ML20217M209

Comment on Proposed Rule 10CFR50 Re Industry Codes & Stds Amended Requirements. NRC Justification for Avoiding Backfit Analysis,Nonstantial.Backfit Analysis,As Required by Law as Mandatory for Proposed Rule Changes

ML20217M209
Person / Time
Site:	Callaway
Issue date:	03/19/1998
From:	Naslund C UNION ELECTRIC CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
FRN-62FR63892, RULE-PR-50 62FR63892-00040, 62FR63892-40, ULNRC-3768, NUDOCS 9804080040
Download: ML20217M209 (18)

Text

/

Union El:ctric PO Box 620 Cd#d* # Ad"' Fulton, MO 65251 DOCKE'ED USNRC

% APR -6 A8 :01 OFT

[fhj, a March 19,1998 DOCKET PROPOSED NUMBER RULE3,0 -a2 {

U. S. Nuclear Regulatory Commission 4 Attn: Document Control Desk Mail Stop Pl-137 gggg9 YO l l

MI.

d Washington, DC 20555-0001 l ULNRC-3768 UE Gentlemen:

DOCKET NUMBER 50-483 CALLAWAY PLANT UNIT 1 UNION ELECTRIC CO.

FACILITY OPERATING LICENSE NPF-30 PROPOSED REVISION TO 10 CFR PART 50 (RIN 3150-AE26)

This letter submits AmerenUE's comments on the proposed revision to 10 CFR Part 50 dated December 3,1997. Comments are included as an attachment to this letter.

l The proposed rulemaking is extensive, and provides evidence that NRC has devoted substantial resources to its development. The changes proposed in the rulemaking are recognized to be of v'.tal importance to the nuclear industry. The iniportance of these changes is the primary basis for the extensive resources AmerenUE devoted to development of the attached comments.

AmerenUE is prepared to work with NRC to resolve each issue raised. Please contact us with any questions concerning these comments.

l g40 040 9E0319

\'

i 50 62FR63892 PDR , Cf a,rles D. Nasi d Manager, Nuclear Engineering CDN/J WDSH/BRN/ts Attachm nts: I g,p 3 4

,,as,mn en-e cn-a.a

E ULNRC-3i68 iw Page 2 t AmerenUE Comments on NRC-Proposed Changes

'To 10CFR50.55a

1. ~

Background:

No Comments.-

- 2. Section 2.3.1 (ASME Section XI) 5 A) Eng'i neiring Judgement (Page 63694, paragraph 2.3.1.2.1):

1) It is neither possible nor advisable for the ASME Code to directly address every aspect of

. every activity within its scope. Since the Code is not written as a handbook, this proposed regulation impacts virtually all ASME Code activities. The following list identines several activities not speci5cally addressed within the Code, yet requiring engineeringjudgement. This list is not comprehensive-it provides only a few typical examples.

a) Selecting materials b) Determining corrosion allowance c) Determining design pressure and temperature d) Determining the number of cycles e) Determining impact test temperature and impact test acceptance criteria f) Designing nonstandard fittings g)- Using component conngurations other than those depicted in the Code -

- h) Using IWA.2240 to select alternate NDE methods and techniques

2) NRC is required by law to use consensus s:andards such as the ASME Code { Reference Public Law 104-113}.- NRC participates extensively in the ASME Code meetings where these standards are developed. Through this participation, NRC offers comments, suggests changes, and addresses concerns. Whe.1 NRC or any other participant produces credible arguments, Code changes are made. Conversely, non-credible comments are -

dismissed. While the Code does not incorporate all NRC comments, it reflects the ASME's consensus on issues raised. Engineeringjudgement is mandated by the ASME Code. Since engineering judgement is a Code requirement, it is a provision open for

- comment during the Code development process. It is likely, then, that either ASME dismissed NRC concerns regarding Code endorsement of engineeringjudgement, or NRC did not voice these concerns during meetings of the ASME. If these concerns were raised, it is incumbent upon NRC to providejustification beyond that rejected by ASME to substantiate the proposed restrictions. If these concerns were not raised,NRC has not satisGed its obligation to participate in consensus standards development. It is recognized that not every NRC concern can be voiced during Code committee meetings.

An exception to a ftmdamental Code premise such as engineeringjudgement, however, certainly merits consensus committee discussion. Rulemaking is an inappropriate method of changing this fundamental ASME philosophy.

p, 3) Sound engineering judgement is the cornerstone of ASME Code philosophy, upon which rests the safety and reliability of each nuclear plant. Existing Code rules and NRC regulations compel Owners to implement only safe, prudent engineering decisions. The proposed change would impede an Owner's ability to make engineering decisions by transferring part of this decision-making responsibility to NRC. NRC is neither qualified nor staffed to make these er.gineering decisions.

4) This generic restriction will universally regulate engineering judgement. This universal approach is costly and ineffective. Misuse of engineering judgement by

' individuals / organizations should be regulated on a case-by-case basis. NRC should

ULNRC-3768

~

Page 3 initiate specific enforcement actions when engineering judgement is misapplied (i.e.,

when it is used in a manner contrary to that specified in the ASME Code). Restricting the industry's prudent use of engineering judgement is not an effective way to prevent case-specific misuse of engineeringjudgement.

5) Prudent use of engineering judgement has been an ASME Code requirement since publication of the first Code Edition. From that publication forward, no ASME Code Section has contained all details of all permissible engineering configurations. Prudent use of engineering judgement, therefore, is a long-standing ASME Code premise. Past uses of this premise have not raised generic safety concerns, yet the proposed rule implies that future uses will. A logical, but ill-advised evolution of the proposed restriction would be NRC assessment of the restriction's impact on a]l past engineering decisions. If,in fact, NRC never endorsed the use of engineeringjudgement in the past, innumerable " violations" of ASME Code requirements exist.

6) The ASME Code provides minimum requirements. NRC, by impeding the engineering judgement process, will discourage actions that exceed minimum requirements. Owners will be advised to follow these minimum requirements step-by-step-a practice neither intended not advised by the ASME Code. It is difficult to reconcile this likely outcome with NRC's intent to improve plant safety. It is likely that the final result of this change will be contrary to what was intended by the authors.

7) NRC-proposed limitations on " Engineering Judgement" apply to the ASME Code 1992 Addenda through 1996 Addenda. The 1992 Addenda is the current IWE/lWL Code Addenda (ref.10CFR50 RIN 3150-AC93). Several Code interpretations addressing 1992 Addenda IWE/lWL concerns have been generated. Each involves " topics not directly addressed by the ASME Code" and will, under the revised rule, require NRC approval prior to use. The proposed rule imposes an undue hardship in requiring each of 109 facilities to seek NRC approval prior to use of each of these inquiries. This concern also applies to non-lWE/lWL inquiries. Ir. fact, the proposed restriction invalidates many ASME Code interpretations. The proposed restriction, therefore,is inappropriate.

8) Engineering judgement is required to determine if, when, and how engineering judgement is to be used. There is little credence in the idea that one can draw a clear line describing the point at which use of engineeringjudgement becomes unacceptable and/or requires NRC notification.

9) This change will divert future Owner / regulator discussions from safety-significant topics to the topic of whether or not engineeringjudgement was used. Since time scent discussing whether or not engineeringjudgement was used adds little or nothing to plant safety, a reduction in both plant safety and efficiency is likely.

B) Quality Assurance (NQA-1)(Page 63694,63695; par. 2.3.1.2.2): No comments.

C) Class 1 Piping (Page 63895, par. 2.3.1.2.3):

The reference to Code Case N-324 is inaccurate. The correct reference is probably Code 1)

Case N-334.

i

2) With this limitation, NRC cites rejection of three ASME Code Cases: N-322, N-334, and '

N 1981. There is clearly disagreement between the ASME and NRC on the issue of required Class 1 piping nondestructive examinations. It is apparent that NRC has not convinced the ASME consensus body that these examinations are warranted. It is not 1

apparent, however, that the bases documented in the proposed rule are more compelling or credible than those dismissed by the ASME. If the bases in the rule are new, NRC is obligated to initiate Code changes through the consensus process. If the bases are j unchanged, NRC must (in accordance with Public Law 104-113) address why arguments rejected in the consensus process now warrant endorsement through the rulemaking l

process. If NRC bases have not changed since the ASME approved these Code Cases, the proposed restrictions exceed what is prudent. As such, these changes will impose I

ULNRC-3768 Page 4 unwarranted examinations and unnecessary repairs. These unneeded actions expend limited Owner resources that would be better spent performing tasks with true safety signincance.

D) Class 2 Piping (Page 63895, par. 2.3.1.2.4):

1) This limitation mandates that Owners acquire specific NRC approval of programs that identify and specify Class 2 examinations. NRC approval is required "to ensure that safety signincant components...are not exempted from appropriate examination requirements." Owners routinely comply with requirements that contain limitations.

This requirement is similar to others, and is less complex than many. It is an inef0cient use of resources to stipulate this NRC approval. It is a common NRC practice to invoke specine restrictions, then to assure compliance through the normal regulatory process.

Regulatory Guide 1.147 contains numerous examples of this practice. The requirement for individual submittal constitutes an unnecessary regulatory burden, wasting both Owner and NRC resources. This requirement for submittal to NRC is unwarranted and should be dropped.

2) NRC is using the regulatory process to restrict use of published ASME Code Case N-408-3. The ASME Code Case revision process is available to NRC, and should be used to add these restrictions to the Case itself. As currently published, N-408-3 demonstrates that the ASME does not consider these NRC-imposed restrictions warranted, if the ASME consensus process rejected these NRC-proposed restrictions, the regulation should explain why the restrictions remain valid in spite of ASME rejection. If the NRC did not propose these restrictions to the ASME, NRC should initiate action to do so.

Since this requirement exceeds that imposed by the consensus standard, a backfit analysis -

must be used to document the NRC basis for this signincant increase in Owner requirements and costs.

3) NRC endorsement of ASME Code Cases has historically been accomplished through Regulatory Guides, not through 10CFR50.55a. Regulatory Guides are, and should remain the NRC method of Code Case approval. ASME Code Case endorsements should be deleted from the proposed rule and moved to Regulatory Guides.

4) Code Case endorsements in the proposed rule raise several issues. First, will the rule be

' revised as reference Code Cases are revised, annulled, incorporated in;o the Code, or permitted to expire? Second, will NRC require Owners to request approval of Code Cases that are in the rule, but not in Regulatory Guides? Third, will NRC revise l

Regulatory Guides to authorize Code Case endorsement in 10CFR50.55a revisions?

Fourth, if these Cods Cases are eventually incorporated into Regulatory Guides, will Regulatory Guide restrictions supersede rule restrictions?

E) Reconciliation of Quality Requirements (Pages 63895,63896, par. 2.3.1.2.5): No comments.

3. Section 2.3.2 (OM Code and Background)

A) Classification of Safety-Related Valves (Page 63896, par. 2.3.2.1): No comments.

B) Limitations - Quality Assurance (Page 63897, pa r. 2.3.2.4): No comments.

f.

C) Modifications - Stroke Time Testing (Page 63897, par. 2.3.2.5.1): No comments.

4. Section 2.4 (Expedited Implementation)

A)- Appendix Vlli (Page 63897, par. 2.4.1):

1)  : Expedited implementation of the 1995 Edition 1996 Addenda of Appendix Vill includes the expedited implementation of Appendix Vill, Vil, parts of ASME Section 111, and parts of Section XI Subsection IWA. Nojustification is provided for these ASME Section 111 and IWA requirements. As written, the rule ignores the other Code Editions and Addenda indirectly invoked by imposition of Appendix Vill, and therefore avoids

ULNRC-3768 Page 5 consideration of several issues which will result in costs comparable to those of Appendix Vill. Justification for the requirements indirectly imposed by expedited implementation of the 1995 Edition 1996 Addenda of Appendix Vill should be addressed in a backfit analysis.

a) Many Owners are working to the 1989 edition of ASME Section XI and to the original Construction Code (which may be B31.1). As a result, it is unclear which Section lil/XI Editions and Addenda apply, which requirements from these Editions / Addenda apply, and how the user will determine when to apply which Edition and Addenda. NRC must address these issues. For example:

= The 1996 Appendix Vill invokes requirements from the 1996 Addenda of IWA-4000. When repairing a defect detected using 1996 requirements, which Edition and Addenda ofIWA-4000 apply to the repair?

= If the 1996 Addenda ofIWA-4000 applies, where has NRC documented justi6 cation for imposing these new IWA requirements?

= If the 1996 Addenda IWA-4000 applies, do these requirements supersede those in the 1989 Edition? If the 1989 Edition applies, do these requirements supersede those of the 1996 Addenda? Or, does the Owner attempt the impossible task of satisfying both Editions / Addenda at once?

= If the 1989 Edition ofIWA-4000 applies, where has the NRC documented their authorization for Owners to use the 1989 Edition in lieu of 1996 Code requirements?

b) Similar questions apply to Construction Code requirements referenced by the variousSection XI Editions and Addenda. By raising, but not addressing these applicability issues, the proposed Appendix Vlli expedited implementation drastically increases program complexity, c) Expedited implementation is unwarranted (as described below) and should not be mandated. If, however, NRC imposes expedited implementation, its provisions must denne the speci6c Code Editions and Addenda applicable to all inspection, repair, and replacement activities. Regulations invoking these speci6c Editions and Addenda must assess the cost impacts of each and must provide justification for each. j l

B) Appendix VilI (Pages 63899 and 63900, par,2.4.2): l

1) Consensus processes such as that used by ASME, while effective, are imperfect.

Resulting standards, after publication, undergo continuous reassessment and improvement. Newly published standards are " immature" standards; i.e., standards whose workability has not yet been assessed using real-world logistics and capabilities. When real-world assessment identifies problems, the proper course of I j

action is a consensus-process revision of the standard. When industry use of the j

standard is voluntary, the ASME has time to revise the standard without significant i

industry financial hardship. When expedited implementation is enforced, effective recourse in addressing these problems is denied. Real-world assessment has confirmed that Appendix Vill, as published,is unworkable. In an appropriate response to these concerns, high-priority industry /ASME/NRC efforts have been initiated to rectify Appendix Vill problems. Until these problems are resolved,  ;

Appendix Vill is not an effective, workable standard. Since NRC is fully aware of Appendix Vill problems, it is difficult to concede the proposed rule's implication that ,

l Appendix Vill is not only workable, but warrants expedited implementation. Known problems must be addressed before Appendix Vill is mandated. )

i i

2) Several Appendix Vill restrictions proposed in the rule are based on topics addressed by consensus committees during Appendix Vill development. NRC arguments for j these restrictions were not compelling, and consensus committeesjudged them I

l

ULNRC-3768 Page 6 inadequate to warrant Code changes. Argument: dismissed by the ASME have gained no added credibility since this dismissal. NRC must either refute the consensus committee's decision oy adding new technicaljustification (beyond what was offered during the consensus process), or comply with Public Law 104-113 and accept this consensus document as published. If the NRC-proposed restrictions were not discussed during the consensus process, NRC must explain why requirements that did not warrant mention during Code development are now urgent safety concerns.

3) The proposed rule states that examiners quatined to Code Editions prior to 1989

" generally han marginal UT skills" This statement is not supported by referenced test data. Standards have changed to reDect new information and technology. These changes have made upgrade tests more difficult, and this increased difficulty corresponds with an increased test failure rate. It is incorrect to state that an examiner capable of passing the previous tests yet incapable of passing the newer, more stringent tests has " marginal UT skills" This NRC assertion is analogous to creating a new test substantially more difGcult than the upgrade test, administering it to those currently upgraded, and then citing each who fails as a person with marginal skills.

4) The proposed rule arranges F olic comments on the draft Appendix Vill Regulatory Guide into five general categories, then provides NRC responses to each category.

This dilutes and misrepresents public comments. Significant industry resources were expended in cornment development. These categorizations are highly subjective, lending themselves to simplined answers that overlook many of the issues raised during the public comment process. Backfit analysis is necessary to effectively address Appendix Vlli issues. Simplified responses to summarized concerns inaccurately portray public comments.

C) Appendix Vill Personnel Qualification (Page 63899, par 2.4.1.1.1):

1) 10 CFR 50.55a(b)(2)(xvii) increases Appendix Vll-4240 annual training from 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> to 40 hours4.62963e-4 days <br />0.0111 hours <br />6.613757e-5 weeks <br />1.522e-5 months <br />. ASME Subcommittee XI determined, using the consensus process, that 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> is the acceptable minimum annual training requirement. NRC provides inadequate factual evidence to justify increasing this time period. Barring empirical evidence to the contrary, the existing 10-hour minimum training requirement should be retained. In accordance with current requirements, certifications are revoked when examination personnel do not fully satisfy all skill requirements. Mandating

• additional training imposes cost and regulatory burdens without any assurance of increased examiner competency.

2) NRC stipulates that 40 hours4.62963e-4 days <br />0.0111 hours <br />6.613757e-5 weeks <br />1.522e-5 months <br /> of training be distributed throughout the year. No definition cS&quate distribution is provided (e.g.,3.3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> per month,10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> per quarter, ca. Furthermore, no guidance details the topics to be covered during the 30 additional training hours. If NRC imposes this requirement, specific provisions should be included to avoid subjective future interpretations and the bases for these provisions should be clearly de0ned.

3) is training required to be conducted in a " blind" environment? If so, is the ttainee to be informed of the results of this training? If the trainee is informed, how is the

" blind" environment to be maintained?

4) Owners recognize the benent of UT practice using flawed specimens. One possible alternative to the proposed rule would be to require each site to maintain a sample set of flaw specimens. Incoming NDE examiners would be required to practice on these specimens under the supervision of a Level 111 NDE examiner. This practice would sharpen NDE examiner skills immediately prior to being sent to the field.

D) Appendix Vill Specimen Set Cracks (Page 63899, par. 2.4.1.1.2)

1) NRC states that " notches are not realistic representations of service induced cracks."

It is certainly possible to create notches that are not realistic representations of service-

ULNRC-3768 Page 7 induced cracks. Notches in PD1 coupons, however, are established by design. Notch locations and orientations are selected specifically to assure that they accurately represent known types of service-induced flaws. It is incumbent on NRC to provide evidence that the specine notches used in PDI coupons are not realistic representations of service-induced Daws.

2) The proposed rule inappropriately states that "only cracks should be used in vessels" because "only cracks are used in piping". Vessel reDectors are created using notches to enable coupon fabrication and to better challenge procedure essential variables.

While installing cracks in piping has become routine, it is far from routine in vessels.

This requirement imposes substantial costs without commensurate safety benents.

E) Appendix Vill Specimen Set Microstructure (Page 63899, par. 2.4.1.1.3)

1) 10CFR50.55a(b)(3)(xix) addresses material microstructure effects on one-sided examinations. Presently, ultrasonic testing technology is very limited in its ability to detect unfavorably oriented Daws under these conditions. The proposed rule imposes requirements that exceed current technological capabilities.

2) PDI is, as yet, unable to qualify far-side length or depth sizing for single-sided-access austenitic welds. The proposed rule mandates such qualiGcation. The proposed rule imposes quali6 cation requirements that cannot be satisfied.

3) The proposed rule will add flaws that cannot be detected. Considerable expense will be incurred in fabricating these test specimens. The result will be examination blocks for which neither personnel nor procedures can be quali6ed.

4) The industry has already proven the ability to create flaws that neither equipment or personnel can detect. Current acceptance criteria and industry experience demonstrate that these Haws are not relevant to safety. Inability to detect these Daws, therefore, does not pose a safety concem. The revised rule requires detection of these undetectable Daws, yet provides no evidence to demonstrate that these flaws are safety-significant. It is injudicious for NRC to impose these onerous requirements without first demonstrating a commensurate increase in plant safety.

5) No definition is provided as to what constitutes a test specimen that "contain(s) microstructures like the components to be inspected." Compliance with this requirement is likely to impose considerable costs, since materials (including welds of dissimilar materials) are unique to ;ndividual plant sites. The likely result is a requirement that individual plants design, fabricate, and perform qualifications on their own test specimens. Time and expenses associated with these types of activities are prohibitive, as demonstrated by PDI efforts. The proposed restriction, when coupled with expedited implementation, invokes a requirement that cannot be satis 6ed within the timeframe specified.

6) If the requirement to use test specimens that "contain(s) microstructures like the components to be inspected" is strictly interpreted, compliance is impossible.

Duplication of material microstructures may be conceivable on a microscopic scale, but not on a macroscopic scale.

7) This proposed test will reduce the number of qualified examiners in the industry to zero. Since NRC does not address this impact in the proposed rule, estimated cost i savings and dose reductions are inaccurate. NRC should also reassess the rule's impact on self-employed examiners and on small businesses that employ examiners.

Page 63906, par. 6 of the proposed rule inaccurately states that " ..this rule will not, if promulgated, have a signiGeant impact on a substantial number of small entities.. "

F) Acceptance of PDI Methodology (Page 13900, par. 2.4.2):

1) Current PDI methodology precludes simp!c addition of transducers and scopes to the essential variables list. An NDE Examiner must use each component as part of his

ULNRC-3768 Page 8 qualification test to satisfy essential variable requirements. Expensive PDI qualification is the only means of adding new equipment to the essential variables list.

No cost-effective alternative exists. This restriction will inhibit use of technological advancement due to prohibitive certification costs.

2) Examination equipment technology has not changed for many years-only the methodology of applying that technology has improved. An alternative to the proposed rule is a generic letter mandating the use of PD1 NDE techniques. PDI procedures provide a solid and consistent industry standard using carrent NDE methodologies. Industry standardized procedures will ensure availability of knowledgeable and experienced examiners. Standardized procedures will constitute a baseline to simplify regulatory assessment.

G) NRC Supporting Document Titled " Documented Evaluation for Appendix Vill Expedited Implementation":

1) The Documented Evaluation cites 1GSCC crack detection as a primary difficulty for NDE examiners. The industry has acknowledged the need for special IGSCC certification and examination requirements. The industry also, however, recognizes 1GSCC as a mechanism found primarily Boiling Water Reactors. Requirements for IGSCC certification and examination should be limited to BWRs, or credible evidence should document the need for imposition of these requirements on PWRs. A backfit analysis is the appropriate vehicle for documenting this evidence. Note that this concern has already been addressed by the Triparty Agreement (BWROG, EPRI, and NRC) which was recently turned over to PDI's IGSCC certification examination. 1

2) The Documented Evaluation cites "round robin" examinations where examiners were unable to find all test specimen flaws. Most materials examined in these tests were thick-wall wrought or centrifugally cast stainless steels. It is inappropriate to invoke new requirements on aji Section XI piping and examiners based on evidence limited to these large, heavy-wall pipes. Evidence demonstrates that these types of heavy-wall examinations are limited more by equipment capabilities than by examiner capabilities. The Documented Evaluation inappropriately cites these round robin tests as evidence of the need for expedited implementation of Appendix Vill. The primary lesson learned from these round robin tests is that there is a need for continued technological advancement.

3) The Documented Evaluation equates the skills of an NDE Examiner with those of a reactor operator or airline pilot. Reactor operators and airline pilots are required to make time-critical decisions that may have immediate and severe impacts on public health and safety. It is misleading to suggest that an NDE Examiner makes decisions under similar circumstances or bearing similar consequences. A more accurate analogy would be to equate the NDE Examiner's responsibilities to those of a nurse who checks the blood pressure of the reactor operator or airline pilot. The nurse records the resulting information, compares it against known acceptance criteria, then reports the results to someone else. The person to whom the results are reported then assumes responsibility to determine what (if any) actions are warranted. The nurse's responsibility ends when the results are reported.

II) NRC Supporting Document Titled " Draft Regulatory Analysis"(Table 2, Note 2):

1)

Table 2 Note 2 states "Over 300 examiners for piping are qualified for parts of Supplements 2 and 3 (including IGSCC), and all five teams for vessels are qualified for Supplements 4 and 6." This statement falsely implies that PD1 has certified ample ultrasonic examiners. Only 75 (or fewer) of these examiners have a full set of certifications. Only half of these 75 (approximately 40) are certified for IGSCC examinations. Many of these 40 examiners work at specific facilities and are therefore unavailable to the industry as a whole. The result is a handful of available examiners with full certifications. If(as is likely) use of examiners with partial certifications is

ULNRC-3768

~

Page 9 necessary, each exam will require more than one examiner. Considering the likelihood that more examiners per weld will be required, the NRC-projected cost savings and dose red:ctions are most likely inaccurate.

2) Plant experience and PD1 involvement indicate that Appendix Vill implementation will double the size of each ISI crew. The result will be twice the demand for examiners-a demand that cannot be filled.

5. Section 2.5 (Vof u,wy implementation)

A) Section 111 (Page 63900, par. 2.5.1)

1) Limitations (Page 63900, par. 2.5.1.1)

2) Engineering Judgement (Page 63900, par. 2.5.1.1.1): See item 2A.

3) Section 111 Materials (Page 63900, par. 2.5.1.1.2): No comments.

4) Weld Leg Dimensions (Page 63900 and 63901, par. 2.5.1.L3): No comments.

5) Seismic Design (Page 63901, par. 2.5.1.1.4): No comments.

6) Quality Assurance (Page 63901, par. 2.5.1.1.5): No comments.

7) Independence of Inspection (Page 63901, par. 2.5.Lt.6): No comments.

8) Modification (Page 63901, par. 2.5.1.2)

9) Applicable Code Versions for New Construction (Page 63901, par. 2.5.1.2.1): No comments.

B) Section XI (Voluntary implement t w (Pqe 63901, par. 2.5.2)

1) Subsectior. IWE and Sun % 3 L (Page 63902, par. 2.5.2.1):

a) Imposition ofIWE/th sequirements is unwarranted. These requirements do not provide safety improvements commensurate with costs. The design-specific problems NRC cited as the basis for IWE/lWL did not justify industrywide IWEllWL enforcement. NRC did not demonstrate that IWE/lWL would rectify the isolated concerns used tojustify IWE/lWL enforcement. Neither did NRC perform the mandatory backfit analysis. NEl submitted these and other industry comments to NRC during initial rulemaking. NRC responses to these comments were inadequately substantiated. The comments submitted by NEl remain valid.

Imposition of IWE/lWL imposed significant costs accompanied by marginal "fety benefits. The proposed rule should be used to revise IWE/lWL requirements from " mandatory" to " voluntary" b) IWE and IWL are " immature" ASME standards (see comment 4.B.1). The workability of these requirements had not been fully assessed when they were NRC-mandated in 1997 (61 FR 41303). Since mandated. Owners have identified numerous flaws in the Code requirements themselves. Using the consensus process, Owners haw eacted numerous IWE/lWL changes to rectify these flaws. Changes resulting trc,m this industry effort are to be incorporated in the 1998 Edition of the ASME Code. The 1995 Edition through 1996 Addenda ofIWEllWI- resolves only a few of these issues. While 1995 Edition /1996 Addenda approval provides some benefit, it is recommended that NRC focus resources on expeditious approval of the 1958 Edition ofIWE/lWL.

2) Flaws in Class 3 Piping (Page 63903, par. 2.5.2.2):

a) NRC endorsement of ASME Code Cases has historically been accomplished through Regulatory Guides, not through 10CFR50.55a. Regulatory Guides are, and should remain the NRC method of Code Case approval. ASME Code Case endorsements should be deleted from the proposed rule and moved to Regulatory Guides.

. ULNRC-3768 Page 10 b) Code Case endorsements in the proposed rule raise several issues. First, will the rule be revised as reference Code Cases are revised, annulled, incorporated into the Code, or permitted to expire? Second, will NRC require Owners to request approval of Code Cases that are in the rule, but not in Regulatory Guides?

Third, will NRC revise Regulatory Guides to authorize Code Case endorsement in 10CFR50.55a revisions? Fourth, if these Code Cases are eventually incorporated into Regulatory Guides, will Regulatory Guide restrictions supersede rule restrictions?

c) NRC proposes five N-513 restrictions. Note, however, that N-513, if approved, will apply only to moderate energy ASME Class 3 piping. N-513 requires case-by-case structural integrity and leakage evaluations. N-513 requires an

" evaluation for suitability" in accordance with the Owner'sSection XI Repair / Replacement Program, which assesses the cause of the leak and actions to prevent recurrence. No steam is involved in these moderate energy systems, nor is system pressure sufficient to warrant personnel safety concerns.

Considering these Code Case limitations, several of the restrictions imposed in the rule are unwarranted. Unwarranted restrictions include the exclusion of:

• Flaws remaining from original construction: Most through-wall leaks in Class 3, moderate energy systems are the result of corrosion. Corrosion usually destroys the defect initiation site. As a result, it is impossible to determine whether the flaw was one that remained from original construction. In any event, the origin of the defect has minimal bearing on whether or not a temporary repair is feasible. This restriction is unwarranted and should be removed.

= Leakage through a flange gasket: Leakage through a flange gasket is not within the scope of N-513. IWA 4000 clearly exempts gaskets from the ASME Code. Leakage through gaskets is addressed by the ASME Code only when performing a Code-required pressure test, and N-513 does not apply to leakage detected during Code-required pressure tests. Leakage in ASME Class 3 systems is normally permitted by plant Technical Specifications. This restriction is unwarranted and should be removed.

• Seal welded threaded connections: A pinhole leak in a seal weld is not a reasonable basis for plant shutdown. For each leak, an Owner is required to perform an evaluation for suitability of repair. This evaluation is required to assess the cause of the leak (e.g., was the leak due to a seal )

weld defect, was it due to an external impact, was it due to thread failure, etc.). If this evaluation fails to adequately address the cause of the leak, it fails to satisfy ASME Section XI requirements. N-513 requires an evaluation of structural integrity, which cannot be done without assessment of thread degradation. The proposed restriction, therefore, assumes that an Owner will not satisfy existing N-513/ASME Section XI requirements.

Enforcing existing requirements seems more effective than duplicating existing requirements through rule-imposed Code Case restrictions. This l restriction is unwarranted and should be removed.

= Degraded socket welds: A pinhole leak in a low-energy Class 3 socket weld is not a reasonable basis for plant shutdown. N413 requires the Owner to quantify defect size. Once defect size is known, N-513 requires  !

a structural integrity evaluation to assure component acceptability. This evaluation assures that the defect will not compromise socket weld integrity. The proposed restriction is unwarranted and should be removed.  ;

C) O&M Code (Voluntary implementation)(Page 63903, par,2.5.3) l l

I) Code Case OMN 1 (Page 63903, par. 2.5.3.1): No Comments.  ;

1

ULNRC-3768

~

Page1I

2) Appendix lI (Page 63903, par. 2.5.3.2):

a) The proposed Appendix 11 check-valve monitoring program will allow more flexibility regarding test method and frequency. Acceptance of this provision in the proposed rule is encouraged.

b) Proposed Section 50.55a(b)(3d)(iv)(A) will mandete bi-directional testing of check valves. Section ISTC 4.5.4 will require both open and closed tests. Many Owners have valves with only a single position safety function, and these are generally tested in only one direction. The proposed change will require a signincantly higher rumber of tests and will require changes to the associated procedures. The rule should be revised to exempt these valves from unwarranted bi-directional testing.

3) Subsection ISTD (Page 63904, par. 2.5.3.3):

a) This proposed change permits voluntary use ofISTD, since snubber programs are currently included in technical specincations. This change overlooks the fact that many Owners have removed snubber programs from their technical ,

specifications and placed them in the plant's Final Safety Analysis Report (FSAR). This portion of the rule should be revised to refer to "other Operating License Documents" rather than strictly technical specincations. Overall, the proposed change will benefit the industry, since these later OM Code editions i j

define more appropriate snubber examination and test programs.

b) ISTD 7 describes inservice functional test frequency as " performed every refueling outage" Many plants have made FSAR changes (via the 50.59 process) to allow on-line snubber testing. The proposed rule should be revised to acknowledge the acceptability of on-line snubber testing.

c) Many Owners have used the 50.59 process to gain NRC approval of plant- 2 specific practices. While use oflater OM Codes provides several advantages, these advantages will be negated if previous Owner-NRC agreements are voided. The regulation should describe NRC's acceptance of prior Owner-NRC agreements when adopting later OM Codes. j Containment Isolation Valves (Page 63904, par. 2.5.3.4): {

4) a) OM-10 subsection 4.2.2.2 dictates that leakage rate testing of Containment isolation Valves is controlled by 10CFR50 Appendix J while leakage rate testing of all other valves is controlled by OM-10 subsection 4.2.2.3. The current 10CFR50.55a(b)(2)(vii) invokes OM-10 subsection 4.2.2.3(e) and (f) for Containment isolation Valves. The proposed rule removes this modification, and is considered a beneficial change. l l  ;

b) This proposed amendment would allow utilities to use the rules specified in '

l 10CFR50, Appendix J for Containment isolation Valve leakage analysis and corrective action. Appendix J was specifically written for leakage analysis and j corrective action regarding Containment isolation Valve leakage rates. The l  ;

I proposed amendment to remove 10CFR50.55a(b)(2)(vii) is beneficial, since adequate guidelines are already provided in Appendix J.

6. ASME Code Interpretations (Page 63904, par. 2.6)

A) The proposed rule states that NRC " generally agrees with (ASME) interpretations." Since, however,NRC"has a certain understanding" of the Code Editions and Addenda in the regulations, NRC reserves the authority to reject ASME interpretations contrary to this "certain understanding". Since it is this "certain understanding" that becomes the NRC's Gnal basis for determining what the ASME Code requires, it is crucial that Owners be granted access to this understanding. NRC is asked to define the formal NRC internal process through which this l

ULNRC-3768 Page 12 "certain understanding" is achieved. NRC is requested to define where this " understanding" is documented, and how Code users are to be provided access to this documentation.

B) It is possible that situations will arise where the NRC's "certain understanding" of a specific Code requirement was not clearly documented during the review process. The proposed rule should be revised to state that, in these cases, the official ASME interpretation will be accepted as the authoritative, final interpretation of the specific Code requirement. This rule revision is essential since, without documentation, NRC is not likely to reproduce the "certain understanding" achieved during the actual Code review.

C) It would appear that the "certain understanding" achieved during NRC's ASME Code review must be the consensus of the reviewers, as opposed to the "certain understanding" of any one individual. For this reason, the "certain understanding" must be documented to assure that the "certain understanding" of any one individual is not used to supersede the "certain understanding" achieved by group consensus. Clearly, without formal NRC documentation, the "certain understanding" is a moving target, and subjects Owners to arbitrary NRC interpretations of ASME requirements. This portion of the rule should not be implemented until documentation of the consensu: understanding is produced.

D) Without NRC documentation, Owner actions that met NRC's "certain understanding" when the 1989 Edition of the Code was endorsed may not meet the "certain understanding" of an individual NRC employee who was not involved in the review. NRC should provide both the program used to document this consensus and the resulting consensus documentation. Without this documentation, the industry is incapable of determining how to comply with the ASME Code.

E) When NRC disagrees with an ASME interpretation, one of the following alternatives should be implemented:

1) Challenge the interpretation using the ASME inquiry process defined in the ASME Codes, or:

2) Identify the specific regulation that takes exception to this Code requirement, or;

{ 3) Revise the regulation to take exception to this Code requirement.

F) With this 10CFR50 revision, NRC attempts to transfer responsibility for interpreting a consensus document from the consensus body to the NRC. In transferring this responsibility, l

NRC would effectively negate a fundamental means of resolving Owner /NRC disagreements i over ASME Code rules. This transfer of responsibility is unacceptable and is without technical justification.

G) This revision contradicts Public Law 104-113, Section (12)(d)(l), which " .. directs federal agencies to achieve greater reliance on technical standards developed by voluntary consensus standards development organizations"(quote taken from proposed rule, paragraph 2.7,"DSI-l l 13").

DSI-13 (Page 63905, par. 2.7):

A) NRC has been strongly encouraged to simplify and expedite the consensus standards endorsement process'. Mr. Hoyle identifies " ..the need to streamline and simplify the NRC's intemal process for endorsing codes and standards within a year after they are issued by a professional society." The proposed rule mentions NRC's " active consideration" of this issue, yet fails to incorporate approval process improvements. The proposed rule revision is an ideal opportunity for NRC implementation of Mr. Hoyle's recommendations. The industry strongly supports and encourages use of this opportunity to implement a revised NRC approval process.

' Memorandum, March 71997, John C. Hoyle (Secretary) to L. Joseph Callan (EDO) entitled " Staff Requirements - COMSECY 96-062 - Strategic Assessment issue Paper: The Role of Industry (DSI 13)"

ULNRC-3768 Page 13 B) NRC is strongly encouraged to adopt a six-month, automatic process for endorsing Codes and Standards. Bases for this recommendation were discussed in a November 16,1994 NEl/NRC meeting, and included:

1) The majority of ASME Code changes incorporate improved technology and methodology. Few Code changes result in significant plant safety improvements. Use oflater Code Editions and Addenda should be optional unless formal NRC backfit analysis documents a specific plant safety improvement. Code revisions rarely incorporate significant plant safety improvements, as evidenced by the fact that; a) Licensees are currently required to update only once every 10 years, b) Licensees use various editions / addenda of the Code, c) The process wherein NRC revises 10CFR50.55a to adopt a later Edition and Addenda is a slow one.

2) NRC personnel serve on ASME Code Committees. NRC, therefore, has a signi6 cant influence on Code rule development. NRC participation enables NRC to raise specinc concerns and to participate in the consensus process wherein these concerns are resolved. NRC participation reduces the likelihood that NRC will, during the six-month review period, identify concerns that have not been addressed and resolved during Code development.

3) Prior to final approval, proposed ASME Code revisions are published in the ASME

" Mechanical Engineering" Magazine and ANSI" Reporter" for public comment. This public comment process provides an opportunity for additional NRC review and comment. This public comment period also provides time, in addition to the proposed six-month review period, for NRC review.

4) NRC is authorized, through backfit analysis, to impose safety signiGcant requirements at any time. The proposed six month time period will not delay NRC imposition of safety-significant requirements. It will, however, speed up industry access to the experience and process improvements incorporated into Code revisions.

5) OMB Circular No. A-119," Federal Participation in the Development and Use of Voluntary Standards," encourages adoption and use of private standards developed by voluntary standards bodies. A-119 identifies a preference for standards based on performance criteria rather than design, material or construction criteria. The ASME Codes are largely based on (and continue to move toward) performance-based criteria, thereby satisfying this condition.

6) Timely endorsement has no effect on the NRC's option to take exception to any Code change. if the proposed six-month time period does not provide adequate time for thorough review of a specific requirement, NRC can take exception to the requirement pending completion of a more thorough review.

7) Timely endorsement is not new. A precedent for automatic approval without specific response from the Commission is prescribed in 10CFR50.54(a)(3)(iv). This paragraph states:" Changes to the Quality Assurance Program . . shall be regarded as accepted by the Commission upon receipt of a letter to this effect . . or 60 days afler submittal to the Commission, whichever occurs first."

8) On occasion, NRC has directed licensees requesting NRC rule changes or Code relief to pursue these changes through ASME Codes and Standards. In support of this response, it is incumbent upon NRC to expeditiously approve consensus Code updates.

9) NRC itself proposed a six-month endorsement timeframe in a May 18,1994 letter from Mr. Russell to Mr. Beckjord.

ULNRC-3768 Page 14'

10) NRC cndorsement of the proposed automatic si.:-month Code update process may

~

warrant revising the NRC internal review process. Elimination of public and CRGR reviews would facilitate automatic approval within six months.

C) Benefits associated with six-month automatic endorsement of consensus Codes and Standards include:.

1) Automatic endorsement would permit timely implementation of cost beneficial Code changes. As discussed in the October 21,1993 Entergy request, few if any Code changes have a potential for significant public health and safety improvements Many Code changes, however, provide substantial industry benefits by providing improved technologies, methods, practices, and criteria.

2) Automatic endorsement would reduce the need for development (and NRC approval) ofindividual Code Cases. Currently, important Code revisions are acce.mpanied by identical Code Cases. The Code Case is a duplicate effort, necessary only because of i the slow process by which Code Editions and Addenda are endorsed. Timely endorsement of Code Editions and Addenda would greatly reduce the effort spent reviewing duplicate Code Cases.

3) Automatic endorsement would allow licensees to update ISI and IST program plans more frequently, gradually incorporating new requirements if and when desired. This would levelire the burden on plant stafTing and reduce reliance on consulting services for ISI and IST program updates.

4) Timely endorsement would enable licensees to integrate changing Code requirements as they are published. By keeping pace with Code changes, utility resources would be expended where the greatest gains in safety and cost efficiency would be realized.

D) Code updates, when automatically cndorsed, would be available to licensees as " acceptable attematives", not as requirements that automatically supersede existing Code provisions.

7. Steam Generators (Page 63905, par. 2.8): No Comments.

8. Finding of No Significant Environmental Impact (Page 63905,63906, par. 3.):

A) Imposition of ASME Appendix Vill requirements and the added restrictions in the proposed rule will have environmental impacts beyond those addressed in the Rule. Examples include:

1) The rule proposes that " Implementation of(Appendix Vill) ... should reduce repeat ultrasonic testing (UT), which could reduce occupational exposure." This statement is incorrect. Implementation of Appendix Vill as modified by the proposed rule will force significant increases in examination sensitivity. As sensitivity increases, the number of detected / suspected flaws will increase. Each flaw must then be evaluated for acceptability. Trending of each flaw is likely to increase examination frequency.

Repetitive examinations on greater numbers of flaws will equate to increased occupational exposure. While significant resources will be consumed in these activities, no corresponding increase in plant safety will result because these small flaws are not safety-significant. The result, therefore, is added costs to push detection capability to its furthest limits, without credible evidence that current practices result in inadequate safety. The quoted NRC statement overlooks the fact that the changes proposed in the regulation are unrealistic and cost-prohibitive.

2) The rule proposes that " .. flaws should be detected at an earlier stage of growth resulting in less extensive repair operations, which could further reduce occupational exposure."

This argument is incorrect in the following respects:

a) Ultrasonic examination effectiveness is limited primarily to the inner (lD) surface of a component. Repair of an ID defect normally requires excavation from the outside (OD) of the component. An OD excavation will encounter a large (e g.,

ULNRC-3768

~

Page 15 deep) ID defect sooner than it will encounter a small (e.g., shallow) defect. The argument that smaller Daws result in less extensive repairs is weak Consider also that replacements are often used to climinate flaws, negating the argument that Oaw size determines repair difficulty.

b) The requirement to use more sensitive examination requirements will increase the number of unnecessary repairs. Because examiners will be required to quantify smaller Daws with greater accuracy, any Daw that cannot be definitively quantified is likely to be considered rejectable. Acceptable Daws will be rejected when examination capabilities / access will not permit conclusive assessment of flaw size and orientation. Use of equipment and personnel at the extreme limits of their detection capabilities will increase sizing uncertainties, thereby increasing rejection (and subsequent repair) of acceptable flaws. The result will be significant increases in both manpower and radiation exposure. The net effect will be a safety reduction, since removal of a non-safety-significant Daw constitutes an unwarranted repair.

9. Paperwork Reduction Act Statement (Page 63905, par. 4.): No Comments.

10. Regulatory Flexibility Certification (Page 63906, par. 6):

A) The NRC assenion that "this rule will not, if promulgated, have a siguificant impact on a substantial number of small entitics" is incorrect. The assertion that "this proposed rule affects only the licensing and operation of nuclear power plants" is also incorrect. The rule modifies Appendix Vill such that PDI quali0 cations for at least 300 individuals are immediately invalidated. Many of the 300 examiners are self-employed, thereby qualifying as small entities.

The rule, therefore, imposes significant hardship on small entities.

11. Backfit Analysis (Page 63906,63907, par. 7.):

A) Page 63907 states: "The NRC believes that there is a basis for reasonably cor.cluding that such degradation could occur in virtually all PWR's." "Because of... the belief that additional occurrences of noncompliance with GDC 14 . . will be reported, the NRC has deternined that imposition of Appendix Vill . . is appropriate, therefore, a backfit analysis is not required."

The:e conclusions are inaccurate, as described below.

1) The technical NRC basis for " reasonably concluding" that all PWR's are at risk is minimal This " belief that additional occurrences" will come to pass is a nonsubstantial basis for avoiding the backfit process. This NRC statement provides evidence as to the highly subjective nature of the NRC decision to avoid backfit )

analysis. As evidenced by the proposed rule, NRC frequently mandates rules more conservative than those adopted in industry standards. Conversely, NRC evades the j

conservative approach to rulemaking,i.e., the backfit process. This is evidence of a double standard, and is unacceptable.

2) Owners believe that a backfit analysis is required, and that this analysis would enable significant improvements in the proposed rule. Backfit analysis is also likely to 4 moderate adverse impacts (both anticipated and unanticipated) resulting from implementing the proposed rule. NRC has expended considerable effort in explaining how to avoid backfit analysis for the proposed rulemaking, yet these explanations fail to address issues crucial to the backfit decision. These resources would be better used in the backfit analysis itself, NRC is implored to implement the backfit analysis mandated by their own rules.

B) in the Section 7 discussion on Backfit Analysis, it is stated that a backfit analysis is not required because," licensees understand that 50.55a requires that they update their inservice inspection program every 10 years to the latest edition of Section XI. " Whether or not licensees

" understand" this 50.55a requirement has no bearing on whether or not back0t analysis is required. Backfit analysis of the changes proposed in the rule is mandated by 10CFR50.109.

a ULNRC-3768 Page 16 C) Section 7 states that," endorsing and updating references to the ASME Code, a national consensus standard developed by the participants (including NRC) with broad and varied interests, is consistent with both the spirit and intent of the back6t rule (i.e. NRC provides for the protection of the public health and safety, and does not unilaterally impose undue burden on applicants or licensees)". This statement is erroneous in its implication that the proposed rule, by endorsing the ASME Code, imposes no undue burdens. The rule modifies the Code by invoking numerous exceptions and additions, many of which impose undue burdens. These changes have not been accepted by the consensus process and are not, therefore," consistent with both the spirit and intent of the backGt rule." The NRC justincation for avoiding backGt analysis is nonsubstantial. Back0t analysis, as required by federal law, is mandatory for the proposed rule changes.

ULNRC-3768 Page 17 cc: U. S. Nuclear Regulatory Commission (Original and I copy)

Attn: Document Control Desk Mail Stop PI-137 Washington, DC 20555-0001 Mr. Ellis W. Merschoff Regional Administrator U.S. Nuclear Regulatory Commission Region IV 611 Ryan Plaza Drive, Suite 400 Arlington, TX 76011-8064 Senior Resident Inspector Callaway Resident Office U.S. Nuclear Regulatory Commission 8201 NRC Road Steedman, MO 65077 Mr. Barry C. Westreich (2 copies)

Acting Licensing Project Manager, Callaway Plant Office of Nuclear Reactor Regulation U. S. Nuclear Regulatory Commission Mail Stop 13E16

- Washington, DC 20555-2738 Manager, Electric Department Missouri Public Service Commission PO Box 360 Jefferson City, MO 65102 Mr. Thomas A. Ba: er .

Shas ', Pittman, Potts, & Trowbridge 2300 N. Street N.W.

Wash.ngton, DC 20037 i

.m ULNRC-3768 k Page 18 :

bec: G. L. Randolph w/a J. V. Laux/G. A. Hughes w/a

. A. C. Passwater/D. E. Shafer (470) w/a -

M. P. Barrett (100) w/o P. L. Reynolds -(470) NSRB w/a H. D. Bono w/a CDN Chrono w/a . .

Senator Christopher Bond w/a -

Senator John Ashcroft w/a Representative Kenneth Hulshof w/a E210.0001 -

- A160.0761

. Manager, Plant Support w/o

. Wolf Creek Nuclear Operating Corporation PO Box 411 -

Burlington, KS 66839 I

_