ML20035B693

Operating Experience Feedback Report - Pressure Locking and Thermal Binding of Gate Valves.Commercial Power Reactors

ML20035B693
Person / Time
Issue date:	03/31/1993
From:	Caroline Hsu NRC OFFICE FOR ANALYSIS & EVALUATION OF OPERATIONAL DATA (AEOD)
To:
References
TASK-AE, TASK-S92-07, TASK-S92-7 AEOD-S92-07, AEOD-S92-7, NUREG-1275, NUREG-1275-V09, NUREG-1275-V9, NUDOCS 9304020327
Download: ML20035B693 (30)

Text

{{#Wiki_filter:- _ - _ _ - - - - - - - - - - - - - - - - - - - - - - NUREG-1275 Vol. 9 Coerating Ex;perience Feecback Repor': Pressure Locring ancL Taermal Bincing of Gate Va:ves Commercial Power Reactors ~ U.S. Nuclear Regulatory Commission Office for Analysis and Evaluation of Operational Data C. Hsu v" "*'% j s f 78R438!!?' 1275 R PDR

~ _ _. 6 g ('! 'i I l 1 AVAILABILITY NOTICE Availability of Reference Materiais Cited in NRC Publications Most documents cited in NRC publications will be available from one of the following j sources: I I 1. The NRC Public Document Room, 2120 L Street, NW. Lower Level, Washington, DC l 20555 l 2. The Superintendent of Documents, U.S. Government Printing Office, P.O. Box 37082, Washington, DC 20013-7032 i i 3. The National Technical Information Service, Springfield, VA 22161 Although the listing that follows represents the majority of documents cited in NRC publica-i tions, it is not intended to be exhaustive. 6 i Referenced documents available for inspection and copying for a fee from the NRC Public i Document Room include NRC correspondence and intemal NRC memoranda: NRC bulletins, I circulars, information notices, inspection and investigation notices; licensee event reports; l vendor reports and correspondence: Commission papers; and applicant and licensee docu-ments and correspondence. j The following documents in the NUREG series are available for purchase from the GPO Sales Program; formal NRC staff and contractor reports, NRC-sponsored conference proceed-l ings, international agreement reports, grant publications, and NRC booklets and brochures. j Also available are regulatory g.aJes, NRC regulations in the Code of Federal Regulations, and Nuclear Regulatory Commission Issuances. '\\ Documents available from the National Technical information Service include NUREG-series I reports and technical reports prepared by other Federal agencies and reports prepared by l the Atomic Energy Commission, forerunner agency to the Nuclear Regulatory Commission. j i Documents available from public and special technical libraries include all open literature items, such as books, journal articles, and transactions. Federal Register notices, Federal - l and State legislation, and congressional reports can usually be obtained from these libraries. Documents such as theses, dissertations, foreign reports and translations, and non-NRC l conference proceedings are available for purchase from the organization sponsoring 'the publication cited. Single copies of NRC draft reports are available free, to the extent of supply, upon written request to the Office of Administration Distribution and Mail Services Section. U.S. Nuclear Regulatory Commission, Washington. DC 20555. Copies of industry codes and standards used in a substantive manner in the NRC. regulatory process are maintained at the NRC Library. 7920 Norfolk Avenue Bethesda, Maryland, for use by the public. Codes and standards are usually copyrighted and may be purchased from the originating organization or, if they are American National Standards,- from the American National Standards institute, le30 Broadway, New York, NY 10018. I i i ~

V NUREG-1275 Vol. 9 i Operating Experience Feedback Report - Pressure Locking and i Thermal Binding of Gate Valves Commercial Power Reactors Manuscript Completed: February 1993 Date Published: March 1993 C. lisu V Division of Safety Programs O!Tice for Analysis and Evaluation of Operational Data U.S. Nuclear Regulatory Commission Washington, DC 20555 ,p* "*% e 1..

.. - - =- ~ I 1 I l l r ABSTRACT f The potential for valve inoperability caused by pressure This report provides a review of ope ating events involv-locking and thermal binding has been known for many ing these failure mechanisms. As a result of this review years in the nuclear industry. Pressure locking or thermal this report: (1) identifies conditions when the failare binding is a common-mode failure mechanism that can mechanisms have occurred,(2) identifies the spectrum of I prevent a gate valve from opening, and could render safety systems that have been subjected to the failure 'I redundant trains of safety systems or multiple safety sys-mechanisms, and (3) identifies conditions that rr'ay intro-f d ce the failure mechanisms under both normal and acci-tems inoperable. In spite of numerous generic communi-p dent conditions. cationsissued in the past by the Nuclear Regulatory Com-mission (NRC) and industry, pressure locking and On the basis of the evaluation of the operating events, the thermal binding continues to occur to gate valves instr. led Office for Analysis and Evaluation of Operational Data in safety.related systems of both boiling water reactors (AEOD)of the NRC concludes that the binding problems [ i (BWRs) and pressuriz.ed water reactors (PWRs). 'Ite ge-with gate valves are an important safety issue that needs { neric communications to date have not led to effective priority NRC and industry attention. 'Ihis report also industry action to fully identify, evaluate, and correct the provides AEOD's recommendation for actions to effec-( prob!cm. tively prevent the occurrence of valve binding failures. l F i ? 1 i t 1 i i L-I } i I i t t b L i r I I I i I iii NUR EG-1275. Vol. 9 f

1 i a CONTENTS i Page i ABSTRACT 'iii AB B R EVIATI O N S.......................................................................... vii 4 INDIO D UCH ON............................. 1 i DISCUSSION.............................................................................. 1 A. Purpose............................................................................... I t B. Background............................................................................ 1 C. Thermal Binding Phenomenon............................................................. '2 j D. Pressure Imcking Phenomenon........................................................... _2 i E. Con sequ ences of Locking.............................................................. 4 l F. Preventive M ethods................................................................... 7 t G. S u rvey Findin gs.......................................................................... 7 H. Synopsis............................................................................ 8 SAFETY S I G NI FI CAN CE................................................................ 9 FINDINGS.............................................................................. 10 CONCLUSION... 10 RECOMMENDA'IlONS...................... 10 i REFERENCES................. 11 1 APPENDIX A PRESSURE LOCKING EVENTS............................. -.................... A-1 4 A. Iow-Pressure Coolant Injection and Low-Pressure Core Spray System Injection Valves............. A-1 B. Safety I nj ection System................................................................... A-2 { C. Containm ent Spray Syst em................................................................ A-2 [ f D. Residual Heat Removal Shutdown Cooling Suction Isolation Valve.............................. A-2 E. Residual Heat Removal Hot Irg Cross.Over Isolation Valve................................... A-3 i F. Residual Heat Removal Containment Sump Suction Isolation Valve............................. A-3 + G. Residual Heat Removal Suppression Pool Suction Valve....................................... A-4 j H. Residual Heat Removal Heat Exchanger Outlet Valve......................................... A-4 L High-Pressure Coolant Injection Steam Admission Valve...................................... A-4 1 J. Emergency Feedwater Isolation Valve...................................................... A-4 r APPENDIX B: THERMAL BINDING EVEbrPS................................................... B-1 i -) A. Reactor Depressurization System Isolation Valve............................................ B-1 : B. Residual Heat Removal Inboard Suction Isolation Valve............. B-1 i C. High. Pressure Coolant Injection Steam Admission Valve....................................... B-1 D. Power-Operated Relief Valve Block Valve.................................................... B-1 l 'i E. Reactor Coolant System Letdown Cooler Isolation Valve...................................... : B-1 l F l k F. R esidual Heat Removal Suppression Pool Suction Valve..................................... B-1 'j l G. Containment Isolation Valves........................................................... B-2 H. Condensate Discharge Valves........................................................... B-2 l t v NUREG-1275. Vol. 9 - k I ,s m x

t i 1 1 1 1 L CONTENTS (continued) j i ) l l Puae i I 1 APPENDIX C:

SUMMARY

OF AEOD PIANT SURVEY C-1 James A. FitzPatrick: .................................C-1 Ginna:.................................................. ................................C-2 Nine Mile Point:.......................................................................C-2 I Salern:............................................................-.................. C-2 Hope Creek: ......................C-3 i S usque hanna:.................................. .....................................-C-3 i 5 i FIGURES i '? t Figure 1 Flexibl e. Wedge G at e Valve....................................................... 3 i Figure 2 - Double-Disc Parallel-Scat Gate Valve............ 3 i I Figure 3 Pressure locking Flexib!c-Wedge Gate Valve. 5 I r ) i 1 I l i i l-1 1 i l I 1 i l i l l 1 1 1' i l NUREG-1275, Vol. 9. vi

ABBREVIATIONS ASP Accident Sequence Precursor LER licensee event report LLRT localleak-rate test LOCA loss-of-coolant accident BWR boiling water reactor LPCI low-pressure coolant injection CCDP conditional core damage probability LPCS low-pressure core spray CS containment spray MOV motor-operated valve ~ NPRDS Nuclear Plant Reliability Data System diffe cnt pr sure PORV power-operated relief valve ECCS emergency core cooling system PWR pressurned water reactor EFS cinergency fecewater system GL Generic Letter SCSS Sequence Coding and Search System HELBA high-energy line break axident SDC shutdown cot, ling HPCI high-pressure coolant injection SI safetyinjection IST inservice testing TOL thermaloverload v vii NUREG-1275, Vol. 9

= I i i INTRODUCTION DISCUSSION A. Purpose This study was u...utiated based on a report of an event at g the James A. FitzPatrick plant, described in licensee lhe purpose of this studyis to provide a review of operat-event report (IER) 333/91-014, which described pres-ing experience to: (1) identify conditions under which the sure locking of flexible-wedge gate valves. Although the phenomenon of pressure locking or thermal binding has specific problem was revealed during a hydrosta'ic test occurred,(2) identify the spectrum of safety systems that which press'irized the lxmnet area of the flexible-wedge have been subjected to pressure locking or thermal bind-gate valve, it was evident that these valves could also ing, and (3) determine what conditions may introduce the become pressure locked during normal plant opemtion failure mechanism under both normal and accident condi-and thus, may not function during an accident. Subse-tions. This study also provides an assessment of the safety quent investigation of the industry operating experience impact due to valve operator motor or valve internal showed that similar events, caused by the phenomenon, damage as a result of the associated valve being locked in continue to occur and challenge the operability of safety the closed position due to pres;ure h>cking or thermal binding. Based on findings from this study, several recom-systems. mendations are developed. B. USCkgr0Und The operating experience shows that double-disc and flexible-wedge gate valves in many safety applications The problem has been addressed by the NRC and incus-have not been operable due to pressure k)cking or ther-try since 1977 (Reference 1). Particularly, throughout the i mal bindmg.'Ihese valves,as a result of ther, design, can 1980's, the industry issued a number of event reports become pressure locked by being placed m operatmg concerning safety-related gate valve failures due to disc-configurations which subject them to high-pressure fluids binding lhese failures were attributed to either pressure in the bonnet. Sucn forces oppose movmg the valve disc locking or thermal binding. Binding of gate valves in the i from the seat and these forces were not considered when closed position is of safety concern because gate valves sizing the valve's motor operator. These valves generally have a variety of applications in safety-related systems have their thermal overloads (TOLs) bypassed and could and may be required to open during or immediately fol-i d fail as a result of experiencing this phenomenon. Thermal lowing a postulated design-basis accident (DBA). During binding occurs because the flexibic-wedge gate body con-such events, valve performance is severely challenged by [ tracts a greater amount dunng cooldown than the valve the rapid cooldown and depressurization rates and the l disc and pinches the disc in the valve seat. Consequently, valves are exposed to the largest differential pressures i when the valve is closed hot and allowed to cool, the (Ap) across their discs. Valve operators are generally not i difference in thermal contraction can cause the seats to sized to open a valve against binding forces generated by a bind the disc so tightly that reopening is either extremely high pressure fluid trapped in the bonnet cavity (pressure difficult or impossible until the vatve is reheated. The locking) or when the disc was seated hot and subsequently 1 operating experience has also shown that valve surveil-cooled due to differential thermal movement (thermal [ lance testing is usually conducted under conditions that binding). Pressure locking or thermal binding of gate F would not detect or identify valve susceptibility to pres-valves represents a nonrevealing common-mode valve sure locking and, in some cases, thermal binding. As a failure mechanism since normal surveillance tests may result, safety-related valves subjected to the phenome-not detect or identify them. non would stay undetected in the locked state. The licensees of the plants with recent pressure locking and thermal binding events indicated that they did not address all of the potential operating conditions in their 'lhe potential for valve inoperability caused by pressure evaluations. Onc licensee has recently discovered that the i kicking and thermal binding has been known for many root-cause of a failed motor operator previously attrib-years in the nuclearindustry. In spite of numerous generic uted to a broken torque switch was due to pressure kick-l communicationsissued in the past by the Nuclear Regula-ing. Several licensees have indicated either to the author tory Commission (NRC) and industry, pressure kicking or stated in an LER that they planned to reevaluate gate i l and thermal binding ccmtinues t9 occur to gate valves valve susceptibility to pressure locking or Ihermal binding installed in safety-related systems of both boiling water under all postulated system tests and operating condi-reactors (BWRs) and pressurized water reactors (PWRs). tions. It appears that the generic communications to date have -{ not led to cifective industry action to fully identify. evalu. Desenptions of pressure kicking events invohing gate ate, and correct the problem. valves installed in safety-related systems are listed in i 1 NUREG-1275, Vol 9 [ t i

Appendix A of this report. De safety-related systems in documented in the AEOD report, " Survey of Licensee which valves have become pressure locked include the Actions Taken to Address Pressure Locking of Double high. pressure coolant injection (HPCI) system, low. Disk and Flexible Wedge Gate Valves"(Reference 2). A pressure coolant injection (LPCI) system, low-pressure summary is provided in Appendix C and the survey find-core spray (IPCS) system, safety injection (SI) system, ings are described in the text (Section G) of this report. containment spray (CS) system, emergency feedwater gstem (EFSk and the residual heat removal (RHR) sys-C. Thermal BindinE Phenomenon tem. Thesc events occurred under different operational modes in the time period from 1969 to 1992. The safety-If a wedge gate valve is clos.d while the system is hot, related gate valves involved in these pressure locking thermal binding can occur as the system cools. The valve a events were LPCI and IECS system mjection valves, CS body and discs mechanically interfere because of the dif-valves, RilR shutdown cooling (SDC) isolation valves, ferent thermal expansion and contraction characteristics RHR hot leg cross-over isolation valves, RHR contain-of the valve body and the disc. The difference in thermal ment sump and suppression pool suction valves, HPCI contraction can cause the seats to bind the disc so tightly steam admission valve, and emergency feedwater isola-that reopening is either extremely difficult or impossible tion valve. until the valve is reheated. This is particularly true for valves with internals which have reduced clearances due A review ofIhe,,e events shows that there were two pot en-to improper maintenance or alt erations. Excessive closing tial causes of pressure locking; liquid entrapment in the force can contribute to thermal binding because excessive lxmnet and ap across the disc while in the closed position. closing force causes the disc to be driven into the seat Most of the events occurred during infrequent plant evo-more tightly and, on cooling, the thermal binding effect is lutions such as heat-up, qooldown, and testing. The phe-increased. Several potential remedies have been sug-nomenon can also occur during rapid depressurization. gested to alleviate this situation, including slightly open-Several events resulted in motor operator failures. All ing and reclosing a valve during cooldown, limiting valve pressure lockings which occurred in these events have actuator closing forces, and using compensating spring adversely affected the operation of motor-operated packs to reduce valve initial closing forces. In general, valves (MOVs), and rendered the associated safety train neither ac nor de valve motor operator sizing analyses unavailable. account for the extra force needed to unscat a vah c when it is thermally bound. A search of the nuclear plant reliability data system (NPRDS) database was conducted for thermal binding on D. Pressure Locking Phenomenon gate valves. The search, which covered the period from 1983 to the present, identified a number of events involv-Pressure locking in flexible-wedge (Figure 1) and double-ing thermal binding prob! ems that posed a potential disc gate valves (Figure 2) generally develops because of safety problem to plant operation. He descriptions of the nature of the design in combination with characteris-thermal binding events are listed in Appendix B. He tics of the bonnet and specific local conditions at the safety-related valves involved in these events were reac-valve. The essential feature to develop pressure locking is tor depressurization system isolation valves RHR in-the presence of fluid in the bonnet cavity including the board suction isolation valves, HPCI steam admission area between the discs. De fluid may enter the bonnet vahes, pressurizer power-operated relici valve (PORV) cavity during normal open and close valve cycling at what-block valves, reactor coolant system letdown isolation everline pressure exits at the time. Also, fluid may enter valves, RHR suppression pool suction valves, contain-the bonnet cavity of a closed valve which has a ap across ment isolation valves (sample line, Ictdown heat ex-the disc. ne pressure differential causes the disc to move changer inlet header), condensate discharge valves, and slightly away from the seat creating a path so that the reactor fcedwater pump discharge valves. Thermal bind-bonnet cavity becomes filled with high pressure fluid. ing occurred when valves were closed while the associated Whether these situations lead to a valve pressure locking systems were hot and allowed to cool. scenario depends upon the fluid pressure when the bon-net cavity was filled, temperature changes from when the in order to understand the extent and adequacy of the fluid entered the bonnet cavity, and hx:al line pressure past and recent evaluations and corrective actions taken compared with bonnet cavity pressure at the time the by bcensees in solving the pressure locking problem, MOV is called upon to operate. AEOD recently conducted a survey of six selected plants. The primary areas addressed in the survey were: (1) meth-Various plant operational sequences could introduce ods of analyzing pressure locking potential, (2) methods conditions conducive to pressure locking. Irrespective of of testings for Icakage and operability, and (3) training initial bonnet cavity fluid pressure (low or high)and tem-programsprovidedforplantoperatingstaff tounderstand perature, it is clear that a subsequent temperature in-the failure mechanism. The complete survey results are crease of the fluid will cause an increase in bonnet ca ity NUREG-1275 Vol. 9 2 =

/ / u o w //,/ / s f' / A w m -- ;zy ~ w 7 7 a r wyma/ y Sy mm V Figure 1 Flexible-Wedge Gate Valve i ~y- - n W // c //L D Stem-m ~ +- y i i / / / Parand seat sgs y / N VYl ffflll } Figure 2 Doubic-Disc Parallel Scal Gate Vahe 3 N UREG-1275, Vol. 9 l )

pressure due to thermal expansion of the fluid. He tem-because the inservice tests are normally performed dur-perature increase can occur as fluid on either side of a disc ing refueling outages when the trapped pressure in the heats up during various modes of plant operation or possi-valve has time to decay, no longer causing double-disc ble changes in ambient air temperature caused by plant drag forces during valve testing. For the IST conducted operation, leaking pumps or valves, or in the event of a during normal plant operation at some plants, high pres-high-energy pipe break. In these situations, the rate of sure nitrogen gas is introduced to the upstream side of the temperature increase, which may be relatively slow to valve to reduce the op across the valve when performing very high, controls the bonnet cavity pressure and valve a test. His is because the downstream side is at reactor susceptibility to pressure locking. Conversely, a bonnet pressure due to the check valve leakage. Introduction of cavity filled with high pressure fluid, such as leakage from high pressure nitrogen pas to the upstream side reduces the primary reactor coolant system, becomes a pressure the ap to the design basis magnitude for the valves. locking candidate should a loss-of-coolant accident Testing under this condition will not detect pressure lock-(LOCA) or other transient cause pipe line depressuriz. ing. ation. E. Consequences of Locking Additional consideration of kical valve conditions illus. trates tbe effect of a relatively small 1emperature increase These phenomena can delay the valve stroke time or when a gate valve is closed with fluid trapped m the cause the valve motor actuator to stall. The events at bonnet cavity or area between the discs. As the system FitzPatrick and Susquehanna indicate that the RHR/ temperatures mercase, the valve is subsequently heated, IEC1 and IECS injection valves of a BWR are suscepti-the trapped fh'id expands causmg pressure to mcrease in ble to the cavity pressurization condition of pressure lock-the valve bonnet and between the wedges of the valve ing. In these two systems, the motor-operated injection discs (hgure 3). The pressure increase inhibits opening of vdves are normally shut and are required to automatically the valve by causmg the discs to press tightly against the open upon an actuation signal. Because the testable valve seats, resulting m binding of the valve. The valve can M VMye between the reactor (or the recirculation line) be heated up and the trapped fluid expands as a result of and the injection valve is not a leak-tight valve, leakage heating either during nonnal plant startup or should a past the check valve over time can pressurize the piping high-energy line break accident (HELBA) occur. He between the valves and the injection valve cavity to reac-valve does not have to be in a high temperature system tor pressure. Near leak-tight seating surfaces or the injec-but only m close proximity where hcat conduction tion valve may allow the valve cavity to remain pressur-through the pipe or via the surroundmg air will heat the ized and become subject to pressure locking when trapped fluid in the bonnet cavity. The rate of pressure njection is needed during a LOCA. Under this condition, rise can be as high as 100 psi per 1 "F temperature m-the bonnet pressure is more than 1200 psi, while the crease in a solid filled bonnet for system temperature downstream pipe suddenly depressurizes to between 400 above 450 *F. For lower system temperature (approx - and 500 psi. He upstream pipe has a pressure around 300 mately 100 *F), the rate is reduced to 33 psi per 1 'l-psi. His high internal-to-external op across both seating temperature rise. This type of pressure kickmg could surfaces would result in double-disc drag forces, which if cause failure of the valve pressure boundary parts unde

• they exceed the available thrust of the actuator, will pro-zero-leakage valve ccmditions.

doce pressure locking. Over time, bonnet pressure will decay, due to leak ige past the seating surfaces or packing, For the situation where the bcmnet cavity is pressurized at a rate dependent on the leak tightness of the valve.The when the valve is closed and the valve is leak tight, the valve will not stroke until the cavity pressure decays to a pressure will be retained, even when the rest of the sys-level less than the maximum allowable bonnet pressure. tem is depressurized, and then force the discs against the If the depressurization time for the valve is longer than valve seats. causing double-disc drag forces.The resultant the system response time to initiate valve actuation dur-drag force can be excessively high such that the valve can ing a LOCA, the valve willlock up and the valve actuator not be opened when actualed. His type of pressure kick-motor will stall. ing does not require that the bonnet cavity be filled solid with fluid. In fact, this condition can be enhanced if a gas When a valve disc becomes locked in a closed position due or steam bubble exirts in the top of the bonnet cavity to pressure kicking or thermal binding, actuation of the because it will act as a pressurizer as the fluid eventually motor vdil result in locked-rotor current which will rap-leaks past the seat. His action would prolong the effect of idly increase the temperature of the motor internals. the pressure locking. Within 10 to 15 seconds, the heat buildup can degrade the motor's capability to deliver a specified torque, damage ne ITCl and 11CS injection valves are susceptible to the motor, or both. These characteristic responses illus-potential pressure locking during LOCA actuation. How-trate that MOV susceptibility to pressure locking needs to ever, regular IST can not detect the potential problem be assessed under all plant operating conditions. This can . NUREG-1275. Vol. 9 4

k 3 A h i ExbESS \\ b. PRESSURE / // INTERCONNECTED / ~ (EQUALPRESSURE) U / / ~ ~ ~ ~~ f s /,(///// p/ ~ ////// z 2 Figure 3 Pressure locking Flexible-Wedge Gate Valve 5 NUREG-1275, Vol. 9 ~ j

be appreciated by considering anticipated extremes for an will most likely result in motor damage with little chance MOV that may have one safety function to open but is that bonnet cavity pressure decay could lead to successful subject to pressure locking prior to its one time operation MOV operation. and another MOV t hat could be subjected to intermittent pressure locking followed by valve stroking (perhaps in-service testing IIST)). Foi the FitzPatrick event, based on leak rate testing re-sults, two of the four low pressure emergency core cooling system (ECCS) valves could fail to open if the reactor The primary issues for the MOV subjected to piessure vessel was rapidly depressurized as it would be following a locking for one time operation are the force needed to large-break LOCA. He conditional core damage prob-overcome the effectsof bonnet pressure, bonnet pressure ability (CCDP) for this event wa1 estimated (Reference 6) changes with time, and the MOV opening time. ne usual to be 9.5 x 10 5. He core damage probability was gener-stroke speed of a motoroperator is 12 inches per minute. ated by utilizing the Accident Sequence Precursor (ASP) Hus, MOV stroke times would generally vary from 15 program event analysis. He event was modeled as an seconds to 2 minutes for valves ranging fmm 3 inches to unavailability of two of the four LPCI and LPCS injection 24 inches in diameter. With this information it is evide, t valves for a 1-year time interval. Both LPCI valves were that the 10 to 15 second heat buildup time becomes the assumed to be unavailable. Conditional failure probabili-critical element. If valve bonnet pressure is sufficiently ties of 0.3 and 0.5 were assigned to the two potential high, pressure locking will result in operator motor burn-operable LPCS injection trains. He impact of the valve out. The physical configuration of the bonnet and valve failure on large-break LOCA sequences was included in discs offerslimited opportunity for the bonnet pressure to the analysis. A sensitivity study was then performed as-decay. When considered with the 10 second time for mo. suming all four ECCS valves were failed and the core tor heat buildup at locked-rotor current, this suggests that damage probability was estimated to be 3.9 x 10 d.His is a depending on pressure decay to permit MOV opening is factor of 4 higher than the nominal conditional probabil-not a viable option. Also, the heat buildup could degrade ity estimated for this event cited above. He sensitivity the motor torque capacity so that the motor operator study difference is primarily a resa!! of the conditional would not open the valve even under design basis loads, probabilities assumed for the two LPCS trains, given the Thus. the concem relates to misinterpretation of poten-failed LPCI trains. tial beneficial effects from bonnet cavity pressure decay. For example, a calculation may show that the sum of the ne potential for pressure locking also exists for the time delay to initiate the stroke (due to pressure decay) ECCS low head injection valves of PWR plants except for and the MOV stroke time may still meet the minimum those valves which are normally open (this is the case for time required to inject fluid in the accident analysis. How-most Westinghouse plants). Operating experience has ever, if the MOV received a signal to open and drew shown that interfacing isolation check valves between the - locked rotor current for greater than 10 seconds while the reactor vessel and the injection valve are subject to leak-bonnet pressure was decaying, then the valve would most age. Similar to the scenario described in the preceding likely fail to ever open because of either motor burnout or paragraph for the RHR/LPCI injection valve and the degradation of motor torque capacity. LPCS injection valve, leakage past the isolation check valve can pressurize the injection valve to reactor pr::s-One aspect previous!y discussed was valve operator mo. sure. If this pressure is trapped in the valve bonnet, then for damage due to locked-rotor currents during pressure the injection valve may be subject to pressure locking locking. TOL devices are frequently used to protect the when attempting to open during sudden reactor depre-motors of MOVs in safety systems from the effects of ssurization, such as for injection during a DBA-LOCA overheating. Regulatory Guide 1.106 (References 3 and and probably during SDC.

4) provides staff guidance for the use of such devices. He intent of the guide was to balance protection of the motor Hree events (Vogtle Unit 1, Ginna, and Grand Gull) in comparison with assurance of system function. A re-have demonstrated that pressure locking can occur at a view of operating experience, AEOD report S503, relatively low temperat ure increase.He lowest tempera-

" Evaluation of Recent Valve Operator Motor Bumout ture at which this type of failure was reported in these Events" (Reference 5) found : hat most TOL devices events was slightly below 200 *F.his indicates that ther-were either permanently bypassed or were significantly mal conduction can heat up fluid impped in the bonnet oversized. Hus, the practical impact was that TOL de-and cause pressure locking failures in valves which would vices would not protect the valve actuator motors from not normally be considered as having tempemtures ex-locked-rotor current (several events have been identified ceeding 200 *F. Also, the event at Vogtle Unit 1 indicates where the TOL device did not trip until the motor burned that evaluation for pressure locking potential should be out). It is criticel that the valve operations which require extended to valves which are norma'ly open and to valves excessive loads be analyzed to avoid the sustained locked-which would not be subject to thermal binding or pressure rotor conditions. Motor operation at locked-rotor current locking under ordinary circumstances to avoid the NUREG-1275 Vol.9 6

i possibility that some safety-related valves having the po-Methods to Prevent Thermal Binding tential for failures are overlooked. 1. Double-disc, parallel-seat valves are less susceptible to thermal binding than flexible-wedge gate valves. Replacing existing flexible-wedge gate valves in-F. Prevent.ive Methods volves additional cosi. It is best suited for plants in i the design or construction phase. Here are several preventative and corrective measures 2. While cooling a system, periodically open the valve for pressure locking and thermal binding. Each method has limitations with respect to applicability, safety, effec. slightly and then reclose it several times to allow tiveness and cost. All of these measures have been util-uniform cooling and contraction of discs and bodies. i ized in the past. He following lists summarize these his will involve changes of operating procedures methods (many of them have been described in the previ. and operator actions. ous generic communications listed in Reference 1). 3. Ensure that the valve actuation or actuating medium l is properly adjusted to prevent excessive closing forces on the valve disc. his may not be an effective Methods to Prevent Pressure Locking means to prevent thermal binding if the tempera-ture transient is large. 1. Drill a small hole on the upstream side of the valve disc (or upstream disc for a double-dise valve) to 4. Installation of compensating spring packs on motor f relieve pressure buildup in the bonnet and between operators to absorb inertial closing forces after the the discs. His method makes the valve unidirec-motor has de-energized will avoid excessive closing l tional in sealing against high pressure. He drilled forces on fast acting valves. ? side of the disc should always be towards the high pressure. It takes away the option sometimes used to G. Survey Fm. dings prolong useful seat hfe by reversing the disc after a period of time. An alternative is to drill a hole in the AEOD recently conducted a survey of sr.x selected plants bridge between the seat rmg and the valve bonnet on pour BWRs and two PWRs). He purpose of these six site the upstream side of the valve. This method also visits was to:(1) understand the past and recent hcensee makes the valve unidirectional and should be in-evaluations and corrective actions concerning gate valve stalled accordingly. His is the simplest and a very pressure locking and (2) identify any nonconservative or effective method requiring no operator actions, and mcomplete aspects of these licensee assessments. Ap-the cost is minimal. pendix C describes a summary of the survey report (Ref-erence 2). Itased on this summary, the following findings 2. Install a pressure relief or vent valve in the txmnet to are provided: automatically relieve the txmnet pressure. His rnethod requires the use of external components. If 1. Pressure kicking is a potential generic problem to a manual vent valve instead of automatic relief valve both flexible-wedge and double-disc, parallel-seat is used to release pressure when the system heats up, gate valves. Pressur e locking has occurred to various i operator action would be needed to position it. sizes of both gate valve types at the plants surveyed as well as others. The valves involved were manufac-tured by Anchor-Darling, Alloyco, Crane, Lu.tken-3. Install an external bypass line with a manual valve heimer, Pacific, powell, Rockwell, Velan, Westing-from the bonnet to the upstream side of the valve. house, and Copes-Vulcan. Gate valves are also j Manually open the bypass valve during heatup to susceptible to thermal binding. Different types of relieve pressure from the bonnet. A relief valve ora gate valves vary m terms of susceptibihty, the relieving check valve also has been used. His am unt of disc flexibility being the pnmary determi-method provides an alternate to method No. I above nar4. From most susceptible to least, they are solid when isolation in both directions is required. wedge, flexible-wedge, spht wedge, and double-disc, parallel-seat. Double-disc, parallel-seat gate valves j 4. For valves not required to provide complete isola-are one solution to the thermal binding problem tion, stopping the valve disc travel by position limit because they are the least susceptible to interfer-switches rather than motor torque can keep the ence due to thermal ccmtraction. His method of valve from going completely closed and thereby, prevention may be an optimal solution because it prevent high pressure fluid from being trapped in does not rely on either operator action or devices to the bonnet. prevent thermal binding. 7 NURiiG-1275, Vol. 9 i

I l 1 1 i 2. Valve surveillance testing may not detect or identify experience, the potential for the valve locking valve susceptibility to the problem. His is because mechanism was not fully assessed on a case-by-case the surveillance tests are normally conducted during basis at some plants. either refueling outages or normal operating condi-tions. Pressure locking or thermal binding phenom-6. Two basic types of modification were used in the - ena generally do not exist in these conditions. Most plants surveyed to prevent valve pressure locking. i of them occur during infrequent plant evolution One modification was to drill a small hole in the such as heatup, cooldown, and rapid depressuriz. Upstream disc to relieve bonnet pressure. He sec-l ation. ond type was to install an external bypass line with a blocking valve between the valve bonnet and the 3. The scope of licensee reviews varied widely and upstream side of the valve to equalize bonnet pres-there were no uniform guidelines. sure with the valve inlet pressure. Some completed their review based on engi. 7. One licensee indicated they had surveyed several neeringjudgment without analysis. BWR plarits and found that many plants have not modified the LPCI and LPCS injection valves to i Although others used engineering analysis, the prevent pressure locking. e methods may not have been comprehensive i nor conservative. 8. Most licensees surveyed did not initially consider the problem credible based on a limited assessment ef-Leakage rate assumption was not based fort. Some even decided not to implement corrective on actual operating conditions. and valve actions after the potential problem had been identi-configuration used in the assumption was fied. l not for valves with new, rebuilt, or re-( worked valve discs and seats. E Synopsis Interface heatup such as heat from ther-mal conduction through the adjoining De operating experience illustrates that pressure locking l pipe was not considered. and thermal binding have occurred under a wide range of i plant operating conditions.Thus, MOVs could be subject l 1 conditions involving accumulation of damage to the l Environmental temperature and pressure changes induced such as during a IIELil A valve actuator motor due to pressure locking or the load were not included in the analysis. magnitudes could exceed the design basis loads of the l operator. In addition, a LOCA could cause the conse-Motor-operator thrust capability assump-quential failure of a safety system installed to mitigate tions were based on vendor-supplied data that accident (the IfCA creates the pressure locking in - rather than actual diagnostic testing re. the MOV which renders it inoperable). He systems in sults. For conservatism, the results should which MOV pressure locking has occurred include HPCI, { have considered the effects of diagnostic ITCI, LPCS, SI, CS, EFS, and RHR. errors, live-load packing, stem and stem nut drag, stem rejection load. etc. Also,in Pressure locking occurs when the valve bonnet cavity some cases, degraded voltage conditions pressure creates loads that exceed those which the valve should be considered. operator was intended to overcome. High bonnet cavity pressure can result from valve open or close cycles where 4. No licensees surveyed had provided a training pro-fluid fills the cavity and is subsequently heat ed by conduc-gram on the subject for their engineering staff.nis tion from plant operation or an increase in ambient tem-may lead to incorrect conclusions during root cause peratures such as a postulated high energy line break. A findings. differential pressure across the valve discs can also result. in a high bonnet cavity pressure. Pressure locking that 5. Previous generic communications may not be spe-prevents movement of the valve discs off the seat causes etfic enough to guide the licensees to identify valves the motor to experience kicked-rotor conditions with very susceptible to either pressure kicking or thermal high motor current. Within 10 to 15 seconds, the motor binding. internals heat up which may either cause motor burnout or degraded motor torque capacity so that the valve may Although the phenomena which can cause pressure not be able to perform during design basis conditions. locking or thermal binding on gate valves were de-scribed in the generic communications, information li pressure locking occurs, the mechanisms available to i was not provided based on system-specific condi-limit damage to the MOV motor appear to be 'K)L i tions. Also, duc to lack of system-specific failure protective devices or bonnet cavity pressure decay due to l l NUREG-1275. Vol. 9 8 l l i

leakage. Previous AEOD studies have shown that TOL modification is needed to effectively prevent occurrence devices frequently are either permanently bypassed or of pressure locking. Two basic types of design modifica-oversized. As a result, it is imperative that plant operating tion are widely used in operating plants, one is to drill a modes that potentially cause valve motor operator loads weep hole in the upstream disc to allow trapped fluid to in excess of designed operator sizing be analyzed. Al-flow from the bonnet cavity. He second method is to though a maximum valve leakage rate is frequently speci-install a vent line from the bonnet with a blocking vane to fied, the actual leakage rate may vary with maintenance the upstream side of a valve to equalize with valve inlet practices between zero and a maximum level, but in effect pressure. it is indeterminate for a given condition. Further, if leak-age cannot reduce the bonnet cavity pressure within about 10 seconds, then MOV performance is doubtful. SAFETY SIGNIFICANCE ne potential beneficial effects of leakage to bonnet cav-ity pressure decay are diminished when pressure locking Flexible-wedge and double-disc gate valves are used in a occurs because the increased cavity pressure will reduce variety of applications in safety systems. Many of these ~ the leakage through the valve discs. valves are required to open to perform their safety func-tions during or immediately following postulated design Binding of the valve disc in the closed position due to basis conditions. Valve binding due to pressure locking or thermal binding or pressure locking appears to be a result thermal binding is often synonymous with MOV failure to of valve design characteristics. Hermal binding occurs operate and inability of the associated safety systems to because of the way a gate valve is designed, the seats move perform their sa. :, functions. He safety implications of inward an amount that is proportionally greater than the safety-related gate valve pressure locking are that it rep-disc shrinkage when the valve cools after closing. His resents a potential common-cause failure that could ren-causes the seats to pinch the discs tightly. Consequently, der redundant trains of cenain safety-related systems or when the valve is closed hot and allowed to cool, the multiple safety systems inoperable. These locking condi-difference in thermal contraction can bind the disc so tions generally develop from normal plant evolutions tightly that reopening is impossible until the valve is (Vogtle 1 and Grand Gulf events),leakmg check valves reheated. Pressure k)cking in flexible-wedge and double-(which is a well documented situation, FitzPatrick and' disc gate valves generally develops because of the nature Susquehanna 1 events), or actual line depressunzation of design in combination with characteristics of the bon-(San Onofre 1 and FitzPatrick containment spray events). net and specific local condition at the valves. An operat-Hermal binding also constitutes a potential common-ing characteristic of these valve designs is that fluid enters cause failure mechanism for safety-related valves. He the bonnet cavity during normal open and close cycling valve disc binding generally is a result of normal plant operations. As system temperature increases, the bonnet evolution involving system temperature change (l2Salle fluid temperature eventually increases, resulting in po-1, RHR inboard suction valve event). Therefore, MOV tentially high pressure. Another design characteristic is failure to operate for any of these conditions represents a that when a valve has a differential pressure across the failure to comply with General Design Criteria 1. 4,18, disc in the closed position, the pressurized side of the disc and 21 of Appendix A to 10 CFR Part 5d and Quality can move away slightly from its seat, allowing high pres-Assurance Criterion XI of Appendix B 1010 CFR Part 50. sure fldid to enter the bonnet cavity. If the pressure in the valve body is subsequently decreased, the bonnet pres-Some safety systems are more vulnerable than others to sure will result m doub!c-disc dmg forces against its seats. pctential unavailability depending on the associated valve susceptibility to the binding mechanisms. For the Fitz-Valve surveillance testing may not detect or identify the Patrick event, the CCDP was generated by utilizing the susceptibility of inadequately designed valves for these ASP program event analysis. He event was modeled as failure mechanisms.His is because the surveillance tests an unavailability of two of the four 13'Cl and IJ'CS injec-are normally conducted during either a refueling outage tion valves for a 1 year time interval. Both LPCI valves or normal opemting conditions. De phenomena which were assumed to be unavailable. Conditional failure can cause the valve binding generally do not exist during probabilities of 03 and 0.5 were assigned to the two po-these testing conditions. A comprehensive evaluation is tentially operable IJ'CS injection trains. The impact of needed to determine susceptibility to the binding mecha-the valve failure on large-break LOCA sequences was nism for a gate valve. nis often requires various detailed included in the analysis. The CCDP for the event was analyses involving transient conditions and some un-estimated (Reference 6) to be 9,5 x 10 5 A sensitivity known factors which may need conservative assumptions study was then performed assuming all four ECCS valves to demonstrate the analyses are adequate and complete. were failed and the CCDP for this event was estimated to be 3.9 x 10 *. We would expect that valve binding in On the basis of AEOD site visits and a review of recom-different safety systems would exhibit varying core dam-mended modification from valve suppliers, valve design age probabilities. 9 NUREG-1275, Vol. 9 l 1

FINtMNGS valve motor will result in locked-rotor current which will rapidly heatup the motor internals. Within 10 to 15 sec-2. ( xnprehensive evaluation and extensive analysis, onds, it may either cause motor burnot't or degraded including consideration of plant system conditions motor torque capacity so that the valve may not be sole to and ambient conditions during all modes of plant perform during design basis conditions. operation, are needed in order to identify the valves susceptible to binding and determi> the impact on Although 'IOL protection devices or cavity pressure de-safety system function. cay due to leakage can prevent or limit damage to the MOV motor, neither of these is reliable. Most TOL de-2. 'The binding problem is a result ofinadequate design vices are citMr permanently bypassed or significantly consideration and the most severe impact has not oversized. L. ;ty pressure decay is dependent of valve been accounted for in the accident analysis, or in leakage rate wnich is indeterminate for any given condi-typical probabilistic risk assessment. tion. In any case, if leakage cannot reduce the casity pressure within 10 seconds while the motor runs at 3. Although the valve binding problem and corrective locked-rotor current, then MOV safety function could actions have been known in the industry for many not be assured. years, the potential for this type of valve failure has not been fully evaluated or corrected-Based on the history of industry problems with pressure locking and thermal binding, the following conclusions 4. The inadequacy in design or installation will not are provided: necessarily be found during subsequent plant startup testing or regular surveillance testing. 1. Pressure locking or thermal binding can occur to gate valves used in safety-related systems of both 5. Two types of valve modification are used in the in-PWR and BWR plants. These failure mechanisms dustry, Jepending on several considerations; main-have prevented safety-related systems from func-tenarve, radiation, and leak rate test. These two tioning when called on. Pressure locking is a poten-mcJifications are: (a) drill a small weep hole in the tially significant common-mode valve failure mecha-npstream or downstream dise, and (b) install a vent nism with consequent ECCS system unavailability line, with a block valve, from the bonnet to the main during accidents or severe depressurization tran-pipe. sients. 6. Reliance on calculations of bonnet pressure reliefin 2. Industry and NRC feedback on pressure locking has lieu of physical modification of the valve is not a not been effective in precluding the potential prob-reliable solution to the pressure locking problem. Iem. This is because several parameters used in the calcu-lation are not constant and are subject to change. 3. Valve susceptibility to pressure k)cking may not be For instance, leak rate can change following mainte-detected or identified during normal valve surveil-nance, repair or adjustment on the valve. lance testing. A comprehensive evaluation to iden-tify all gate valves that are susceptible to pressure CONL.,LUSION locking or 1h ermal binding is n ceded. 'Ihe evaluation should include a detailed analysis to assure that all As a result of this study, we ccmclude that previous NRC plant operating and accident modes are considered. and industry efforts to eliminate the gate valve pressure Also,it appears that design modification is a simple, locking and thermal binding problems have not been suc~ effective, and necessary step to assure proper opera-cessful. Licensees continue to experience valve pressure tion of valves susceptible to pressure locking. locimg or t hermal binding events /Ihe reviews conducted by licensees in response to the previous generic comm uni-RECOMMENDATIONS cations were not adequate. 'Ihe scope and extent of licen-see resiews varied widely. Some licensees comp!cted The design characteristics of double-disc and flexible-their reviews based on engineering judgment without wedge gate valves make them susceptible to pressure analysis; although others performed engineering analy-kickmg and thermal binding under specific plant operat-ses, the analyses were neither comprehensive nor conser-ing modes. This condition poses common-mode failure vative. Pressure locking and thermal binding prevent mechanisms that can render multiple ECCS systems in-valves from operating. 'Ihe primary c(mcern is that valve operable. 'Ihis situation has been known within the nu-operators can not be sized to account for the extra force clear industry since 1977, but corrective action has been needed to unseat the pressure kicked valves. When a inadequate.'lherefore, AEOD recommends that the Of-valve becomes h>cked in a closed position, actuation of the fice of Nuclear Reactor Regulation address the following: - NURiiG-1275, Vol. 9 10

1, Licensees should evaluate all safety-related gate f. Circa 1982-Foreign Utility Orders Valve De-valves to determine potential susceptibility to pres-sign Changes at All Power Reactors to Prevent sure locking or thermal binding. The evaluation-Pressure Ix)cking. should employ indepth engineering analyses to cover all plant operating and accident modes. g. November 30,1983-Industry Communication on RHR Suction _ Valve Pressure Locking at a 2. For those valves identified as potentially susceptible foreign reactor, to the binding mechanisms, licensees should imple-ment effective valve modifications and appropriate h. July 6,1984-AEOD Reissue of Study on Pres-procedures to prevent the binding from occurring. sure 12)cking of Flexible Disc Wedge-Type Gate Valves. We suggest this action could be accomplished with either i. December 14,1984-Industry Communication i a supplement to Generic letter (GL) 89-10 or a new NRC Bulletin. Recommended action items a. and c. of on Pressure 12)cking and Thermal Binding.. GL 89-10 should involve the types of analyses to satisfy recommendation (1) above to determine susceptibility to j. March 25,1988-Industry Communication on i pressure locking or thermal binding. However, we recog. Pressure 12)cking of Both RHR Crossover Iso- { lation Valves, nize this option would represent Supplement 5 or 6 to that GL.which may detract industry effort from specific concerns currently being addressed. Thus, a new bulletin .14,1988-General Electric Service k. October may be a preferred optior Although selection of the Information Letter No. 368, Rev.1, "Recircu-generic communication method is optional, the operating lation Discharge Isolation Valve 12)cking." i experience strongly supports the need to assure correc-1. April 2, 1992-NRC Information Notice i tive action is implemented. 92-26, " Pressure Locking of Motor-Operated Flexible Wedge Gate Valves." j Since most of pressure locking sind thermal binding occur-rences are a result of plant evolution, system transients' I 2. C. Hsu, A. L Madison, " Licensee Actions Taken to - or unusual system alignment, we suggest that an NRC/In-Address Pressure I2)cking of Double Disk and Flex-dustry workshop would be a constructive method to com-ible Wedge Gate Valves," Office'for Analysis and municate the importance of system ccmdition analyses in Evaluation of Operational Data, U. S. Nuclear order to identify those valves susceptible to the binding Regulatory Commission, July 1992. I mechanisms. l 3. U.S. Nuclear Regulatory Commission, Regulatory Guide 1.106, "Ihermal Overload Protection for REFERENCES Electric Motors on Motor-Operated Valves." 1. Geneac communications on pressure locking and 4. U.S. Nuclear Regulatory Commission, " Thermal thermal binding issued by the NRC and industry. Overload Protection for Electric Motors on Safety-Related Motor-Operat'ed Valves-Generic Issue a. Circa 1977-Westmghouse Issues Design II.E.6.1," U.S. NRC Repon NUREG-1296, June Change to All Westinghouse Supplied Gate 1988~ Valves Requiring Vent Lines at Salem. j 5. E. Brown," Evaluation of Recent Valve Operator l b. March 9,1977-IE Circular 77-05," Fluid En-Motor Burnout Events," AEOD/S503, September trapment in Valve Bonnets." 1985. c. October 8,1981-IE Information Notice No. 6. J. W. Minarick, J. W. Cletcher, D. A. Copinger, B. 81-31, " Failure of Safety injection Valves to W. Dolan, Oak Ridge National l;tboratory, *Precur-l Operate Against Differential Pressure." sors to Potential Severe Core Damage Accident: 109. A Status Report," U. S. NRC Report d. Circa 1981-Two Recirculation Discharge Iso-AEG/CR-4674 (ORNL/NOAC-232), Vol.15 lation Valves Fail to Open After a Scram at a and 16. September 1992. BWR. 7. L T. Gucwa, Georgia Power Company, " Plant i e. December 1,1981-General Electric Service Vogtle-Unit 1 NRC Docket 50-424, Operating lj- + Infonnation Lettet No. 368, " Recirculation cense NPF-68, Special Report on Pressure Lxicking Discharge Isolation Valve 12 ding." of Motor-Operated Gates Valves," March 31,1988. { l 11 NUREG-1275, Vol. 9 i i l

1 APPENDIX A PRESSURE LOCKING EVENTS ne following pressure locking events occurred from reactor pressure into the bonnet. Pressure of the order of 1969 to the present. Information relating to these events 1000 psig could be reached in the bonnets of all four low was obtained from if.Rs, licensee failure analysis re-pressure ECCS injection valves (LPCI and IECS sys-ports, and international reports. IERs were identified by tems). The licensee's calculations showed that bonnet a search of the Sequence Coding and Search System fluid pressure in the range of 600 to 700 psig could be (SCSS). sufficient to lock the affected valves shut, preventing low pressure ECCS injection; (i.e., the valve operators would not provide sufficient thrust to overcome the double-disc A. Low-Pressure Coolant Injection drag forces caused by the bonnet pressure.) and Low-Pressure Core Spray A second hydrostatic test was performed on July 28,1991. ~ System Inject,on Valves Instrumt ntation installed during this test confirmed that i pressure locking was taking place. During this test, as pressure was increased t 800 psig, the rate of pressuriza-FitzPatrick-LER 333/91--014 tion dropped to zero for approximately 30 minutes, mde Susquehanna Unit I-LER 387/91--013 cating compression of air in the valve bonnet. Target test pressure of 2100 psig was held for 10 minutes and re-IIR 91-014 for FitzPatrick describes an event invohing leased. Birty minutes after depressurization, operators pressure locking of a flexible-wedge gate valve.The plant attempted to open the valve from the control room.The was shut down on May 7,1991, to repair valves m both actuator motor line current went to locked rotor current ITCI lines. On J uly 17,1991, following corrective maint e-and the circuit breaker was manua;1y opened by an elec-nance for valve and actuator problems with the outboard trician monitoring line current. He bonnet was vented LPCI injection valve, a hydrostatic test of the piping was through the stem packing gland. Coincident with the bon-performed between the mboard (MOV-25B) and out-net depressurization, valve position indication in the con-board (MOV-27B) LPCI injection valves. He hydro ~ trol room changed from closed to intermediate.ne valve static test pressure was around 2100 psig. Upon comple-then stroked normally from the control room. tion of the test, the pipmg between the valves was n pre r or eturn g the 1 to se c i e to u oin at 1 SDC mode. high pressure side of the valve. Approximately 9 to 10 hours after completion of the The ITC! and ITCS injection valves at the Susquehanna hydrostatic test, the loop had been filled to the inboard plants are also susceptible to potential pressure locking LPCIinjection valve.ne operatorattempted to open the during LOCA actuation. On October 18,1991, with both 24-inch flexible-wedge gate valve (MOV-25B) from the Units 1 and 2 in Mode 1 at full power, an extensive control room. The actuator remained energized for ap-evaluation of all safety-related air operated valves and proximately 30 seconds after which the motor actuator MOVs was conducted to determine susceptibility to pres-circuit breaker tripped. The normal stroke time for this sure locking and/or thermal binding.He licensee deter-valve is 120 seconds. He licensee determined the root mined that a susceptibility to the pressure locking phe-cause of the motor actuatorfailure to be fluid at pressure nomenon exists for the RHR/LPCI and the LPCS system trapped between the discs of a flexible-wedge gate valve. injection valves. His potential has existed since plant This phenomenon is known as pressure locking. His type construction. of valve arrangement is common in BWRs for both IECI and LPCS injection lines, rendering both systems suscep-Based on the licensee's evaluation, the pressure locking tible to the failure. phenomenon could occur if inboard isolation check valves leaked, allowing the LPCI and core spray injection valve The ITCl and LPCS systems have a testable check valve bonnets to potentially pressurize to recirculation loop between the reactor and the inboard isolation valve - pressure. Following a rapid depressurization during a (flexible-wedge gates).The check valves are only required DH A-LOCA, this trapped high pressure fluid could lead to reduce res erse flo e to less 'han 10 gallons per minute. to pressure locking of the valve, increasing the force re-Small amounts of leakage past the check valves will even-quired to open the valve. The evaluation indicated that - tually bring reactor pressure to one side of the flexible-opening times for IPCI and IPCS injection valves will wedge disc. The wedge will then flex, allowing fluid at delay I second and 4 seccmds, respectively. Although A-1 NUREG-1275, Vol. 9

these times remain within the required ECCS response Prior to the test which resulted in failure of the CS valve, times, the margin of safety is reduced. the RHR/IECI system had been operating and then shut-down with only the pump minimum flow valve open for ne licensee modified the Unit I valves by drilling a small cooling of the pump internals. When the RHR/IECI hole in the upstream discs and planned to make similar pumps are shutdown, system pressure at the elevation of ' modification to the Unit 2 valves during the next refueling the CS isolation valve decays from approximately 240 psig outage. to the normal operating pressure of the " keep full" sys-tem, which is approximately 100 psig. Due to the leak-tightness of the valve design, the fluid trapped between B. Safety Injection System the valve discs stays at 240 psig. Pressure upstream of the - valve was approximately 100 psig and pressure down-stream f the valve was approximately zero psig. His San Onofre Unit 1-LER 206/81-020 resulted in a op across the upstream disc of approxi-Both trains of the SI system were inoperable during an m tely 140 psi and approximat ely 240 psi across the down-actual SI signal when the injection valves could not be stream disc.ne valve was designed to open against a Ap opened as required on September 3,1981.Here were no f 325 psi across only the downstream valve disc. He adverse consequences in this event since there was no f rce required to open the valve Ap of approximately 240 LOCA. De reactor pressure remained above the SI Psi across one disc m addition to approximately 140 psi pump's shutoff head and therefore no actual injection of across the other disc is more than a design op of 325 pst water would have occurred if the valves had opened. across one disc. However, if reactor pressure had decreased and actual mjection been required, injection flow would not have ne valves were modified by drilling a hole approximately been automatically available as designed. 1/8 inch in diarneter through the upstream disc to relieve the pressure in the bonnet and also preclude op across the disc with the hole so that the valve motor would only Subsequent evaluations concluded that the valves failure need to overcome Ap across a single disc when opening to open was due to the pressure locking phenomenon-the valve. De high pressure water trapped in the bonnet cavity caused an excessive bonnet-to-body Ap when the pres-D. Residual Heat Removal Shutdown sure in the injection line decreased. De high op m-creased the disc-to-seat contact forces and locked the Cool.mg Suct. ion Isolat. ion Valve valves closed. All of the injection system valves had been successfully tested periodically during mld shutdown. Foreign Reactor (August 7,1982) However, these tests were not performed while the Rese 900-MWe PWRs have two separate suction paths bonnet-to-body ap conditions existed. for RHR. With the primary system at 175 *C (347 *19and 2.5 MPA (362 psig), two valves in series in one of the RHR ne valves were 14-inch double-disc gate type manufac-suction paths failed to open. No problems resulted from tured by Anchor-Darling. these failures because the two valves in the second RHR suction path opened as designed. De two malfunctioning valves could not be opened until the insulation was re-C. Containment Spray System moved from the valve bodies and the temperature was decreased to 44 *C (111 *19 FitzPatrick-LER 333/88-013 ne RHR valves were flexible-wedge gate valves. ne During a refueling outage at the FitzPatrick plant on E" E** "E

1. " all wed the bon-November 17,1988, the operator motor of the A CS valve ne a rea PMay Feme.W,pmage>

tripped on TOL when attemPti"8 to stroke the valve as part of a post-work test. Subsequent investigation re-sure was reduced, the high pressure water in the bonnet vealed that the valve operator motor had failed on over-cavity was trapped, causing the disc to press more tightly load due to trapped pressurtred fluid in the bonnet cavity. against the valve seats. The utility also believes thermal %c trapped fluid resulted in Ap across the valve disc in binding between the valve disc and seat contributed to the excess of the design value for openmg the valve. The failure. The high friction created between the disc and investigation also found that the same excessive Ap con-seat prevented the valve operators from opening the valves. dition muld exist for the corresponding valve m Imop B. Here was a possibility that both loops of CS would have been moperable when requtred.De valves mvolved were Turkey Point 4-LER 251/89-004 the double-disc gate type manufactured by Anchor-On May 23,1989, with the unit in Mode 4, the RHR Darling. normal suction isolation valve MOV-4-751 failed to open NUREG-1275, Vol. 9 A-2

with its control switch. The valve was a Copes Vulcan been closed under ambien' temperature conditions.The double-disc gate valve. The reactor was being cooled temperature at the valves then increased to approxi-down from Mode 2 in order to test the ability of the plant ma:ely 300 *F due to heat conduction over a time period. to shutdown the reactor using the alternate shutdown 1he temperature increase caused expansion of the fluid panel. This valve was to be opened from the alternate in the valve bonnets, and the increased bonnet pressure shutdown panel to place RHR in service for its normal forced the discs against the seats with such force that the heat removal function. Following the valve failure to motors stalled and failed before unseating the discs. The open, the licensee used the steam generators to continue valves are Westinghouse 8-inch double-disc gate valves cooling the reactor. Two subsequent attempts to open the with Limitorque SMB-00 actuators. valve with the control switch were unsuccessful. Both attempts resulted in breaker tripping on TOl_ The valve To prevent recurrence of pressure locking, the valves was then manually opened. were modified by drdling a 3/16 inch diameter hole through t' e upstream disc.1his hole provides pressure The licensee's evaluation concluded t hat the valve failure equalization between the bonnet area and upstream pip-was caused by pressure locking. Leakage past the up-ing. In this case the drilled upstream disc for the valves are stream disc allowed the bonnet cavity to reach pnmary on the opposite sides of the header such that leak tight pressure. When primary pressure was reduced, the up-isolation of a passive failure of the header can always be stream disc seated tighter than expected due to the cavity obtained. pressure. This trapped the high pressure water in the txmnet cavity and between the discs causing the discs to press more firmly against the valve seats. The motor op-The licensect review of the system operating history re-realed that a motor on one of the actuators failed m erator was not able to overcome the friction force which i October 1987. The failure was attributed to a broken was higher than expected. Valve MOV-4-750 down-stream of MOV-4-751, was also determined to be sus-torque switch, but the failure conditions were sumlar to ceptible to pressure locking. thow for the January 1988 event. It is now beheved that the motor failure m the 1907 event was also the result of The licensce's corrective action included installation of a P"""" "E' pressure equalizing ime on cach valve.1hc same problem was also identified to occur in the corresponding valves in The pressure locking phenomenon in the stated operat-Unit 3.1hese valves were also modified by installing an ing ccmdition was not identified in the review that the equalizing line. licensee conducted in response to an industry generic communication issued in 1984 (Reference li).1he licen-E. Residual Heat Removal Hot Irg Cross-Over Isolation Valve F. Residual Heat Removal Vogtle Unit 1 (Reference 7) Containment Sump Suction Isolation Valve On January 28,1988, at Vogtle Unit 1, both RHR hot leg crossover isolation MOVs failed to open during surveil. lance testing. The umt was in Mode 4 (approximately Ginna 320 F and 350 psi). Prior to entering Mode 3, a routine quarterly RllR pump and valve test was planned. This On December 16,1969, the Ginna plant crperienced a test required opening the valves that had been closed pressure locking of the RHR containment sump suction during the outage when system temperature was approxi-isolation valve. '1he valve could only be opened one-third mately 100 *F to ensure isolation between the two RHR of its total travel during regular operability testing. The trains. During performance of the test, both valves were valve was a 10. inch double-disc Anchor-Darling gate sequentially given a signal to open from the c4mtrol room. valve. Upon disassembly both discs were found to be Within about 30 seconds, an area smoke alarm was re-bowed outward, preventing full opening. The direction of ceived, and the circuit breakers of the two motor opera-the bowing indicated that there was a buildup of pressure tors tripped on overcurrent. I oth motors had overheated between the discs. It was detcrmined that the pressure and failed, and the valves remained in the fully closed increase was caused by beatup of the water entrapped in position. the bonnet and between the discs during plant operation. Vent lines were subsequently installed on both the sump The licensee's subsequent evaluation concluded that isolation valves to preclude pressure k)cking. This modifi-pressure kcking was the most likely cause of both motor cation was recommended by Westinghouse and Anchor-operator failures. Prior to January 28, both valves had Darling. A-3 NUREG-1275, Vol. 9

G. Residual Heat Removal H. Residual Heat Removal Heat Suppression Pool Suction Valve Exchanger Outlet Valve LaSalle Unit 1-LER 373/83-117 Grand Gulf-LER 416/92-001 At LaSaue Unit 1, on September 20,1983, and again on November 12,1983, with the unit in cold shutdown, the "B" RHR heat exchanger outlet valve failed to open by he Division 1 RHR system pump suction valve from the either the motor operator or manually. His valve was a suppression pool failed to open when operators were flexible-wedge gate valve manufactured by Anchor-realigning the system from the SDC mode to the standby Darling. He valve failure made the H SDC loop and B LPCI mode as reactor heatup commenced following a suppression pool cooling loop inoperable. It is believed plant outage on January 8,1992. The motor operator that the valve became inoperable in the closed position TOL circuit tripped on the first attempt to open the valve due to high pressure water trapped in the bonnet cavity. from the control room. The overload device was reset and ne bonnet cavity does not have a mechanism to vent the immediately tripped again. When the trips occurred, re-trapped high pressure water and, therefore, the valve actor coolant temperature was approximately 175 *F and discs became pressure locked. At the recommendation of suppression pool temperature was 74 *F. The valve is a the valve manufacturer, the valve limit switches were 24-inch Powell flexible-wedge gate valve. temporarily adjust ed such that the disc travel was stopped by position and not by torque. nis change kept the valve from going completely closed and, thereby, prevented Operators reestablished Division 1 RHR SDC operation high pressure water from being trapped in the bonnet and reduced reactor coolant temperature to approxi-cavity. His mode of operation was allowed on these par-mately 135 F. The TOL circuit tripped again during ticular valves because they are not required to provide another attempt to open the valve. Operators then at-zero leakage isolation. tempted unsuccessfully to open the valve manually. The temperature difference across the valve was 43 1. The permanent modification was to drill a weep hole in M& &a h & bom cavity. He valve was successfully opened via the motor operator g..Ieetion when reactor coolant temperature was further reduced to 6 a value corresponding to a 20 *F difference across the Steam Admission Valve valve. The valve failure to open was determined to be caused by pressure kicking. As a result of heating the Brunmick Unit 1-LER 325/88-012 trapped water volume in the valve bonnet during plant On May 28,1988, while Brunswick Unit 1 was at full startup, the trapped water pressure increased, causing power, the HPCI steam admission valve failed to open additional forces on the valve wedge.ne licensee's sub-when the switch was energized during a periodic test.ne sequent evaluation concluded that both Division 1 and 2 failure caused a total loss of system function. The cause RHR suppression pool suction valves have the potential was determined to be pressure locking. He valve was ~ to become inoperable due to pressure locking when sub-previously closed with fluid trapped in either the bonnet jected to a temperature ddferential of approximately cavity or between the discs. When the valve was subsc-15 F or greater across the valve. Such conditions may quently heated, the trapped fluid expanded or flashed to exist with the respective RHR train in the SDC mode steam and caused pressure to increase. The pressure in-c(mfiguration. An RllR train in SDC mode operation crease inhibited opening of the valve by causing the with this limitation is inoperable for the LPCI safety func-wedges to press tightly against the valve seats, resulting in tion-binding of the valve. To prevent recurrence the licensee drilled a small hole in the upstream disc to provide a pressure relief path. The valve involved was an Anchor-In a previous evaluation, the licensee, based on engineer-Darling flexible-wedge gate valve. ing judgment, had considered these two suction valves not to be susceptible to thermal binding or pressure locking J. Emergency Feedwater Isolation because of their distance from potential heat sources. g.g Hoth RHR suppression pool suction valves were modified Foreign Reactor (March 22 and August 16, with vent lines to prevent pressunzation of the water 399g trapped in the bonnet. He vent lines were connected His event occurred at a foreign reactor. The utility's from the valve bottem to the upstream pipe. insersice test procedure for the IIFS required that the NUREG-1275, Vol. 9 A--4 L

isolation gate valve should be tested prior to running the shutoff, and pressed the double-discs strongly against the pump. On March 22, and August 16,1990, during two valve seats. He actuator motor torque was insufficient to insenice tests, the utility revised the procedure, running unscat the valve. A calculation, without considering the the pump before exercising the valva in both instances, safety factor, showed that the motor torque needed to the isolation gate valve failed to open. %e valve was a open the locked valves would be about 355 Nm (262 double-disc gate type. A subsequent evaluation indicated Ibs-ft) which is higher than the torque limit switch set-that the valve failure to open was due to pressure locking. point of 265 Nm (195 lbs-ft). The reported insenice test During the pump test, water was pumped against the prececture will not detect the pressure locking condition, closed valve's disc on the pump side, forcing this disc away because the valve is exercised before the pump test is run, from the valve seat, which permitted high pressure water and under such conditions, no pressurized water is en-to enter the valve lxx!y. He water at pump discharge trapped inside the valve. pressure remained inside the valve following the pump i 1 A-5 NUREG-1275, Vol. 9 1 g

APPENDIX B THERMAL BINDING EVENTS De thermal binding events listed below were obtained down. Subsequent evaluations found that the valve fail-from a search of the NPRDS and SCSS database files. ure to open was due to thermal binding and that effons to The search covered the pcriod from 1983 to the present. open the valve had damaged the motor windings. De Some of these events found through the NPRDS search valve was a flexible-wedge gate valve manufactured by were not reported via the LER process. Anchor-Darling. A. Reactor Depressurization System D. Power-Operated Relief Valve Block Isolation Valve Valve At a domestic plant on June 16,1990, while performing Big Rock Po. t-LER 155/84-001 the PORV cycle test, the PORV isolation valve RC-11 in faHed to open from the closed position on demand from At Big Rock Point, three of four reactor depressurization the control room. When this occurred the plant was in system isolation valves failed to open during surveillance preparation for start-up following a refuelmg outage and testing on February 22,1984. The plant was in Mode 4 had to cooldown for a valve replacement. Following the with reactor pressure at 50 psig. The valves involved were plant cooldown, the valve stem was found separated from air-operated flexible-wedge gate valves manufactured by the discs when the operators managed to open the valve Anchor-Darling. De failure was caused by the disc wedg-manually. Subsequent inspection showed a visible crack ing tightly into its seat so that the air operator spring force on the face of the valve seat.The cause of the valve failure was insufficier't to open the valves. De valves had been was determined to be dus to thermal binding. The valve closed while hot. When the valves cooled during plant involved was a Vulcan gate valve. shutdown, thermal binding occurrect. E. Reactor Coolant System Letdmvn B. Residual Heat RemovalInboard Cooler Isolation Valve Suction Isolation Valve Davis-Besse-LER 346/90-002 LaSalle Unit 1-LER 373/83-142 On November 4.1983, at LaSalle Unit 1, while the unit MU-2B at Dav. Besse was closed to isolate reactor cool-is-was in hot shutdown for planned maintenance, the m.- ant letdown following a reactor trip. On January 29, ocard suction isolation valves could not be opened ty MU-2B was opened but no flow was observed. The op-cither the motor operator or manually. He valve was a crators then closed the valve and observed the following: fl b!c-wedge gate valve manufactured by Anchor-(1) erratic position indication (2) motor operator went to "E-stall current, and (3) stem to disc separation. The valve failure was determined to be caused by thermal binding. He valve was manually seated to stop leakage during the When the valve was closed during the trip, flow through previous operatmg penod. Subsequently, with the plant the valve stopped, the valve cooled relative to its tem-m hot shutdown and temperatures significantly less than perature when there is flow through the valve. The rigid when the valve was manually seated, the valve failed to disc wedged into the seat as the valve cooled. When op-. open due to thermal binding. The valve was opened by crators attempted to open the valve, the stem separated externally heating the valve body. from the disc. The valve was a Velan solid wedge gate valve. C. High-Pressure Coolant Injection Steam Admission Valve F. Residual Heat Removal Suppression Pool Suction Valve Brunswick Unit 1-LER 325/88-017 Six thermal binding events occurred to four of the RHR On July 1,1988, at Brunswick Unit I with the unit at 68 suppression pool suction valves at both units at a domestic percent power, while performing the operability test of site. Two events occurred at Unit 1 on January 28 and the unit HPCI system, the HPCI turbine steam supply March 7,1991.ne four other events occurred at Unit 2 isolation valve would not open.The unit was then shut-on June 26, 1989 and August 17,1990. At Unit 1, the B-1 NUR EG-1275, Vol. 9 4

m-_ f i - t l RHR suppression pool suction valves 1-E11-F004B and binding.nc valve was complet ely rebuilt and returned to -i E11-R)04D would not open and the operator motor normal service. nc valve was a Rockwell gate valve. l tripped on TOLs w hen an open signal was given on two separate occasions. The unit was at zero power when the At another domestic plant site, on October 23,1988, with j event occurred. At Unit 2 with the unit at 7ero power,on the plant in cold shutdown, it was discovered during sur-l June 26,1989, during periodic testing, the RHR suppres-veillance testing that the letdown heat exchanger inlet sion pool suction valves 2-E11-R)D4B and 2-Ell-R)D4D header containment isolation valve would not stroke to would not open when the signal to open was given.ne the closed position. De cause of the failure was deter-valves were then found to be stuck in the seats. He mined to be thermal binding. An investigation revealed failures were due to thermal binding. He valves were that the valve discs expanded greater than the body seats disassembled, cleaned, and satisfactorily tested following when heated. A design change was implemented to re- ' } t reassembly. On August 17,1990, these two suction valves place the existing gate valve with a globe valve. nc valve again would not open from the reactor turbine gauge involved was a WKM valve. l room due to valve operator motors tripping on the TOL. j ne root cause of the failures was thermal binding.%c II. Condensate Discharge Valves system operating procedures were revised to allow the valve to be manually cracked off the seat prior to motor Thermal binding occurred on the two condensate dis-operation to alleviate thermal binding. Ihe valves m-charge valves (MO-2998B and MO-2998A) at a plant on I volved were Anchor-Darling flexible-wedge gate valves. two separate occasions (April 2 and April 26,1991). With l the plant in the cold shutdown condition for refueling, personnel performing preventative maintenance activi-l G. Containment Isolation Valves 'S I "h* "^* "cre un ble to get the valves to stroke l open. Since the system was not in service at these times, j the failures had no effcct on plant operation.The cause of At a domestic plant on Dectmber 18,1988, with the unit the failures was seat binding as a result of thermal cycling. 1 in cold shutdown and preparing to heatup, the chemistry The valves were closed when the valves were still hot. l technician attempted to sample the pressurizer steam This was contrary to the plant procedure. As they cooled, 1 space and found that the containment isolation valve for the valve bodies contracted, causing the disc to bind. A-I this sample line would not open. His is a solenoid-contracting firm was brought in to heatup the valve to - I operated valve and no electrical problems could be found. allow it to be opened. Hey unstuck when the tempera-The licensee determined the failure was due to thermal ture reached approximately 225 'F. t 3 1 i. 1 a 4 NUREG-1275 Vol.9 B-2 P - - ~ -m-,=

APPENDIX C

SUMMARY

OF AEOD PLANT SURVEY Despite actions taken by licensees in response to previous b. The water trapped in the valve bonnet expands generic comm unications issued by the NRC and industry, as a result of heating during a HElliA. the pressure locking problem still occurs to gate valves installed in safety-related systems. In order to understand c. One side of the valve is initially pressurized by the extent and adequacy of the corrective actions taken by check valve leakage and then suddenly depre-licensees, AEOD conducted a survey of six selected ssurized as a result of a IDCA. plants (four BWRs and two PWRs).nc site visit at one plant (Nine Mile Point) was canceled from the survey The analysts concluded that none of the 11 valves is process because of an ongoing outage and other NRC subject to motor stall curren: under the first two conditions. His conclusion is based on the expecta-inspections. However, the evaluation documents from the plant were reviewed. Survey results are documented tion that Icakage rates, as derived from the current in AEOD report " Survey of Ijcensee Actions Taken to localicak-rate test (IlltT) leak rate measurements, Address Pressure Locking of Double Disk and Flexible are greater than the bonnet water thermal expan. s n rates for all eleven valves. Among the 11 valves, Wedge Gatc Valves" (Ref crence 2).His Appendix pro-vides a summary of the survey results. four valv are affected by the third condition. LPCI: 10 MOV-25A 123 see 10 MOV-25B 55.4 min James A. FitzPatrick: IECS: 14 MOV-12A 0.0 sec 1. Ilad not considered problem credible because: 14 MOV-12B 233 min a. lack of plant spectfic pressure k>cking events. All four valves will not function until the pressure of the fluid trapped in the valve bonnet decays to a b. 1.ack of system-specific failure experience. level less than the maximum allowab!c bonnet pres-sure the valve actuator can overcome to open the c. He specified phenomena in industry commu-valve ne depressurization times for these valves nications are too general, not specific to LPCI are estimated to be from 12 seconds to 55 minutes, system. These time periods are larger than the required valve response times during a LOCA (two valves: 10 d. Engineering could not provide a comprehen-MOV-25B AND 14 MOV-12B). sive analysis and calculation to convince man-agement to initiate corrective action. Although the analysis provided analytical justifica-tion for delaying installation of a bonnet vent for the c. Net able to identify valves with the potential remaining seven valves, the contractor recom-for pressure h>cking or thermal binding condi. mended modifications during the next outage to tions. climinate the possibility that maintenance or repair on the subject valves could result in a reduction of leak rates and an increase in leak-off times ne f. Engineers did recommend modification to the 1.PCI injection valves in 1985. Management licensee is still evaluating reactor core isolation cool-declined to take action due to lack of actual ing, HPCI, ccmtainment coolmg, and sump isolation valves for modification. However, they have mds-operating event. cated that modification for these valves may not be necessary due to opening the valves during rapid 2. Following the 1991 cvent (LER 91-014) depressurization being unlikely and the op across the HPCI valve being low during a small break A consulting firm (Failure Prevention Inc.) was LOCA. (lhis may not be true for the c(mditions hired to evahtate impact on other motor-operated. during a large break LOCA). flexible. wedge gate valves. Eleven valves were se-lected to be examincd for the following ibrec scenar-3. Training Progmm ios: The licensee originally did not provide training on a. The water trapped in the valve bonnet expands this failure mechanism. Following generic commu-as a result of heating during a normal plant nicationss the issue was placed in required reading startup. and maintenance lesson plans. ne item was not C-1 NUREG-1275. Vol. 9 l J

= I l t i incorporated in training courses for operations valve showed that the valves were leak-tight with l personnel until 198S; In 1989, a training plan was near zero leak rate. The low head injection valves j provided for maintenance personnel.The item was are 6 inch Velan gate valves. then added to the continuing training program for engineering support groups following the 1991 The licensee planned to initiate a review of all other event. gate valves for potential pressure locking. Subse-quent to our visit, the licensee hired an independent 4. Surveillance Test consultant to assess the problem. The assessment concluded the low heao pressure injection valves are ne LPCI valve IST surveillance was performed susceptible to pressure locking. - j quarterly. Operability testing was performed 4. ,Fra..mmg Program monthly. Before cycling the valve, nitrogen gas was injected into the upstream side of the valve to reduce the Ap across the valve disc if downstream pressure The failure mechanism concern was provided in the was too high. The nitrogen pressure could go up to training instructions for operations and mainte-j 400 psig.The licensee believed the downstream high nance personnel. However, no such training was pressure would decrease or diminish during a provided for the engineering staff. ] LOCA; therefore, a large op condition was not pos-5. Surveillance Test sible. Surveillance testing of the low head injection valves Ginna: was performed by cycling the valves within 48 hours after plant cold shutdown. 1. An RHR containment sump suction isolation valve j experienced a pressure lockup during a monthly test Nine Mile Poinu in 1969. This was a pressure locking scenario due to expansion of entrapped water in the valve bonnet 1. He licensee provided documents that revealed it I cavityas a result of heating. The valve was a 10 inch had reviewed all double-disc and flexible-wedge gate double-disc gate type manufactured by Anchor-valves manufactured by Anchor-Darling, Rockwell, Darling. Vent lines were subsequently installed on Crane, and Powell in response to industry communi-both sump isolation valves. This modification was cationson pressurelockingissues. Additionally,the recommended by Westinghouse and Anchor-engineering organization supporting Nine Mile Darling. Following the event, the licensee initiated Point had recommended modifications to prevent an inspection of all other valves of the same type valve pressure locking. However, plant managers provided by Anchor-Darling and Alloyco. Thirty did not approve the implementation of the recom-valves were motor-operated and fourteen were mended modifications because valve pressure lock-manual. Modifications were recommended for sev-ing was not perceived as credible since they lacked eral normally closed MOVs which are provided with site-specific experience with the phenomena. heat tracing. Dese valves are: (a) refueling water Therefore, no modifimtions were performed at storage tank supply to Si pumps, (b) boric acid sup-Nine Mile Point. ply to Si pumps. and (c) emergency boration line. De results of this inspection were used in their 2. The licensee had provided training to operations and response to IE Circular 77-05 in 1977, maintenance personnel but did not appear to have provided training for engineering personnel. 2. Icdustry Communication: The licensee considered that valves manufactured by a company other than Anchor-Darling were not 1. Allof thedouble-disc,orflexible-wedge,gatevalves susceptible to pressure locking.The evaluation done installed in other than cold water systems at Salem for IE Circular 77-05 was determined to be ade-have been previously surveyed for susceptibility to quate for their response to the industry communica-thermal binding and pressure locking by the licensee tion. and Westinghouse during the plant design and ccmstruction phase (commercial operation: Unit 1 in 3. De licensee was not aware of the pressure locking June 1977, Unit 2 in October 1981). A 3/4-inch ex-potential far the low head injection valve until the ternal pressure vent line, with a block valve, was AEOD visit.ney had never considered the possibil-installed between the valve bonnet cavity and the ity of check valve leakage.The recent LLRT for both upstream pipe in susceptible valves. Examples, are the low head injection valve and downstream check the RHR and Si systems. NUREG-1275. Vol. 9 C-2

et w 2. In their evaluation in 1986, the licensee concluded entrapment, pressure locking or thermal con-that the previous modifications in the plant design traction. and construction stage were adequate to prevent c. There is a low risk of liquid entrapment or pressure locking or thermal binding from oaurring. There is a very low probability for the problem to bonnet pressurization because the fluid will ~ occur to these gate valves without modification.his leak past the seat, or possibly through the stem is based on: (1) documented history of no valve ' packing to relieve bonnet pressure. failure due to pressure locking or thermal binding, d. Although General Electric had recommended and (2) all gate valves in safety syFlems were pro. cured to a design specification that cites an allowable modifications to several gate valves in the seat leakage of 2 cc/hr/in of nominal valve size for its recirculation and RHR system (CS, RHR injec-hydrostatic seat leak test. (This is the maximum leak. tion), the licensee chose to use maintenance age rate at hydrostatic test pressure.The actual out. and operating pmcedures to present the prob-- leakage rate would be less. A consenative assess. Iem from occurring. such as removal of insula-ment should assume an out-leakage rate based on a tion to allow valve cooling, loosen the valve gate valve with new, rebuilt, or reworked valve discs packing to relieve bonnet pressure, or alter-and seat under an actual test with higher Ap across nately open and close the affected valves dur-the valve). He licensee indicated that generally, ing cooldown or heating up to prevent thermal there is a low probability that liquid entrapment or binding. (The measures chosen by the licensee pressure locking of gate valves will occur since the may not be effective in preventing occurrence trapped liquid will leak past the seat or possibly of pressure locking or thermal binding). through the packing gland area to prevent pressure 2. Training programs on this item had been pmvided buildup. for maintenance and operations personnel. The li-censee plans to provide the same training to engi-3. Training Program neermg personnel. The item has been included in the training program 3. He licensee admitted that depending on engineer-foi enaintenance and operations personnel since the ing judgement to determine depressurization rate time they performed their evaluation. In process of for fluid entrapped in the bonnet is not adequate. developing training mstructions for engmeermg ney said they plan to initiate a complete reevalu-staff subsequent to our survey. ation of the plant gate valves. 4. . Surveillance Test g. Surveillance test of low head injection valve was 1. An initial review in 1985 identified that only RHR conducted during plant cold shutdown. SDC valves HV-F008 and HV-F009 were suscepti-ble to thermal binding and/or pressure locking. As to Si valves, the licensee considered that the associated Hope Creek: check valves would release the pressure in these valves, therefore pressure locking would not occur 1. Based on the engineering evaluation conducted in to them. Although the review recommended modifi-response to an industry communication, the licensee cat n to valves HV-F008 and HV-F009, manage-did not consider the gate valves at the plant to be ment chose to modify some procedures to avoid oc-susceptible to pressure locking or thermal binding. currence of the problem due to lack of operating Therefore, the licensee chose not to modify any of expnience. the gate vaives. 2. Following the 1991 FitzPatrick valve pressure lock-The licensee evaluation indicated that: ing event, the licensee. re-evaluated all motor-operated flexible-wedge and double-disc gate valves a. Here was no history of gate valve failure due to with tecnnical consulting from an engineering firm. this mechanism. He evaluation determined susceptibility to'the pressure locking phenomenon existed for the LPCI b. Generally, the gate valves function as equip. and LPCS injection valves (R115 A & H. F005 A & ment isolation maintenance valves and thus, B) on both Units 1 and 2. The licensee modified the they are locked open, or normally in the open Unit I valves by drilling a small hole in the upstream position during plant operation. Gate valves discs. He Unit 2 valves will be modified during the operated in the open position preclude liquid next refueling outage. C-3 NUREG-1275 Vol.9

3. After the discussion of the assumptions used in the the valve bonnets; (2) pressure decay rate of en-licensee's re-evaluation during our visit, the licensee trapped fluid in the bonnet due to valve leakage was considered that some assumptions may not be con-based on as-found 11RT data, which is higher than servative and planned to review those assumptions, that of actual conditions; (3) the available actuator thrust was based on vendor-supplied data rather 4. The nonconservative assumptions given in the licen-than actual test data. see's evaluation of potential pressure locking due to system depressurization were: (1) pressure increase 5. Training programs for maintenance and operations . as a result of bonnet volume heatup was not consid-personnel were provided. Ilowever, no such training cred in the calculations for the maximum pressure in was provided for the engineering staff. l. NURiiG-1275, Vol. 9 - C-4

U.S. tfJCLEAR REGULATORY COMMISSION

1. REPORT NUMBER NRO FORM 335 1 Assignnd by NRC. Add Vol..

(2-49) $upp., Rev., arid Addendarn Narn-NRCM 1102, ta'$- d *avd 3m, 32a2 BIBLIOGRAPHIC DATA SHEET NUREG-1275, Vol. 9 (See instructions on tne reverse) AEOD/S92-07

2. TITLE AND SUDTITLE

3. DATE REPORT PUBLISHED Operating Experience Feedback Report-Pressure locking and 'Ihermal Binding you;g i

y, y, I of Gate Valves March 1993

4. FIN OR GRANT NUMDER Commercial Power Reactors

6. TYPE OF F4EPORT fi. eult40HLS)

Technical C. Hsu

7. PERIOD COVERED (inclusive Dates)

September 1981-December 1992

8. Pf.RFORMING ORGANIZATION - NAME AND ADDRESS (if NRC, provide Divison. Office or Regm, U.S. Nuclear F$agulatory Commission, arar

~ analling address; if contractor, provide name and fnalling address.) Division of Safety Programs Office for Analysis and Evaluation of Operational Data U.S. Nuclear Regulatory Commission Washington, DC 20555

9. 6PONSORING ORGAt4ZATION - NAME AND ADDRESS (if NRC, type *Same as above"; if contractor, provide NRC D6 vision, Office or Region, U S. fAclear Regulatory Commissiori, and mail 6ng address.)

Same as 8. above.

10. SUPPLEMt~NT ARY NOTES

11. ABCTRACT (200 words or less)

The potential for valve inoperability caused by pressure locking and thermal binding has been known for many years in the nuclear industry. In spite of numerous generic communications issued in the past by the Nuclear Regulatory Commission (NRC) and industry, pressure locking and thermal binding continues to occur to gate valves installed in safety-related systems of both boiling water reactors (BWRs) and pressurized water reactors (PWRs). The generic communications to date have not led to effective industry action to fully identify, evaluate, and correct the problem. This report identifies: (1) conditions when the failure mechanisms have occurred, (2) the spectrum of safety systems that have been subjected to the failure mechanieme; and (3) conditions that may introduce the failure mechanisms under both normal and accident conditions. On the basis of the evaluation of the operating events, the Office for Analysis and Evaluation of Operational Data (AEOD) of the NRC concludes that the binding problems with gate valves are an important safety issue that needs priority NRC and industry attention. This report also provides AEOD's recommendation for actions to effectively prevent the occurrence of valve binding failures.

13. AVAILABluTY STATEMENT ~

12. KEY WORDB/DESCRPTORS (List words or pesases that will assist researcters in locating the regart.)

Unlimited

14. SECURITY CLASSlFICATION pressurelocking W' P'84 thermal binding Unclassified MOV (n sepero gate valve Unclassified valve binding is. NuusER oF eAGES low-pressure coolant injection low-pressure core spray residual heat removal

16. PRICE NRC FORM 335 (2-49)

i l t i l ~ 4 i i-f .1 i Printed On recycled paper l a Federal Recycling Program

NUREG-1375, Vol. 9 OPERATING EXPERIENCE FEEDHACK REPORT - PRESSURE LOCKING AND MARCH 1993 TIIERMAL BINDING OF GATE VALVES UNITED STATES nRsT CLASS Mall NUCLEAR REGULATORY COMMISSION roSTAGE AND FEES PAfD WASHINGTON, D.C. 20555 0001 USNRC 12'5591?'2531 1 laNI) S i l f 1'? n ' OFFICIAL BUSINESS US NCC-3A'" FENALTY FOR PRIVATE USE, $300 cIy Fn?a r, U n t i r

• T I r r, ^

Vr ipeo r p p. *! ; ! a r r. r,* 1

• h
• 3 H } *Q ', T C I, CI

'i~ f i _}}