ML20195J762

Appeals NRC Denial of Backfitting Claim Re Containment Integrated Leak Rate Testing.Util Maintains View That Two NRC Positions Not Supported by App J Constitute Backfits for Facilities

ML20195J762
Person / Time
Site:	Oconee, Mcguire, Catawba, McGuire, 05000000
Issue date:	06/20/1988
From:	Tucker H DUKE POWER CO.
To:	NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM)
References
NUDOCS 8806290222
Download: ML20195J762 (9)

Text

, -

Duxn POWER GOMPANY P.O. HOX G3180 CitARLOYM4 N.O. 28242 HALB. TUCKER ress enown -

7303 PSESIDENT (704) 373 4 501 ROOLEAR PBODtTTIOt.

June 20, 1988 U. S. Nuclear Regulatory Commission Attention: Document Control Desk Washington, D. C. 20555 Re: Oconee Nuclear Station, Docket Nos. 50-269, -270, -287 McGuire Nuclear Station, Docket Nos. 50-369, -370 Catawba Nuclear Station, Docket Nos. 50-413, -414 Appeal of Denial of Backfitting Claim Regarding Containment Integrated Leak Rate Testing Gentlemen:

By lettgr dated September 1, 1987, Duke Power Company, pursuant to 10 CFR 550.109 , presented a claim thaz two NRC Staff positions on containment integrated leak rate testing requirements under Appendix J to 10 CFR Part 50 constitute backfits. These two Staff posit permituseoftheMass-Plot (orMass-Point){onswere:

(i) the Staff's refusal to analysis technique for a test duration of less than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, and (2) the Staff's insircence, based on <

Information Notice 85-71, that Type A test results be adjusted by certain leakage measured during local leak rate tests. The Staff denied the backfitting claim by letter dated October 29, 1987.

Duke Power Company hereby appeals the Staff's denial of its backfitting claim.

It remains Duke's view that the two Staff positions are not supported by Appendix J and constitute backfits for Duke's facilities.

DISCUSSION i A. Test Duration For Mass-Point Analysis Technique This issue concerns the duration of the Type A test for which tha Mass-Point technique for evaluating test results may be used. In its April 1, 1986 review of leak rate testing methodology, the Staff stated that the only acceptable test of less than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> is that using the "Total Time" analysis technique, as Duke understands that despite the decision in Union of Concerned Scientists

v. NRC, 824 F.2d 108 (D.C. Cir. 1987), the NRC is continuing the adhere to the policy and procedures of Section 50.109 and Manual Chapter 0514 u4.til the new backfitting rule is finalized.

2 Duke has used the term "Mass-Plot" while the Staff uses the term

, "Hass-Point"; the meaning of the terms is identical. For consistency's sake, Duke will use the term "Hass-Point" throughout this letter.

8806290222 000620 DI PDR ADOCK 05000269<

p PDR .

U. S. N:: lear Regulatory Commission June 20, 1988 Page 2 justified in Bechtel Corporation Topical Report BN-TOP-1, dated November 1, 1972.

For this technique, the Staff has approved a test duration of six hours. It is Duke's position, however, that a test duration of less than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> using the Mass-Point analysis technique is consistent with Appendix J requirements. In fact, this technique has been accepted by the NRC in previous integrated leak rate tests with durations less than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. The current Staff position, therefore, constitutes a backfit.

Before proceeding to a discussion of the merits of this irsue, Duke notes that it is willing to continue to work with the Staff in an effort to reach agreement under Paragraph 7.6 ANSI N45.4-1972 on a test duration of less than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> using the Mass-Point analysis technique. SeeDukeLetterdatedSepgember17, 1987. Given the fact that the current version of the ANSI standard adopts the Mass-Point technique with an 8-hour minimum test duration, the Staif should be prepared to consider such an agreement. Duke's appeal on this issue may be rendered moot if a timely, and favorable, response to our September 17, 1987 letter is received, or if the issue is favorably resolved in the Staff's Proposed Rule. However, Duke needs a resolution of this issue prior to April, 1989, as that is the next scheduled ILRT.

Turning to the merits, the Staff's view that Appendix J requires a 24-hour test duration is not supported by the regulation. Appendix J itself is completely silent on the subject of test duration.Section III.A.3(a) of Appendix J, which governs "Test Methods", simply provides that "[a]Il Type A tests shall be conducted in accordance with the provisions of the American National Standards N45.4-1972 . . . ." The Staff relies upon the fact that paragraph 7.6 of ANSI N45.4-1972 states that the test period shall be 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> of retained internal pressure, ur..ess "it can be demonstrated to the satisfaction of those responsible for the acceptance of the containment structure that the leakage rate can be accurately determined during a shorter test period ...." Althougn Duke relies primarily upon other arguments herein, it would point out that "those responsible for acceptance of the containment structure," as used in the ANSI standard, normally refers to the owner and operator of the facility and the authorized representative of the ANSI.

It is also not clear that the Commission intended that paragrsph 7.6 be included within the scope of the incorporation by reference under Appendix J $III.A.3(a).

It seems apparent that the Commission did not intend to incorporate as a requirement every provision of the ANSI standard. Because the incorporation by reference is in the "Test Methods" section of Appendix J, it may be that the Commission intended to incorporate the ANSI standard only as far as Section 5 of the standard provided acceptable test methods -- namely, the absolute and A good description of the three data analysis techniques is provided in the Staff's recent proposed rule to amend Appendix J by expressly allowing the use of the Mass-Point technique (though for a test duration of 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />). 53 Fed. Reg. 5985 (February 29, 1988). As the Staff notes, the Mass-Point analysis technique is of nore recent development and has been accepted by the NRC as the prefarred data analysis technique at this time. Ibid. The more recent versions of the ANSI standard, beginning in 1981, adopt this technique. ANSI /ANS56.8-1987 (at paragraph 5.4) specifies a minimum test duration of 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />.

U. S. Nuclear Regulatory Commission June 20, 1988 Page 3 reference-vessel methods. The 24-hour test duration provision is not part of the ,

"Test Methods" section of the ANSI standard, and therefoge may not have been

, intended to be incorporated by reference into Appendix J The rece.t proposed rule on the Mass-Point technique would prescribe an explicit 24-hour test duration requirement. The fact that the NRC sees a need to add an explicit provision on test duration strongly suggests that such a requirement is not now clearly reflected in the regulation. The proposed rule provides no basis for the 24-hour provision but merely states that this reflects "prior exemption approval." 53 Fed. Reg at 5985. No exemption approval of that nature has been issued for Duke. Thus, any such positions in exemptions for other facilities do not establish a Staff position applicable to Duke. Manual Chapter 0514 at 21.

Even assuming the 24-hour test provision has been properly incorporated as a binding requirement,_the NRC has previously accepted the use of the Mass-Point technique for Duke for test periods less than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. As explained at page 3 of nur September 1,1987 claim, NRC inspectors have previously observed two integrated leak rate tests at Oconee where test durations of less than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> were used with the Mass-Point analysis technique. The inspectors observed the tests and reviewed the ILRT summary test report and found that the tests were condugted in conformance with Appendix J with no related violations or unresolved items . The Inspection Reports are consistent with the fact that the later revisions to the ANSI standard (see ANSI /ANS-56.8-1981), which adopted the Mass-Point technique, provide for a minimum test duration of eight hours.

5 In its October 29, 1987 denial, the Staff dismissed this fact on the ground that approval in an NRC Inspection Report does not suffice to grant a deviation from an explicit regulatory requirement. As explained above, it is not clear that the 24-hour test duration provision of ANSI N45.4-1972 was incorporated as a specific requirement of Appendix J. Assuming for the sake of argument that it was, nothing in the ANSI standard precludes an NRC Inspection Report from being an 4

For a publication to be incorporated as a binding requirement, the NRC must comply with the regulations of the Office of the Federal Register governing incorporation by reference. Those regulations provided that "[t]he language incorporating a publication by reference shall be as precise and complete as possible...." 1CFR 551.9(a). Where a document is not properly incorporated by reference, it cannot be considered binding. See PPG Industries Inc. v. Costle.

659 F.2d 1239, 1250 (D.C. Cir. 1981) (citing Administrative Procedure Act 5552(a)(1)). It has been held that "an incorporation by reference must give one affected enough knowledge so that he may easily and certainly ascertain the conditions by which he is to be bound". Applachian Power Company v. Train 566 F.2d 451, 457 (4th Cir. 1977).

The reference to ANSI N45.4-1972 in Section III.A.3(a) of Appendix J arguably does not give licensees the reasonable and certain notice regarding test duration that is required by law.

5 l' Inspection Report No. 50-270/83-35, dated December 15, 1983 (test duration of 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> 25 minutes); Inspection Report No. 50-287/81-04, dated April 6, 1981 (test duration of 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> 45 minutes). l,

- ,-n.- .--r,- . , - - , , - , - , - - , . , -. - - - - - - . - - , -w y, ,7., ,-,w-,.,,,-,,-- -,--.nwm,w --.,,,,,,.,..w,,-,,. ,.,,n,m,-,., .,,-n

U. S. Muclear Regulatory Commission June 20, 1988 Page 4 appropriate means for an agreement on a shorter test duration by "those responsible for the acceptance of the containment structure" within the meaning of Section 7.6 of the standard.

Moreover, the Staff cannot so easily disavow the findings of NRC Inspection Reports. Manual Chapter 0514 makes clear that a backfit arises when something previously accepted by the NRC as adequate is later preclub d. Manual Chaptar 0514 at 22, 28-29. According to NRC Staff policy, Inspection Reports are to receive management review and approval as to content and are to reflect only 6 approved, NRC positions as opposed to an individual inspector's personal vieus .

In view of the Staff policy, the NRC Inspection Reports from 1981 and 1983 accepting test durations of less than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> with the Mass-Point technique should be considered to establish the "applicable Staff position" r Duke. A contrary requirement now must therefore be identified as a backfit In conclusion, Duke contends that the Staff position precluding the use of Mass-Point technique for tests of less than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> represents a backfit for its facilities. Nevertheless, Duke is willing to work with the Staff in an effort to reach an agreement on test duration. In addition, Duke will commit to a minimum test duration of eight hours, for the Mass-Point technique, which is consistent with Duke's past practice as approved in NRC Inspection Reports and with ANSI /ANS-56.8-1987.

B. "As Found" Type A Leakage This issue concerns the Staff position in Information Notice 85-71 on the need to correct Type A test results for certain Type B and C leakage to derive an "as 6

-See, e.g., NRC Inspection Manual Chapter 0610, at page 6 (contents of inspection reports to receive appropriate management approval); NRC Inspection Manual Chapter 0611, at page 1 (inspection reports not to reflect an inspector's l personal opinion, but rather an "NRC position, opinion or recommendation or the )

inspector's presentation of technical positions on items and areas inspected"). I The Staff's October 29, 1987 derdal (at page 2) lists three documents as reflecting a "long-standing policy" on this issue, namely, a transcript of a workshop held in San Diego in 1982, a Region I letter to a licensee, and the EPRI report entitled "Containuent Integrated Leax-Rate Testing Improvements," dated November, 1982. At least two of these ( ,.uments support Duke's porition. The Regior I correspondence with another liccasee indicated that, because the plant's i Technical Specification required a 24-hour test duration, a Technical Specification enange would be necessary prior to conducting a test or less than i 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. However, the enclosure to the letter indicates that with such a l change, a test duration of 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> would be acceptabic using the Mass-Point technique. Duke has no similar Technical Specification requiring a 24-hour test.

Thus the Region I letter actually supports Duke's position here. Similarly, the EPRI report, while noting the Staff position here. Similarly, the EPRI report, while noting the Staff position, supports Duke by noting (at 3-32) that "[t]here is no engineering or scientific reason for a minimum leak test duration of 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />."

U. S. Nuclear Reguletory Commission June 20, 1988 Page 5 found" overall containment leakage rate. In its October 29, 1987 denial, the Staff concluded (1) that the Staff position in Information Notice 8571 "is not subject to the provisions of the backfit rule because [the Notice] was issued before the effective date of the rule"; and (2) that the Staff position is supported by the explicit requirements of Appendix J. Duke respectfully disagrees with these two conclusions.

With respect to whether the Staff position is subject to the backfitting rule, the critical point in time is not the date of publication of the Information Notice (August 22, 1985), but rather the date the Staff position reflected in the Notice was specifically imposed on the licensee. See 10 CFR 550.109(a); Manual Chapter 0514 at 21. Indeed, it is Staff policy that an Information Notice does not establish requirements or "applicable Staff positions" for purposes of Section 50.109. See Manual Chapter 0514 at 21-22.

For Duke, the Staff position was first imposed in 1986, when one unit of the McGuire Nuclear Station was ruled to have failed an ILRT. Inspection Report 50-370/86-16. Because the positio:. was imposed af ter the effective date of the ravised Section 50.109 (October 21, 1985), it is subject to the backfitting standards of that rule. Even if one assumes that Information Notice 85-71 could have established an applicable Staff position predating (by two months) the revised Section 50.109, it is still subject to the NRC backfitting policy. NRC Manual Chapter 0514 is expressly applicable to backfits imposed on or after May 1, 1985. Manual Chapter 0514 at 3 . Thus, contrary to the view expressed in the October 29, 1987 denial,thetimingofInformatgonNotice85-71doesnotshield it from review under the NRC backfitting policy We now turn to whether the position in Information Notice 85-71 is actually supported by Appendix J. In our view, the Staff positloa is inconsistent with the language, and history of Appendix J.

8 Staff management has made clear that even for backfits imposed prior to the effective dates of Section 50.109 or the Manual Chapter, the Staff will not stand on ceremony cnd refuse to apply the backfitting policy to these items. Region I Backfitting Workshop at TR. 81 (statement of Mr. Olmstead); Region II Backfitting Workshop, Vol. 2, at TR.48-49 (statement of Mr. Sniezek). Indeed, the 1985 backfitting rule was merely a revision to the backfitting rule that had been en the books since 1970 and which required a showing that the backfit would produce (substantial, additional protection...." The NRC also adopted an interim backfitting policy in Generin Letter 84-08, which also antedated Information Notice 85-71.

9 None of the four cases listed in Information Notice 85-71 in which the Staff applied its positions prior to the Information Notice involved a Duke facility. Thus these plant-specific cases cannot have established a Staff position "applicable" to Duke,. Manual Chapter 0514 at 21. Indeed, as explained in Duke's September 1, 1987 backfitting claim, in Inspection Report No.

50-769/80-06, dated March 20, 1980, the NRC did not require incorporation of as-found Type B and C leakage into Typo A test results. This NRC Inspection Report -- not inspection findings at other licensees facilities - establish the "applicable" Staff position for Duko.

i U. S. Nuclear Regulatory Commisoion June 20, 1988 Page 6 Appendix J does not require licensees to adjust Type A test results in the manner required by the Staff. Appendix J $1II.A.1(a) states in pertinent part as follows:

Containment inspection in accordance with V.A. shall be performed as a prerequisite to the performance of Type A tests. During the period between the initiation of the containment inspection and the performance of the Tyre A test, no repairs or adjustments shall be made so that the containment can be tested in as close to the "as is" condition as practical.

This section clearly prohibits "repairs or adjustments" during the period between initiation of the containment Inspection (as described in Section V.A. of Appendix J) and c.mpletion of the Type A test, so that the containment can be tested in as close to the "as is" condition (at the time of the Containment Inspection) as practicable. Appendix J permits "repair or adjustment" during this period if valves / penetrations are "maloperating" or leaking, or if there is evidence of structural deterioration which may affect containment integrity or leak-tightness. See, e.g., Section 111.A.1(b) and V.A. of Appendix.

However, the regulation does not restrict activities outside this period or require that licenseas factor Type B and C leakage from tests outside this period intotheT6Pa^testresutstoderivean"asfound"overallcontainment condition The Staff apparently interprets the "as is" provision in Section III.A.1(a) to require that leakage rates from Type B and C Lests, including those taken for any required repairs or adjustments, performed before the Containment Inspection be determined and added to the Type A test results. However, the Staff cites no specific language from, Appendix J to support this interpretation. In fact, the Staff's interpretation is inconsistent with the history of the Appendix J rulemaking. The proposed Appendix J published for comment in 1971 would have contained a provision consistent with the Staff position. The proposed rule stated as follows (proposed Section III.A.2.(a), 36 Fed. Reg. 170053, 17054 (August 27, 1971)):

, If leak repairs of testable componenta are performed prior to the conduct of the Type A test, the reduction in leakage shall be measured and added to the Type A test result. If this measured reduction in leakage exceeds the acceptance criteria of Section III.A.7., this information shall be included in the report submitted to the Commission as required by Section V.B.

The Commission, however, did not adopt this proposed provision in the final rule. l Thus, it must be concluded that the Commission did not intend such a provision to l be a binding part, of the regulation. This is clear from a review of the NRC's i recent proposed rule to revise Appendix J. In the proposed rule issued in October 1086, the Commission noted that the Staff was considering an additional requirement (outside the scope of the proposed revisions) that:

10 Appendix J 5III. A.1(a) requires the reportir4 of the change in leakage rate from repairs and adjustments during the type A test.

l l

U. S. Nuclear Regulatory Commission June 20, 1988-Page 7 .

All Type B and C tests performed during the same outage as a Type A test, or performed during a specified time period (nominally 12 months). prior to a Type A te be factored into the determination of a Type A test "as found" condition Given the Commission's statement, it is clear that such a requirement is beyond

, the present' scope of Appendix J and would constitute a backfit. The NRC's Cost Analysis for the Appendix J-rulemaking also noted that the proposed requirement to determin nd report "as found" Type A test results represented a new requireme t Furthermor,, the Staff position is inconsistent with the present Type A acceptance criteria.Section III.A.5(b) of Appendix J specifies an acceptance critorion of .75 of allowable leakage (La) for Type A tests. This 25% margin is intended to allow for possible degradation of containment leakage during the future period of operation until the next Type A test. The requirement to derive Type A "as found" leakage is intended to reveal leakage degradation during past operation. For this purpose, a 1.0 La acceptance critorion is sufficient. To combine an "as found" leakage requirement with a .75 La acceptance criterion would result in double-counting.

Again, the Commission itself has recognized this. The 1986 proposed rule would expressly require a determination of "as found" Type A leakage but would adopt an acceptance criterion of 1.00 La for the "as left" condition. 51 Fed. Reg. at 39543. The Commission, thereiore has clearly recognized the inherent inconsistency of requiring a determination of the "as found" Type A condition

, with Appendix J's current .75 La criterion. This excess conservatism has already l resulted in an unnecessary failure of the ILRT at one McGuire unit.

In its October 29, 1987 denial, the Staff relied upon paragraph 4.2 of ANSI I N45.4-1972 as the regulatory basis for its position. As the Staff stated, that i paragraph simply provides "that, for periodic CILRTs, a test be conductei before- l any preparatory repairs are made to the containment boundary in order to disclose the normal state of the containment." However, for the reasons explained above, it is not at all clear that paragraph 4.2 has been incorporated by reference into Appendix J since it does not relate to "test methods" within the ::eaning of Section III.A.3 of Appendix J. Paragraph 4.2 concerns pretest requiremeu s and sequence of testing. Indeed, paragraph 4.2 uses terms such as "initial record proof test" and "preparatory repairs" which are not defined and seem to conflict ,

with the mere specific Appendix J provisions. In short, the provisions of  ;

Section 4.2 of the ANSI standard are vague and confusing and can hardly be '

consideree a clear regulatory basis for the Staff position.

ll i Leakage Rate Testing of Containments of Light-Water-Cooled Nuclear Power Plants, 51 Fed. Reg. 39538, 39539 (October 29, 1986).

12 Final Report, Cost Analysis of Revisions to 10 CFR Part 50, Appendix J, Leak Tests for Primary and Secondary Containments of Light-Water-Cooled Nuclear Power Plants, dated April 1985, at 30.

U. S. Nuclear Regulatory Commission June 20, 1988 Page 8 In summary, the Staff's position that Type A test results must be adjusted for Type B and C tests, including necessary repairs and adjustments, prior to initiation of the Containment Inspection is inconsistent with Appendix J and previous 6taff positions applicable to Duke facilities. It should therefore be identified as a backfit. Nevertheless, as noted above Duke commits to continue its practice of supplying information on such Type B and C tests as a supplement to its ILRT reports.

CONCLUSION Duke is confident that its integrated leak rate testing progrsm complies with applicable requirements and is adequate to ensure containment leakage intepf.ity.

As shown by our past practices and the commitments herein, Duke has been willing to go beyond the express requirements of Appendix J to provide additional information to the Staff in a spirit of cooperation. However, we believe our previous backfitting claim was fully justified and ask that the Executive Director for Operations reverse the Staff denial and grant the claim.

Very truly yours, t&

Hal B. Tu ker

\#

PGL/29/sbn xc: Dr. J. Nelson Gt s i. Regional Administrator U. S. Nuclear '.egulatory Commission Region II 101 Marietta Street, NW, Suite 2900 Atlanta, Georgia 30323 Dr. K. Jabbour Office of Nuclear Reactor Regulation U. S. Nuclear Regulatory Commission Washington, D. C. 20555 Ms. Helen Pastis Office of Nuclear Reactor Regulation U. S. Nuclear Regulatory Commission Washir.gton, D. C. 20555 Mr. Darl Hood Office of Nuclear Reactor Regulations U. S. Nuclear Regulatory Commission Washington, D. C. 20555 Mr. P. K. Van Doorn NRC Resident Inspector Catawba Nuclear Station

U. S. Nuclear Regulatory Commission Junn 20, 1988 Page 9 Mr. W. T. Orders NRC Resident Inspector McGuire Nuclear Station Mr. P. H. Skinner NRC Resident Inspector Oconee Nuclear Station Mr. Victor Stello, Jr.

Executive Director for Operations U. S. Nuclear Regulatory Commission Washington, D. C. 20555 Dr. Thomas E. Murley Director, Office of Nuclear Reactor Regulation U. S. Nuclear Regulatory Commission Washington, D. C. 20555 l

I l

l l

!