ML20196E308

Summary of 981112 Meeting with Nei,Epri & Industry Re Technical Issues Involving Guidance in Draft Reg Guide DG-1074, Steam Generator Tube Integrity. Meeting Agenda & Slides Provided at Meeting Encl

ML20196E308
Person / Time
Issue date:	12/01/1998
From:	Tim Reed NRC (Affiliation Not Assigned)
To:	Sullivan E NRC (Affiliation Not Assigned)
References
TASK-*****, TASK-RE NUDOCS 9812030116
Download: ML20196E308 (44)

Text

_ _ . . - -

p%g t UNITED STATES

- [q j t

NUCLEAR REGULATORY COMMISSION -

WASHINGTON, D.C. was "1 ,- ..

k # December 1, 1998

./

MEMORANDUM TO: Edmund Sullivan, Acting Chief h; F -

Materials and Chemical Engineering Branch Division of Engineering Office of Nuclear Reactor Regulation THRU: f mmett E Murphy, Acting Chief

Y Section B Materials and Chemical Engineering Branch Division of Engineering Office of Nuclear Reactor Regulation FROM

Timothy A. Reed, Senior Project Manager Materials and Chemical Engineering Brane Division of Engineering Office of Nuclear Reactor Regulation

SUMMARY

OF THE NOVEMBER 12,1998 TECHNICAL MEETING WITH NEl/EPRl/ INDUSTRY TO DISCUSS ISSUES INVOLVING DG-1074 On November 12,1998, the NRC staff met with representatives of Nuclear Energy Institute (NEI), Electric Power Research Institute (EPRI) and industry to discuss technical issues involving the guidance in draft regulatory guide DG-1074 " Steam Generator Tube Integrity."

Meeting attendees are identified in Attachment 1. The meeting agenda and slides are provided as Attachment 2.

The meeting began with NEl presenting a proposed regulatory framework for implementing NEl 97-06 which involves a technical specification (TS) amendment that removes steam generator (SG) TS information to the updated final safety analysis report (UFSAR) and other licensee controlled documents. The remainder of the meeting consisted of presentations and discussion of technicalissues previously identified by the staff at the October 7,1998 technical meeting.

//

//

Industry provided feedback on each of the previously identified technicalissues where there exist differences between the DG-1074 guidance and the guidance supporting implementation of 8/p,3 NEL 97-06.

At the conclusion of the meeting, NElindicated they would formalize their draft response on the technicalissues. it was also agreed that it would be beneficial to continue the technical discussion and that the next meeting could either be when industry has completed its Steam CONTACT: T. Reed, EMCB/DE 415-1462

~

9812030116 981201 PDR REVGP ERGNUMRC

) ' l AIP1

@o y olM naypyd,

'n f 2 l Generator Tube Integrity Assessment Guidelines or when industry develops a draft submittal l . supporting implementation of NEl 97-06 via a TS amendment /UFSAR changes.

' Attachments: As stated CONTACT: T. Reed, EMCB/DE 415-1462 l DISTRIBUTION: PUBLIC . PDR EMCB RF SMagruder JStrosnider GClainas OGC Document Name: G:\ REED \NOV1298. MIN WDCATE W BOX: *C"= COPY WiO ATTACHMENTIENCLOSURE, "E"= COPY WIATT/ ENCL, *N"=NO COPY OFFICE EMCB:DE , h EMG7pE, #

NAME Tree [f y DATE 11 19l/98 OFFICIA;. RECORD COPY t

! 030070

b

. 2 Generator Tube Integrity Assessment Guidelines or when industry develops a draft submittal supporting implementation of NEl 97-06 via a TS amendment /UFSAR changes.

Attachments: As stated l

?

NEl/EPRl/ INDUSTRY TECHNICAL MEETING SG TUBE INTEGRITY:DG-1074 ISSUES NOVEMBER 12,1998

, LIST OF ATTENDEES

~'

l l NAME ORG/ POSITION

1. Tim Reed NRC/NRR/DE/EMCB

2. Clive Callaway NEl

3. Mati Merilo EPRI

l. 4. Ron Gamble Sartrex Corp

5. Dave Modeen NEl

! 6.- Rich Barrett NRC/NRR/DSSA/SPSB

-7. Steve Brown APTEC

8. David Ayres ABB l

9. Don Streinz ABB

10. Emmett Murphy NRC/NRR/DE/EMCB i

11. Steve Long NRC/NRR/DSSA/SPSB

12. Joe Muscara NRC/RES

13. Davis Goetcheus TVA

14. Ted Sullivan NRC/NRR/DE/EMCB l 15. Helen Cothran TVA

16. Phil Rush NRC/NRR/DE/EMCB

17. Bill Cullen - Westinghouse

18. Gary Hanry EPRI

19. Mohamad Behravesh EPRI

20. Daniel Pratt ABB 21 Greg Kammerdeiner Dusquesne Light

22. - Anthony Saccavino BGE

23. Fred Anderson Tetra Engineering

24. Dan Mayes Duke Power

25. Rick Mullins Southern Co.

l l i

I ATTACHMENT 1 i

i f

I l.

c .

l NRC/NEI Task Force Meeting I- November 12,1998 Purposes

1. Industry to provide responses to issues presented at the NRC/NEI Task Force meeting October 7,1998 l 2. Discuss appropriate methodology for calculating performance criteria

3. Discuss ISI validation / qualification  !

Agenda

! 9:00 a.m. Introduction / Purpose NRC/ Industry  ;

j. 9:15 a.m. Industry's Proposed Regulatory Dave Modeen Framework 10:00 a.m. Feedback on Issues StructuralIntegrity Performance Kevin Sweeney Criteria Accident-Induced Leakage Rick Mullins .

Performance Criteria

, Noon 1:00 p.m. NDE Requirements Dan Mayes 2:00 p.m. Balance of feedback on fssues Industry

, presented on October 7,1998, Mtg '

l l

2:30 p.m. Discussion NRC/ Industry 3:00 p.m. Meeting Adjourned i .

t i

ATTACHMENT 2 l

j ' Draft *

l ,

Appropriate Regulatory Framework for Implementing NEI 97-06

]

i NRC's Proposal ,

i Option 1: i l

= Maintain existing TS  :

-=

Utilities commit to implementing industry guidelines endorsed by NRC l

Option 2: i

. Replace existing TS with new TS to include.

i - performance criteria j - condition monitoring and operational assessment level of conservatism must be addressed  ;

}

- . . repair criteria approved repair metht.!s other than plugging i

• Utilities commit to implementing industry guidelines endorsed by NRC  :

?

Industry Pronosal: Technical Specification Revision  ;

1. Revise technical specifications,( e.g., Improved Standard Technical Specifications, ISTS) Section 3.4, Reactor Coolant System (RCS) to incorporate a l 150-gpd primary to-secondary leakage limit through any one steam generator.

I

2. Delete the steam generator program from the technical specifications ( e.g., l ISTS) Section 5.5.9, Steam Generator SG Tube Integrity Program.  ;

1

. . Dr2ft Nov 12,1998 Pege 2 Industry Pronosal: Final Safety Analysis Report Revision Performance Criteria

. Applicability to Methodology

. References Section 2.," Performance Criteria" & Section 3.3 " Tube

, Integrity Assessment."

s a

4 4

6 a

4 4

l 1

6

Draft Nov 12,1998 -

• Page 3 ,

TechnicalIssues

1. NRC Sta// Comments: Structural Performance Criteria The structural integrity criteria should be revised consistent with DG-1074.

Specifically, 4

All tubes shall retain safety factors consistent with stress limit criteria of ASME Code, Section HI, for all service level loadings , consistent with design and licensing basis Factor of 3 criterion for NOP and 1.4 criterion for accident conditions should be specifically spelled out, to ensure consistency ofinterpretation ,

t

- NEI SG Workins Groun Resnonse:

NEI 97-06 currently states that the EPRI Steam Generator Tube Integrity Assessment Guideline will provide evaluation methods that will ensure "... safety factor margins implicit in the stress limit criteria of the American Society of

- Mechanical Engineers (ASME) Code."

r

! In comparing the two (2) statements, certainly " implicit" is as robust as " consistent" '

e and Working Group sees no need to call out ASME Section III as the reference i application. In fact, it is ASME Section XI that applies to degraded components.

ASME Section XI is consistent with design and licensing basis per the Technical Speci6 cations and compliance is required per 10CFR50.55a. ASME Section XI does ,

specify criteria and safety factors for all service levelloadings.

For clarity the performance criteria will be revise to be consistent with current regulations: -

Ensure steam generator tubes will maintain adequate margin against burst' under normal andpostulated accident conditions concurrent with a safe shutdown earthauakefor the operating cycle.

The Working Group also sees no reason to specify safety factors or con 6dence levels in NEI 97-06. The ASME Code, as interpreted by the EPRI Guidelines, is sufficient l for defining application of these safety factors. This position is consistent with other interfaces with the ASME Code. Code requirements with respect to NDE inspection are not included in NEI 97-06. These requirements are instead, i 1 A tube rupture or burst is a gross failure of the tube such that the formation of a primary-to-secondary opening with an area affiliated to that of a double-ended guillotine breakoccurs.

I-r

. . Draft Nov 12,1998 Page 4 interpreted and supplemented by the EPRI PWR Steam Generator Examination Guideline. As such, the Working Group sees no reason to spell out these details or restrict users against future changes to the ASME Code by including these values in NEI 97-06.

For clari5 cation purposes, we will revise NEI 97-06 to include in Section 2,

" Performance Criteria" and Section 3.3 " Tube Integrity Assessments"to include EPRI Steam Generator Integrity Assessment Guideline (Reference 6) offers guidance for the evaluation methods, margin, and uncertainty considerations used to determine tube integrity such as to provide a conservative assessment of the condition of the tubing relative to the performance criteria.

2. NRC Staff Comments:

Definition of tube rupture and burst should be revised consistent with DG- ,

1074. l NEI SG Workinz Groun Resnonse: i The current wording in NEI 97-06 is an interpretation of regulatory documents that i define " gross failure" as either burst or rupture. While DG-1074 attempts to separate the two terms with regard to structural failure versus leakage, the Working Group sees no benefit in defining the two as individual events. In fact, the proposed definitions offered in DG 1074 are inconsistent with previous regulatory positions (e.g., are " testimony of James Knight before the Atomic Safety and Licensing Board" Docket Nos. 50 282 and 50 306 and Regulatory Guide 1.121) and FSAR definitions. It should be noted that Section C.2.1.1 of DG 1074 requires a margin against gross failure of burst, however offers no specific definition for " gross failure" in the context of this requirement. Finally, the definition in DG 1074 appears to be defect ("... crack propagation") and location specific.

Obviously, analytical and test d finitions of burst, specific to the type of defect

.and/or test performed, are needed to provide guidance to the users. This guidance is best provided in the EPRI technical support documents and not in NEI 97 06.

3. NRC Staff Comments:

Alternative criteria may be applied for SGDSM specific applications if reviewed and approved by NRC - represents a change to safety factors in licensing basis and may have potential risk implications NEI SG Workina Groun Resnonse:

Draft Nov 12,1998 . .

Page 5 .

The NEI Working Group interprets this comment in the context of the deterministic performance criteria, and concurs that any performance criteria other than that specified in NEI 97-06 and that are not consistent with current licensing basis (CLB), should be reviewed and approved by the NRC. NEI 97 06 does not imply otherwise.

It should be noted, that if the NRC comments refer to alternate repair criteria (ARC), the Working Group also concurs with the Staff position. Per ASME Code Section XI, IWB-3630, ARC evaluation methodologies are required to be approved by the NRC. It should however be noted that per Section 3.3.1 of NEI 97-06,

... plug-on detection"is not considered an ARC."

For clarification purposes, the following statement will be added to NEI 97-06, Section 3.3.1 Per ASME Section XI, an ARC methodology shall be reviewed and approved by the NRC. Guidance on the required risk analysis may be found in Reference 6 and Regulatory Guide 1.174.

4. NRC Staff Comments:

Licensees should be encouraged to follow the risk-informed guidance in RG 1.1 74.

NEI SG Working Groun Resnonse: I See response to Item 3.

1 NRC Sta// Comments: Probabilistic Structural Performance Criteria Proposed criteria do not allow for unknown defect types and known defect types for which burst probabilities have not been quantified l NEI SG Working Group Response:

The NRC position is based on historical considerations that indicate that ~50% of past tube rupture events were due to unknown defect types. It is the Working Group's position that the defense in depth program elements in NEI 97-06 including, but not limited to, degradation assessment, improved NDE, improved operational leakage criteria, FME control and chemistry improvements indicate that the factor contained in DG-1074 is inappropriately high. A tube rupture causal factor review of NUREG/CR-6365, Steam Generator Failures" indicates that l

l l

'. . Dreft Nov 12,1998 Page 6 most, if not all, of the SGTR events could have been avoided with a steam generator program in compliance with NEI 97-06. Since any allowance for unknown defects is SG design, tubing material, NDE inspection and time-at temperature related, guidance on an appropriate factor is appropriate in the EPRI Guidelines and not NEI 97 06.

Defining burst probabilities for known defects is not considered an issue. The elements of the steam generator program are intended to provide the licensee with sufficient information to assess all known defects. No changes to NEI 97-06 are warranted relative to this comment..

NEI 97 06 will be revised to reflect the following:

a <5 x 10-2 that one or more tubes burst during a design-basis accident,

6. NRC Staff Comments: Accident Leakage Performance Criteria

" Limiting postulated events"is not defined in NEI 97-06. However, NEI 97-06 '

defines " limiting accident"to be that resulting in the largest differential pressure across the SG tubes. This definition does not necessarily result in i most limiting accident in terms of dose consequences or risk. ,

"Limitingpostulated events"should be revised consistent with DG 1074;i.e.,  ;

postulated design basis accidents other than an SGTR.

NEI SG Workina Groun Response: '

NEI 97 06 will be revised to discuss " postulated design basis accidents other than a SGTR" instead of the " limiting postulated events" or " limiting accident."

Ensure that the potentialprimary-to-secondary leak rate during postulated \

design basis accidents other than a SGTR will:

7. NRC Staff Comments: Accident Leakage Performance Criteria Allowing leakage in excess of the LCO operational leakage limits represents a change to theplant licensing basis with potential risk implications.

The accident leakageperformance criteria should be revised consistent with  ;

DG 1074. '

NEI SG Working Group Response: '

Draft Nov 12,1998 -

Page 7 .

True licensing basis is 10 CFR Part 100 and GDC 19, not an assumption, e.g., LCO operationalleakage, in the accident analysis. However, since the leakage may result in risk implications, the use of I gpm best estimate is promising.

DG-1074:

Calculated p-s leakage rate during postulated design basis accidents other than SGTR should not exceed the TS LCO operationalleakage rate limits (in terms of totalleakage from all SGs and leakage from each individual SG).

Alternative criteria may be applied for SGDSM specific applications if reviewed and approved by the NRC.

Alternative accident leakage criteria may be applied to the component of calculated accident leakage associated with implementation of SGDSM programs.

The balance of the calculated accident leakage rate (i.e., leakage rate for defect types not addressed by SGDSM programs) should not exceed the LCO operationalleakage limits.

l l -

As a maximum, the alternative accident leakage criteria should not exceed the value in the licensing basis accident analyses, minus the TS LCO operationalleakage limits.

For plants with flex incorporated in TS, the alternative accident leakage criteria may be determined from the value given in the flex plot as a function of dose equivalent I131 (minus the LCO operational leakage limits).

Licensees are encouraged to follow the risk informed guidance in RG 1.174 when submitting proposed alternative criteria for specific SGDSM program (s) or when proposing to extend the applicability of existing approved criteria to additional SGDSM program (s).

Staff will consider risk when reviewing licensee proposals.

8. NRC Staff Comments: 7hbe Inspections The NEIguidelines should specify that the frequency and level of sampling shallfollow the EPRIguideline approach forperformance based sampling rather than for the prescriptive approach.

. . Dr:ft N:v 12,1998 i Page 8 NEI SG Working Group Response:

NEI 97-06 is not the place for specifying the frequency and level of sampling required. NDE frequency and sampling shall be defined by degradation assessment and confirmed by integrity assessments.

9. NRC Sta// Comments :

The NEIguidelines or EPRI Guidelines should specify guidelines for deviating from the EPRIguidelines.

NEI SG Workina Groun Resnonse:

i The SGMP is developing guidelines for deviating from the EPRI guidelines. Once i complete, these guidelines will be incorporated in NEI 97-06 and/or each guideline document at the next revision of the document.

10. NRC Staff Comments The NEI and EPRI Guidelines should be revised to address the issues and objectives addressed. in the DG-1074 guidelines for NDE validation, i.e.,  ;

quantification of defect detection and sizingperformance of the NDE system (technique and personnel) relative to ground truth expected under actual field conditions. <

For example, NEIguidelines could address the definition and objectives of NDE validation. The EPRIguidelines couldprovide guidelines for validating

{

NDE systems. i i

Approach being developed by TVA/PG&E to quantify NDE detection and sizing performance for PWSCC at TSPs appears promising.

NEl SG Workins Groun Resnonse:

4 Validation of NDE techniques is adequately addressed in the EPRI PWR SG  !

Examination Guidelines Section 6, Ay pendix G and Appendix H. These sections of the NDE Guidelines identify the process by which an eddy current technique is {-

developed and the analyst performance factors associated with the NDE techniques.  ;

Taken as a whole, the process addresses:

. technique performance e analysis performance l

I . .

Droft Nov 12,1998 -

l Page 9 .

l \

• process controls e human performance factors .

. field analysis feedback  !

Appendix H addresses eddy current technique capabilities measured against i l

metallurgically determined flaw depths. Appendix G addresses analyst qualification and analyst flaw measurement variations. Section 6 combines the requirements of both appendices with protocols that maximize inspection effectiveness. The process accomplishes this through the elements identified below.

1. Adequate eddy current technique detection performance is accomplished through Appendix H compliance, which provides :

- a statistically adequate number of flaws in the technique qualification data set (H2.2.2a)

- data sets consisting of pulled tubes or lab flaws that are representative of actual Daws (H2.2.1c&e)

- specific details on hardware and set-up essential variables for eddy l current data acquisition (H.1)

- assessment of extraneous test variables associated with each damage mechanism (H2.2.1d)

- a technique POD (80% at a 90% con 5dence level) applicable to the range of flaws in the data set (H2.3.1, H2.2.2c)

- review of results by qualified personnel (section 6.2.1, H3.5.1)

2. Techniques are veriEed applicable to each site they are to be used at via the site-qualification process. This process verifies that the technique to be used for each degradation mechanism is appropriate by comparing the technique data set and plant data in the following areas:

(section 6.2.4, H2.2.1d)

- tube exterior conditions (dents, geometry changes deposit effects)

- tube conditions (pilgering, noise, signal characteristics, signal amplitude)

3. Analyst capabilities are enhanced and characterized through the requirements of Appendix G. Appendix G specifies personnel training and qualification requirements for SG NDE personnel. Analysts are trained in eddy current theory, SG design , SG tube degradation mechanism and practical data analysis for the degradation mechanisms. They are qualified through the Qualified Data Analyst (QDA) program, which consists of ,

written and practical examinations. The practical exam is a blind test '

containing a statistically signi6 cant number of flawed and non flawed tubes for each degradation mechanism. (App. G)

4. Overall analyst performance is further enhances by utilizing two independent analysts to view all data. This practice statistically improves

. -_ .~ -. ..--- -.--..--- - -.-_-. ._ - - - - .

Draft Nov 12,1998

. Page 10 the capability of the examination analysis process to identify degraded tubes.

In addition, the examination protocol requires any eddy current indication identi6ed by either of the independent analysts that is to be deleted as no defect present must be reviewed and agreed upon by two resolution analysts.

The protocol ensures conservative decision making with regard to flaw identification. (Section 6.3.3) .

5. Site specific performance demonstrations are performed prior to each SG tube eddy current inspection. The SSPD's purpose is to ensure, on a site specific basis, that analysts are familiar with and capable ofidentifying the site's SG degradation mechanisms using the prescribed eddy current techniques . The SSPD, in conjunction with the analysts QDA qualification, provides statistically based confidence that the analysts can effectively identify SG tube degradation. (Section 6.6.1)

6. Analyst feedback on eddy current indications not identified by either one of the two independent analysts is provided during the examination. This process raises the standard of performance during the examination and provides continuous on the job training on a site specific basis. (Section 6.6.1)

7. When depth sizing of cracks is to be used to return tubes with cracks below the technical specification repair limit to service the SG Examination Guidelines provide for adequate quantification of technique capabilities through the following requirements. These requirements are in addition to 4 the items listed above.

an additional number of flaw samples is required to statistically establish the techniques capabilities. (H2.2.3a, H2.3.2) e the flaw specimen data set shall cover the range of Daw depths of interest (H2.2.3d) e the additional sample set and eddy current technique must provide enough data to quantify measurement uncertainty to the degree required by the Condition Monitoring and Operational Assessments.

(H2.2.3d) e the total NDE System measurement uncertainty is used in the Condition Monitoring and Operational Assessments. This accounts for eddy current technique and analyst variability. (sections 5.2,6.1)

8. In support ofits CM & OA, qualification of NDE uncertainty beyond those supplied by the PWR Examination Guidelines, Rev. 5 and how the uncertainties are integrated will be determined on a plant-specific basis.

Draft Nov 12,1998 .

Pagall .

11. NRC Staff Comments De5nition of" active degradation mechanism"in EPRI Guidelines, Appendix F, should be revised consistent with deBnition of " active defect types"given in DG 1074, Definitions.

This is consistent with the objective of being performance based.

NEI SG Working Groun Respount:

Per the EPRI guidelines, an active degradation mechanism is:

A combination or ten or more new indications of degradation (220% TW) and previous indications of degradation which display an average growth rate equal to or greater than 25% of the repair limit per cycle in any one SG Or One or more new or previously identified indications of degradation, inclu 'ing cracks, which display a growth greater than or equal to the repair limi' one cycle.

This defir ' on is used in section 3.3.1, Examination of Tubes, to determine the .

frequency C required steam generator inspections. The dennition is used to determine if an inspection is required at the end of each operating cycle or if inspection is required every other operating cycle.

Section 3.6, Expansion of the Inspection, provides guidance on expansion of the sample. Expansion of the sample is dependent on the prevalence and seriousness of the degradation detected. For volumetric flaws, if a single defective tube (indication exceeds the repair limit), the sample is expanded to a minimum of 20% in the affected steam generator. For cracking, a single defective tube results in an expansion to 100% in the affected SG and a minimum of 20% in the remaining steam generators.

Consequently, the definition of active degradation mechanism is used to determine the frequency of the steam generator inspections, not the expansion of the i inspections.

In the March 30,1998 version of DG 1074, active degradation mechanisms and active defect types are defined to mean that new indications associated with these mechanisms and defect types have been identified during inservice inspection or that previously identified indications associated with these defect types have exhibited growth since the previous inspection of the subject tubes.

This broad, regulatory definition of active degradation mechanism is used primarily

. . Draft Nov 12,1998 Page 12 in Section C.1.1.4, Expanded Inspection Sample, to require the sample expansion for each active defect type identined in the initialinspection.

Consequently, there are two different definitions for active degradation mechanism which are used in two different approaches for inspection of the steam generators that arrive at essentially the same results.

From the industry perspective, there is no beneSt to redefining active degradation mechanism. A redefinition would require a re write of the PWR Steam Generator Examination Guidelines with a new logic path. The result would be increased confusion with a decrease in margin of safety as a result of the confusion.

It should be noted that active degradation mechanism is not used in Section 4 of the PWR Steam Generator Examination Guidelines. Section 4 is the performance based inspection guidelines referenced in question 13 above.

Additionally, use of the DG 1074 could result in AVB wear scars being considered as active degradation mechanisms after the AVBs have been replaced. The apparent growth of the wear scars is the result of NDE uncertainty and the variability of the eddy current measurement between cycles. If the AVB wear scar depth changed by 1%, AVB wear would be an active degradation mechanism even though the source of the degradation, i.e., AVB wear, had been removed.

12. NRC Staff Comments The EPRIsub tierguidelines should be directive guidelines, not non-directive.

Note: NRC staff does not have copy of these sub tier guidelines.

NEI SG Workins Groun Resnonse:

A goal exists to make the sub tier EPRI guidelines directive in nature. However, based on the number of variables in the steam generator arena, e.g., materials, design, plant operating characteristics, water quality, degradation, flexibility must be built into any guidance document. To issue a new document directing how testing or analyses should be conducted without field testing would be detrimental to a utility's ability to successfully manage its steam generator program.

Consequently, non-directive documents will become directive when the SGMP is satisfied that the documents are capable of accomplishing what was intended without causing unnecessary perturbations within the industry.

13. NRC Staff Comments

~

Draft Nov 12,1998 Paga13 .

Formal NRC endorsement of these sub-tier guidelines may not be necessary given development of a satisfactory top tier NEIguideline. However, the staff would like the opportunity to look at these guidelines prior to reaching a position on this point.

NEI SG Workine Groun Resnonse:

When th3 sub tier guidelines are ready to be issued to the industry, copies will be provided to the NRC Staff. Providing copies of documents to the NRC Staff with which the industry is not satisfied typically causes more problems than it addresses.

14. NRC Staff Comments The NEI Guidelines should provide guidance concerning the treatment of uncertainties and variabilities associated with each of the input parameters affecting the outcome of these assessments. The level of conservatism to be achieved with these assessments should be identilied in qualitative or quantitative terms, consistent with staff draft guidance (sample TS, DG-1074).

If NEIguidelines specify level of conservatism in qualitative terms, then the sub tier guidelines should quantify level of conservatism to be achieved, consistent with DG 1074.

NEI SG Working Groun Resnonse:

NEI 97-06 is a high level document not intended to provide detailed guidance. As such, quantitative guidance on the treatment of uncertainties and variabilities are l better left to the more detailed EPRI topical reports. However, a qualitative

! description similar to the following will be added to NEI 97-06:

Analyses should account for all significant uncertainties. Potential significant sources of uncertainty include uncertainties associated with the projected limiting defect or indication size, material properties, and structural model. Conservative assumptions should be employed to account for uncertainties not directly treated in the assessment. Additional guidance on the level of conservatism can be found in l the Steam Generator Tube Integrity Assessment Guidelines.

These assessments shall account for all significant uncertainties such as to provide a conservative assessment of the condition of the tubing relative to the performance criteria.

Specific levels of conservatism will be included in the sub tier EPRI topical reports.

l

=' . .

. . Draft Nov 12,1998 i

. P:ge 14

15. NRC Staff Comments:

) NEIguidelines should state that tube integrity assessments may be based on i NDE sizing defect sizing measurements only if the sizing performance of the NDE system has been validatedper definition in DG 1074for the subject i defect type.

NEI SG Workine Groun Resnonse:

1 j Per Steam Generator Tube Integrity and the PWR Examination Guidelines, technique utilized shall support the requirements of condition monitoring and  ;

operational assessment. Section 6 and Appendices G & H of the EPRI PWR '

l Examination Guidelines are capable of suppling data that quantifies detection and sizing capabilities for eddy current techniques and analysts. Quantification of NDE i uncertanties beyond those supplied by the PWR Steam Generator Examination

) ~ Guidelines and how the uncertanties are integrated will be determined on a plant-j_ specific basis in a manner that supports the plant's integrity assessments.

1 Recent industry NDE and in situ results suggest that this approach is effective in assuring steam generator tube integrity. Consideration will be given to .

standardizing treatment of NDE uncertanties as part of the Tube Integrity

] Guidelines; however, the present approach is considered adequate.

)

i 16. NRC Staff Comments i

l

• NEIguidelines should state that condition monitoring and operational assessment is to beperformed for each active defect type.

4 NEI SG Workine Groun Resnonse:

NEI 97 06 will be revised to require each degradation mechanism known to exist in l a steam generator be assessed.

d

17. NRC Staff Comments:

t NEIguidelines should address need for corrective actions in event condition monitoring indicates performance criteria have been exceeded.

NEI SG Workins Groun Resnonse:

E A qualitative description of possible corrective actions will be added to NEI 97-06.

4 4

4 I

_ . _ _ _ _ ~ _ . _ . - . _ . _ _ _ _ _ _ _ . _ . _ _ . _ . . . _ . _ . _ _ _ .

?

Draft Nov 12,1998 .-

Page 15 ,

18. NRC Sta/f Comments: Tube Repair Limits NEIguidelines should state that the tube repair criteria are contained in the plant TS.

NEI SG Working Group Response:

Tube repair criteria should be contained in the steam generator program.

19. NRC Staff Comments: ,

NEIguidelines should state that " plug on detection" applies when NDE systems are not validated for given defect type.  :

NEI SG Workins Groun Resnonse: ,

The intent of this requirement already exists in NEI 97-06, Section 3.3.1, Repair Limits. The section states in part: ,

"Forplants experiencing a damage form or mechanism for which no depth sizing capability exists, tubes identified with such damage are

" repaired / plugged-on-detection"and integrity should be assessed."

20. NRC Staff Comments:

The TS may be amended to include new ARCS to be implemented as part of SGDSMprograms referenced or described in the TS subject to NRC review '

and approval.

Implementation of new ARCS may have potential risk implications.

Licensees are encouraged to follow the risk informed guidance in RG 1.174  !

when submittingproposed ARCS.

The staff will consider risk when reviewing licensee proposals. l l

NEI SG Workinr Groun Resnonse:

Per ASME Code Section XI, IWB 3630, ARC evaluation methodologies are required i i

to be approved by the NRC. It should however be noted that per Secticn 3.3.1 of NEI 97-06, "... plug on-detection"is not considered an ARC."

~~ '

. . . Drz.ft Nov 12,1998 i Page 16

21. NRC Staff Comments: Tube Repair Methods I

The NEIguidelines should also state that repair methods other than plugging \

shall be in accordance with theplant TS.

NEI SG Workins Groun Resnonse:

Tube repair methods shall be evaluated in accordance with the requirements of f I 50.59 to determine the need for NRC review, and incorporated in steam generator program. l 1

l l

4

iG 1

A Proposed Regulatory Framework for Steam Generators NRC-NEI Task Force Meeting November 12,1998

, Rockville, MD V'

i 1 Framework Objectives a Industry's

. Establish an industry standard applied consistently at all PWR sites l . Adopt performance-based concepts

. Provide flexibility to adapt program i

elements to new information/ techniques V'

l l

l Framework Objectives l l = NRC's (an industry perspective)

. Assure regulations are met l . Establish a consistent industry approach consistent with NRC licensing, inspection, assessment and enforcement policies

. Maintain appropriate level of NRC involvement for actions encroaching on current licensing basis

7' l

l l

Framework Structure &

implementation a Multi-Level Approach

. Regulations l

. Tech Specs /FSAR

. Licensing Commitments

. Plant Procedures a Multiple opportunities for regulatory oversight .

QEI 2

I Regulatory Requirements a 10 CFR 50 Appendix A

= 10 CFR 50 Appendix B

= 10 CFR 50.65

= 10 CFR 50.72 & 50.73 l

l 1

)

'it '  !

1 Technical Specifications  !

. Overhaul consistent with Q50.36 (1995)

. Only one LCO/ Surveillance necessary:

. Primary-Secondary Operational Leakage reduced to 150 gpd per S/G

. All other LCOs/Surveillances relocated to licensee-controlled documents

. Change package development will follow the already established iSTS protocol (travelers) YI 3

i FSAR

. Principal reference document containing design bases and limits on SSC operation required by the license

. New addition: Performance Criteria &

Reporting Requirements

. Clarifies the design bases

. Changes made through Q50.59 with on-site review board oversight

. Content consistent with NEI 98-03 NEl 97-06, Steam Generator Program Guidelines

. Establishes consistent, industry-wide expectation for each S/G program element

. Each PWR. licensee committed to implement via the formal industry position

. Industry will request formal NRC review ofNEI 97-06

. To the degree appropriate, key aspects will be reflected in licensee-specific documents V'

• N,*.

l t

EPRI Guidelines

. Contains industry best practices

. NEI 97-06 specifies:

. Mandatory use of directive documents via i the formalindustry position

+ Formal change protocol (controlled by EPRI Steam Generator Management Project)

. Guidelines and any subsequent revisions will be provided to NRC for .

information YI Plant Procedures

. Contents: Reflect NEI 97-06 & EPRI Guidelines

. NEI 97-06 Section 3, " Licensees shall document the program through plant procedures."

. Changes made through 50.59 with on-site review board oversight

. Opportunity for NRC review for 4

program & procedure compliance I l

5 I

o

Structural Performance Criteria T'

Discussion Agenda

- Industry position Wording in NEl 97-06 for Deterministic Performance Criteria is  ;

satisfactory

- Meets licensing basis - ASME Code

- Revision to include " concurrent with safe shutdown earthquake"

- Safety Factors and Confidence Levels Requirements in DG-1074 overly restrictive for non-ARC applications

- Exceeds intent of ASME Code

- Example for piug-on-detection

- Final industry position on Probabilistic Performance Criteria subject to agreem-ent on implementation of Deterministic Criteria O

• O QE' I ASME Code ASME Section XI not Section 111 is the applicable Code section for degraded components l - Supported by regulation L

ASME Section XI - Applicable paragraphs I

• lWB 3521.1 l - Defines allowable flaw size to be left in service

- 40% for r/t <8.7 IWB 3640 i

- Addresses flaws left in service > IWB 3521.1

- Requires analyses acceptable to NRc

- Wording to be added to NEl 97-06

- Appendix C - Evaluation of Flaws in Austenitic Piping Guidance deemed reasonable per NUREG/CR-6365

- Safety Factors ~3 for Service Level A&B (NOP), - 1.4 for Levels C&D ,

- EPRI GLs adopt 3NOP and 1.4P% (Integrity Assessment & In situ)  !

i I

m '!

i

?

~

Repair Criteria j

Repair Scenario ASME Code NB9748 DG-1074  !

., i a

1 Plug at 40% IWB-3521.1 - Allowable 40% meets original Tech Same as NEl 97-06 (Section Flaws for U-tube Steam Spec (RG 1.121) C.5.1)

Generators-40% of tube Licensee required to  ;

wall for r/t<8.7 -no perform EOC Integrity ,

requirement for flaw Assessment to r'neet .

evaluation , Performance Critetia  :

~

Plug on Detection No specife guidance Same as 40%. NEl 97-06 Not specircally addressed.

states that " plug-on- All repair scenarios require  ;

detection"is not an ARC EOC Integrdy Assessment l .

i Alternate Repair IWB-3640 requires that ladustry position concurs ARCS must be submitted and Criteria flaws left in service with ASME requirements. approved by NRC and i exceeding IWB-3521 shall Only specdic cases supported by risk j be evah:sted by analyses require risk assessment assessment -

acceptable to NRC

t l Operational Assessment l

Deterministic Structural Performance Criteria l

Bases NEl 97-06 DG-1074

' ASME Code industry position that ASME Section XI Performance criteria based on ASME '

Applicability applies to degraded components Code Section ill ,

i i Safety Factors Safety factors implicit in stress limit criteria Safety factors of 3 against gross failure in ASME Code-Code provides service or burst under NOP - NB-3225 of loading safety factors (~3 for normal / upset, Section til for postulated accdonts

~1.4 for emergericylupset) (3 and 1.4 adopted in EPRI GLs ,

Acceptance Criteria ASME Section XI NEl 97-06 Most limiting projected defect satisfy performance criteria with a probability Best Estimate of EPRI SG integnty of 0.95 evaluated at 95% confidence ,

projected limiting Assessment GL (details of projectirig maximum defect flew to limit criteria Mean of St1 Burst up to licensee on premise that NDE is correlation at 3AP validated for sizing)

Uncertainty ASME Section XI NEl 97-06 Assessment should consider all .

significant uncertainties NDE - NR NDE -mean NDE +95%

Material-mean-1a Material-95/95 LTL Material-95/95 LTL '

Growth - mean Growth - mean Growth -account for tail of distribution 1

e

• es.

Summary Operational assessments which compare the best~ estimate limit flaw to the limit criteria

- Meets the intent of the ASME Code

- Conservative application of safety margins since:

NDE uncertainty is considered

- Results in some uncertainty double counting Structural limit defined with 95/95 LTL material properties

- Detailed interpretation and application performance criteria to be included in EPRI GLs and not NEl 97-06 Position consistent with NDE requirements Requirements in DG-1074 overly restrictive

- Lead to restricted operating cycles for reasonable plug-on-detection ,

scenarios using appropriate NDE techniques ABB-CE example l

- . . - . , . - - - - - - - - - . . - - - . . , , . - .._,.~,_...aa.-_.~......__..,.... -

~,.n.,... ---s-.. -m..a.,u-E

~,

• O c

-e _

'o E o m C up e 3

. *g @

&) -

E f/) E i.t.

O4 y D) c c M i.u 'C

.g * . -

c *

.O e ..E E'c ma 6 M wG G' eQ QO at at

- et e

Q sruu,e w wxwwwsassa wn:mmne w Operational Assessment

=

Determine the structural limit as a function of the projected flaw size (not NDE measurement). .

4:

= Determine the end of cycle (EOC) flaw i distribution. S

= Determine the projected limiting flaw for each degradation mode.

. Compare the limiting flaw to the structural limit.

= From the EOC flaw distribution, determine total i

leakage under peak accident conditions. .

I -

x=s s t

I

t  : !  ;: !i i!!!jl ll!ll!!!;I l ,Il!

l 5 t s 4 9 a et

_ e aht mn e

_ h i t s ,t d

_ e i e3i c t necf r n o c i

mee a

'm rhd ek -

_ r et f i

n u a

" t h se L eco di,h c sp e

r x s

d w%

l p4

_ a r h ow5 o9th 1r -

e ea wo i

l t t

_ t u m )b de rP t

s

(

(s t uA

_ c l o r r ab g

_ u r Cea aeut t l oi n t emv a el l

nt a r

n r ywp e t

_ S oai t i _

M tpi bble o

_ l a

gcau i

nebt f

oe oP m.

r) serh Udpt nA Em Re n

e

= Ae t

e ~

w

'?$*fgn:.:,){.*.-'

/

o,  ;._...u..

.#.: in;,.'z Flow Stress Distribution: Room Mg, j _.

Temperature Values for TwoSGs d*'  % $ $? . % $, (_ bk

'/5,%5"U-t j gm g; n ,,

4)-

g (7 '

Oy,, . !$J 1

w,,:..4..

Nh..

0.8 .l f g -

J.y'  !

4 A g a 0.6 .-  ; .--i , _ _.- -

- SG1 1.- a  ;

l i 3 O '

0.4 . . _

i SG2

~

0.2 -- g

-.y , .j i '

i 4

0. -

i 125 130 135 140 145 150 Sy + Su, Ksi ABB MEipp

/ '

am m wee a:;cmme.gwyss w;pmesgesurwg i r

r

~

Structural Limit forOpew. Gonal 4

Assessment AxialFlaws .

j 100 .

. 90 *

$ 80  !

'A-~~~--------

-- 70 gQ.~)g So , ----

-- - - 99th PM e .

95th Percentile OE# 30 - - - -50th Percentile S

.c w

m 20 LL , 10 i i

0 . . , .

0 02 0.4 0.6 0.8 1 1.2 1.4 1A 1A 2 Flaw Length, Inches I

ABB 5

l! l! ' !l!l lil! 1l l t!i!;i l

/

e

?:

e dw ea n

- ~

t yz h 6

/ di dl t

z f eis e

g i o i gl nuiev r odz i h

t s

s w

y t ucds a g

u dhnt dioh cu a d.etic nl f o .

g b nhf t r a i uwr ef o ~ oaiB ri cO

@ r f sb neio nf pd .

g.

g t nw oa uid mor h c

l i

4 i s il t f af ncn t

d n eot a o

r

% co et eami D p 9 i r u g drh ed fnibr i p g net Dn a em i

bs u ot s

_ 5 f

g w d e ud m )i f ooci 0-y yr l a mnnis r

e ui(

t l

y) l l

aez gi d

5 9

p F i

f nhf ob e ns .L ani G

ot,t c n o o bo rd .

t end ee

&p a C h o ci(Ph nr eda i srit nh fs sg O acn t eep ah esu i

t nems obr i

. ;e n

B r osmt Firb n e yimBnd n yueh cmtes i I BW B

_ d i

L = =

= AN e

r

_ a i  :: ii',

- .lI

O sa 4, 0

. 8

- 32 OO l 4(

88 v >  ; l 8

f5 ll"I

+.a -(U o g ..

4 1 g

8 E/

3 22 i

& l

- o G g@ . w a

50 e z

i E  ;

i w .

. 1 Q3  : + .

c$

O

.r

-a

.9s le E 8 LAJ o

j OQ

• s Ez 8 8 .

8 a- -

O .

vi'41 dea wexa qe7

.s.*

i k?f...h,h*Wf.us.yf.m.<. .

hkhh???hhN2???DW+%=%:'

ra%'Th'9p,TgffdV?.fa*$s r.g,.: ) ~ ',

2 v${

t

%. c N*..' .h i ,

t. 5

.: uM;de~h.t 3;werp ;in.gj.

9 7 , f .,3 2 . ,

,p -

' ' * ; ,,,f'

.. ,/ 5.y-:

.y . ..>.,

,'e" 1

l l

at 0 ,

8 \ d 5,-O

+

lll3 8 w .m - t -

0 M .,

44, '

)

e h

C k R E

.G- . s g x a. a (U y 3 1__ t

. m o

o 2

+ ,

1 i3

! l i

i

+

A>-

o -

L.

+

8W . o Q. O m o e o -

Z. 4 0.

v-

. e- o N 6 $

o M you! 'q16uel wex3 ge 1 t

ihNE$$$N$hhkh$hWW'b"'OW'*'"""

~

.;f/e .n A

\ V.u; !% .

,$,. mh g', , ? e:r E

'.{r. c . d. .,. .f.p.y? 4 *. g.e .. , , s 3

.s .'f-g

m.  ;&. .:.e -

%.".?$;};l_;;}&,$';l.3

,.a N }

v.'

, I. ' '

, , , _ _ _ _,,,.-o.. s--~.+ ~ ~ - - ~ ~ ~ ~ ~ '

.- -- w - r --

ummsawaasermemmermams EOC Flaw Distribution .

t a

The EOC flaw distribution consists of the numberand size distribution of. flaws projected to be present at the end of the next operating cycle.

! The number is made up of the flaws present at the l beginning of cycle (BOC) plus the flaws that will initiate during the cycle.

The EOC size is the size at the BOC plus the growth during the cycle.

o i

A Monte Carlo combination of the BOC flaw distribution, the initiating flaw distribution, and the growth rate distribution produces the EOC number of flaws and the - -

size distribution. '

I MS ~ 7 i -

4

. . - '*l i

EOC Limiting Flaw

~ ~

= Each trial in the Monte Carlo process considers '

l the growth of all flaws.

=

l L For each trial there is an EOC flaw which would l produce the lowest burst pressure (the worst flaw for this trial).

= The worst flaws for every trial are ranked in severity and form a distribution of worst flaws.

= The EOC limiting flaw bound is the locus of points bounding the best estimate of the. worst -

E flaw.

4 amm I

Mme g

4._a4,.a iS c. alm d..J A 4Ja mu--e 4M .m mmm .heaM44AeOmg.a w apM 4. m Jum A6 4a. m s,n he Wenumm4WA 4 dup 46- &J 45d 4 F.=M4.le4-.=m'a.Mm.La-..J,.hm,=am._

= a.-,a;maam4e.er m -smwwypmy..m4-e-_-w_m.w_4---amm.-s.w.-. - -- .mm A.

iO

.je M

~

s

4 h gh

• a

• [" g 5 .c. . E 4

! 5

.?$

' c 3b OC

%w k

W m j.og Q

i g m m .e 1 e C @ "

f.;t.

y . .,

m Q i  % U) O "U O i

i. .O c

-* .C

%e j T e Q e i 1 O ,o O e=

(U i

%e "" C d d i

g3 U

3 "3

! i W U s,

e 6@ .C

-~ a. .-

.g .g

==.C=g

!'sn -

wog -

m E %s# 0

~u (U .tEu -(U d (g y j

Q- .-

3 -

i I

a w b 2DeD m .c a N

>&W 2

• u

- W $$

o '

at

} .r. s a 4 i

4 j <t Af

.- y a ---- ,

M7I E Eir L 6 M ill3 HJ2 E B E E E E i

l Comparison of EOC Flaw Bounds ~

' 100 .

go

' E Skuctirallint(

80  !

uw / 95/95)

=: 70

..7--

, . M

,,g- m 60 - - LimEngRuw Bound

• 3c2 50 jy

'~-:

a (BestBt) e 40

• i  !

_..,. 95th percense d O ':s 30 # I. .

. worstflaws I

> g . ,g- 20 ,

e 10 2-

---

99th percenSe W al  !

g,g, 0 1i .

flaws 0 0.20.40.60.8 1 12 1.4 1.6 1.8 2 - - 95thpercenSe d al t

l flaws  !

Flaw Length, inches I

1 l;1o - . O ooST A.. yg 10 t

o, l

Summary i

Operational assessments which compare the best estimate limit 4

flaw to the limit criteria

- Mee' ts the intent of the ASME Code

- Conservative application of safety margins since:

NDE uncertainty is considered

- Results in some uncertainty double counting Structural limit defined with 95/95 LTL material properties

~

- Detailed interpretation and application performance criteria to be included in EPRI GLs and'not NEl 97-06 Position consistent with NDE requirements Regitirements in DG-1074 overly restrictive

- Lead to restricted operating cycles for reasonable plug-on-detection .

scenarios using appropriate NDE techniques ABB-CE example l t

t