ML20086H943
| ML20086H943 | |
| Person / Time | |
|---|---|
| Site: | McGuire, Mcguire  |
| Issue date: | 12/04/1991 |
| From: | Mcmeekin T DUKE POWER CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| 91-24, NUDOCS 9112100200 | |
| Download: ML20086H943 (9) | |
Text
{{#Wiki_filter:. ,1: T C ALWrm - Duke hw Ccmpany kkGuire Nuclear Gennation Depar*rnent Vice President 12700 Hagers Ferry Road ( AIGotA) 604)875-4800 ' Huntersville, NC280781985 (704)STS 4309 Fa DUKE POWER December 4, 1991 U.S. Nuclear Regulatory Commission Document Control Desk Washington, D.C. 20555 Subject - _McGuire Nuclear Station Unit I and 2 -Docket No. 50-369 Voluntary Special Report Gentlemen: Attached is a Voluntary Special-Report concerning the Unit I and Unit 2 containment spray check. valves. These check valves which are located in the annulus serve as a vacuum breaker assembly. This report is being submitted as a Voluntary Special-Report. This event-is considered to be of no significance with respect to the health and safety of the public.- Very truly yours, C/n 2 n%y g T. C. McMeekin LJK/cbl Attachment-xc: Mr. S. D. Ebneter Admit 1strator, Region II U.S. Nuclear _ Regulatory Commission 101 Marietta St., NW, Suite 2900 Atlanta, GA 30323-INPO Records Center Suite 1500 1100 circle 75 Parkway Atlanta, GA 30339 Mr. Tim Reed U.S. Nuclear Regulatory Commission Office of Nuclear Reactor Regulation Washington D.C.-20555 Mr. P. K. Van Doorn NRC Resident Inspector McGuire Nuclear _ Station 9112100200 911204 j l [{p' ' ) PDR ADOCK 05000369 S PDR __z_
- -~ - ~= i l McGUIRE SAFETY REVIEW GROUP INPLANT REVIEW REPORT ) 1.0 REPORT NUMBER: 91-24 2.0 DATE OF REVIEW November 4, through November 29, 1991 3.0 SUBJECT DESCRIPTION: This report is being submitted by McGuire Nucinar Station (MNS) af ter the determination was made by Station Management that the event described below is not reportable under 10CFRSO section 73 a2(1)B. An investigation was conducted to address the circumstances described in Problem Investigation Report (PIR) 0-M91-0158. The investigation provided the root cause and revealed the preferred long term resolution desired by McGuire Engineering and System Engineering personnel to resolve the circumstances surrounding-the event. 4.0 EVALUATION AND COMMENT: PIR 0-M91-0158 was originated by McGuire Engineering personnel to addrens the function of several Unit 1 and 2 Containment. Spray (NS) system check valves, and their effect on the operability of the NS and Annulus Ventilation (VE) systems. 4.1 -Background The NS 'mystem is an Engineered Safety Feature (EST). which serves to remove thermal energy from the Containment in the event of a Loss of ' Coolant Accident (LOCA) or secondary system high energy line break. It performs this function in conjunction with the Emergency Core Cooling System (ECCS), which subcools the Reactor by direct injection. The system will maintain the Containment pressure below the design pressure of 15 psig,af ter all the ice from the Ice Condenser -,has melted. The NS system consists of two redundant trains. Each train contains a pump and heat exchanger with associated piping,. valves and spray headers.. A series of check valves is provided for each NS train that allows air to enter the Containment spray line and help mitigate water hammer damage to the' containment spray piping and.inside Containment check valves. The Code of Federal Regulations (10CFR50) Appendix J Part II, defines a Containment. isolation valve as any valve which is relied on to perform a' Containment isolation function. It further defines the primary Reactor Containment as -- any. structure or vessel that' encloses the components of the Reactor pressure boundary and serves as - an essentially = leak-tight barrier against the uncontrolled release of radioactivity to the environment. 4.2 Description of Event on September 18, 1991, -' PIR 0-M91-0158 was originated by McGuire-Engineering personnel to address the function of several Unit I and 2 NS system check valves and their effect on the operability of the NS and VE systems. There was concern that leakage of these valves into the' Annulus, while uhe NS system was in servir, could cause an offsite radioactive dose concern. The valves in .astion were INS-83, 84, 85, 86, 87, and 88 on Unit 1, and 2NS-l'.s, 111, 112, 113, 114, and.115 on Unit 2 (reference pg 6 of 11 ). These valves are
' ?. s.. DPC/MNS Inplant Review No. 91-24 Page 2 described as Containment Penetration Cross Connect Check Valves. The PIR was originated as a result of an initial review of the Design Basis Documentation (DBD) for the NS system performed by System Engineering personnel. The review discovered the aforementioned check valves located in the Annulus appeared to be Containment isolation valves but were not listed as such in the Final Sifety Analysis Report (FSAR). At the time of the event discovery, Units 1 and 2 were in Mode 1 (power operation) at 100 percent power. The NS check valves were r.ot part of the original design of the NS system but were installed during the construction phase of McGuire Nuclear Station (14NS), prior to initial startup. They were designed and installed to - function as vacuum breakers to protect the NS system piping from the possibility of water hammer. Without the check valves, the possibility of water hammer would be present upon restart of the NS pumps af ter swapping over to the Containment sump from the Refueling Water Storage Tank (FWST), or when the system was cycled by the Containment Pressure Control System (CPCS) during a Derign-Basis Accident (DBA). These valves also perform a system pressure boundary isolation and may be required to perform a Containment penetration isolation function in some accident scenarios. However, no documentation has been identified whlen shows that the check valves have ever been tested for valve movement or leakage since their installation, up until this time, because they were not identified as requiring testing. A review of the maintenance history on check valves of the same type revealed no work done on the valves for significant leakage. The maximum leakage found from testing was 0.2 gallons per minute (gpm), and the test was performed at a pressure greater than the NS system operating pressure. -Once the concern with the check valves was identified, "as found" partial stroke (valve movement) and leakage tests were performed by Performance (PRF) personnel on Unit 2, October 6,1991, and on Unit 1, October 24, 1991. The tests were performed using procedures PT/1,2/A/4208/12, NS Check. Valve Verification
- Test, and PT/1,2/A/4208/13, NS Check Valve Leak Test.
The objective of the tests were to verify that when the check valves performed their function as vacuum breakers, they partially opened, to allow air to enter the NS piping while a vacuum was being applied on the downstream side of the inside valves. This was accomplished by pressurizing the valves to open at </= 6.0 peig. To test the check valves in their function as Containment penetration and ECCS pressure boundary isolations, the NS system was pressurized with water. -The leakage from each check valve was not to exceed 1.0 gpm, -tested individually in either train, with the total leakage not to - exceed 2.0 gpm for two trains of the NS system. The test acceptance criteria was based on General Design Criteria (GDC) 19 - Control Room habitability, this was the most conservative criteria. The check valves successfully met the test acceptance criteria. During the "as found" testing of the Unit 1 NS system check valves, it waa noted, by System Engineering personnel, that even though the check valves had successfully passed the test acceptance criteria, the valves showed greater leakage than the Unit 2 check valves because of the diff erence in valve seat material. Subsequently, the Unit i valve seate were replaced with a soft seat (rubber), similar
e e. DPC/MNS Inplant Review No. 91-24 Page 3 to the Unit 2 check valve seats. When the valves were retested on November 6,1991, they passed the test acceptance criteria; however, it was noted, by System Engineering personnel, that one of the newly replaced valve seats leaked 0.7 gpm. Not being satisfied with this minimal leakage, System Engineering personnel had the valve disassembled and re-lapped. Retesting of the check valve revealed 0.0 leakage. In a McGuire Engineering Management proposal for the resolution of PIR 0-M91-0158, dated October 31, 1991, it was decided by McGuire Engineering and System Engineering personnel that the NS system check _ valves in question were Containment isolation valves and should racet leakage criteria to limit 1oses As Low As Reasonably Achieveable (ALARA) for offsite and : ital areas of MNS during postulated accidents. It was noted in the proposal that arguments could be made in support of or against the NS ch3ck-valves being classified as Containment isolation valves. The decision to classify the check valves as containment isolation valves was considered a conservative stance by ncGuire Engineering and System -Engineering personnel. A -review-of the FSAR reveals that the Containment piping penetrations and isolation valves associated with the check valves of concern are shown in Figure 6.2.4-1 and are listed in Table 6.2.4-1 of the FSAR and have been exempted from 10CFR50 Appendix J test requirements. The containment penetration and associated valves were reviewed and approved by the NRC in the original plant licer. sing process. The actual NS check valves are not listed in the FSAR and are, in fact, in non conformance with GDC 56 - Primary Containment Isolation, which states that a simple check valve cannot be used as an outside Containment isolation "alve. The operability of the NS and VE systems for Units 1 and 2 is not, however, in question since the check valves satisfactorily passed the partial stroke and leakage tests. The.ohort term resolution to this event, as determined by McGuire Engineering and System Engineering personnel, is to continue performing partial stroke and leakage testing of the NS s)atem check valves.at specified intervals to ensure offsite and vital areas of MNS are ALARA during postulated accidents. The preferred long term resolution desired by McGuire Engineering and System Engineering personnel is to remove the NS check valves in question, install a 10 inch check valve downstream of the NS pump discharge, and modify Electrically Motor Operated (EMO) Containment isolation valve: control logic. The selected check valve is an Anderson-Greonwood wafer style eneck valve with a soft-seat (reference pi 7 of _ ). The 10 inch check valve would be added to the vertical'.vn of-the NS piping. The current check valves and associated hangers would be removed and the piping capped where the check' valves were removed. The benefits to choosing this resolution are as-follows:
- 1) By eliminating the current check valves, the periodic leak testing is eliminated,
- 2) Sinca the resolution is similar to the Catawba Nuclear Station (CNS) NS check valve design, it makes the licensing process less complicated and,
'. ~ DPC/MNS Inplant Review No. 91-24 Page 4'
- 3) By modifying the EMO Containment isolation valve control logic to close the valves when the NS pump is not operating, this. ensures adequate leak tight protection which prevente pulling a vacuum in the NS piping since the EMOs meet the Containment isolation valve leak rate criteria.
Nuclear Station Modifications (NSMs) MG-12398 and -22398 have been originated to irnplement the preferred long term resolution to the NS check valve. concern. 4.3 Conclusion This event has been assigned a cause cf Design Deficiency because of a functional design deficiency. The NS check valves were not identifed as valves which should have fallen under some form of valve testing program. Not until the concern was raised by System Engineering personnel regarding the function of the check valves and their ef fect on the NS and VE system operability was the need identified. Even though the Containment penetration pipings and isolation valves associated with the check valves of concern are exempted from Appendix J testing, the check valves still f all under the ASME_IWV valve testing program. The testing of the check valves by PRF personnel subsequently followed, even before the decision was made - to identify the valves as contairuuent isolation valves. The check valves have been classified as Category A,C valves under IWV-2200. Category A is defined as a valve'for which seat leakage is limited to a specific maximum amount in the closed position for the fulfillment of its function. Category C is defined as a valve which is self actuating in response to some system characteristic, such as - pressure (relief valves) or flow direction (check valves). The check valves fall under test requirements for IWV-3420, valve leak rate test, and IWV-3520, tests for check val'ies. As a result of this event, it has been determined by McGuire Engineering and System Engineering personnel that the desired long term corrective-action on both trains of Units 1 and 2 is to delete the NS check valves, install a 10 inch check valve downstream of the NS pump discharge piping, and modify the EMO Containment isolation valve control logic. This modification'will eliminate periodic leak testing of the check valves and will prevent pulling.a vacuum in the - NS piping. A review of the Operating Experience Database for twenty-four months prior to this event revealed three events which were classified as 1 Design Deficiencies because of. a - f uncticinal design deficiency; however, the Design Deficiencies were not as a result of a failure to identify valves which should fall under a valve testing program and involved different equipment and systems. This event is, therefore, not considered to be recurring. This event is not Nuclear Plant Reliability Data System (NPRDS) reportable. There were no personnel injuries, radiation overexposures, or uncontrolled releases of-radioactive material as a' result of this event. 4.4 SAFETY ANALYSIS: The NS system is an ESF which serves to remove thermal energy from the Containment in the event of a LOCA. Removal
'DPC/MNS Inplant Review No.:91-24 Page 5 of the thermal energy maintains the containment pressure below the design pressure of 15.0 psig af ter all the ice -in the Ice condenser has melted. As part of the NS system, the check valves of concern were determined to perform a containment isolation function in some accident scenarios, as well as their original design function as vacuum breakers for-the NS system piping. The valves had not been included in any leak rate testing program. Therefore, with no proof of acceptable leakage, the potential existed for an unevaluated leakage path to increase the dose consequences to the control Room and C e sion Area Boundary. Therefore, the established test acesguce criteria was based on GDC-10, Control Room habitability and .W. mad the Control Area Ventilation (VC) system to be in servi n and operable. Ueing this test acceptance criteria, leakage from tha NS check valves was limited to a maximum of 1.0 gpm per-valve tested individually in either train, with a total leakage not to exceed 2.0 gpm for both trains of the NS system. The consequence of any ' leakage from the NS check valves, up tp. 2. 5 gpm can be expected to be mitigated by the VE system which would filter the leakage. prior to its monitored release to the atmosphere. In tho event of excessive leakage from one train of the NS system check valves,: the system, which is designed with two redundant - trains, could be switched over to the-second train. In addition, if the control Room became uninhabitable, Control Room Operators would wear self contained breathing apparatus (SCBAs), provided in the control Room area. Radiation monitors in the Control Room would alert control Room personnel of high radiation levels. The NS check valves.were successfully tested in their function an vacuum breakers and Containment isolation valves. Therefore, the NS systems on Units 1 and 2 would have been able to perform their design function if required. At no time was the system required-to mitigate the consequences of an accident during this event or has it been required for the life of the plant. The. health -arrt safety of the public were not af fected by this event. 5.0 CORRECTIVE ACTIONS: -Immediate: ~None Subsequent: PRF personnel successfully tested valva-1NS-83, 84, 85, 86, 87, and 88 using procedures PT/1/A/4' d 12, NS Check' Valve Verification Test, and PT/1/A/4208/1 ~ NS ' Check Valve Leak Test; and PRF personnel also successfully tested 2NS-110,111, 112, 113, 114, and ll5. using procedures PT/2/A/4208/12, NS Check Valve Verification Test and PT/2/A/4208/13, NS Check Valve Leak Test. Committed: NSMs MG-12398 and -22398 have been originated by Projects Services personnel to implement the long term preferred option of deleting the NS check valves and installing 10 inch wafer style check valves in the discharge piping of the NS pumps on: Units 1 and 2. l
DPC/MNS Inplant Review No. 91-24 Page 6 UNIT 1 (UNIT 2) TRAINS A and B of the NS CHECK VALVES (VACUUM BREAKER ASSEMBLY) 0.*TAINMENT VES$1L l ggggtgggy gygg,pggg i 1
- 5. $ il
~ i jL i I IN5 14 INS-il l (2NS.14) (/NN ll) I INS +84 Id ( 1 A) i p. I -E IINS-8) l I(2NS Ilo) g I -(;NS.ll2) -{ (I NS 2 9 A INS-29Al i I I -~ INS-)I I N S-10 l I (2NS-ll) (2NS-)0) 8 ANNut.V3 CONTAllOG ET VESSEL
- Atf3tLIARY SUllDING 1
A0111,lARY BltllSING U WTAINMPNT VF55Fl. g i I I N T-I k I b I INS-li I (2NM-12) (2hS 16) I I INS-87 g,3,gg, 3 !2"8'II ,,._((2NS-l$$) g5 I I HS 4 g INS-86 1
- (7N1-il3)
IN1-83, ins-12e i '~l(2NS*L26) l %( 2NS a ( 4 4) l \\ g - -q }_ J v INS 14 lN%=ll I I ( 2 NS-14 ) (2NS-l1) A NNtfl.US CWTAIMWFWT VFSSFl. AUIll.lARY BullDINC
DPC/MNS Inplant Review No. 91-24 Page 7 ANDERSON-GREENW-3D WAFER STYLE CHECK VALVE C 8 ];)
- ~
-A u--- --= ~~J{Q Q ~m.. f_ x,. 1 ,I M.,9 ed dd ~ [1.- .t i w f k \\ I\\ lb A ps, '7 i--- - s \\ / / y/////////gQ ' ~ ~^ [_ r____.___m.h2,_3 [ Otrocoon of Flow $ Meta! Seat Soft Seat 7 l 1 l l l i
o Volunt_ary Special Report - December 4,01991 Page 2-bxci T.S. Barr B.W. Bline-R.C. Futrell (CNS) R.L. Gill P.F. Guill-R.E.-Hall P.R. Herran'- T.L. McConnell R.C. Norcutt M.E.. Patrick (ONS) G.H. Savage R.O.-Sharpe G. B.' ' Swindlehu rst ~ H.B. Tucker -R.L. Weber QA-Tech. Services NRC Coordinator (EC 12/55) 4 b 4 5- ' 4 - ..-..-..,,...._,.,..-.;.,...,-,:~---- -. ~.}}