ML18026A272
| ML18026A272 | |
| Person / Time | |
|---|---|
| Site: | Susquehanna  |
| Issue date: | 05/16/1996 |
| From: | Eugene Kelly NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | Byram R PENNSYLVANIA POWER & LIGHT CO. |
| References | |
| NUDOCS 9605280003 | |
| Download: ML18026A272 (100) | |
Text
May 16, 1996 Nr. Robert G. Byram Senior Vice President - Nuclear Pennsylvania Power E Light Company 2 North Ninth Street Allentown, Pennsylvania 18101
SUBJECT:
BWR SAFETY RELIEF VALVE PERFORMANCE MEETING
Dear Hr. Byram:
This refers to the public meeting conducted in King of Prussia, Pennsylvania on April 24, 1996.
This meeting was held to discuss safety relief valve performance issues with Boiling Water Reactor (BWR) licensees.
Special emphasis was placed on reviewing safety relief valve leakage, corrective actions that have been implemented to address setpoint dri.ft, and associated issues.
The meeting was attended by 37 individuals, including those representing your organization.
Copies of the presentations are attached.
In accordance with Section 2.790 of the NRC's "Rules of Practice,"
Part 2, Title 10, Code of Federal Regulations, a copy of this letter will be placed in the NRC's Public Document Room.
We appreciated the responsiveness and turnout for this meeting, and will continue to review SRV performance issues and discuss areas of concern with BWR licensees.
Should you have any questions concerning any of the topics
- raised, we will be pleased to discuss them further with you.
Sincerely, Docket Nos. 50-387;50-388 Attachments:
1.
Attendees 2.
Agenda 3.
Licensee Presentations 4.
NRC Presentations Eugene H. Kelly, Chief Systems Engineering Branch Division of Reactor Safety 9605280003 9605ib PDR ADOCK 05000387 P
PDR aooo34 4r-'
Hr. Robert G,
Byram cc w/encl G. T. Jones, Vice President - Nuclear Engineering G. Kuczynski, Plant'anager J.
M. Kenny, Supervisor, Nuclear Licensing G.
D. Miller, Manager - Nuclear Engineering R.
R. Wehry, Nuclear Licensing H. H. Urioste, Nuclear Services
- Manager, General Electric C.
D. Lopes, Manager - Nuclear Security W. Burchill, Manager, Nuclear Safety Assessment H.
D. Woodeshick, Special Office of the President J.
C. Tilton, III, Allegheny Electric Cooperative, Inc.
Nuclear Safety Information Center (NSIC)
Commonwealth of Pennsylvania
Nr. Robert G.
Byram Distribution w/encl:
Region I Docket Room (with concurrences)
K. Gallagher, DRP NRC Resident Inspector PUBLIC W. Dean, OEDO C. Poslusny, Project Manager, NRR J. Stolz, PDI-2, NRR Inspection Program Branch, NRR (IPAS)
DRS file '(2)
DOCUMENT NAME:
A:FOLLOWUP.SRV To recehe a copy of this document, indicate in the box:
C' Copy without attachment/enclosure E'
Copy with attachment/enclosure 'N'o copy OFFICE DATE RI/DRS KKOLACZYK 05/14/96 RI/0RS GKELLY 04/
/9 05/
/96 OFFICIAL RECORD COPY 05/
/96 05/
/96
f
ATTACHMENT 1 SRV NEETING ATTENDEES Nuclear Re ulator Commission NRC Re ion I Gene Kelly Walt Pasciak Kenneth Kolaczyk John Caruso Art Burritt NRC - Head uarters Gary Hammer Nary Wegner Frank Cherny Joe Sebrosky George Thomas NRC Re ion II Joseph Lenahan
,NRC - Re ion III Zelig Falevitis John Neisler Philadel hia Electric Com an Brian Kozemchak Jim Nitman Greg Curtin Northeast Nuclear Ener Com an Brian Smith Don Clemans Donald Cleary Penns lvania Power 3 Li ht Com an Gerald Naertz Otis Osborn Boston Edison Com an Bill Carrol New York Power Authorit Rick Plasse Gerry Ottman Sherard Anderson De artment of Environmental Protection Rich Janati Dave Ney
Attachment 1
General Electric John Dale Willard Roit Electric Power Research Institute EPRI Vick Varma State of New Jer se Ariadni Kapsalopoulou Nebraska Public Power District Coo er Nuclear Station John Lindinger Tar et Rock Cor oration Andrew Szeglin Ventre Kevin Heormen W le Laboratories Patrick Turrentine ALTRAN - Materials En ineerin Incor orated Ockert Van Der Sebijff Other James White
1 ATTACHHENT 2 BWR SAFETY RELIEF VALVE PERFORMANCE SHERATON VALLEY FORGE APRIL 24, 1996 AGENDA 8:00 a.m.
8:15 a.m.
9:00 a.m.
9:45 a.m.
10:00 a.m.
10:45 p.m.
Welcome and Opening Remarks by James Wiggins, Director, Division of Reactor Safety "SRV Setpoint Drift Fix Committee - Status" by Gerry Ottman, BWR Owners'roup "NRC Perspective Concerning SRV Performance" by Gary Hammer, Hechanical Engineering Branch, Office of Nuclear Reactor Regulation, NRR and Nary Wegner, Reactor Analysis Branch, AEOD Break "SRV Performance Honitoring at FitzPatrick," by Sherard
by Kenneth Kolaczyk, NRC Region I ll:30 a.m.
Lunch 12:45 p.m.
1:30 p.m.
"Limerick September 1995 'Event,"
by Jim Hitman, PECO Energy "EPRI SRV Haintenance manual,"
by Vick Varma, Electric Power Research Institute 1:45 p.m.
Break 2:00 p.m.
3:00 p.m.
"Results of the Hillstone Unit 1
SRV Setpoint Drift Study," by Brian Smith, Northeast Nuclear Energy Company Closing Remarks by Eugene H. Kelly, DRS Systems Engineering Branch 4:00 p.m.
Adjourn
ATTAC8MfNT 3 BWROG SRV DRIFT FIX.
DEVELOPMENTCOMlNITTEE Presentation for BWR SAFETY RELlEF VALVE PERFORMANCE MEETING April24, 1996 VALLEYFORGE, PA Gerry Ottman Committee Chairman NYPA
SRV DriftFix Development Committee 2
Committee Objective:
~ RESOLVE TARGET ROCK TWO-STAGE SRV SETPOINT DRIFT PROBLEMS THROUGH EITHER:
ELIMINATIONOF PILOT DISC TO SEAT STICK(NG, OR USE OF PRESSURE SWITCHES TO ACTUATETHE SRV ATTHE DESIRED PRESSURE
~
REDUCE PILOT LEAKAGETO IMPROVE SRV PERFORMANCE Presentation Objective:
~ PRESENT THE CURRENT DIRECTION, PROGRESS AND SCHEDULE OF THE COMMITTEE
SRV DriftFix Development Committee
Background:
~ THE COMMITTEEWAS FORMED WITHTHE INTENTTO RESOLVE THE TRC 2-STAGE SRV SETPOINT DRlFT PROBLEM.
. ~ lDENTIFlEDTWO MECHANISMS CONTRIBUTINGTO SRV DRIFT:
LABYRINTHFRICTION:
.PILOT DISC TO SEAT BONDING LABYRINTHFRICTION ADDRESSED BY IMPROVED MAINTENANCE INITIALATTEMPTTO ELIMINATE DISC TO SEAT BONDING USING PH13-8 Mo DISC WAS UNSUCCESSFUL THE COMMITTEE IS PRESENTLY'NSTALLING ANDTESTING Pt ALLOY PILOT DISCS
SRV DriftFix Development Committee THE USE OF PRESSURE SWlTCHES, A SEPARATE MECHANISMTO OPEN SRVs, IS AN ALTERNATEMETHOD OF RESOLVING SETPOINT DRIFT EVALUATFTHE SUCCESS OF THE ALLOYDISC (PHASED APPROACH)
A} TOTALFIX, OR
- 8) ADD CATALYSTPLATING, OR C} INSTALLPRESSURE SWITCH
0
SRV DriftFix Development Committee Issues:
~ CATALYSTSTATUS ALLMEMBERS EXCEPT DECO (NEW MEMBER) HAVE PURCHASED AND INSTAI LED Pt ALLOYDISCS CURRENTLY EVALUATING THE SUCCESS OF THE ALLOY DISC THROUGH INSTALLATIONOF DISCS IN SRYs IN OPERATiNG PLANTS PRELIMINARYDATAINDICATE EXCELLENTRESULTS WITH Pt ALLOYDISCS
~
PRESSURE SWITCH/TRANSMITTER SER RECEIVED, 10/95 IN OPERATION AT HATCH PLANTS FOR 2 CYCLES, RECENTLY INSTALLEDAT lNILLSTONE
SRY DriftFix Development Committee SRV PILOT LEAKAGE CONCERNS EXPERIENCE AT LIMERICK INDICATESGROSS PILOT LEAKAGEMAYRESULT IN SRV BLOWDOWN
. INVFSTIGATINGEFFECT OF PILOT LEAKAGEON SRV PERFORMANCE
SRV DriftFix Development Committee Actions:
~ IMPLEMENTATIONOF Pt ALLOYDISCS TO FIX SETPOINT DRIFT IS PROGRESS(NG FINALDECISION ON EFFECTIVENESS OF Pt ALLOYOISCS WILLBE MADEIN SUlNMER, 4997
~ COMMITTEEPURSUING ACTIONS TO REDUCE PILOT LEAKAGE ELIMINATIONOF STROKING SRVs DURING STARTUP RECOMMENDATIONSTO IMPROVE MAINTENANCEOF SRV PILOTS IMPROVED LEAKAGEDETECTION/SRV MONITORINGTO QUANTIFYLEAKAGE
Participants:
. DECO TVA NUSCO
SRV DriftFix Development Committee
'8
SUMMARY
OF SRV DATABASE NUMBER OF DISCS:
33 Pt-STELLITE 24 STELLITE 0
AVERAGE DRIFT:
Pt-STELLITE STELLITE 2.7%
5.3%
NewYorkPower Authori
. SRV PERFORMANCE MONITORING AT FITZPATRICKPLANT GERRY OTTMAN SHERARD ANDERSON r
PRESENTED AT:
BWR SAFETY RELIEF VALVE PERFORMANCE MEETING VALLEYFORGE, PA.
APRIL 24, 1996
NewYorkPower Authori PLANT DESCRIPTION:
g~R - 4, MARK 1 CONTAINMENT RATING: 840 MWE YEAR COMMERCIAL: 1975 SRV DESCRIPTION:,
TARGET ROCK 2 STAGE (PILOTED) 11 SRVs / 7 ADS
NewYorkPower Authori TAILPIPE CONFIGURATION:
TAILPIPES ARE FULLYINSULATED THERMOCOUPLES STRAP ON OUTSIDE OF PIPE LOCATED 4 6 FEET FROM SRV
'~
NewYorkPoorer Authori HISTORICALPERFORMANCE CYCLE 11 (FEB,1993 NOV, 1994) 10 SRVs WITH STELLITE PILOT DISC 1 SRV WITH PT ALLOYDISC AVERAGE DRIFT: 4.1 %, 6 PILOTS LEAKING
NewYorkPower Authori CYCLE
SUMMARY
STELLITE LEAKING STELLITE LEAKTIGHT PT ALLOY LEAKING QUANTITY 5
AVERAGE SETPOINT DRIFT 2.5 5.1 1.5 MAXIMUM SETP OINT DRIFT QUANTITY WITH )
3% DRIFT 5.8 %
1.5%
0
NewYorkPower Authori OBSERVATIONS FROM CYCLE 11:
TAILPIPE.TEMPERATURE OBSERVED FOR VALVESWlTH PILOT LEAIMGE POOR CORELATION BETWEEN LEAKAGE AND TEMPERATURE INCREASE LEAKAGEREDUCED SETPOINT DRIFT FOR
.PILOTS WITH STELLITE DISCS TOTAL SRV LEAIMGE200 0/HR TORUS HEATUP 0.8 F/DAY SRV LEAKAGEDID NOT NECESSITATE EXTRA OPERATION OF TORUS COOLING
NewYorkPower Authori EARLIER PLANT EXPERIENCE:
OPERATING CYCLE 10 (1990-1991):
ONE SRV PILOT SUBSTANTIALLY STEAM CUT CALCULATEDLEAKAGEWAS: 325 0'/HR TORUS HEATUP WAS 1.35 F/DAY TORUS COOLING WAS RUN ONCE EACH 10 DAYS
NewYorkPoier Auihori CURRENT PLANT STATUS:
SRV LEAKAGE 9 SRV TAILPIPE TEMPERATURES STABLE 1 INCREASING 1 DEG. F / DAY ( D SRV )
1 SLOWLY TRENDING UP ( F SRV )
ESTIMATED TOTAL LEAIWGE: 60 ¹/HR CURI&NTTORUS HEATUP:
.25 DEGREES F /
DAY
. SHUTDOWN:HEATUP,:
.:20.DEGREES-.F./.
DAY TORUS COOLING OPERATED IN SUPPORT OF PLANT SURVEILLANCETESTING
~
~
STARTUP SRV TESTING
NewYorkPower Authori CURRENT TAILPIPE TEMPERATURES:
02RV-71A:.
217'F 195'F 240 F 297'F 230'F 275'F 215'F 277'F 255 F 267'F
. 190'F
NewYerk Power Authori CURRENT PLANT STATUS:
LEAKAGEINITIATION:
TAILPIPE TEMPERATURES OF ALL7 LEAXING PILOTS INCREASED AS A RESULT OF STARTUP CYCLING
~
FOUR IMMEDIATELYFOLLOWING TESTING OTHERS WITHINONE WEEK
'RECENT REPLACEMENTS
~
THREE IN 9/95
~
SAME THREE IN 2/96 TWO OF THREE REPLACED IN 2/96 REFURBISHED WITH IMPROVED INSPECTION AND MAINTENANCEPRACTICES:
1 TIGHT LONGTIMEPOOR LOCATION 1 EXHIBITINGLEAID!AGE
NeviYorkPoorer Authori PERFORMANCE MONITORING:
METHODOLOGY BULKTORUS HEAT-UP MEASURED HEAT BALANCEPERFORMED TORUS WATER TEMPERATURE CHANGE SRV STEAM LEAIMGE ALLTORUS HEAT-UP ASSUMED TO BE FROM SRVS FOR CONSERVATISM OTHER POSSIBLE CAUSES:
~
AMBIENTEFFECTS DRMVELLTEMP:
110 DEG. F TORUS AIR TEMP:
85 DEG. F NORMALTORUS WATER TEMP: 65-82 DEG. F TORUS ROOM TEMP.
75 DEG. F
~
INPUTS FROM OTHER SYSTEMS
NewYorkPower Authority PERFORMANCE MONITORING OTHER OBSERVATIONS NO CORRELATIONS WITH TAILPIPE TEMPERATURES EXIST SRV LEAIMGE TORUS HEAT-UP RANGE OF TAILPIPE TEMPERATURES:
NORMAL:
160 TO 240 DEG. F LEAIGNG:
INSULATED TAILPIPES SUPERHEATING OF STEAM EXITINGPILOT
0
NewYorkPower Authori PERFORMANCE MONITORING COMPARISON TO TEST RESULTS:
3 PILOTS REMOVED IN SEPT, 1995 TOTAL CALCULATED: 105 ¹/HR ACTUALLEAKAGEOF 3 PILOTS:
54 ¹/HR DIFFERENCE ATTRIBUTEDTO:
~
MINOR LEAIWGE FROM OTHER SRVs NOT QUANTIFIED
~
CONSERVATISM IN METHODOLOGY
NewYorkPower Authori OPERABILITYDETERMINATION:
FITZPATRICKTECH SPEC ALLOWS 2 SRVS OUT OF SERVICE TARGET ROCK LIMITIS 200 ¹/HR (OPEIVV3ILITY)
AT 400 ¹/HR TOTALLEAKAGETWO SRVS CONSIDERED OOS A PLANNED SHUTDOWN TO OCCUR PRIOR TO REACHING 600 ¹/HR MONITORINGGUIDANCE IS CONTAINED IN PLANT PROCEDURES TARGET ROCK LIMITFOR INADVERTENT BLOWDOWN IS 1000 ¹/HR CU&&NTMETHODOLOGYBOUNDS THIS LIMIT
NewYorkPower Authori FUTURE DIRECTION OF SRV ISSUES:
BWROG ACTIVITIES:
ACTIVELYPARTICIPATING ON SRV COMMITTEE INSTALLINGPT ALLOYDISCS
. PURSUING IMPROVEMENTS IN MAINTENANCEOF SRV PILOTS
NewYorkPower Authori SRV LEAKAGEMONITORING:
CONSIDERING PERFORMANCE OF TEST TO CORI&LATELEAKAGETO TAILPIPE TEMPERATURE:
VARIOUS THEKMOCOUPLE LOCATIONS DIFFERENT LEAKRATES TEST WOULD BE USED TO:
EVALUATEREMOVALOF INSULATION FROM TAILPIPES OPTIMIZE THERIVIOCOUPLE RELOCATION
NewYorkPoler Authority START UP TESTING:
EVALUATINGGE REPORT ON ELIMINATIONOF STARTUP CYCLING OF SRVs:
TEST INVOLVES TESTING CIRCUITRY AND ACTUATINGPILOT PRIOR TO PLANT STARTUP MAINVALVEOPERATION DEMONSTRATED BY TESTING AT STEAM TEST FACILITY.
BENEFITS:
STARTUP TESTING IS OFTEN AN INITIATOROF PILOT LEAk&GE REDUCES NUMBER OF CHALLENGES TO SRVs AND PIPING
. 8%R SAFKTYRELIEF VALVE PERFORMANCE KINGOF PRUSSIA, PENNSYLVANIA APML24, 1996
~ BRIEF
SUMMARY
OF TRC 2-STAGE SRV PERFORMANCE AT MILLSTONEUNIT 1 I
~ OBSERVATIONS FROM TWO VISITS TO %%LE LABS INAUGUST2 SEPTEMBER 1995
~ POSSIBLE CAUSES CONTRIBUTINGTO PILOT VALVESTICKING ANDPILOT VALVELEAKAGE
~ COURSE OF ACTIONTO IMPROVE THE PERFORMANCE OF THE TRC 2-STAGE PILOT VALVES
Summary of2-stage SRV Performance @ MP1
~
2-stage installed June 1979
~ Cycle 7: No "as-found" data taken at Wyle for pilots. Only 1/2 removed.
~ Cycle 8: 5 ofthe 6 valves removed were turned over to GE for diagnostic testing 4, no "as-found" results are available. Leakage present on 5 Ofthe 6 pilots.
~ Cycle 9: Leakage acceptable 8'c one SRV drifted high 8.2%.
~
Cycle 10: Leakage acceptable.
One SRV popped very high (10%)..3 SRV's popped at >3%, but <4%,.
~ Cycle 11 Leakage acceptable.
One SRV stuck Ec 2 SRV's popped >3%, but <4.6%.
~ Cycle 12: Leakage acceptable.
One SRV popped very.
high (17.9%) A one SRV popped at 4.6%. (Performance
, drops significantly after this cycle!)
Summary of2-stage SRV Performance @ MP1
~ Cycle 13: Leakage is a@i acceptable.
One SRV is stuck Ec 4 other SRV's popped >3%. SRV 0 1219, 1220, 1221, 1222 are new~!
~ Cycle 14: No leakage on any pilot valve, but 2 SRV's are stuck and 3 more popped at >6%.
~ Cycle 15A: Shutdown aAer 8 months ofoperation due to excessive pilot leakage.
Also, 3 SRV's popped at >6%.
~ Cycle 15B: Shutdown after 7 months ofoperation for planned LNP testing gives MP1 an opportunity to test the pilot valves.
Some leakage on 4 SRV's 2 2 SRV's with zero leakage.
One SRV is stuck A 3 SRV's popped at
'6%
V
~ Cycle 15C: Shutdown after 45 days of'operation to determine the condition ofthe SRV's. Two pilots not tested in order to obtain "as-found" seating surface replications. 2 SRV's leaking Ec 2 with zero leakage. One SRV popped 4.8%.
I
~ Cycle 15D: Shutdown after 2 months ofoperation for refueling outage.
Most ofthe SRV's are leaking Ec 3 popped at >3.7%.
OBSERVATIONS I'ROM TWO VISITS TOVIE LABINAUGUST R SEPTEMBER 1995
~
Steam cutting was evident in nearly every seating surface with a distinct pattern. Two areas 180 apart were leak tight with evidence ofleakage between these two points. (Photo 01)
~ A gray to black oxide was observed on the steam-side ofall pilot discs. However, there was no indication ofoxide bonding.
Instead, the damaged areas appeared to be caused by metal deformation dne to mechanical binding or galling. (Photo ¹1)
~ The seat machining process uses a single point tool bit mounted to a cantilevered bar. (Photo ¹2) (Drawirig¹ I)
~ "Critical"dimensions for the pilot body are not clearly identified.
~ The seat angle is checked by use ofa test piece and;the disc angle is checked by a shadow comparator.
~ Dimensional check ofthe labyrinth seal area and the depth
&om the seat to the top ofthe flange are performed.
~
Other "critical"dimension checks may be performed, but are usually not recorded.
OBSERVATIONS FROM TWO VISITS TO mH,E LABINAUGUST 0 SEPTEMBER 1995
~ Many ofthe pilot seats have been weld repaired.
(Photos ¹3&4)
~ No post-weld heat treatment is conducted after field welding process to stress relieve the surrounding base metal. This is performed on new pilot valves.
~ The stress reliefgroove is not reestablished after the weld repair.'
O.D. ofthe seating area is no longer uniform after.weld repair.
~ Inspection ofthe seating surfaces (by replication and the use of a scanning electron microscope) aQcz lapping revealed that the surfaces are not consistent &om valve to valve and, in many cases, the lapping failed to remove the machining marks &om
~the seat cutting operation.
1
~ The contact bandwidth is not measured or controlled.
POSSIBLE CAUSES CONTRIBUTINGTO PILOT VALVESTICKINGANDLEAKAGE
~ Weld repair ofthe pilot seats.
~ No post-weld heat treatment is conducted after field welding process to stress relieve the surrounding base metal. This may result in a non-uniform expansion/contraction ofthe pilot seat as temperature in the pilot chamber changes.
~ The stress reliefgroove is not reestablished after the weld repair which may also result in a non-uniform expansion/contraction ofthe pilot seat as temperature in the pilot chamber changes.
li
~ The O.D. ofthe seating area is no longer uniform after the weld repair which may also result in a non-uniform expansion/contraction ofthe pilot seat as temperature in the pilot chamber changes.
~ Insufficient lapping ofthe disc and seat mating surfaces resulting in minute leaks which become larger over time.
POSSIBLE CAUSES CONTRIBUTINGTO PILOT VALVESTICJGNG ANDLEAKAGE
~ Degraded valve insulation. From SIL 196, Supplement 16, Revision 1: Saturation temperature ofsteam at 1000 psi is approximately 545'F. Poor insulation willcause a decrease in temperature in the pilot chamber ofthe SRV. Because the pressure remains the same and the temperature decreases, dissolved gasses (Hz 8'c Oz) are released and accumulate in the chamber.
This wet, hot, high-oxygen atmosphere can be very corrosive and can increase the likelihood ofcorrosion-induced bonding between the pilot seat and disc.
~ Poor insulation may also increase the potential forpilot valve leakage.
~ "Critical"pilot valve dimensions may. not be maintained.
~ Critical dimensions are described in the TRC Technical Manual, but these measurements are not recorded.
(Drawings ¹ 2 2 3)
Concentric
~ Perpendicular
~
Parallel
~ Bandwidth
COURSE OF ACTIONTO IMPROVE THE PERFORMANCE OF.THE 2-STAGE PILOT VALVES
~ New, custom-fit mirror type insulation has been installed on each valve.
~ The automatic electric liftcapability has been added.
This is not a solution to the set point driftproblem, but willprovide a back-up function to assure that the pilot valves liftwhen required.
~ Work with TRC to determine "critical"dimensions ofthe pilot valve.
~ Determine how to accurately measure the "critical"dimensions ofthe pilot valve.
~ Provide Wyle Lab with the dimensional specification.
~ Return to Wyle Lab to disassemble spare set of 6 pilot valves for dimensional checks-and additional inspections.
~ Ifnecessary (and ifpossible), machine surfaces to obtain proper dimensional tolerances.
~ Ifnecessary (and ifpossible), machine out pilot seat and weld repair.
~ Heat treat any pilot valves with weld repaired seats.
~ Ifmachining/weld repair is not possible, then purchase new pilot bodies.
RENEW OF S Y ON SRV PERFORMANCE
~
These pilot valves age capable ofperforming the function for which they were designed.
~ Apilot valve which performs well~ continues to perform
'ell until a change is made internally such as:
~ Weld repair to the pilot seat
~ Installation ofa'new disc
~ Machining seating surfaces
~ A new pilot valve which performs poorly willcontinue to perform poorly.
The platinum-alloyed discs held no advantage over. the standard Stellite 6 discs..However, all Pt-alloyed discs experienced some degree ofleakage, which defeats the purpose ofthe platinum.
Therefore, the results from MP1 are inconclusive.
Nuclear Maintenance Applications Center Safety & Relief Valve Testing & Maintenance Guide Safety Relief Valve Performanco Meeting Sheraton Valley Forge VloVarrhe tr04) srr4054 EPRr Icharlorra
.NMACSafety Relief Valve Guide
~ Contents
-Tochnical descriptions ofvarious types of valves
-Failure modos and cause analysis
-Testing methods
-Goneral Information on maintenance
-Guidelines on personnel training
-Industry dataivendor contacts KIIAc EPRI Icharrono NMACSafety Relief Valve Guide EpRIIcharrorta NMACSafety Relief Valve Guide
~ Oocument No. TR-105872
~ Presently In layout and printing
~ Expected date of distribution Juno 1, 1996
~ Available to all NMACMembers
~ Appendices
-Reforencos
-Selection 8 sizing
-ASMBCode tosting requirements
-Auxiliaryliftdovlccs Test Benches and Test Systems NarAC NlrAC EPRI ICharrorre NMACSafety Relief Valve Guide
~ Topics covered in the Guide
-BWR Main Steam Pressure Relief Valves
-PWR Pressurizor Pressure Relief Valves
-PWR PORVs
-PWR Secondary Main Steam Safety Valves
-BOP Safety Relief 8 Relief Valves NarAC
EPRI ICharlotte NMACSafety Relief Valve Guide
~ Appendices
-References
-Selection 8 sizing-
-ASME Code testing requirements
-Auxiliarylift'devices Test Benches and Test Systems NMAC EPRl ICharlotte BWR Failure Modes NMAC
EPRI/Charlotte PWR Pressurizer Safety Valve Failures NMAC EPRI / Charlotte PWR PORV Failures LiftLow 8.7%
Fail to Lift 21.7/o Fall to Reset 21.FYo Seat Leak 46.8%
NMAC
INADVERTENTOPENING OF AN SRV AT LIMERICKGENERATING STATION UNIT 1
BACKGROUND LIMERICKUNIQUE VERTICALDISCHARGE DESIGN chronic main seat leakage problem due to condensate anywhere from 7 to all 14 main discs replaced each cycle tailpipe temperatures at end of cycle as high as 295F (leaktight = 145F) no direct safety implication valve body modification to eliminate main seat leakage (1992 [Unit 1] and 1993 [Unit 2])
modification effective (only 1 valve with main leakage after modification for two cycles prior to Unit 1 sixth operating cycle)
LIMERICKPILOT LEAKAGE EXPERIENCE PRIOR TO BLOWDOWN EVENT no valves with significant pilot leakage very few (if any) pilot discs replaced good simmer margin (125 to 145 psi)
EVENT during 1R05 (2/94) aH 14 SRVs replaced 5 ADS valves stroked during startup after startup, 3 SRVs had elevated tailpipe temperatures 2 ADS valves (M 8 S) and 1 non-ADS valve (F)
M & S valves quickly stabilized at 220F (typical of main seat leak).
F valve temperature was 170F M 8 S valve tailpipe temperatures tracked very closely during operating cycle (295F just prior to lift) on 9/11/95 with Unit 1 @ 100% power - indication that 1M SRV was open
INVESTIGATION TESTING actuator, logic, SOVs satisfactory SRV sent to Wyle for inspection/testing S valve pilot leaktight, main seat/disc severely steam cut M valve pilot leaking profusely (maybe 2500 Ib/hr), main leaktight quick press up - internal parts shifted {R 870 psig, leakage significantly reduced slowly increased pressure to 1250 psig, valve did not actuate
- test terminated DISASSEMBLY pilot disc nose completely separated pilot disc shank cut away 50/o of circumference pilot rod severely eroded and buckled pilot rod to bellows weld steam cut pilot seat had radial steam cuts pilot seat
&. all bores concentric
INVESTIGATION continued METALLURGICALEVALUATION material confirmed to be Stellite 6B hardness HRC 36 EQUIPMENT ROOT CAUSE pilot leakage initiated during ADS test (leakage initiator unknown) prolonged leakage eroded pilot disc 8 pilot rod causing structural failure of pilot rod prolonged blowdown is expected due to inability to repressurize main piston chamber
CORRECTIVE ACTIONS IMMEDIATE all 5 Unit 1 SRVs with elevated tailpipe temperatures were replaced SRV tailpipe temperature action plan developed (assume all leakage is from pilot) 225F (alert level) log temperature every 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> trend temperature to project to 275F begin preparation for planned outage to replace valve 250F (action level) engineering recommend when to schedule replacement (before 275F) engineering to evaluate any sudden and significant temperature increases Basis:
225F is minor leak (approx. 20 lb/hr) 250F is more significant leak (500-1000 Ib/hr)
M SRV took 9 months to go from 250F to 275F
CORRECTIVE ACTIONS continued 250F (1000 Ib/hr) for all 14 SRVs is well within pool cooling capability 275F provides margin to 295F (M lift point) - 6 months FOLLOW UP testing on D, F and L SRVs F pilot eroded severely at 247F,-D pilot leaked 7k/hr at 215F, and L had cracked main seat action plan revised due to F valve inspection results-severe pilot disc erosion. Setpoint self actuation, blowdown potential, and manual/ADS capability to be evaluated at 225F due to potential loss of disc geometry thermocouples to be installed on all SRV top works to detect pilot leakage (Unit 1 complete, Unit 2 3/97)
GE recommended bench leak testing main valves at 200 psig with nitrogen review ADS stroking during startup for possible elimination data base has been created to improve trendability of main and pilot leakage to attempt to zero in on root causes
POST 1R06 EXPER(ENCE after stroke during startup, 2,ADS valves leaked but cleared after additional stroke D, E, G, and N leaked after 20% power. E tailpipe temp stabilized at 235F plant shutdown after 1 month to replace 4 SRVs E SRV main valve leaked 294/hr, pilot was tight testing D, G, and N to be completed end of April experience base improving with additional leakage/temp data, pilot temp data, and tailpipe temp trending engineering and management sensitive to eventual outcome of pilot leakage concurrently, 2 SFiVs with minor leakage (215F) are being tracked on Unit 2 tech spec change to eliminate ADS SRV stroking submitted
ATTACHMENT 4 R
P r e
iv nc min SRV P rf rman
-Gary Hammer, NRR, Mechanical Engineering Branch (Meeting with BWR Owners on April 24, 1996)
~
Several NRC offices including NRR, RES, AEOD and the
, Regions are following SRV performance issues and events.
~
Target Rock 2-stage SRVs continue to have significant generic performance issues primarily setpoint drift and operational leakage.
~
The NRC staff is in general agreement with BWROG regarding the parallel approaches to resolve setpoint drift issue:
~
Installation of platinum alloy pilot disks for trial evaluation.
~
installation of additional pressure actuation switches if platinum disks do not significantly reduce drift.
P Staff believes schedule needs to be updated for BWROG recommendation and individual licensee decisions regarding need for additional pressure switches.
Should it be in the near term (i.e. Spring/early Summer 1996) because significant amount of platinum disk setpoint data will be available?
I
~
Increased focus on SRV leakage issues as result of 9-95 Limerick event.
~ Staff concerned about SRV performance and effects on overall plant safety:
Reduced SRV setpoint Inadvertent blowdown Increased challenges of other safety systems Adequacy of methods for detection of leakage
~
Staff encouraged that several plants which have experienced leakage following the 9-95 Limerick event have established criteria and have taken action to stop leakage.
~
Potentially related NRC research effort involving friction factor measurements for new and corroded Stellite 6 surfaces:
Initiated because of MOV issues.
~
Gerry Weidenhamer of RES is contact (301-415-6015) for more information;
HEHORANDUH TO:
FROM:
SUBJECT:
Brian W. Sheron,
- Director, DE:NRR Gary H. Holahan, Director, DSSA:NRR Dennis H. Crutchfield, Director, DRPM:NRR Lawrence C. Shao, Director, DET:RES M. Wayne Hodges, Director, DST:RES James T. Wiggins, Director, DRS:RGN-I Albert F. Gibson, Director, DRS:RGN-II Geoffrey E. Grant, Director, DRS:RGN-III Thomas P.
Gwynn, Director, DRS:RGN-IV Charles E. Rossi, Director Safety Programs Division Office for Analysis and Evaluation of Operational Data TECHNICAL REVIEW REPORT TARGET ROCK TWO-STAGE SRV PERFORMANCE UPDATE The enclosed Technical Review Report updates the history of the 2-stage Target Rock safety/relief valves (SRVs). failure to open at the expected setpoint and discusses a new problem:
pilot disc leakage.
On September ll, 1995, at Limerick Unit I pilot disc leakage resulted in a stuck-open SRV and an extended reactor blowdown into the suppression pool.
A review of the generic implications of significant leakage of Target Rock 2-stage SRVs (particularly pilot valve leakage) is being conducted by the NRC.
The staff plans to coordinate its review efforts with the Boiling Water Reactor Owners Group.
The corrosion bonding issue is being addressed by the Owners'roup and is being monitored by NRC.
The updated information on overall setpoint testing shows a gradual increase in the average liftpoint of the SRVs.
This report also provides information currently available regarding setpoint testing results on platinate discs.
The pressure switch fix has NRC approval for a topical report, requiring a
plant-specific submittal for the licensee to adopt it.
Should the platinate disc experiment fail, the pressure switch fix is, therefore, ready for immediate use.
Attachment:
As stated cc w/att:
V. K. Chexal, EPRI Pat Lewis, INPO Debbie W. gueener, NOAC CONTACT:
Mary S.
- Wegner, AEOD/SPD/RAB (301) 415-6369
AEOD TECHNICALREVIEW REPORT UNIT: Multiple DOCKET: Multiple LICENSEES: Multiple NSSS/AE: Multiple TR REPORT NO.: AEOD/T96-02 DATE: April 15, 1996 CONTACT: Mary S. Wegner
SUBJECT:
TARGET ROCK TWO-STAGE SRV PERFORMANCE UPDATE
SUMMARY
Beginning in 1982, the NRC issued several information notices (IN) related to the 2-stage Target Rock safety/relief valves (SRVs) failure to open at the expected setpoint:
IN 82-41, "Failure of Safety/Relief Valves to Open at a BWR;" IN 83-49, "Failure of Safety/Relief Valves to Open at a BWR - Interim Report;" IN 83-82, "Failure of Safety/Relief Valves to Open at a BWR - Final Report;"'IN 86-12, "Target Rock Two-Stage SRV Setpoint Drift;"
IN 88-30, "Target'Rock Two-Stage SRV Setpoint Drift-Update;" and IN 88-30, Supplement 1, "Target Rock Two-Stage SRV Setpoint Drift-Update."
The cause of the problem was identified as one or both of: (1) binding in the labyrinth seal area caused by tolerance buildup during manufacturing, or (2) disc-to-seat bonding caused by oxides of the disc and seat material forming a continuous film and inhibiting disc movement.
Some time after 1990, the Target Rock 2-stage owners'roup submitted two potential solutions:
(1) the preferred solution was the use of platinum alloy disc to cause radiolytic oxygen and hydrogen in the valve to recombine, reducing oxygen available to cause corrosion and (2) the alternate solution was the installation of a pressure switch and control circuitry to operate the valve electrically when the set pressure of the pressure transmitter was reached.
On September 11, 1995, an event occurred at Limerick Unit 1 which resulted in a stuck-open SRV and an extended reactor blowdown into the suppression pool. Pilot disc leakage was identified as the cause of the stuck open SRV. This event was especially noteworthy in that sludge and fibers in the pool, roiled by the blowdown, settled on the emergency core cooling system strainers nearest the SRV tailpipe, clogging them.
The NRC issued. IN 95-47, "Unexpected Opening of a Safety/Relief Valve and Complications Involving Suppression Pool Cooling Strainer Blockage," IN 95-47, Rev. 1, and Bulletin 95-02, "Unexpected Clogging of a Residual Heat Removal (RHR) Pump Strainer While Operating in Suppression Pool Cooling Mode" to address issues resulting from this event.
Attachment
DISCUSSION:
Background:
The original General Electric Company's (GE) BWR-4 design employed the 3-stage Target Rock safety/relief valve (SRV) for over-pressure protection for the reactor coolant system (RCS).
In the early 1970s, the valves exhibited a tendency to liftfor no apparent reason, and on occasions, to stick open.
About half of the plants with the 3-stage Target Rock SRVs experienced this problem.
The solution to the 3-stage problem was the 2-stage Target Rock SRV, which used a 3-stage body, modified to accept the new top works.
The plants that experienced problems with the 3-stage valve installed the 2-stage modification. In 1982, a problem with the 2-stage valves became apparent:
failure to open at the expected setpoint.
On July 2, 1982, at Hatch Unit 1, all 11 Target Rock 2-stage SRVs failed to open at their setpoints of 1080 psig, 1090 psig, and 1100 psig.
Pressure in the RCS rose to 1180 psig before three SRVs on one steam line opened and relieved RCS pressure.
II Georgia Power, GE, and Target Rock initiated a study of the cause of the event at Hatch Unit l. Other utilities that had installed 2-stage Target Rock SRVs joined Georgia Power in an owners'roup to look at the nature of the problem and its solution.
The problem was ultimately identified as one or both of two,situations:
(1) binding in the labyrinth seal area caused by tolerance buildup during manufacturing, or (2) disc-to-seat bonding caused by oxides of the disc and seat material forming a continuous film and inhibiting disc movement.
A material, PH13-8 Mo, whose oxide would not form a continuous film with the oxide of the seat material, was chosen for new discs.
Trial tests of valve performance, after installing the new discs on about 50 percent of valves at each plant, were conducted.
After a promising first set of test results, Hatch Unit 1 reported in October of 1989 that one of the valves with the PH13-8 Mo disc lifted'at 10.54 percent above its setpoint and a second at 9.18 percent above its setpoint.
In November of 1989, Brunswick Unit 2 reported a valve with the PH13-8 Mo material disc lifted at 10.4 percent above its setpoint.
Subsequently, the PH13-8 Mo discs were replaced with the Stellite 6B discs, until an appropriate solution to the corrosion bonding problem could be identified.
1 The NRC issued several INs which describe the foregoing events in greater detail.
They are IN 82-41, "Failure of Safety/Relief Valves to Open at a BWR"; IN 83-49, "Failure of Safety/Relief Valves to Open at a BWR - Interim Report"; IN 83-82, "Failure of Safety/Relief Valves to Open at a BWR - Final Report"; IN 86-12, "Target Rock Two-Stage SRV Setpoint Drift";IN 88-30, "Target Rock Two-Stage SRV Setpoint Drift-Update"; and IN 88-30, Supplement 1, "Target Rock Two-Stage SRV Setpoint Drift-Update." IN 88-30, Supplement 1 was issued on February 2, 1990, and discusses the Hatch Unit 1 and Brunswick Unit.2 problems occurring in 1989.
Current Events:
The Owners Group subsequently submitted two,solutions which were to be pursued simultaneously.
The preferred solution was the use of platinum (first as an inlay, also as a coating, and finally as a trace element in the disc alloy) to cause radiolytic oxygen and hydrogen in the valve to recombine, reducing oxygen available to cause corrosion.
Initialplatinate discs were installed at Cooper, Millstone Unit 1, and Hatch Unit 2 in, 1994:
Brunswick Unit 1 and Unit 2 are using a platinum-coated disc, first installed in 1993 on Unit 2 and 1994 on Unit 1. Pilgrim is using Stellite 21 for their disc material which they first used in 1984.
To date, setpoint test results involving SRVs using the platinate discs have been received from Millstone, Hatch, and Cooper.
Also, results from Brunswick for as-found testing of their SRVs which have the platinum coated discs have been received.
Figure 1 shows the results of the setpoint testing of platinate valve at each of the plants reporting.
The shading indicat~ valve 1, 2, 3, etc.
% Deviation from Nominal 10
-2 4
Millstone 1 94MIIIstone 1 II5 Hatch 2 Hope Creek Cooper Brunswick 1" Brunswick 2~
~ Platinumwoated
.figure 1: Setpoint Test Results for Hatinum Discs The alternate solution proposed was the "Pressure Switch/Transmitter for Two-Stage Target Rock Safety/Relief Valve," NEDC-32121P.
The Target Rock SRV is'a dual function valve which can be actuated by spring force to open when the system pressure reaches its setpoint (the Code safety mode), or it can be actuated by an electrical signal from a pressure sensor
4, l
V
to its electro-pneumatic actuator.
The actuator removes the spring force so that full steamline pressure lifts the pilot disc.
This mode of operation was originally installed for manual operation (the Code relief mode) and automatic depressurization system operation (the nuclear safety-related function). The pressure switch fix would extend the pressure sensor operation to the Code safety mode.
GE recommended that the pressure switch sensor setpoints be set at the same values as used for the spring-actuated mode.
The NRC found the proposed modification to be acceptable.
The topical report was approved on October 24, 1995.
Some licensees made plant-specific submittals prior to the topical submittal.
Hatch Unit 1 and Unit 2 and Brunswick Unit 1 and Unit 2 have installed the pressure switch fix in addition'o changing the disc material and Millstone Unit 1 has committed to installing the pressure switch fix during the 1996 refueling outage.
About the same time as the platinate discs were first installed, another problem with the SRVs was identified, pilot disc leakage which could lead to a stuck open SRV as occurred in the September 11, 1995, event at Limerick Unit 1. The stuck open SRV caused an extended blowdown into the suppression pool.
Sludge and fibers in the pool, roiled by the blowdown, settled on the emergency core cooling system strainers nearest the SRV tailpipe, clogging them.
The Limerick Event
~
~
Limerick Unit 1 and Unit 2 have 2-stage Target Rock SRUs unlike any other nuclear plant.
While the pilot stage is identical to those used by other nuclear plants, the main stage is quite different. Figure 2 compares the Limerick orientation to the standard orientation.
PILOT STAGE PILOT STAGE IKlET MAIN STAGE MAIN STAGE DISCKARCE Figure 2: Limerick SRV Orientation SRV Orientation The difference in orientation is thought to be the cause of the leakage seen at Limerick main-stage leakage.
Pilot-stage leakage can eventually cause the main-stage to lift, but main-stage leakage can not cause the main-stage to lift. After the operability demonstration
following the refueling outage in February of 1994, five SRVs were found to be leaking.
Because of their history of main-stage leakage, Limerick personnel assumed that the five leaking SRVs had main-stage leakage.
Four of them did, but not the M-SRV.
On September 11, 1995, the M-SRV lifted while Limerick Unit 1 was operating at 100 percent power and remained open.
The reactor was manually scrammed.
First the M-SRV and the S-SRV were sent to Wyle Laboratories for examination and testing, then the remaining leaking SRVs were sent.
The physical condition of the M-SRV disc and pilot rod is shown in Figure 3.
The disc was eroded 360'n the seat area such that the nose separated from the remainder of the disc.
Slots were worn in the body of the disc in three places.
The disc material was verified to be Stellite 6B by part serial number, material certification, and by Rockwell C hardness testing.
The pilot rod was eroded and deformed as shown.
/
I I
I I
Nose Fragment Seating Area Stellite Disc Eroded Slots The stuck open SRV caused an extended blowdown into the suppression pool.
The NRC issued IN 95-47, "Unexpected Opening of a Safety/Relief Valve and Complications Involving Suppression Pool Cooling Strainer Blockage," IN 95-47, Revision 1, and Bulletin 95-02, "Unexpected Clogging of a Residual Heat Removal (RHR) Pump Strainer While Operating in Suppression Pool Cooling Mode" to address issues resulting from this event.
Pilot Rod ANALYSIS The 2-stage SRV issues remaining are Figure 3: The Disc and Pilot Rod From the M-SRV
, 1) "setpoint drift" as the result of disc-to-seat corrosion, and 2) leakage past the pilot disc.
The results of setpoint tests of 2-stage SRVs conducted since 1987 (begins near the beginning of the testing of the valves containing PH13-8 Mo discs) until the present are summarized in Figure 4.
Pct Deviation from Nominal Setpoint ttt 4
~,
+
3K 3K r
Kl
.+
vl X
Z
+
+
'/
ts
/
Y
+
hT 1987 1 988 1 989 1990 099'$
'$992 0 993 0 994 0 995 Browne Fer 2
Brunswick 1
Brunswick 2 Coo sr Fsnnl 2 Fttxpatrtck Hatch 1
Hatch 2 Ho Creak Llmsrlck tatttstons 1
Ptt rim 3.2 0.2 2,0 3.4
~.2
'IA 27 1.5 1 7 27 O.T 1 0 01 O.'7 2.5 2.T 4,0 5.1 1.5 4A 4.4 2.$
4.5 4.5 d.d 1.5 4.5 2.5 4.5 2.6 1.$
0.7 1 2 4,3 4.6 47 2.4 Avtt Utt Fermi 2 Llmerlck Browne Fsny 2 Fttzpatrtek altttslone 1
+ Srunswlek 1
Hateft 1
+ Pttdrtm Brunswick 2 Hatch 2
> coop
~
Hope Creek Figure 4: SRV Set point Test Averages
By design, setpoints for SRVs were a nominal figure such as 1100 psig P 1%.
The expectation was that the deviation from the nominal would be minimal and the average deviation would be 0.
In fact, the deviation from the nominal has ranged from the minimal to values in excess of 67 percent, the majority being positive.
The solid line in Figure 4 is the linear regression fitof average setpoints of all SRVs tested during the year.
The average setpoint has trended upwards from about 2.5 percent above nominal in 1987 to about 3.5 percent above nominal in 1995.
The various points represent the average setpoint for SRVs tested by a particular plant for the year.
Of these 61 data points, 7 or 11.5 percent were within g 1%, 29 or 45.5 percent were within + 3%, and 5 or 8.2% were 5 percent or more above the nominal.
There are no negative setpoint test averages.
It was common to stop setpoint testing at or near 1250 psig when an SRV would not lift; however, the raw data shows liftpoints in excess of 1250 psig for some valves.
For representation in Figure 4, all valves which did not liftat or before 1250 psig were called stuck and their setpoints were called 1250 psig whether they lifted or not and whether they were pressurized above that. pressure or not. The tolerance was then calculated using that number. It should be noted that the SRVs in Figure 4 do not all have the same disc material.
Figure 4 includes SRVs having Stellite 6B, PH13-8 Mo and platinate materials.
A review of the generic implications of significant leakage of Target Rock 2-stage SRVs
{particularly pilot valve leakage) is being conducted by the NRC.
Factors being considered're:
effects of significant leakage on valve operability, effects of operability testing on valve performance, risk significance of more frequent opening or multiple SRV opening, adequacy of methods (tailpipe monitoring, torus heatup rate) for detection of leakage for the range of BWR configurations, possible concerns about increased suppression pool cooling operation, and what leakage limitmonitoring and reporting requirements apply or are needed.
The staff plans to coordinate its review efforts with the Owners Group.
FINDINGS AND CONCLUSIONS The corrosion bonding issue is being addressed by the Owners'roup and being monitored by NRC. The updated information on overall setpoint testing shows a gradual increase in the average liftpointof the SRVs.
This report also provides information currently available regarding setpoint testing results on platinate discs, but at least one full set of setpoint testing results are necessary to evaluate the test program.
When sufficient information is available on the platinate test results to warrant an update, one willbe issued.
n The pressure switch fix has NRC's approval for the topical report, requiring a plant-specific submittal for the licensee to adopt it. Should the platinate disc experiment fail, the pressure switch fix is therefore ready for immediate use.
Millstone Unit 1 is. also installing the pressure switch fix, alleviating concerns for the adequacy of overpressure protection at that site.
10
Target Rock Two-Stage SRV Issues
The issues remaining are 1)
"setpoint drift" as the result of disc-to-seat corrosion which I willaddress and 2) leakage past the pilot disc which Gary, Hammer has addressed.
0 L>
oe
The "Setpoint Drift" Issue
The problem:
corrosion binding of the disc and seat of the pilot valve The solution:
- a. platinate discs to recombine radiolytic oxygen and hydrogen, thereby reducing oxygen necessary for corrosion
- b. pressure switch fix
Pct Deviation from Nominal Setpoint
~~Ax Z
A
~
A
+
3KL /
4'
+
/+
+
/
R 7
- 4 X
/
A
//
Z 0
-1 1987 1988 1989 1990 1991 1992 1993 1994 1995 Av Uft 8rowns Fer 2
Orunswlck 1
Brunswick 2 Coo sr Fermi 2 Fltzpatrlck Hatch 1
Hatch 2 Ho e Creek Llmerlck Millstone 1 Pll rim 2.0 2A 0 2 20 4,2 2 '7 1.6 1 7 3.5 3A 2.7 0.7 1.8 0,1 0.7 2.6 2.7 3.6 1.5 1,7 6 1 1 8 3,4 4A 4.4 2.5 4.2 4.6 2.3 5.5 1.8 4 8 2.6 4.6 2,8 1 5 0.7 1.2 4,6 26 4.7 2.4 3.8 4.0 1,3 4,2 Avg Urt Fermi 2 Llmerlck Srowns Ferry 2
'runswick 1
FltzPetrlck Hatch 1
Millstone 1 Pllgrlm Brunswick 2 Hatch 2 Cooper V Hope Creek
,Figure 4: SRV Setpoint Test Averages
Of the 61 data points on the graph, 7 or 11.6/o were within + 1/o, 29 or 45.5/o were within + 3/o, and 5 or 8.2/o were 5 '/o or more above the nominal.
There are no negative setpoint test averages.
[Allvalves which did not liftat or before 1250 psig were called stuck and their setpoints were called 1260 psig whether they lifted or not and whether they were pressurized above that pressure or not. The tolerance was then calculated using that number.]
Test Results for SRVs with Platinate Discs
Initially, platinate discs were installed at Cooper, Millstone 'I, and Hatch 2 in 1994.
Brunswick 1 6 2 are using a platinum-coated disc.
Figure 1 shows the results of the setpoint testing of platinate valves and platinum coated valves tested by December 31, 1995.
12
% Deviation from Nominal 40
-2 4 Mlllstono1 94Millstone1 95 Hatch 2 Hope Creek Cooper Brunswick 1'runswick 2' Plat(num-coated H~me 2: Setpoint Test Results for Platinum Discs
e ion ae eie a ve er ormance ssues lear Regulatory Commission 5/13/96
Oiscuss SRV performance issues
....,,-,.~~, Oiscuss findings/expectations
'"<~" Outline courses of action
~ Summary lear Regulatory Commission 5/1 3/96
I'
Pilot/main seat valve leakage r
Setpoint drift
~~'oreign material
~ Surveillance of SRVs from remote panels
" lear Regulatory Commission 5/1 3/96
Determine the source Quantify the leakage
'ssess the affect
-On SRV performance
-On plant support systems
~ Develop a mitigation strategy iVQc
"- lear Regulatory Commission 5/13/96
oint
ÃÃ8 PYYPpr, Oetermine consequences Review maintenanceltesting plans Implement corrective actions a
lear Regulatory Commission 5/18/96
rei n 0
ems ateria ontro Assess consequences Oevelop implement corrective action
-Use approved lubricants
-Minimize use of lubricants
-Keep nitrogen lines clean lear Regulatory Commission 5/1 3/96
iti ation etos Consider the affect of annunciated alarms on operators
<~ Procedures revised
~ Use PRA/FSAR to consider leakage affects on systems
~ Look at big picture 7
lear Regulatory Commission 5/13/98
t
mote R
erations gypsy Y8gi Procedures in place Equipment periodically surveilled Operators trained 5
6 C
Alpslear Regulatory Commission 5/13/96
SRV performanceissues continue to
,:, be a concern They can be managed
~ Remember the SRVs are part of a larger system
~ Treat problem but protect plant s gs8 Q6M lear Regulatory Commission 5/13/96
'~, il o~N i
I I
It A
I