View source for B12071, Requests Temporary Exemption from Single Failure Requirements Applicable to Small Break LOCAs Per 10CFR50.12. Exemption Would Apply Only to Proposed Interim Response Mechanism

{{Adams
| number = ML20204A362
| issue date = 04/22/1986
| title = Requests Temporary Exemption from Single Failure Requirements Applicable to Small Break LOCAs Per 10CFR50.12. Exemption Would Apply Only to Proposed Interim Response Mechanism
| author name = Opeka J
| author affiliation = CONNECTICUT YANKEE ATOMIC POWER CO.
| addressee name = Grimes C
| addressee affiliation = NRC OFFICE OF NUCLEAR REACTOR REGULATION (NRR)
| docket = 05000213
| license number = 
| contact person = 
| case reference number = TASK-2.K.3.03, TASK-2.K.3.05, TASK-2.K.3.30, TASK-2.K.3.31, TASK-TM
| document report number = B1207, B12071, NUDOCS 8605120121
| package number = ML20204A363
| document type = CORRESPONDENCE-LETTERS, INCOMING CORRESPONDENCE, UTILITY TO NRC
| page count = 18
| project = 
| stage = Other
}}

=Text=
{{#Wiki_filter:3 0 0 CONNECTICUT YANKEE ATOMIC POWER COMPANY BERLIN CONNECTICUT PO BOX 270   H AHT FOHD. CONNECTICUT 06141-0270 T E LE PuckE 2oses-sooo                                                A Pril 22,1986 Docket No. 50-213 B12071 Of fice of Nuclear Reactor Regulation Attn:        Mr. C. I. Grimes, Director Integrated Safety Assessment Project Directorate Division of PWR Licensing - B U. S. Nuclear Regulatory Commission Washington, D. C. 20555

==References:==
(1)  J. F. Opeka letter to Christopher 1. Grimes, "Haddam Neck Plant Probabilistic Safety Study - Summary Repor t and Results," dated March 31, 1986.
                         '2)  3. F. Opeka !ctter to Christopher 1. Grimes "lladdam Neck Plant - Probabilistic Safety Study LOC A Analysis," dated April 10,1936.
Gentlemen:
Haddam Neck Plant Request for Temporary Exemption from Single Fai!ure Requirements - Probabilistic Safety Study LOC A Analyses Further to Reference (2), and pursuant to 10 CFR 50.12, Connecticut Yankee Atomic Power Company (CYAPCO) hereby requests a temporary exemption from the single failure requirements applicable to pmall break loss of coolant accidents (SB LOC A's) at the Haddam Neck Plant.ll) The exemption would apply only to a proposed interim response mechanism to certain SB LOCA's pending completion of additional analyses to identif y and establish a permanen t resolution of the issue. The Staff will be kept informed of the status of CYAPCO's in <e gation regarding a permanen t resolution and schedule for implementation.
(1)        As discussed below, single failure requirements applicable to Emergency Core Cooling System (ECCS) analyses are imposed by General Design Criterion (LOC) 35, " Emergency Core Cooling," and by the "loterim Acceptance Criteria for Emergency Core Cooling Systems for Light Water Power Reactors" (I AC), 36 Fed. Reg.12247 (June 29,1971), Appendix A.
Without conceding its applicability to Haddam Neck, CYAPCO requests this exemption urvler GDC 35 to facilitate Staff review.
(2)        Haddam Neck is currently down for refueling.                CYAPCO anticipates                      _.
restarting the f acility (ig, attain criticality) the week of April 21, 1936.                 ir)g' CYAPCO will continue to keep the Staff verbally informed regarding a I
more precise schedule. The next refueling outage is scheduled for the summer of 1937.                                                                             J
                                                            \                                          (Of 8605120121 860422                                                                     ' f PDR    ADOCK 05000213 P                     PDR

f 2
BACKGROUND in Reference (1), CYAPCO submitted a Probabilistic Safety Study (PSS) in conjunction with the Integrated Safety Assessment Project (ISAP) for the Haddam Neck Plant, which included best estimate LOCA analyses for that facility. Through these analyses, CYAPCO identified a small range of break sizes in one loop of the reactor coolant system (RCS) for which safety injection flow in the high pressure recirculation mode may be insufficient to provide adequate core cooling absent modification of the facility emergency operating procedure and/or system (e.g., valve) re-alignments. As explained herein, CYAPCO had not previously identified the consequences of such accidents because the Commission-approved ECCS evaluation model used at Haddam Neck did not evaluate this precise scenario.
It is important to recognize that CYAPCO has heretofore satisfied NRC requirements applicable to ECCS analyses, in fact, only because of CYAPCO's voluntary performance of in-depth state-of-the-art LOCA analyses as part of the plant-specific PSS for Haddam Neck, was the subject condition identified. The ECCS evaluation model approved in the IAC, which serves as the licensing basis for the Haddam Neck plant, coupled with accepted analytical assumptions regarding the need to model the long term recirculation mode, was essentially insensitive to break location for the breaks at issue here and, thus, did not identify the breaks recently determined through the PSS to pose a potential challenge to ECCS response capabilities. Of course, these facts do not temper the need to address this issue. They do, however, underscore that ISAP-related ef forts can and will contribute to improved plant safety and protection of the public health and safety.
In view of the above, CYAPCO has elected to implement measures to demonstrate continued compliance with the Interim Acceptance Criteria (IAC) for Emergency Core Cooling Systems (3) even though the probability of the subject breaks occurring is extremely remote, and the code used in the PSS which identified this condition has not been formally approved by the NRC Staff.
CYAPCO has already initiated an interim response mechanism, and is exploring possible long-term measures, to assure adequate core cooling will be maintained in accordance with the LAC. The subject breaks, and a possible permanent response, are described in Attachment 1.
l l
(3)            See Fed. Reg.12247 (June 29,1971). The Interim Arceptance i                  Criteria govern ECCS performance at the Haddam Neck plant because the operating license was issued prior to December 28, 1974, and, with the exception of four zircaloy clad Lead Test Assemblies (LTA's), only stainless steel clad fuel is used at the f acility. (The NRC previously approved the use of the zircaloy assemblies by letters dated October 2 and 15,1984.).
i t
J

i As described in Attachment 1 (and Reference (2)), the interim measures undertaken by CYAPCO involve revising the Haddam Neck emergency operating procedure regarding a LOCA to provide an alternate flow path for core cooling during the high pressure recirculation mode. (A copy of the revised procedure is provided with Attachment 1.) The revision provides for the use of the HPSI pumps to provide long-term recirculation in the event of a SB LOCA of the size for which the CYAPCO analysis shows the present mode of high head recirculation (using charging pumps) could be inadequate. To line up the HPSI pumps for this recirculation path, certain valves require realignment under specified conditions. Two of those valves, both outside containment, do not satisfy the single failure criterion.      Accordingly, CYAPCO has established measures to provide added assurance of valve operability (a conforming license amendment to Haddam Neck Technical Specifications will be requested shortly (g Attachment I at 4)). These measures assure there will be no undue risk to the public health and safety. Nonetheless, CYAPCO requests a temporary exemption from the single failure requirements for these valves pending implementation of permanent response measures.
REQUEST FOR EXEMPTION Regulatory Provisions CYAPCO requests a temporary exemption from NRC requirements concerning consideration of single failures in ECCS evaluations.           Provisions requiring consideration of single failures in this context are set forth both in GDC 35 and the Interim Acceptance Criteria. GDC 35 provides in applicable part as follows:
Suitable redundancy in components and features .... shall be provided to assure that for on'.ite electric power system operation (assuming offsite power is not available) and for olfsite electric power system operation (asmJng onsite power is not available) the system safety function can be accomplished, assuming a single failure.
(10 CFR Part 50, Appendix A, Criterion 35.)
As a general proposition, the General Design Criteria are themselves not directly applicable to Haddam Neck.(4) However, the Interim Acceptance Criteria provide with respect to, intei alia, the Westinghouse ECCS Evaluation Model to (4)    Haddam Neck received a provisional operating license on June 30, 1967, and a full term operating license (FTOL) on December 27, 1974. The General Design Criteria set forth in 10 CFR Part 30, Appendix A became effective May 21, 1971, and were not made formally applicable to plants previously licensed such as Haddam Neck. The Staff has however, assessed Haddam Neck against the intent of prcposed General Design Criteria, published July 11, 1967, and found the riant satisfied the intent of those criteria (s_ce e  FTOL Safety Evaluation Report for Haddam Neck Section 3.4, dated July 1,1971).
J

e .
which Haddam Neck was originally evaluated, as follows:
The combination of systems used for analyses should be derived from a failure mode and ef fects analyses, using the single failure criterion.
(Interim Acceptance Criteria for Emergency Core Cooling Systems for Light Water Power Reactors,~ Appendix A, Part 3,36 Fed. Reg. at 12249.)
Thus, whether by GDC 35 or the IAC, a single failure requirement is imposed on ECCS evaluations of light water power reactors, including Haddam Neck.
Accordingly, CYAPCO seeks an exemption from the single failure criterion as it applies to ECCS evaluations. For administrative purposes, CYAPCO has framed this as a request for exemption from GDC 35. In the event the Staff may view this request as also (or alternatively) requiring any exemption from the IAC, CY APCO asks that this request be treated in such manner.
This exemption is requested specifically with respect to two valves, both outside containment, which would be used under procedurally defined conditions to respond to a SB LOCA. Because the valves are not single failure proof, CYAPCO has taken measures to assure operability of these valves. (See Attachment I at 4.)
CYAPCO submits that in these circumstances, and for the reasons discussed below, the requested exemption is authorized by law, provides adequate protection of the public health and safety and is consistent with the common defense and security.
No Undue Risk to Public Health and Safety The proposed exemption satisfies the requirements of 10 CFR 50.12(a)(1) in that it does not present an undue risk to the public health and safety. Specifically, granting the proposed exemption is consistent not only with the intent of the IAC and single failure requirements, but is consistent with the goals of the Integrated Safety Assessment Program, in connection with which CYAPCO conducted the PSS that identified the range of SB LOCA's at issue here. In addition, because this request seeks only a temporary exemption, the incremental risk which may be associated with utilizing a non-single failure proof response mechanism is significantly less than would be involved if a permanent exemption were requested. Importantly, probabilistic analyses demonstrate that the overall level of protection to the public health and safety is actually increased by the implementation of this response.
Regulatory Purposes Served: Authorizing the temporary use of the proposed response mechanism is consistent with the Commission's policy set forth in the IAC regarding consideration of new technology app!!cabic to ECCS performance.
The IAC reflected the results of the Commission's review of then-available information relevant to establishing the basis for accepting ECCS designs for light-water reactors (s_ce c  36 Fed. Reg. at 12247). The Commission specifically noted that the IAC were formulated "on the basis of today's knowledge". In applying the use of that information, however, the Commission noted that
      " improvements are expected in analytical techniques, and experimental programs are expected to provide increased and improved knowledge about ECCS J

performance." The Commission stated that it anticipated modifying the criteria from time to time to reflect such new information. (36 Fed. Reg. 12248).
Further, on at least one occasion, the Commission indicated that it expected its Staff to consider new technology applicable to ECCS evaluations as it became available.l5)
It is apparent that the Commission expected new technology concerning ECCS performance to be considered as it was developed. This expectation would, of course, be furthered by permitting reasonable application of new technology such as the ECCS evaluations employing sta te-of-the-ar t analytical techniques utilized in the PSS for Haddam Neck, i.e., the NULAP 5 computer code.
CYAPCO submits that its proposed approach to addressing this matter, including implementation of the measures which give rise to the instant request for temporary exemption, is reasonable and consistent with the Commission's intentions to integrate new technology in ECCS performance evaluations.
CYAPCO recognizes, however, that the NULAP-3 code has not been formally approved by the NRC. Thus, the proposed interim response mechanism does not itself rely on the NULAP-5 code to demonstrate acceptable ECCS performance.
Rather, as described in Attachment 1 (at 4), it is apparent by inspection that the use of the HPSI pumps in the recirculation phase would yield acceptable results in the IAC model. Significant increases of flow delivered to the core occur using the HPSI pumps (as contrasted with the charging pumps assessed in the IAC model).
Further, the proposed response to the identified SB LOCA, although not strictly satisfying the single failure criterion, is consistent with the regulatory intent of the single failure requirements. Those requirements have served as a licensing review tool to assure system reliability as one element of the defense in depth approach to reactor safety.          The Staff has acknowledged that use of "probabilistic methods of the type used in the Reactor Safety Study will gradually come into increasing use and supplement the " Single Failure Criterion" (NRC Staff Information Report," Single Failure Criterion," SECY-77-439 (August 17,1977) at 1). Consistent with this consideration, and as discussed more fully below and in Attachment 1 (at 5-6), CYAPCO compared the total annual core melt frequency (CMF) associated with small and medium break LOCA's at Haddam Neck assuming implementation of the proposed response option (use of HPSI pumps during recirculation even in the absence of a single failure proof flow path), to the total annual CMF without that response option but including the existence of the adverse break. CYAPCO determined that the total CMF would be reduced by a minimum of 27% Further, even if one were to assume the original response mechanism (use of charging pumps) was adequate with respect to the adverse break, the addition of the HPSI pump option decreases core melt frequency by 16% in summary, even though the interim response is not single failure proof, it provides greater assurance of the ability to mitigate (3)    See in the Matter of:       R ulernaking Hearing, Acceptance Criteria for Emergency Core Cooling Systems for Light Water Cooled Nuclear Power Reactors,6 AEC 1085,1123(1973)(Commissioner Ariders, concurring).
J

e ..
                                                                                                                                 . (, .
1.OCA's than previously utilized response mechanisms which were found to satisfy applicable regulatory criteria whether or not those mechanisms provided the level of protection originally assumed to exist. Consistent with the position described in SECY-77-439, this is an appropriate factor to consider in assessing the adequacy of the interim response.
Finally, this request for a temporary exemption is consistent with the intent of the Integrated Safety Assessment Program (ISAP) as it applies to the Haddarn Neck Plant. In its letter of July 31,1935 (from H. L. Thompson (NRC) to J. F.
Opeka (Northeast Utilities)), at 3-4, the Staff indicated that for certain issues to be addressed in ISAP, there may be a need to request exemptions, justifying continued operation and describing compensatory plant design features which provide a basis for continued plant operation, pending a decision with respect to a more ef fective resolution and establishment of an implementation schedule.
The Staff indicated it would act on such requests promptly.
Period of exemption: CYAPCO requests this exemption only on a temporary basis.                                     CYAPCO is presently performing analyses to identify permanent measures to address such breaks, CYAPCO will appraise the Staf f of the results of these analyses and schedule for implementation as soon as practical. Our objective is to implement a permanent resolution by not later than the end of the 1937 refueling outage. In view of the temporary nature of the exemption, and the existence of appropriate response measures in the interim, it is clear that no undue risk to the public health and safety is presented by this request.
In summary, CYAPCO's identification of the subject breaks and implementation of the proposed interim response mechanism satisfies the intent of Commission re6ulations regarding ECCS evaluations, single failure criterion, and ISAP evaluations. Further, the exemption is temporary and poses no incremental safety concerns. Thus, the requested exemption poses no undue risk to the public health and safety.
Special Circumstances As demonstrated below, special circumstances, as defined by 10 CFR 50.12(a)(2),
exist for granting the requested exemption.                                                          First, pursuant to 10 CFR 50.12(a)(2)(iv), the requested exemption benefits the public health and safety, compensating "for any decrease in safety that may result from the grant of the exemption." Second, as provided in 10 CFR 50.12(a)(2)(s), the exemption seeks only temporary relief and CYAPCO has made good faith ef forts to comply with the regulation. Granting the exemption is justified on either or both of these grounds.
Benefit to the public health and safety: Benefit to the pub!!c health and safety may be measured in several ways. One measure of such benefit is whether the proposal results in an overall reduction in the probability of adverse consequences frorn reactor operation that could af fect the public and health and safety, in this instance, the original high head recirculation mode using charging pumps has been found to be deficient for a narrow spectrum of breaks, whereas the safety benefits derived from utilizing the HPSI pumps represent a credit for 2

i a much broader range of postulated breaks. Therefore, implementation of the proposed interim response measure (use of HPSI pumps during recirculation) measurably decreases the overall core melt frequency associated with small and medium break LOCA's at Haddam Neck.
As discussed in Attachment 1 (at 5-6), CYAPCO evaluated the change in core melt frequency due to utilizing a single HPSI punip during the recirculation mode following a LOCA event. The Haddam Neck PSS event tree models for medium and small break LOCAs (submitted by Reference (1)) were modified to reflect the use of a HPSI pump in conjunction with the RHR System for high pressure recirculation. High pressure recirculation using HPSI pumps and using charging pumps result in a redundant means of providing recirculation flow except for the adverse breaks in Loop 2 or in the charging iiiection line. For these breaks, recirculation using HPSI was considered to be the only means of providing high pressure recirculation flow. Assuming these chinges, a reduction in the Haddam Neck CMF of approximately 27% was calculated. The revised Haddam Neck CMF ip value.)t6)3.93 E-4/ year, versus the 5.43 E-4/yr calculated in the PSS (mean Further, CYAPCO also calculated the Haddam Neck CMF assuming no adverse breaks and using only charging for high pressure recirculation, i.e., the original design conditions assumed to exist prior to discovery of the adverse breaks.
Even under these assumptions, the CMF was calculated to be 4.74E-4/yr,16%
less than the CMF originally calculated in the Haddam Neck PSS. Therefore, the proposed modifications to use either HPSI or charging for high pressure recirculation results in a reduction of core melt frequency even conservatively assuming no adverse breaks existed with the original design. (Attachment I at 6.)
It should be noted that the above probabilistic evaluation examined only the small break and medium break LOCA events. The potential benefit from use of HPSI pumps for high pressure recirculation following a consequential LOCA (stuck open unisolated PORV) or following feed and bleed operation was not evaluated but would in all cases lead to an additional reduction in core melt frequency. The 27 and 16 percent reductions should, therefore, be viewed as minimums.
In addition, CYAPCO has taken certain independent measures to provide added assurance that the breaks of concern will not occur in the first-instance, and that the two valves subject to this request will operate as required. CYAPCO has not only conducted ultrasonic examination of welds in the region of piping of concern and performed testing of these valves for operability, but has initiated routine surveillance and operability checks of these valies to be performed as long as this method of core cooling is employed. Each of these measures are described in Attachment 1.
(6)   Note that consideration of HPSI pumps as the only means of providing high pressure recirculation could have a negative impact (increase in core melt frequency). This is due to the fact that HPSI in the recirculation mode is less reliable than charging in the recirculation mode, thereby having a negative impact on LOCA events in loops 1,3, and 4.

3 In summary, the interim procedure to use HPSI pumps in the recirculation phase benefits the public health and safety. Thus, approval of an exemption request pursuant to 10 CFR 50.12(aX2)(iv) is warranted. Of course, CYAPCO recognizes that even further improvement in the protection of the public health and safety may be achieved by implementing measures consistent with the single failure criterion. Accordingly, as already indicated, CYAPCO is pursuing identification and implementation of permanent measures which satisfy all relevant licensing requirements to address this range of breaks.
Temporary Relief / Good Faith Efforts:         The instant exemption request also satisfies 10 CFR 50.12(aX2Xv), which authorizes exemptions providing temporary relief from a particular regulation where the applicant or licensee has made good faith efforts to comply with the regulation.      As previously noted, CYAPCO is seeking only a temporary exemption. Further, as discussed below, CYAPCO has made more than good faith efforts to assure satisfaction of IAC requirements.
IrAed, the breaks presently being addressed were identified by analyses wtuntarily undertaken by CYAPCO which go beyond those required by the IAC.
Thus, this request is also authorized by 10 CFR 50.12(a)(2Xv).
Regarding satisfaction of the IAC, CYAPCO performed, and received Staff approval of, all analyses required by the IAC. CYAPCO submitted to the Commission on December 30,1971, and in a response to Commission questions in a {{letter dated|date=May 19, 1972|text=letter dated May 19, 1972}}, evaluations of ECCS performance using analytical techniques expressly authorized by the IAC. These analyses, performed by Westinghouse for CYAPCO, used IAC-approved analytical techniques. These techniques included applying the SLAP computer code (which calculates the transient depressurization of the reactor coolant system and describes the mass and enthalpy flow through a break). Such analyses represented the best available analytical tools then available for evaluating ECCS performance. Use of these analyses by CYAPCO in and of itself constitutes good faith elforts to comply with the subject requirements. Indeed, the Commission previously approved the IAC analyses originally performed for Haddam Neck.
Further, it is significant that the analytical methods previously employed were incapable of identifying the present concern. In the first instance, the SLAP code analytically represents the entire reactor coolant system (RCS) as two regions, and all four steam generators with one region. Because of the simplicity of the model, it was not sufficiently refined, absent modification, to evaluate system response sensitivity to the precise break locations at issue. (Nonetheless, the analyses presented at that time ({{letter dated|date=May 19, 1972|text=May 19,1972, letter}}) ostensibly considered a spectrum of breaks in the cold leg, from 0.75 inch (0.003 f 2t ) to 6 inch (0.02 f t2 )
equivalent diameter. The review extrapolated the response to smaller break sizes.)    in addition, as was the common practice with such evaluations, the analyses were terminated af ter the fluid level in the reactor vessel recovered and the core appeared stabilized. In all cases presented, the core was recovered and appeared stable during the injection phase of the accident, when injection was by the HPSI pumps taking suction from the refueling water storage tank.
Consequently, long term decay heat removal was simply evaluated by comparing the delivery capability of the recirculation pumps with that required to remove decay heat. In any event, for the reasons explained above, the fact that the high

9 head recirculation mode utilized a single injection path would not have had an impact on an analysis using the SLAP code .tven if the analysis had been performed out to the recirculation phase.(7)
In contrast to the above, the pn:sent SB LOCA condition was identified as a result of CYAPCO's own ef forts. CYAPCO has an ongoing ef fort to upgrade accident analyses for the Haddam Neck Plant. This upgrade involves reanalyses of large and small break LOCA events, and non-LOCA events, including analyses performed in connection with the response to TMI Action Plan Items II.K.3.5, II.K.3.30 and II.K.3.31.(3) CYAPCO identified the ECCS system sensitivity to breaks in the loop 2 cold leg or the charging line during the recirculation phase of a SB LOCA in the recently completed PSS for the Haddam Neck Plant (Reference (1)). It is apparent, therefore, that only as a result CYAPCO's good faith elforts was the present condition even identified.(9)
In summary, CYAPCO performed in good faith ECCS evaluations to evaluate Haddam Neck conformance to IAC requirements. The analytical techniques employed were the best then available and were expressly approved by the Commission. However, those analyses were inherently incapable of identifying the present concern and, in any event, then-accepted ECCS analytical assumptions did not indicate a need to analyze long term system response. These breaks were identifed as part of a compreshensive reevaluation of LOCA analyses performed on CYAPCO's own initiative. These circumstances clearly demonstrate CYAPCO's good faith efforts to satisfy ECCS performance requirements, including the IAC. Accordingly, this request is autho:ized by 10 CFR 50.12(a)(2)(v).
(7)    CYAPCO's assessment of the SLAP code is premised on the description thereof in the {{letter dated|date=May 19, 1972|text=May 19, 1972, letter}}. A more comprehensive review was not possible because Westinghouse has informed CYAPCO that the IAC analytical codes used at Haddam Neck are no longer available, in any event, those codes do not reflect the best available technolog".
(8)    Descriptions of the small break LOCA calculation methods and applications of these methods to TMI Action Plan Items !!.K.3.5, ll.K.3.30, and II.K.3.31 have been submitted to the NRC for review and approval. See CYAPCO letter from W. G. Counsil to W. A. Paulson, dated August 23, 1934; CYAPCO letter from W. G. Counsil to 3. A. Zwolinski, dated November 17, 1934; and CYAPCO letter from W. G. Counsil to 3. A. Zwolinski, dated December 20, 1984.
(9)    It should be noted that in view of the significant advances in analytical techniques, and more comprehensive analyses, it is not surprising that CYAPCO would identify concems not identified in earlier, less sophisticated analyses. Of particular note is the degree to which the small break LOCA model utilized in support of the PSS, employing the NULAP-5 computer code, is more sophisticated than the original analytical model.
For instance, this state-of-the-art computer code represents the RCS with 39 nodes and the steam generators with 18 nodes. In contrast, the SLAP code utilized a two and one node approach for the RCS and steam generator models, respectively.

Related Staff Approvals In the course of its normal review cycle for this matter, CYAPCO identified certain actions necessary to implement the revised procedure which require amendments to the Haddam Neck Technical Specifications. These amendments are discussed in Attachment 1. They will be submitted for Staf f approval             !
shortly. They involve (1) changes to the surveillance and inspection provisions related to the two valves for which this exemption is requested (also discussed above), (2) revisions to the operability criteria conceming containment high range radiation monitoring, and (3) establishment of alternatives to monitoring RCS . system pressure.       (An Administrative Technical Specification change regarding items (2) and (3) will be made prior to restart.) Each of these measures will impose more stringent requirements than currently exist and do not require modification of existing equipment. Thus, Staf f approval of these amendments will not be required prior to restart. Nonetheless, we will request that the Staf f act on these requests promptly when received.
Part 100 Compliance As discussed in Attachment 1 (at 5), and as indicated to be under review in Reference 2 (at 5), CYAPCO has assessed the possible implications of the use of the HPSI recirculation option for satisfaction of Part 100 criteria. CYAPCO recognizes that if one were required to assume worst <ase fuel failure for this SB LOCA, as is assumed for large break LOCA's, release paths creating the potential for exposures in excess of Part 100 radiation limits could exist in that the HPSI recirculation option involves reci culation of primary fluid in systems outside containment which were not previously analyzed for those conditions.
However, such an assumption would require postulation of non-mechanistic scenarios with consequences beyond the e resulting from non-mechanistic assumptions made for purposes of Part 100 evaluations associated with large break LOCA analyses. CYAPCO does not consider such assumptions to be required or appropriate. The analyses performed with respect to this scenario demonstrate that not only is such an accident highly unlikely, byt that the proposed response mechanism precludes the possibility of fuel failure.tl0)
(10) In this regard the Staff should note that in contexts analogous to SB LOCA analyses the Staff has accepted use of analytical determinations of the degree of potential fuel failure to serse as the basis for excluding consideration of consequences beyond the design basis assumptions for Part 100. Specifically, the Sttif has accepted such analyses for Haddam Neck in the context of both control rod ejection and steam generator tube failure accidents. Such an approach is consistent with the methodology described in NUREG-0800, the Standard Review Plan. (See eg, Chapter
             !$.4.8 (at 15.4.8-6, 7) and Chapter 15.6.3 (at 15.6.3-4).) Similarly, in the context     of equipment qualification determinations pursuant to 10CFR50.49, it is not necessary to postulate conditions associated with maximum fuel failure where, by analysis, such conditions are shown not to exist. In summary, not requiring postulation of maximum source term conditions for purposes of analyzing the subject SB LOCA's against the requirements of 10 CFR Part 100 is consistent with accepted regulatory I
practice.
1

  . o In any event,if the Staff considers additional non-mechanistic assumptions to be required in connection with the present SB LOCA event analysis, CYAPCO requests an exemption from the requirements of Part 100 to the extent deemed necessary. Such an exemption is warranted on the basis that the postulated event is highly unlikely to occur in the first instance, and even if it were to occur analyses have shown that fuel failure will not occur, and because the requested exemption would also be temporary.
CONCLUSION For the above reasons, the Staff should grant this request for temporary exemption. CYAPCO is, of course, prepared to meet with the Staff at its convenicace to discuss this request. Based upon previous discussions with the Staff, we understand that final Staf f action on this request is not required prior to restar t. In any event, CYAPCO requests that the Staff apprise us of the results of its review promptly to minimize any potential impact on startup.                                                                          ;
                                                                                                                                                           )
Pursuant to 10 CFR 170.21, CYAPCO encloses herewith a check for $150.00 to
                                                                                                                                                           ~
cover the fee for processing this request.
Very truly yours, CONNECTICUT YANKEE ATOMIC POWER COMPANY                                                                            l 1 F/1,JL 3.F.N3phka                                                           O Senior Vice President Attachment

o   .
Attachment 1 HADDAM NECK PLANT PROBABILISTIC SAFETY STUDY LOCA ANALYSES SMALL BREAK LOCA RESPONSE MEASURES INTRODUCTION In conjunction with the Integrated Safety Assessment Program (ISAP) for the Haddam Neck Plant, Connecticut Yankee Atomic Power Company (CYAPCO) performed a Probabilistic Safety Study (PSS) which included best estimate LOCA analyses for that facility. Through these analyses, CYAPCO identified a small range of break sizes in one loop of the reactor coolant system (RCS) for which safety injection flow in the high pressure recirculation mode may be insufficient to provide adequate core cooling absent modification of facility operating procedures and/or system (e.g., valve) re-alignments.
CYAPCO reported the identification of this potential condition to the NRC on March 25, 1986, pursuant to 10 CFR 50.72. A detailed description of the condition was transmitted with LER 86-013 on April 1,1986, pursuant to 10 CFR 50.73. By {{letter dated|date=April 10, 1986|text=letter dated April 10, 1986}}, CYAPCO transmitted additional information to the NRC Staff. In that letter CYAPCO described its response to the postulated breaks. CYAPCO provides additional information regarding its response below. Possible iong term response measures are also discussed.
BACKGROUND Northeast Utilities Service Company (NUSCO) recently completed, on behalf of CYAPCO, a plant-specific Probabilistic Safety Study (PSS) for the Haddam Neck Plant. The PSS included a best estimate LOCA analysis to be used to assess possible plant backfits in the context of the Integrated Safety Assessment Program for the Haddam Neck Plant. The PSS Summary Report and Results were transmitted to the NRC Staff on March 31, 1986.
The Haddam Neck PSS utilized current state-of-the-art computational and modeling techniques. The study included a comprehensive analysis of best estimate plant-specific LOCA and transient responses; plant-specific operating, reliability and maintenance data; and the latest pertinent plant design changes.
The PSS employed the NULAPS computer code in modeling LOCA and transient responses. The NULAP5 code is based on RELAP5 Mod I code which is a transient and hydraulic non-equilibrium blowdown and heat up code developed by EG&G of Idaho Falls, Idaho. NUSCO modified the code to apply the methodology to both ECCS best estimate and licensing analyses of the Haddam Neck Plant.(U (1)   CYAPCO originally submitted analyses using NULAP5 for NRC Staff review in conjunction with the analyses responding to TMI Action Plan Items II.K.3.5, II.K.3.30 and II.K.3.31. Those analyses were transmitted to the NRC by letter from W. G. Counsil to 3. A. Zwolinski, dated December 20, 1984.

2 CYAPCO is implementing certain revisions to its emergency operating procedure to establish an appropriate interim response to the subject range of break sizes, pending identification and implementation of a permanent resolution.            This interim response provides reasonable assurance that Haddam Neck will continue to satisfy the IAC peak cladding temperature limits.(2) Specifically, these measures assure adequate core cooling during the high pressure recirculation phase is prcvided for all postulated break sizes and locations to maintain peak cladding temperatures below the limits established by the IAC.(3) However, the interim response involves realignment of various valves, two of which are not single failure proof. Accordingly, CYAPCO is seeking a temporary exemption from the single failure requirements applicable to ECCS evaluations with respect to these two valves.
DISCUSSION Postulated Transients: A3 described in LER 86-013, the results of the PSS indicated that, with the exception of a small range of break sizes in the loop 2 cold leg reactor coolant pump discharge and the charging line, medium and small break LOCAs during the recirculation phase can be satisfactorily mitigated by the high and low pressure safety injection systems. Cold leg breaks of 0.02 to 0.045 f t 2 in the loop 2 discharge may result in insufficient safety injection flow in the high pressure recirculation mode to provide adequate core coolin Adequate core cooling is maintained for breaks smaller than or equal to 0.02 f tg because of the lower break flow rates and because the injection is sufficient to maintain the core in a covered condition.        Breaks larger than or equal to 0.045 f t2 depressurize rapidly enough so that low pressure recirculation can be activated, assuring required core cooling.
Breaks located directly in the charging line for sizes 0.038 f t2 and smaller (a specific minimum size was not predicted) may also produce results analogous to the loop 2 cold leg discharge breaks discussed above. For these breaks, the RCS pressure also remains above the shutoff head of the RHR pumps, thereby preventing activation of low pressure recirculation, while the charging flow provides insufficient flow to maintain core cooling.
(2)    The ECCS performance analyses for Haddam Neck were originally docketed in 1971. The NRC first determined that Haddam Neck satisfied the IAC in 1974. The Staff has also since approved several revisions and changes to the IAC-related analyses. Specifically, analyses were revised in 1973 to account for fuel densification, changes in the input parameters were made in 1977 (upper head fluid temperature) and again in 1982 (core inlet temperature). The Staff approved each of these revisions and changes.
(3)    The IAC specifies a peak cladding temperature of 23000F, whereas the acceptance criteria in 10 CFR 50.46 (applicable to zircaloy clad fuels and plants which received operating licenses subsequent to December 31, 1974) specifies a peak cladding temperature limit of 22000F. Except for four Lead Test Assemblies using zircaloy cladding, the use of which has been approved by the Staff, Haddam Neck employs stainless steel clad fuel, in any event, the proposed interim measures assure that both the stainless steel and zircaloy cladding temperature limits are satisfied.
s

   ,    , - - .                               . ~   .       -    -    . - .              -   -      _ -  _
3-All breaks in all locations, with the exception of the loop 2 cold leg at the pump discharge, demonstrate acceptable ECCS performance because the charging flow (injected into the loop 2 cold leg at the pump discharge) is able to maintain core cooling.
Transient Response: CYAPCO is implementing a revision to the Emergency Operating Procedure (EOP) for Haddam Neck to establish a means to provide adequate core cooling in the event of a SB LOCA. Under prescribed conditions indicative of a small break LOCA of the size and location for which CYAPCO's analysis shows charging high head recirculation to be inadequate, the revised procedure directs the operators to use the HPSI pumps to provide long-term recirculation in the event of a SB LOCA. A copy of the draft EOP revision is attached as Attachment A. Plant operators will be qualified on this revision consistent with plant procedures.
* As depicted in the attached flow chart, (Attachment B(4)), the revised EOP provides that in the recirculation phase, upon receipt of a specific RCS pressure i
indication, the HPSI pump suction will be lined-up to the RHR pump discharge.(5)
Above this pressure, adequate core cooling using the charging pumps is assured.
CYAPCO has determined that if two of the four HPSI cold leg injection valves are closed prior to starting the HPSI pump, an RHR pump aligned to the suction of the HPSI pump provides adequate NPSH for operation of the HPSI pump down                         i to an RCS pressure of 0 psig. In addition, to assure that the HPSI pump will not be dead headed in the recirculation mode after the HPSI pump mini-flow line has been isolated, the revised procedure prevents dead heading of the HPSI pump by opening a PORY prior to starting the HPSI pump. _ It will be necessary to maintain the PORV open until the HPSI pumps are turned off and flow can be provided m the RCS via RHR pumps and the core deluge pathway (low pressure
* recirculation). Low pressure recirculation conditions will be achieved within a maximurn of 10 hours with at least one PORY open. The air supply to the PORVs is designed such that a PORV can be maintained open for 30 hours. Therefore, the revised procedure precludes operation of the HPSI pumps in a mode which could potentially result in pump degradation or damage.
The current recirculation method (use of the RHR pumps to deliver containment sump water to the charging pumps) is adequate for all other LOCAs. It is,
'                              therefore, the preferred approach and will continue to be used in all instances                      ,
unless the above pressure indication is present. Furthei, as will be discussed below, a back-up flow path is identified in the revised EOP. CYAPCO included this path to assist the operators in the event several unlikely events were to (4)      Attachment B supercedes the flow chart attached to CYAPCO's {{letter dated|date=April 10, 1986|text=April 10, 1986, letter}}.
(5)      On April 9,1986, CYAPCO notified the Staff pursuant to 10CFR50.72 regarding wide range pressure measurement uncertainties for the Haddam Neck Plant. The instrument uncertainties could affect operator response to LOCAs.           CYAPCO has evaluated this condition and identified appropriate alternative measures. These measures will be described in a forthcoming LER. These measures are already incorporated in the attached revision to the EOP.
4
     -- r           - - - - -        y       ,,,,c-   w wa    ,,   ,-     -r w ,-rm-----   -  - -              ,,  -
                                                                                                                       <    m

                                                    .tp occur, consistent with the philosophy that emergency operating procedures should identify available alternatives in order of priority. CYAPCO does not rely on this path to support its exemption request.
CYAPCO has made every effort to assure the revised procedure addresses applicable licensing requirements. The principal issues addressed may be categorized into three areas. First, CYAPCO addressed whether the selected flow path provides adequate flow to maintain required core cooling. Second, CYAPCO has identified and implemented certain measures to provide further assurance of valve operability and that the probability of the event occurring in the first instance is remote. Third, CYAPCO evaluated potential measures to provide assurance that there are no adverse radiological consequences which result from the use of the HPSI recirculation option.
Adequate Core Cooling: As already noted, the analyses which identified the SB LOCA condition at issue here were performed in the context of the PSS for the Haddam Neck Plant. The PSS employed state-of-the-art analytical techniques using the NULAP5 computer code. This code is based on the RELAPS Mod I code which is a transient and hydraulic non-equilibrium blowdown and heat up code developed by EG&G of Idaho Falls, Idaho. The code was modified to apply the methodology to both best estimate and ECCS licensing analyses for the Haddam Neck Plant. CYAPCO recognizes that the NULAP5 code has not received NRC approval as a licensing basis code. Accordingly, CYAPCO has assured itself that the measures selected to respond to the range of SB LOCAs of concern would satisfy approved licensing basis ECCS evaluation methodologies.
To this end, CYAPCO compared the coolant flow rates of the proposed HPSI recirculation scheme and the existing recirculation mode (using the charging pumps) which had been assessed in the original ECCS analysis. By inspection, it was clear that the HPSI recirculation mode provided significantly greater flow to the core than had previously been accepted as providing adequate core cooling.
CYAPCO concluded, therefore, that this flow path would satisfy its existing approved licensing basis ECCS performance model.
Single Failure Requirements: Prior to initiating HPSI pump recirculation it is necessary to close one valve to prevent contaminated sump water from being recirculated back to the RWST, and to open another valve to allow water pumped by the RHR pumps to reach the HPSI pump suction. Single failure of either valve would preclude the use of the HPSI pump for high head recirculation.
To provide reasonable assurance that these valves will operate, CYAPCO will impose monthly surveillance and cycling requirements on both valves. At present,   the technical to be locked              sp(ecifications in position.              require the manual operators on both valves
: 6) The breakers     to the motor operators on each valve are also to be locked open. This assures these valves are lined up for the injection phase of ECCS. To implement the HPSI recirculation lineup, the valve (6)    A proposed amendment to Haddam Neck Technical Specifications regarding these measures will be submitted shortly. Initial surveillance and cycling has already been performed. The valves were cycled remotely by the motor operators. Both valves functioned properly. CYAPCO intends to seek expedited approval of this request.
1

from the RWST to the HPSI suction (MOV-24) must be closed (presently locked open) and the valve from the RHR discharge to the HPSI suction (MOV-374) must be opened (presently locked closed). (See Attachment C) Either valve may be repositioned remotely upon manually unlocking the breaker to allow use of the motor operators. Either valve may also be repositioned manually upon unlocking the manual operators. Sufficient time exists following receipt of the previously mentioned indications requiring switchover to the HPSI recirculation option to allow the operator actions.
Radiological Protection: Because the HPSI recirculation option involves the use of pumps and piping outside containment, CYAPCO has also considered possible radiological implications of using the HPSI pumps to respond to this transient.
The design basis accident analyses for the Haddam Neck Plant indicate that no fuel failures are expected for small break LOCAs prior to switching to the recirculation mode. The occurrence of fuel failure prior to or during the recirculation phase of small break LOCAs would require consideration of equipment failures which go beyond the design basis assumptions (e.g., failure of the two valves for which the instant exemption is sought). (Severe fuel failure has been considered in connection with large break LOCA's for purposes of Part 100 evaluations. These assumptions are worst-case assumptions which form the design basis for the plant.) Catastrophic fuel failure during these SB LOCA accidents are not required to be assumed unless mechanistic evaluations so demonstrate. Should the Staff disagree, CYAPCO requests an exemption to the extent necessary under 10 CFR Part 100.
In any event, CYAPCO assessed this recirculation path for potential leakage of primary coolant from this flow path. :The potential sources of leakage would be the HPSI pumps themselves and MOV .'i. A leakage test performed on the pump found no detectable leakage. As for the MOV-24 valve, although difficult to test precisely, a test which pressurized the HPSI pump suction with the RHR pumps was run. The test would have shown if there was gross leakage from MOV-24.
No leakage was apparent.
Probability of Events:
Using the methodology described in the Haddam Neck PSS, CYAPCO has determined the frequency of the breaks at issue here. The frequency of these breaks was estimated by multiplying the (generic) frequency of various size LOCAs times the conditional probability that a break of a given size occurred in the adverse location. This conditional probability was derived by totalling up the number of pipe segments in the af fected location and dividing by the total number of pertinent pipe sections of the given size range throughout the RCS. A
: pipe segment is defined as a section of pipe between major discontinuities (e.g.,
l valies, reducers, etc.). The definition of a pipe segment and the methodology used for quantification is consistent with WASH-1400.
CY APCO evaluated the specific change in core melt frequency associated with
          ' the present revision. The Haddam Neck PSS event tree models for medium and small break LOCAs were modified to re flect the use of a HPSI pump in i         .
                                                                                   ;       *E 9 4 h

s 6-conjunction with the RHR System for high pressure recirculation. High pressure recirculation using HPS! or charging pumps were considered to be redundant means of providing recirculation flow except for the adverse breaks in Loop 2 or in the charging injection line. For these breaks, recirculation using HPSI was considered to be the only means of providing high pressure recirculation flow.
Assuming these changes, a reduction in the Haddam Neck core melt frequency (CMF) of approximately 27% was calculated. The new Haddam Neck CMF is 3.93 E-4/ year, versus 5.48 E-4/ year calculated in the PSS (mean value).
Also CYAPCO calculated Haddam Neck CMF assuming no adverse breaks and using only charging for high pressure recirculation, (i.e., the condition presumed to exist prior to discovery of the adverse breaks) to be 4.74E-4/yr. Therefore, the proposed modifications to use either HPSI or charging for high pressure recirculation results in a core melt frequency approximately 16 percent lower than the original design assuming no adverse breaks.
It shocid be noted that the above probabilistic evaluation examined only the small break and medium break LOCA events. The potential benefit from use of HPSI pumps for high pressure recirculation following a consequential LOCA (stuck open unisolated PORV) or following feed and bleed operation was not evaluated but would in all cases lead to an additional reduction in core melt frequency. The 27 percent reduction should, therefore, be viewed as a minimum, in addition to the above probabilistic evaluations, CYAPCO has taken certain measures to provide additional assurance that the likelihood of any breaks occurring in the piping segments involved in these postulated transients is small.
Specifically, CYAPCO conducted a 100% ultrasonic examination of the welds in the piping between the last check valve in the loop 2 charging line and the RCS main loop, including the joint at which those piping segments are connected.
This examination revealed no adverse indications. Because the probability of breaks in this piping is greatest at the weld locations, this examination provides direct evidence that the probability of a break occurring in this piping in the first instance remains remote. Further, both the charging line and the loop 2 cold leg are stain 19ss steel. This material has provided excellent service experience in PWRs.L7) The absence of dynamic loads (water hammer and water slugging), stress corrosion cracking, other corrosion or fatigue provides additional assurance that the chance of cracking is remote.
Additional Alternate Recirculation Flow Path:                                                            In addition to the above described HPSI recirculation option, the revised EOP describes an alternative flow path using charging flow to the RCS loop fill headers. This alternative
!                              involves the use of a single charging pump in series with a RHR pump to provide flow to each of the four loop fill valves. Analyses completed to date indicate that operation of all four valves may be necessary to assure adequate injection of coolant. Bec;tuse evaluations are still ongoing with respect to this flow path (both with respejt to environmental qualification of components and equipment
        ?                                                                                                                                                           *
        =                                                                                                                                                           i
: i. *                                        .                                                                                                                       ,
{ j                            (7)        d) REG-1061, " Report of the U.S. Nuclear Regulatory Commission Piping i*                                        I4 view Committee Evaluation of Potential for Pipe Breaks," Volume 3, l                                     November 1984.
l1 i
7
  , . , , - . - - - -       ,_.,m___._.--          , - - _ - . _ _ - _.-.-.. - -_ _ _ - . - - _.-___-. _ ,- - --. --- . _ - _ _ _ . - ,

and with respect to potential additional adverse breaks in this region),(3)
CYAPCO does not rely on this option to support its request for exemption. It is described here for completeness.
The revised EOP provides that should either of the non-single failure proof valves required to operate to establish the HPSI recirculation path fail to operate, or there are indications of fuel failure, charging flow to the RCS fill headers may be employed.(9) If necessary, fill header flow could be established in the time associated with opening the fill header isolation valves.
     - (8)   CYAPCO has determined that use of the cables and other components in the loop fill header injection path may not satisfy either seismic or environmental qualification requirements.              Also ECCS analysis requirements provide that in order to utilize this option as a licensing basis response mechanism (g requested here), it would be necessary to postulate 9ie subject break location in one of the loop fill lines. Analysis of this scenario is ongoing. CYAPCO has determined, however that even were the subject break postulated to occur in the loop fill lines, the reduction in CMF previously s ; cussed would be altered only slightly, i.e.,
reduction of approximately 26.5 percent rather than 27 percent. Thus, in any event, use of the loop fill option as a third alternative (even assuming the new adverse break location) still results in a significant reduction in the total CMF for Haddam Neck.
(9)    For the Staff's information, the containment high range radiation monitors will be used to provide a positive indication of fuel failure for these purposes. Accordingly, to provide added assurance of the operability of these monitors, CYAPCO will modify the Haddam Neck Administrative Technical Specifications prior to restart to add conservatism to the existing specifications governing this equipment, with respect to operability requirements and action statements.                CYAPCO will subsequently amend the Haddam Neck Technical Specifications, as necessary. Again, however, CYAPCO is not relying on the loop fill option as a design basis accident response mechanism.            Nor are the above conditions that would procedurally direct use of that mechanism presented as design basis assumptions.}}