR is under consideration

Class 1 Fatigue TLAA Evaluation Process - LRA Section 4.3.2

Class 1 Fatigue TLAA Evaluation Process

• Identify TLAAs and Affected Components

- SRP Guidance, Industry Experience, CLB Search

• Retrieve Component Current Licensing and Design Basis Documents (CLB and CDB)

• Evaluate and Summarize CDB TLAAs

- Analysis Results - Analyzed Locations, CUFs, etc.

- Effects of Modification, Analysis, and Op. History

- Effects of Differences in Analysis Methods

- Identify Disposition

Identify TLAAs and Affected Components

• SRP Guidance

- NUREG-1800 Sect. 4.3 and Tables for Class 1

- Based on ASME III Subsection NB and NG and Industry Experience

• Industry Experience - Examples:

- Other LRAs

- NRC and NSSS vendor Bulletins, Notices, etc.

• CLB Search

- Word Search of CLB (FSAR, Docketed Reports, Letters, etc.)

- Confirm applicability to Palo Verde

- Identify other plant-specific TLAAs

Retrieve Component Current Licensing and Design Basis Documents (CLB and CDB)

• Research and Summarize History of CLB

- Confirm Current Licensing Basis

• Research Design and Analysis History

- Confirm Current Design Basis

- TLAAs are analyses, are therefore part of the CDB

- TLAAs are usually bases of the CLB (incorporated by reference) rather than included verbatim

Evaluate and Summarize CDB TLAAs

• Analysis Results

- All TLAA Results (e.g., all reported fatigue summaries)

- All Analyzed Locations

- Tabulated for Complex Components

• Effects of Modification, Analysis, and Operating History - Examples:

- Instrument Nozzle Half-Nozzle and MNSA Repairs (Several LRA Sections)

- Pressurizer Surge Line Thermal Stratification (LRA Section 4.3.2.9)

- Unit 2 Head Vent Repair - high part-life CUF, head replaced Fall 2009 (LRA p. 4.3-34)

Evaluate and Summarize CDB TLAAs (Continued)

• Identify Disposition

- Based on Analysis History, Results, and Methods

• 10 CFR 54.21(c)(1)(i) - If simple design life or cycle ratios, etc.,

confirm adequate margin at 60 years

• 10 CFR 54.21(c)(1)(iii) - Used in most other cases

- Requires Aging Management

• 10 CFR 54.21(c)(1)(ii) - Reanalysis performed

Evaluate and Summarize CDB TLAAs (Continued)

• Identify Disposition (Continued)

- 10 CFR 54.21(c)(1)(iii) monitoring methods are described in LRA 4.3.1, summarized by monitored location in Table 4.3-4

- Details of 10 CFR 54.21(c)(1)(iii) monitoring methods are location and analysis-specific. Examples:

• Global - monitor transients

• Cycle-based fatigue (CBF) - monitor transients, possible CUF impacts

• Stress-based fatigue (SBF) - impact of transients require reevaluation of CUF

NRC Staff Questions

• Have we answered the questions and issues stated in this presentation?

• Does the staff have additional questions with respect to cycle counting or the PVNGS approach to metal fatigue?

• Review action items

• Future topics Thank you for your participation