IR 05000293/1997001
| ML20198H726 | |
| Person / Time | |
|---|---|
| Site: | Pilgrim |
| Issue date: | 09/16/1997 |
| From: | Conte R NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | Boulette E BOSTON EDISON CO. |
| References | |
| 50-293-97-01, 50-293-97-1, NUDOCS 9709220155 | |
| Download: ML20198H726 (2) | |
Text
- _
_ - _ - _ _ _ _ _ _ _ . _ _ _ _ _ - - _ _ _ _ _ _ _ _ _ _ _ _ _ _
i
.
September 16,1997 E. Thomas Boulette, PhD Senior Vice President Nuclear Boston Edison Company Pilgrim Nuclear Power Station 600 Rocky Hill Road Plymouth, Massachusetts 02360 5599 SUBJECT: NRC INTEGRATED INSPECTION REPORT 50 293/97 01 AND NOTICE OF VIOLATION
Dear Dr. Boulette:
This letter refers to your April 17, and August 29,1997 correspondence, in response to our March 18,1997 letter.
Thank you for informing us of the corrective and preventive actions with completion schedules documented in your letters. These actions will be axamined during a future inspection of your licensed program.
Your cooperation with us is appreciated.
Sincerely, Original Signed By:
Richard J. Conte, Chief Projects Branch 8 Division of Reactor Projects
, Docket No. 50 293 cc w/o cv of Licensee Response letter:
L. Olivier, Vice President - Nuclear and Station Director T. Sullivan, Plant Department Manager N. Desmond, Regulatory Relations D. Tarantino, Nuclear Information Manager
\\
9709220155 970916 PDR ADOCK 05000293 G PDR i
.
}
,! ,! ,
. . .
. .
.
.
L
.
.
Mr. E. Thomas Boulette, PhD 2 cc: w/cv of Licensee Response letter R. Hallisey, Department of Public Health, Commonwealth of Massachusetts The Honorable Therese Murray The Honorable Joseph Gallitano B. Abbanat, Department of Public Utilities Chairman, Plymouth Board of Selectmen Chairman, Duxbury Board of Selectmen Chairman, Nuclear Matters Committee Plymouth Civil Defense Director P. Gromer, Massachusetts Secretary of Energy Resources J. Miller, Senior issues Manager J. Fleming A. Nogee, MASSPIRO Oifice of the Commissioner, Massachusetts Department of Environmental Quality Engineering Office of the Attorney General, Commonwealth of Massachusetts T. Rapone, Massachusetts Executive Office of Public Safety Chairman, Citizens Urging Responsible Energy Commonwealth of Massachusetts, SLO Designee Distribution w/ encl:
Region I Docket Room (with concurrences)
PUBLIC Nuclear Safety Information Center (NSIC)
NRC Resident inspector R. Conte, DRP M. Conner, DRP C. O'Daniell, DRP H. Miller, RA W. Axelson, DRA P. Milano, NRR A. Wang, NRR W. Dean, OEDD R. Correia, NRR F. Talbot, NRR DOCDESK Inspection Program Branch, NRR (IPAS)
DDCUMENT NAME: g:\ branch 8\rplyttr\pil9701.rpy Ts ,eceive e copy of thne document. In.hcote in the boa: 'C' = Copy without ettschment! enclosure 'E' = Copy with attachmentfenclosure *N' = No copy OFFICE Rl/DRP Rl/DRP /
, f l NAME EConnerph f RConte /J{/
DATE 08/22/97 A6/f/97/ 08/ /97 08/ /97 08/ /97
') OFFICIAL RECORD COPY
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _
..'
- l Boston Edloon higram Nuclear Pow station Plymouth. Mastschusetts 023c0 H.ory v. oh.kn General Managet Technical section BECo Ltr. #97 089 U.S. Nuclear Regulatory Commission Attn: Document Control Desk Washington, D.C. 20555 Docket No. 50-293 License No. DPR 35 SUBJECT: CHANGE OF COMMITMENT CONCERNING REPLY TO NRC INSPECTION REPORT NO. 50-293/97-01 Boston Edison Company (BECo) responded to violations identified in Inspection Report 50-293/97-01 by letter to the NRC dated April 17,1997. That response contained the following commitment that was to be completed by August 31,1997:
Establish a multi-discipline team to:
1. Review applicable Pilgn'm Station procedures for additional cases where procedural steps may have been established to provide compensatory actions forless than optimal equipment performance (i.e., establish the extent of the proceduralized operator workaround problem).
2. Revise operator workaround list to include any additional operator workarounds identified during the procedure review process.
3. Review and revise as necessary the process foridentifying, prioritizing, scheduling, and completing repairs / modifications associated with operator workarounds.
4. Develop a training module for operators and engineering support personnel to provide guidance on identifying conditions that represent compensatory actions for degraded performance of equipment.
Af $TGNU $ ,
!
__
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
. .*-
.
.e We have established a multidisciplined team to perform the activities listed above. We have found, however, that we are unable to complete this task by August 31,1997, and have rescheduled completion to December 31,1997. This schadute extension does not impact plant safety because the compensatory actions discussed in the violation response remain effective and in place.
Should you require further information on this change of commitment, please contact P.M.Kahler at (508) 830-7939.
. .
~ '
H. V. Ohelm Mr. Alan B. Wang Project Manager Project Directorate 1-3 Office Of Nuclear Reactor Regulation Mall Stop: OWF 14B2 1 White Filnt North 11555 Rockville Pike Rockville, MD 20852 U.S. Nuclear Regulatory Commission Region 1 475 Allendale Road King of Prussia, PA 19406 Senior Resident inspector Pilgrim Nuclear Power Station
!
PMK/dmcIlR97-01
-
- -
- -
_ _ _ _ _ _ _ _ .
,
. . _ _ _--_--_--
.
.
O Pernge w swein per momh. u ohnens omo E. T. Boulette, PhD sw*n vu PreeWent - NW gp,ggq7 g997 BECo Ltr. 2.97. 044 U.S. Nuclear Regulatory Commission Attention: DocumentControlDesk
- Washington, DC 20555 Docket No. 60 293 License No. DPR 35 SUBJECT:
REPLY TO NOTICES OF VIOLATION (REFERENCE NRC INSPECTION REPORT NO. 50 293/97 01)
Endosed is Boston Ediso'n Company's reply to the Notices of Violation contained in the subject inspection report. Enclosuren 1 and 2 respond to Notices of Violation 1 and 2, respectively. *
The following commitments are made:
o implement a design change to provide 00 indication from the full-in switches as a temporary solution to the all rods full-in issue. This will be awTp;ished by the end of September 1997, o
implement a design change to permanently resolve the all rods full-in issue by the end of the first quarter isW8. If the modification cannot be installed on line, the modification will '
be installed the first outage of sufficient duration following March 1998.
Establish a multi-discipline team to complete the following by the end of August 1997:
1. Review applicable Pilgrim Station procedures for additional cases where procedural steps may have been established to provide compensatory actions for less than optimal equipment performance (i.e., establish the extent of the proceduralized opetutor workaround problem).
2. Revise operator workaround list to include any additional operator workarounds identified during the procedure review process.
Q Y
__ _ --
-_ - _ - _ _ _ _ - _ _ _ _ _ - _ _ _ _ _ _ - _ _ _ _ _ _ _
, .
.
.
'
3. Review end revise as necessary the process for identifying, priorttizing, scheduling, and completing repairs / modifications associated with operator workarounds.
4. Develop a training module for operators and engineering support personnel to provide guidance on identifying conditions that represent compensatory adions for degraded performance of equipment.
Please do not hesitate to contad me if there are any questions regarding the enclosed reply.
E Boulette, PhD
.
KRD/dmcMog7-01
'
.
Enclosure 1: Reply to Notice of Violation 1 Enclosure 2: Reply to Notice of Violation 2 cc: Mr. Alan B. Wang, Project Manager Project Directorate I-3 Omco of Nuclear Reactor Regulation Mall Stop: OWF 14B2 1 White Flint North 11555 Rockville Pike Rockville, MD 20852
-
U.S. Nudear Regulatory Commission Region i
.
'
475 Allendale Road King of Prussia, PA 19406 Senior Resident inspector Pilgrim Nuclear Power Station
.
_ _ _ _ _ _ _ _ _
,
.
Boston Edison Docket No. 60 293 Pilgrim Ucense No. DPR 35 Enclosure 1 Reply to Notlee of Violation i Violation No.1
- Pilgrim technical spedfications, secuon 6.8, Procedures, requires that written procedures be established and implemented for acuvities covered In NRC Regulatory Guide 1.33, AppendW A. Appendix A, Sections 6 and 4.g, requires procedures be established and implemented for Abnormal conditions and operation of the RCIC system.
EbEDple No.1 BECo Abnormal Procedure 2.4.49, Loss of Normal Feed and Feeds. ster Valve Malfuncuons, step 4.2.2, specified to monitor vessel level and perform the following as appropriate: if level is incrossing then the feedwater pump minimum flow valves may be
.
cycled to aid in controlling vessel water level. If water level cannet '. e controlled, then manually scram the reactor and enter procedure 2.1.6.
l Contrary to the above, on February 15,1997, the NRC observed operators respond to
,
a FRV malfunction, enter procedure 2.4.49, and stop/ start the electric driven feed
pumps several times to control reactor vessel water level to prevent the need for a
reador scram. Operators were never trahed to include the practice of cycling feed pumps to maintain vessel level in this manner.
Examole No. 2 BECo procedure 2.2.22, RCIC System, and 8.C.13, Locked Valve Ust, lists RCIC valve 130136 as locked open. Further, procedure O.C,13 step 6.0[2] states not to use plastic valve restraints where high temperature may cause the plastic to melt.
Contrary to the above during plant power operations, the NRC identified that a plastic valve restraint on valve 1301-36 had melted and broken. Additionally, a plastic valve restraint was also Installed on the adjacent valve 130137.
This is a Severity Level IV violation (Supplement 1) "
.
. .
,
REASON FOR THE VIOLATION Example No.1 The reason for the Example No.1 vlolation was evaluated as part of an assessment of operator performance during the reactor pressure vessel level transient and subsequerd unit trip occurring February 15,1997. The reason for the violation was arf.:Fvuistion of procedure 2.4.49.
Discussion:
'
in response to the reactor pressure vessel (RPV) level transient that occurred during plant shutdown, the crew initially responded in accordance with procedure 2.4.49 " Loss of Normal Feed and Feedwater Valve Malfunctions * which directed the reador food
-
/ level. This was ineffective due to a large amount of " leak by" through the feedwater rpalating valves (FRVs).
When difficulty was encountered maintaining level controls, operators were dispatched
% manually close 'B' FRV in the condenser compartment. This action was taken because the operating crew had been tniofed, during the pre-shift brief, to expect leakage past the 'B' FRV. The nuclear watch engineer (NWE) knew that if leakage could be minimized, shutdown of the plant would not be complicated by loss of the main condenser heat sink due to a primary containment isolation system (PCIS) group 1 isolation.
As RPV level continued to rise and prior to manual control of 'B' FRV being established in the condenser compartmord, the running RFP was manually tripped to allow a
doorosse of RPV level. The RFP was subsequently restarted when water level decreased to within normal range. This sequence was repeated once again before the
- FRV could be manually closed in the condenser bay. During the subsequent rrdor power reduction, RPV level rose once again which resulted in the insertion of a manual scram by the control room crew. The actions taken regarding the starting and stopping of RFPs to control vessel level is not addressed in procedure 2.4.49.
The FRVc, malfundioned causing the level control problem, Because of a mechanical malfunctiors, the W FRV was unable to fully close. This allowed excess flow to enter the reactor v6sel for the existing pows mittions and led to the overfeed events.
The operating crew idt osat the 'R' FRV was the cause as level stability was achieved by closing the valve in the field. This was only half of the problem. The W FRV was also in a malfunctioning condition such that, as power was decreased, W FRV could not dose down suffiolently to match steam flow. This led to the condition of rising reactor level that was terminated by the manual scram.
The W FRV is a double poppet balanced flow valve. The root cause of the inability of the W FRV to adequately control foodwater flow below approximately 800,000 lbm/hr at low steaming rates is the velve plug failed to make sufficient contad with the lower seat of the valve cage. The inability of W FRV to adequately " shut off" flow was also
- caused by degradation of its actuator. Information gathered during diagnostic testing
. - - _ _ - _ . _ _
- -
i .,
.
.
demonstrated the valve did not operate smoothly. The valve had a tende.wy to stick thenjump during the closing stroke.
- in order to establish local operation of the 'B' FRV in the condenser bay, the crew started and stopped the RFPs within predetermined limits under the direction of the NWE. These actions were based on the Guidance contained in ARP 906R C7,
" Reactor Water Level Hi", that states *1f necessary, take manual control of feedwater control system and restore reactor water level to normal.' The NWE Interpreted this stop as taking manual control of the RFPs since both FRV controllers were already demanding full closure of their respective valves.
The NWE's decision to control vessel level by cycibe RFPs was made following a careful analysis of plant conditions should the unit be manually scrammed prior to regaining controlof vessellevel.
Procedure 1.3.34 * Conduct of Operations * mandates approved, written procedures shall be adhered to by all Station personnel, it also states: "in the event of an emergency situation not covered by an approved procedure / instruction, the personnel involved should take action to minimize personnel irQury and damage to the facility.
Since variations from the expected course may occur, personnel involved should be
'
prepared to manipulate controls as necessary to cope with the problem.' Since procedure 2.4.4g did provide direction for scramming the reactor when level could not be controlled, procedure 1.3.34 was not applicable. Therefore, the decision to use the ARP for taking ' manual control" of the RFPs was inappropriate. The statement about taking manual control of the "feedwater control system * in the ARP refers to the FRVs.
Example No. 2 The reason for the Example No. 2 violation was operator error, in accordance with Procedures 2.2.22 and 8.C.13, plastic valve restraints are prohibited on valves where high temperatures rra ' cause the plastic to melt. Two valves,130136 and 130137, had plastic valve roNints installed by an operator.
CORRECTIVE STEPS TAKEN AND RESULTS ACHIEVED Example No.1 This event was reviewed by the operations department manager as part of the routine
-
Lloonsed Operator Requalification Training (LORT) program. This training was conducted as part of the pre startup preparations. This session was presented in a case study format of this event from the failure and human performance perspectives.
Specific review of the failure to follow procedure 2.4.4g was conducted, and clear expectations for adherence to off-normal procedures were given. Procedure 1.3.34 was reviewed regarding the adherence to procedures and circumstances for which departure from piooedures is appropriate (i.e., only when immediate personnel injury or plant equipment damage is imminent). The training session also included a review of operations management's expectations for using off crew resources (operations management, engineering) in assisting in the resolution of off normal plant conditions
_. l
.
.
,
-.
. efter plard conditions have stabilized. The need for effective crew involvement dudng responses to plant transients was also discussed during the LORT training.
Mechanical repairs to both feedwater regulating valves are complete, p.mamole No. 2 immediate conective action was taken to replace the melted and broken plastic valve restraints on valve 130136 and 130137 with metal restraints. Furthermore, a walkdown of the high temperature valves was conducted to verify valve restraints were metal and not plastic. No other plastic valwe restraints in inappropdate loostions were
'
found. AN operators were briefed about this inappropdate action during the pre startup training session.
,i CORRECTIVE STEPS THAT WILL BE TAKEN TO AVOID FURTHER VIOLATIONS
.
R.xample No.1 Procedure adherence for off-normal plant conditions will continue to be reinforced both site wide and during future simulator training.
. Simulator evalua$on guidance will be modified to focus more heavily on procedural adherence by August 1997, e Effectiveness of corrective actions will be monitored dudng periodic simulator exams and on-shift observations.
-
- A soif-assessment of effectiveness of the above actions will be performed during -
the annual operating exams by December 1997.
- Examnie No. 2 No further actions are required regarding plastic valve restraints on high temperature valves. This procedural adherence event will be tracked and trended as part of our current conective action process.
DATE WHEN FULL COMPLIANCE WILL BE ACHIEVED '
Relative to Example No.1 (Procedure 2.4.49), the training-related corrective actions are on-going and will continue into the future.
Relative to Example No. 2 (Procedure 2.2.22), full compliance was achieved at the time of the event when immediate actions were taken to identify and replace plastic valve -
restraints on high temperature valves.
.
. _ _ _ . -
.
Boston Edison Docket No. 50 293 Pilgrim Uoense No. DPR 35 Enclosure 2 Reply to Notice of Violation 2 VIOLATION No. 2
- 10CFR 50, Appendix B, Criterion XVI, Corredive Action, requires in part, f' ..
measures shall be established to assure that conditions adverse to quality are promptly identified and corrected. In the case of significant conditions adverse to quality, the measures shall assure that the cause of the condition is determined and corrective action taken to preckufe repetition.
Contrary to the above, the following corrective actions were not timely or effective to prevent recurrence of a condition adverse to quality."
Fxample No.1
'
The corrective actions taken by BEco in response to a self identified deficiency involving the standby switchyard batteries as reported in LER 97-01 [ sic] were not scheduled timely enough in the master surveillance test program (MSTP) to avoid the possibility of recurrence of the original condition adverse to apatity. The NRC ldentified that the MSTP had a surveillance test due date of October 31,1997 which could have exceeded the 5 year capacity test window.
Example No. 2
.
On February 15,1997, during the recovery from a low power scram, three significant operator workarounds conditions detracted from the operations ability to stabilize plant conditions during post scram conditions with a main steam isolation valve (MSlV)
closure - a condition adverse to quality. These conditions involved the control rod ,
bottom lights, MO-2203 Main Steam Drain Line isolation Valve, and the Reactor Water--
Cleanup letdown control valve CV 1239. When responding to the April 19,19g6, low power reactor scram also involving MSIV isolation, the same three significant ' work-around conditions had an !mpact on the post scram recovery. The rod bottom lights and CV-1239 degraded equipment conditions were longstanding and had been accepted through compensatory measures in procedures.
-
This is a Severity Level IV violation (Supplement 1)."
_ _ _ _ _ _ _ _ - _ _ _
-_____-
_ -
- - - - - ------ -
'
< .
) .
i REASON FOR THE VIOLATION Example No.1 - Switchward Batterv
!-
- During a May 1996 BECo Quality Assurance Audit (#9603), a quality assurance
) engineer was not able to determine whether the standby battery could be used as an
!
altamate DC power source to the switchyard DC distribution panel with the
! documentation at hand. The engineer wrote problem report (PR) 96.9236 on May 9,
!
1996, to document the problem. On May 19,1996, following a review of all the i
available documentation related to the switchyard standby battery, it was determined I the plant had exceeded the T.S. 3.9.B.5 threc day LCO without an operabin switchyard l "
battery because the standby battery had been placed in service without the required i surveillance testing in effect. LER 96007 was written and lasued to document this
- finding. What must be highlighted is that Technical Specifications do not apply to the j : standby battery until it is placed in servios. At that time, all required surveillance testing
- i must be current.
i A corrective actkm item for the LER required a revision to Master Surveillance Tracking
. Program (MSTP) tasks S005222 and S005223 for the switchyard standby battery
[ testing to reference the appropriate Technical Specification (T.S.) sections. The MSTP ,
l
-
tasks were updated on May 29,1996, to indicate Task Type as Technical Specification !
related in accordance with the LER corrective action. The update of the due and dead '
l -dates was omitted for task # S005223 due to human error. Based on the last test i performanos date of December 1990 for the standby battery, the new due date should ,
have been December 1995, and the new dead date should have been March 6,1997. l
!
,
- However, the reason the standby battery was added to the MSTP to provide e vacking :
- mechanism to allow operations to ascertain whether the standby battery is ready for ;
-
l Tech Spec service. The standby battery is not required to be maintained oporable per t Tech Specs.
>
As a result of this human error, the corrective actions taken by BECo in response to the >
- self-identified deficiency involving the standby switchyard batteries, as reported in LER j 96007, were not scheduled timely enough in the master surveillance test program
(MSTP) to avoid the possibility of recurrence of the original condition adverse to quality.
! The NRC identified during a follow-up inspection of LER 96 007 (as part of Pilgrim inspection Report 50 293/97-01) the MSTP had a surveillance test due date of October
- - 31,1997, which co'.ild have exceeded the 5 year capacity test window required by T.S. l
4.9.A.2.d. Had the standby battery been placed in service following the March 6,1997,
, dead date and before the October 31,1997, test completion date, the original condition f would have reoccurred.
- i
,
i:
.
!
i
,
...r-e,-~ .,amm,w.w-...mmmw.e, ,e* %,-,aw.w---w ,.,-ne.m. w.w--, .-. . - .,ew.-.m-, s- e. ,e t -,c% , -er&-.,.,e,
--- - - - - - -
.
.
.
Example No. 2 - Operator Workarounds A. ALL RODS FULL IN The first operator vmtkaround involves control rod full-in lights that temporarily do not illuminate after a scram. This is a BWR Ircustry issue and the topic of GE SIL 632 (March 27,1991). Pilgrim Station Procedure 2.1.6 * REACTOR SCRAM * leads to a conservative decision to enter EOP 02 * Failure to Scram', until all rods can be verified to be at or beyond position 02. On April 19,1996 and February 15,1997, after the reactor scrammed, the full-in Indication was not received, and EOP 2 was properly entered.
Pilgrim is not designed with and does not have an ali rods full in Indicator. The Indication referred to as 'all rods full-In*, is an Indication of a single rod withdraw permissive while the reactor mode switch is in the REFUEL position. It will only filuminate if all rods are sensed as full-in and a reactor operator moves the mode switch to the REFUEL position to verify the permlasive.
GE SIL 532 recommends a wiring change to the probe buffer to give the full in signal.
The GE design would interfere with the EPIC scram timing function at Pilgrim. An altemative design developed by Pilgrim was avaltable after ihe April 19,1996, scram and is presently used as a temporary repair for failed full in reed switches until they are replaced during an outage. If this design is installed on all 145 probe buffer cards, it will' mask' a failed full-in reed switch and with the subsequent failure of the associated 00 reed switch, there would be no definitive way to prove the rod was full-In.
Improved options and enhanced process computer capability packages exist. These options interrogate the rod position probes for data to determine if a rod has inserted to position 02 or greater to verify that the scram was successful. None of these fixes or systems would have given an all rods full-in Indication given the conditions press,nt after Pilgrim Station's Apt; i9,1996 scram The design priority of these systems is not to give an all rods full in signal. There is no design available to address an enhanced all rods in or beyond 02 feature. Pilgrim engineering will develop a package with a data accuisition system (DAS) as part of the resolution of this operator workaround.
A problem report (PR96.9206) was entered into the corrective action process for the absence of immediate full-in indication following a scram. The timeliness of corrective action after the April 19,1996, scram can be attributed to the delay in finalizing an acceptable final design concept. This delay was caused by the ut.mplexity of the problem and our conclusion that the generic solutions would have introduced new and differerit problems.
9 4
. _ - _
_ _ _ . _ _ - _ - . -
.
B. MO 220 3 MAIN STEAM DRAIN LINE ISOLATION VALVE
.; MO 220 3 is a non safety-related steam line drain valve used to equalize the pressure arvoss MSIVs to allow the MSIVs to be opened if closed at reactor pressure. Due to past steam leakage problems, a maintenance request was written to replace MO-220-3 in RFO #10 (spring 1995).~ The valve was not an operator workaround problem at this time. The valve was replaced in RFO #10 with a new design using a rising rotating stem.
During the April 19,1996 shutdown, MO-220 3 failed to open from the control room switch while equalizing pressure around closed MSIVs. An operator was required to leave the control room and enter the condenser bay to manually open the valve.
~-
During this shutdown, the valve actuator was inspected, and the limit switch pinion was
,
found damaged. This condition could potonilally cause the torque switch to trip on e
lower torque than desired which could prevent the valve from opening. The open torque switch setting was increased and the limit switch was repaired. During an .
t August 1996 downpower, the entire actuator was replaced with a shop rebuilt assembly.-
Subsequently, during a September 1p96 plant shutdown, the valve again failed when
_ trying to equalize pressure around the MSIVs. Although no corrective actions were taken at this time, the decision was made to replace the entire valve with a more conventional design in RFO #11. The valve was replaced in RFO #11 under MR19602183. It has been satisfactorily post-work tested and was cycled during plant heatup and pressurization. Throughout these events, BECo took reasonable steps to resolve probleras associated with this non safety-related valve.
- C. REACTOR WATER CLEANUP LETDOWN CONTROL VALVE CV-1239
'
.
During plant startup or shutdown, it may be necessary to lower reactor water level via reactor water cleanua (RWCU) system to the main condenser or redwaste. Flow must be manually controlled by the operator via retdown control valve, CV-1239. The valve has challenging service conditions in that it must control small flows at both low and high reactor pressures. This results in cavitating control valve conditions that can lead
- to prema'are wear. CV-1239 had a poor performance history and was replaced per PDC 91-12. The new valve had some initial operating success but subsequently experienced the same type of service failure (i.e., V. ability to control flow over a wide nsnge of pressures).
Since opurators have become accustomed over many years to CV-1239 responding erratically, the MOVs downstream of CV-1239 have been used in the throttled condition to try to control letdown flow. This creates a backpressure in the letdown line which often trips the high pressure isolation switch (originally set at 140 psig). ' Over the years,
- this human factors problec.: was worked around by proceduralizing a step that would install a mechanical jumper that effectively blocked the actuation of the solenoid that e causes the high and now pressure isolations. With the mechanical jumper instelled, operators could be confident that they would have the control of the letdown fk,w that-they needed. Installation of the jumper required an operator to leave the control room
. and go to the reactor building.
i
_ _ _ _ _ _ _ _ . _ . - - - - -
- - - __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
.
.
Correcuve acuons to resolve the design and maintenance issues associated with the CV 1239 valve were not timely. The reason for this lack of timeliness was the apparent remplacency with the operating workaround which led to the low priority assbned to the resolution of this issue.
D. OPERATOR WORKAROUND TRACKING AND REVIEW PROCESS The three operatorworkarounds identified as Example No. 2 of the violation detracted from the operator's ability to promptly stabilize plant conditions during post scram conditions during the April 19,1996 and February 15,1997 scrams. These conditions were adverse to quality and required operators to take compensatory actions complicating the post scram recovery with the MSIVs closed dee to a Group 1 isolation.
Operator workarounds at Pilgrim Station are tracked by the Operations Department and classified as being related to running repairs, transient response, outages, and outage transient response. This information is tracked via spreadsheet format with information related to a description of the workaround, work packages prepared orin process, and estimated completion dates for resolution / elimination of the operator workarounds.
The process of tracking and reviewing operator workarounds is inconsistently prioritizing workarounds that have compensatory nieasures established in procedures.
This is particularly true for workarounds that affect operator actions during off-normal conditions.
CORRECTIVE STEPS TAKEN AND RESULTS ACHIEVED Example No.1 - Switchvard Batterv Problem report (PR 97,9129) captured the human performance error (inattention to detall) providing information to help trend human performance errors.
The due and dead dates for the standby battery tests have been recalculated based on December 1,1990 as the last test performed. MSTP Rep, task # S005223 was revised on February 19,1997. The new dead date is March 6,1997.
A tracking LCO (LCO97-70) has been generated for the switchyard standby battery.
The tracking LCO will remain in effect until the switchyard standby battery performance test is performed.
I1
-
.- ___ _. . _ _ _ .
.
,
.
Example No. 2 - Ooerator Workarounds
.. A. ALL RODS FULL-IN Pilgrim is working with various manufacturers on a design to monitor the control rod position after a scram allowing determination that all rods are full-in or beyond 02 position.
B. MO-220-3 MAIN STEAM DRAIN LINE ISOLATION VALVE The valve was replaced in RFO #11 under MR19602183. It has been
-
-
satisfactodly post work tested and will be cycled during plant heatup and pressudzation.
'
/ O. REACTOR WATER C.LEANUP LETDOWN CONTROL VALVE CV-1239 During RFO 11, data was gathered after initial troubleshooting and recalibration of CV-1239. The results indicated the trim was not passing the expected flows.
The valve manufacturer, Copes-Vulcan, was contacted and the service conditions were revisited. An operations / engineering team wss formed to determine actual flow needs at various plant conditions using the plant simulator. A new valve trim was installed to meet the flow service, reduce the litrelihood fu *ignificant cavitation, and extend the life of the plug seat. The valve has be n setup and calibrated using the Framatome air-operated valve diagnostics tool and will be checked at various flow and pressure conditions during plant startup to verify the design.
-
D. OPERATOR WORKAROUND TRACKING AND REVIEW PROCESS
.
An action plan has been developed that will establish a multi-disciplined team to datermine the cause and extent of this problem.
CORRECTIVE STEPS THAT WILL BE TAKEN TO AVOID FURTHER VIOLATIONS Examole No.1 - Switchvard Batterv The problem report trending process captures and trends the type of human performance error (inattention to detail) that led to Example No.1 of this violation, thus providing trending information to help reduce future human performance errors. Any negative trends related to this type of human performance error will be identified and
- resolved through the corrective action process.
Tracking LCO (LCO97-70) has been generated for the switchyard standby battery. The tracking LCO will remain in effect until the switchyard standby battery performance test is performed.
.
_ _ - _ .
~
"y
..
l ~ Example No. 2'- Operator Workarounds A. . ALL RODS FULL-IN -
As an interim solution to the all rods full-in, we will implement a design change which will provide 00 indication from the full-in switches. This will allow the process compuWr to monitor the full-in condition and allow a meaningful rod scan before the rods settle. This will be accomplished by the end of September 1997.
B. MO-220-3 MAIN STEAM DRAIN LINE ISOLATION VALVE None required.
C. . REACTOR WATER CLEANUP LrTDOWN CONTROL VALVE CV 1239 A temporary modification has been installed to bypass electrically the CV-1239 isolations until a high level of confidence from inservice test data is available to assure the operators will have control of the letdown flow line. This inservice test data will be monitored post RFO #11 to ensure the valve design is performing
- as expected and that it will have a long service life.
D. OPERATOR WORKAROUND TRACKING AND REVIEW PROCESS Establish multi-discipline team representing operations, engineering, training, and regulatory affairs groups to do the following:
.
Review Pilgrim Station procedures for additional cases where procedural steps may have been established to provide compensatory actions for less than optimal equipment performance (to establish the extent of the problem).
. Revise operator workaround list to include any additional operator workarounds identified during the procedure review process.
- Re Aew and revise the process as necessary for identifying, prioritizing, scheduling, and completing repairs / modifications associated with operator workarounds.
.- Develop a training module for the operator and engineering support personnel (ESP) training programs that will provide guidance for engineers and operators to identify conditions that repre ent compensatory actions for degraded performance of equipment (to preclude recurrence of conditions cited in this violation).
'
..
.. .
..
..
_ _
_ _ - - _ _ _ _ _ _ _ _ _ _ _ _ - - - _ _ - .
!
.
DATE WHEN FULL COMPLIANCE WILL BE ACHIEVED Example No.1 - Switchvard Batterv PNPS was not out of compliance with this issue; however, the potential non-compliance was averted by entry into tracking LCO 97-70 on March 6,1997.
Example No. 2 - Operator Workarounds A.- ALL RODS FULL-IN The plant design change to install a data acquisition system should be issued
-
for construction by the end of 1997, if the modification can be installed on line, then testing and tumover will be completed by and of first quarter of 1998.
.
B. MO-220 3 MAIN STEAM DRAIN LINE ISOLATION VALVE Full compliance was achieved with replacement of the valve during RFO #11 under MR 19602183.
C. REACTOR WATER CLEANUP LETDOWN CONTROL VALVE CV-1239
- A temporary modification has been installed to bypass electrically the CV-1239 isolations until a high level of confidence from inservice test data is available to assure the operators will have control of the letdown flow line. This temporary modification will be removed by the end of May 1997,
. D. OPERATOR WORKAROUND TRACKING AND REVIEW PROCESS The multi-discipline team will complete the tasks identified to prevent recurrence by August 1997.
.
14-