IR 05000346/1999004
| ML20207G082 | |
| Person / Time | |
|---|---|
| Site: | Davis Besse  |
| Issue date: | 06/07/1999 |
| From: | NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III) |
| To: | |
| Shared Package | |
| ML20207G080 | List: |
| References | |
| 50-346-99-04, 50-346-99-4, NUDOCS 9906110013 | |
| Download: ML20207G082 (21) | |
Text
,
.
U. S. NUCLEAR REGULATORY COMMISSION REGIONlli Docket No: 50-346 License No: NPF-3 Report No: 50-346/99004(DRP)
l Licensee: Toledo Edison Company l Facility: Davis-Besse Nuclear Power Station Location: 5501 N. State Route 2 Oak Harbor, OH 43449-9760 l
Dates: March 23 - May 13,1999 l
l l Inspectors: S. Campbell, Senior Resident inspector
'
K. Zellers, Resident inspector Approved by: Thomas J. Kozak, Chief Reactor Projects Branch 4 Division of Reactor Projects l
l l
l
'
9906110013 990607 PDR ADOCK 05000346 e PDR
!
l
[ ;
,
.
EXECUTIVE SUMMARY Davis-Besse Nuclear Power Station NRC inspection Report 50 346/99004(DRP)
This inspection included aspects of licensee operations, maintenance, engineering, and plant support. The report covers a 7-week period of resident inspectio Operations
+
Plant management exhibited a conservative operating philosophy to plan and execute a mid-cycle maintenance outage to address a number of material condition issues and to start the outage 2 weeks early in anticipation of exceeding the Technical bpecification limit for unidentified reactor coolant system (RCS) leak rate (Sections 01.1 and O1.2).
-
Operators shut the plant down in a controlled and deliberate manner for a mid-cycle maintenance outage per operating procedures. Control room and station personnel responded very well to equipment problems that were encountered during the shutdow J Equipment deficiencies were documented on condition reports, placed on mode restraint lists, and the deficiencies were resolved before plant startup (Section 01.3).
-
Station management personnel took conservative measures to replace three corroded j reactor coolant system RCS drain valve yokes. An RCS deep drain was required for 1 this activity. The work was well planned and executed, operators were particularly attentive to important plant parameters during the drain, and the RCS water inventory was closely tracked (Section O1.4).
-
The inspectors determined that the reactor startup and grid synchronization was conducted in a conservative, well-controlled manner (Section 01.5).
-
About 50 gallons of reactor coolant were inadvertently released to the containment atmosphere because containment vent header valve RC-98 was assumed to be open, when in fact it was closed during reactor coolant pump seal venting activitie ]
Contributing to this event was a weakness in the ' Conduct of Operations" procedure which allowed the transfer of valve lineup information from a previously completed procedure to one being performed during high workload conditions (Section 01.6). ,
Maintenance
+ The failure to ensure the pressurizer spray valve RC-2 hand switch was in the "close" position as specified in test instructions resulted in the unexpected cycling of the valve and a 10 psig drop in RCS pressure during the test. The effect on plant operations was minimal due to the rapid identification and resolution of the pressure decrease by control room operators. This was a Non-Cited Violation of test control requirements (Section M1.1).
The pressurizer spray valve and low voltage switchgear room 2 ventilation systems were made unavailable for up to 7 hours8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br /> when two breakers were inadvertently opened during an emergency diesel generator (EDG) 2 maintenance outage. The licensee determined that this event caused core damage frequency to substantially increase during this period. The most likely cause was inattention-to-detail by workers who transited a
l
,
1 .
.
relatively narrow passage between the breakers and EDG 2 during the EDG 2 maintenance outage and inadvertently contacted the two breakers and caused them to open (Section M1.2).
Enaineerino l
- An original design calculation error led to a condition where the EDG room ventilation system was sized such that it was incapable of meeting its design basis of maintaining EDG room temperature below 120 degrees with ambient outside air temperature of 95 degrees. This was a Non-Cited Violation of design control requirements (Section E2.1).
-
The licensee resolved over 41 mode restraint items prior to startup from the mid-cycle l outage in a comprehensive manner. However, the licensee's documentation for some
- of these items did not provide appropriate justification for issue resolution. The licensee provided acceptable verbal justification for these conditions and indicated that an update to the documented resolution would be performed (Section E7.1).
i
,
l l
_
E
.
I
- l l
Report Details l
Summary of Plant Status l At the beginning of tha inspection period, the plant was operating at nominally 100 percent l l power. Operators started to shutdown the plant on April 23 and opened the main generator !
output breaker at 1:07 a.m. on April 24 to start a mid-cycle maintenance outage. Operators !
l commenced plant startup activities on May 7, took the reactor critical on May 10, and I synchronized to the grid to end the mid-cycle outage at 6:29 p.m. on May 10. Reactor power was then increased until 100 percent power was reached on May 11. The plant was operated at nominally 100 percent power for the remainder of the inspection perio Operations l 01 Conduct of Operations 0 General Comments (71707) l Plant management exhibited a conservative operating philosophy to plan and execute a mid-cycle maintenance outage to address a number of material condition problem The inspectors conducted numerous inspections of control room activities, including operations at power, shutting the plant down for the outage, reactor coolant system (RCS) deep drain evolutions, and plant startup activities from the outage. The inspectors determined that operations personnel exhibited effective communications and effectively utilized procedures during routine and non-routine evolutions. Operators did not appear to be pressured by scheduling considerations during the activities observe Pre-evolution briefs were thorough and included industry lessons learned for higher risk evolutions, and personnel attending the briefs were attentive. Additional oversight by Mation management, quality control and the independent safety engineering group was observed for higher risk or infrequently performed evolutions. The control room environment was maintained at suitable distraction and noise levels despite the many extra oversight personnel present during some evolutions. Operators utilized peer checks frequently as a tool for event-free operations. However, some observed communications, although effective, did not employ three-way communication .2 Reactor Coolant System Leakane Inspection Scope (71707)
The inspectors tracked reactor coolant system leakage throughout the inspection perio Observations and Findinas During the inspection period, unidentified reactor coolant leakage increased at a rate that could have potentially exceed the Technical Specification (TS) 3.4.6.2 limit of 1 gpm. The licensee conducted several RCS leak detection inspections to identify leak locations. The licensee suspected that the majority of the leakage was from leaking pressurizer code safety valves. During previous operating cycles, leakage from these valves could be counted as identified leakage because of a leak-off line configuration
.
.1
-
l
..
l l
that had been established. However, the leak-off line was modified to address a design t concern which'resulted in all valve leakage escaping into the containment atmosphere ,
l as unidentified RCS leakage. The reactor coolant leaking from the valves evaporated l into the atmosphere and concensed onto the containment air coolers degraded their l performance to the point that about every 10-14 days, maintenance personnel entered
! ~ containment to clean the Plant management had previously planned to conduct a mid-cycle outage starting May 8,1999, in order to address equipment problems so that more reliable plant operations for the summer would be achieved. As a result of the increasing unidentified leakage, this date was moved up by 2 weeks, to April 24, so that a slow, orderly plant shutdown could be conducted instead of a more rapid shutdown that would be required I by TSs should the unidentified leak rate be exceeded. The orderly shutdown would,
!' among other things, allow plant personnel to conduct a thorough reactor coolant l pressure boundary walkdown at normal operating pressures in order to locate any leaks before cooling down and depressurizing the reactor coolant system.
l By the time of plant shutdown on April 24, the unidentified leak rate was at .82 gpm.
.
During the outage, station personnel replaced the pressurizer code safety valves and l
had the old ones tested to determine their leak rate. Test results indicated that the i combined leak rate of the two valves was about .75 gpm which nearly matched the leak l rate observed prior to shutting the plant down. Additionally, a modification to the pressurizer code safety valves was implemented during the outage and as a result, all leakage from these valves would be collected and counted as identified leakag Conclusions l Plant management exhibited a conservative operating philosophy when the decision j was made to start a mid-cycle outage 2 weeks early in anticipation of exceeding the l Technical Specification limit for unidentified RCS leak rat ,
01.3 Shutdown of the Unit for a Mid-Cycle Maintenance Outaae l Inspection Scope (71707) l i
On April 23, operators began decreasing reactor power to shutdown the plant for a mid-
'
cycle maintenance outage. The inspectors observed the operating crew perform
,
portions of the shutdown, reviewed applicable operating procedures, and attended j
! associated pre-job briefing Observations and Findinos The operators followed Procedure DB-OP-06902," Power Operations," and began l reducing reactor power by closing the turbine control valves to reduce main turbine generator load. The inspectors observed effective three-part communication and ;
self-checking practices during this activity. When Turbine Control Valve 3 reached its crack point, the target unit load demand module on the main turbine control panel failed, and the target load rapidly increased from 600 to 720 megawatts. Consequently, reactor power increased from 70 to 78 percent power. In response, the operators placed the steam generator / reactor demand in hand (placing the integrated control system (ICS) in track) to stabilize reactor power and reduced main turbine generator
i
.
-
\
load at 5 megawatts / minute. The operating crew then conducted another brief to decrease power with the ICS in track per Operations Policy EO-03, " General Guidance for ICS Control." The operators continued with the plant power reduction until the main turbine was tripped at 1:03 a.m. on April 24. The licensee documented the target unit load demand module failure on Condition Report (CR) 1999-0648 and replaced a defective target unit load demand module prior to the plant startu After the main turbine was tripped, operators continued to reduce reactor power by using the reactor demand hand station to insert control rods. At 1:36 a.m., when reactor power was at 4 percent, Annunciator 5-4-E, " Tilt imbalance insertion Limits,"
unexpectedly alarmed in the control room. The operators verified the alarm was received because of a quadrant power tilt and referred to the proper alarm procedure {
and TS 3.2.4. The operators confirmed that the associated TS limiting condition for j operation was not applicable when power was below 15 percent. Neve.rtheless, the l licensee documented the unexpected alarm in CR 1999-067 Plant operators continued to cool down the RCS. Operators monitored RCS cool-down ,
rates, RCS pressure, and nuclear instrumentation neutron counts every 30 minute l While the motor driven feed pump (MDFP) was operating, operators noted the lube oil !
pump began cycling on and off. Operators addressed the situation by referring to Procedure DB-OP-00016, " Removal and Restordion of Service Equipment," to temporarily throttle MDFP Lube Oil Pump Bypass Valve LO-2295 and maintain lube oil l pressure at 11 psig. The licensee documented the lube oil pump cycling on !
CR 1999-0651 and subsequently corrected the condition by replacing intermittent lifting relief valve PCV-5886 prior to unit startu Operators then manually tripped the reactor and continued the cool-down proces During this activity, operators noticed that Turbine Bypass Valves SP13B1 and 13B2 remained 50 and 15 percent open, respectively. Operators were then dispatched to manually isolate the valves to maintain control over the cool-down. The licensee documented this condition on CR 1999-0661. Subsequently, maintenance personnel disassembled the valves and identified that cracked and/or broken guide pins (used to guide the valve ball into the seat) had prevented the valves from fully closing. The apparent cause was cycle fatigue due to the valve cage contacting the top of the guide pins prior to the guide pin sliding into the valve cage. The valve cages on all the turbine bypass valves were modified to assure that the guide pins would properly engage the valve cage, and newly designed guide pins were installe When RCS pressure and temperature were reduced sufficiently, Decay Heat Train 2 was placed in service to continue the RCS cool-down. Subsequently, the plant pressure, temperature, and reactivity conditions satisfied entry into Mode 5 at 10:00 p.m.
, Conclusions
!
l Operators shut the plant down in a controlled and deliberate manner for the mid-t,ycle ;
maintenance outage per operating procedures. Control room and station personnel
.
responded very well to equipment problems that were encountered during the shutdow Equipment deficiencies were documented on condition reports, placed on mode restraint lists, and the deficiencies were resolved before plant startu l l
'
i l
. ;
01.4 Reactor Coolant System (RCS) Deep Drain Evolution Inspection Scope (71707)
The inspectors observed control room operators conduct a deep drain evolution in accordance with Procedures DB-OP-06002, " Drain and Nitrogen Blanketing of the RCS," and DB-OP-06904, " Shutdown Operations," in order to establish plant conditions favorable for replacing the yokes on RCS Cold Leg Drain isolation Valves RC-32, RC-38 and RC-4 Observations and Findinas Station personnel determined that reactor coolant system cold leg drain isolation valves RC-32, RC-38 and RC-40 were degraded in that the bottom portion of the yoke ears had experienced some boric acid corrosion from valve packing leaks. The yoke ears provide a holding force for the packing gland studs. The packing gland studs hold the packing gland in place. Failure of the yoke ears could cause the valve packing to fail resulting in increased reactor coolant system leakage that would force a plant shutdown and cool- ,
down for repairs. Station personnel determined that the yoke on valve RC-32 required I immediate resolution but that the other two yokes could wait for resolution until the next j refueling outage. Replacing the yoke on each valve required either an upstream freeze seat or a RCS cold leg drain (which required that the RCS be drained to 18 inches above the centerline of the hot legs - a high risk activity). Station management considered a second option for resolution of the corroded ear on RC-32 which involved I installing a modification to the valve that would provide supplemental holding force for the gland nut. After considering the two options, station management determined the best option was to replace the yokes on all three valves during the mid-cycle outage by conducting a deep drain of the RC j Management recognized this activity as a high risk, infrequently performed activity, and therefore, imposed additional administrative and oversight requirements during the evolution. The inspectors observed control room operator performance during the drain down and determined that the operators were in compliance with the requisite drain procedures, and that they were particularly attentive to critical parameters such as RCS water level, RCS temperature, decay heat system suction pressure, source range power level and rate, and reactor coolant water drain collection levels. Additionally, operations personnel effectively tracked and correlated the movement of water from the RCS to j receiver tank levels. This was done by determining the mass change of RCS based on i level changes and correlating this mass change to the corresponding mass change of reactor coolant receiver tanks determined from receiver tank level changes. This process provided an additional check that RCS level indications were accurate and that water being drained was being routed to the desired location.
The plan to replace the valve yokes required that the entire valve bonnet / yoke / stem assembly be taken off in order to perform the work in a low dose area and better j environment. Therefore, the plan included placing blank flanges on the bonnets v:hile
'
the bonnet / yoke / stem assembly was removed in order to restore the water tight integrity of the reactor coolant system boundary. The plan also included provisions to have three separate maintenance teams, one for each valve, to limit the time that the RCS was in the deep drain condition. During execution of the plan, virtually no problems were encountered, and the valve bonnet / yoke / stem assemblies were removed, blank flanges t
i
.
.
installed, assemblies transported to mechanical hot shops, yokes and packing replaced, l blank flanges removed, and assemblies installed in less than 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />. Because of this efficiency, the period of time that RCS level was 18 inches above the centerline of the hot legs was less than 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br />. The total time that the RCS was in a deep drain l condition (below the reactor vessel flange level) was less than 20 hour2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br /> Conclusions Station management personnel took conservative measures to replace three corroded j RCS drain valve yokes. An RCS deep drain was required for this activity. The work was well planned and executed, operators were particularly attentive to important plant parameters during the drain, and the RCS water inventory was closely tracke .5 Reactor Startuo and Grid Synchronization The inspectors observed the conduct of a reactor startup and grid synchronization activities that occurred on May 10. Control room operators maintained positive control of reactivity additions throughout the startup. The reactor startup procedure was adhered to throughout the evolution, operators were not challenged with equipment
. failures and minimal supentisory oversight direction was required. Shift turnovers during the startup were successful in conveying important operational information to the oncoming crews. The inspectors determined that the reactor startup and grid
'
synchronization was conducted in a conservative, well-controlled manne O1.6 Containment Vent Header Relief Valve RC-769 Inadvertent Lift Inspection Scope (71707)
The inspectors reviewed the circumstances surrounding an event on May 7,1999, where containment vent header relief valve RC-769 inadvertently lifted, releasing approximately 50 gallons of reactor coolant to the containment atmospher Observations and Findinas On May 7, operations personnel implemented Procedure DB-OP-06005," Reactor Coolant Pump (RCP) Operating Procedure," to vent the reactor coolant pump seals in preparation for plant heatup. Prerequisite Step 4.2.5.a required an operator to verify that containment vent header valve RC-98 was open in order to establish a vent path for reactor coolant to the gaseous radioactive waste syste A senior reactor operator (SRO) determined that the position of RC-98 could be verified by employing the verification provisions of Procedure DB-OP-00000, * Conduct of Operations."- Step 6.8.7 of DB-OP-00000 allowed operators to verify a valve's position by noting its position documented on a previously performed lineup. The SRO obtained Attachment 1, " Valve Checklist (Filling and Venting the RCS)," to Procedure DB-OP-06000, " Filling and Venting the Reactor Coolant System," that was completed on May 5, noted that the position of RC-98 was open, and signed off Step 4.2.5.a of Procedure DB-OP-06005 as completed. However, unknown to the operators, Attachment 1, " Normal RCS Valve Lineup," to Procedure DB-OP-06900,
" Plant Heatup," had been completed on May 6, during which operators closed RC-9 I
When the operators vented the RCP seals, the piping upstream of RC-98 became pressurized and caused containment vent header relief valve RC-769 to lift. After about three minutes of the venting evolution, personnelin containment notified control room personnel that water was being sprayed into the containment atmosphere. Operators then stopped all RCP seal venting activities, and the valve lineups were verifie Approximately 50 gallons of reactor coolant was discharged into the containment atmosphere. Station personnel verified that RC-769 had reseated and then cleaned up the spilled reactor coolant. Control room operators did not note any change in makeup tank level or pressurizer level during the event because of the relatively small amount of reactor coolant involved. To illustrate this sequ6nce more clearly, note the following time-line of event Date Procedure Used Action Result l Com Performed 5/5/99 DB-OP-06000 Filling and valve checklist RC-98 opened l (filed Venting the Reactor Coolant 5/7/99) System 5/6/99 DB-OP-06900 Plant Heatup valve checklist RC-98 closed (filed 5/6/99)
5/7/99 DP-OP-06005 Reactor prerequisite RC-98 assumed open by Coolant Pump Operating Step 4.2. referring to the position Procedure indicated in DB-OP-06000 completed 5/5/199 /7/99 DB-OP-06005 Reactor Operators With RC-98 actually closed Coolant Pump Operating vented RCP due to the lineup performed Procedure seals to the 5/6/1999 the containment l containment vent header pressunzed and ;
vent heade caused RC-769 to relieve the ;
pressure by releasing reactor !
coolant to the containment I atmosphere The inspectors reviewed the operations department configuration status files and determined that operators could readily utilize completed valve lineups to determine the configuration of systems during mode 1 activities, because the number of operational i
'
activities or lineups during mode 1 operations is relatively low and can be easily tracked by operations personnel. However, during startup activities, the number of plant configuration activities and number of operations personnel on shift significantly l increases. These factors make it difficult to track, coordinate, and manage plant
!
configuration activities during these high workload condition Also, the configuration file system was organized in such a way that it did not provide assurance that an operator would obtain the most recently performed lineup or t correct lineup to verify the position of a component for transfer to another valve lineup during times when the plant was in a dynamic state of configuration char ge (e.g., startup, shutdown, concurrent evolutions in process).
The inspectors discussed with plant management the provision of performing valve lineups by trasferring the status of a valve from a previously performed lineup. Plant 9 .
management indicated that utilizing this provision was not expected during high activity levels such as plant startup activities. The inspectors questioned whether management expectations were consistent with the written procedural guidance. As a result, plant management agreed to review the provisions of the written procedural guidance to determine if it was consistent with management expectation I Conclusions About 50 gallons of reactor coolant were inadvertently released to the containment j atmosphere because valve RC-98 was assumed to be open, when in fact it was close Contributing to this event was a weakness in the " Conduct of Operations" procedure which allowed the transfer of valve lineup information from a previously completed procedure to one being performed during high workload condition O2 Operational Status of Facilities and Equipment O System Walkdowns (71707)
The inspectors walked down the accessible portions of the following engineered safety I features (ESF) and important-to-safety systems during the inspection period: ,
-
Emergency Diesel Generators (EDGs)
Reactor Coolant System Relief Systems in containment
-
Portions of Component Cooling Water System and Letdown System in Containment !
+
Portions of Reactor Coolant System Boundary in Containment No substantive concerns were identified as a result of the walkdowns. System lineups and major flowpaths were verified to be consistent with plant procedures / drawings and the Updated Safety Analysis Report (USAR) Pump / motor fluid levels were within their normal bands. Vibration and temperatures of running equipment were normal. Only very minor oil and fluid leaks were noted on occasion. Local and remote controllers were properly positioned, and attendant instrumentation was functioning correctl Miscellaneous Operations issues (92700)
08.1 Temporary Instruction (TI) 2515/141. * Review of Year 2000 (Y2K) Readiness of Computer Systems at Nuclear Power Plants."
The staff conducted an abbreviated review of Y2K activities and documentation using Temporary instruction (TI) 2515/141, " Review of Year 2000 (Y2K) Readiness of Computer Systems at Nuclear Power Plants." The review addressed aspects of Y2K management planning, documentation, implementation planning, initial assessment, detailed assessment, remediation activities, Y2K testing and validation, notification activities, and contingency planning. The reviewers used NEl/NUSMG 97-07, ' Nuclear Utility Year 2000 Readiness," and NEl/NUSMG 98-07, " Nuclear Utility Year 2000 l
Readiness Contingency Planning," as the primary references for this revie Conclusions regarding the Y2K readiness of this facility are not included in this repor The results of this review will be combined with reviews of Y2K programs at other plants in a summary report to be issued by July 31,1999.
!
,
l
r ,
i
08.2 (Closed) Unresolved item 50-346/98012-01 (DRP): Entry into 10 CFR 50.54(x) (not able to comply with station TS, an emergency exists, and immediate action is required to protect the health and safety of the public) during recovery from a loss of offsite power due to a tornado touching down onsite. At 9:53 p.m. on June 24, the SS directed operators to commence a plant cooldown. Since there was no power available to operate reactor coolant pumps, a natural circulation cooldown was conducted. The maximum cooldown rate for a natural circulation cooldown allowed by the licensee's procedures was 10*F/hr. Technical Specification 3.0.3 was entered at 1:13 p.m. on June 25 due to EDG 1 being inoperable and having no offsite power sources operabl Although a procedure change was considered to allow up to a 50*F/hr cooldown so that the TS 3.0.3 requirement of cooling down to below 280*F within 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> could be met, this change was not made. The licensee's concern was that cooling down below 350*F would reduce the steam pressure available to drive the turbine driven auxiliary feedwater pumps should they be needed in case the motor driven feedwater pump became unavailable. The requirements of TS 3.0.3 were not met and operators invoked 10 CFR 50.54(x) to stop the plant cooldown at approximately 350*F. The licensee's decision to maintain reactor temperature at 350*F did not increase the risk associated with this event, and the licensee's actions were considered acceptable in this case given the circumstances. This item is close II. Maintenance M1 Conduct of Maintenance M1.1 Pressurizer Sorav Valve inadvertentiv Opened Durina Testino Activities Inspection Scope (71707) '
The inspectors reviewed the circumstances surrounding the unexpected cycling of pressurizer spray valve RC-2 during a testing activity which resulted in a 10-pound reactor coolant system pressure dro ; Observations and Findinas 1 Procedure DB-MI-03052, " Channel Calibration of 58A-ISP/TRCO2A2/3A4 RCS Pressure and Temperature to RPS Channel 2 and Pressurizer Power Relief Valve," was initiated on May 6,1999, at 6:15 p.m. to verify the calibration of a reactor protection system temperature module that had been replaced. A prerequisite in the procedure was to ensure that pressurizer spray valve RC-2 would not inadvertently cycle during testing by requiring that the control room hand switch be in the "close" position instead of its normal " auto" positio . A number of concurrent activities delayed completion of the test including, maintenance l activities on RC-2, testing activities on the PORV, fill and vent of the RCS, l establishment of a steam bubble in the pressurizer a valve lineup verification for the l
RCS fill and vent, and a valve lineup verification for the plant startup. Additionally, operators commenced a plant heatup on May 6 at 9:30 p.m. and had reached an RCS temperature of 190 degrees and pressure of 240 psig on May 7 at 9:25 p.m. As instrumentation and controls personnel were varying the pressure inputs to RPS channel two as a pad of the test procedure, a control room reactor operator quickly
l-
I .
! identified that RCS pressure had decreased by 10 psig and noted that the spray valve was open. Control room personnelinstructed the testing personnel to stop the testing, and determined that the pressurizer spray valve was in the automatic position contrary i to the test procedure prerequisite requirements. The test procedure prerequisites were then reverified, the pressurizer spray valve hand switch was placed into the "close" position, and the test was completed without inciden Computer data was subsequently examined which demonstrated that the spray valve had cycled several other times during the test activity. The first inadvertent cycling occurred almost immediately after commencement of the test activity. This fact implied that either the pressurizer spray valve had not been initially positioned correctly or that a valve lineup for plant startup had changed the valve position. Due to the short time period between verifying the position of the spray valve and the time at which maintenance personnel started to verify the pressure input, the inspectors determined that the most likely cause was that the spray valve was verified to be closed, but that the hand switch was not verified to be in the "close" position. Additionally, the inspectors had several other observations on the conduct of the test:
-
No pre-test brief was performed for control room operators prior to the conduct of DB-MI-0305 The performance of the test occurred over several shifts and in the midst of several plant activities that could have affected the spray valve positio However, instrumentation and controls personnel made no effort to reverify that ;
the prerequisites to the test were still valid from shift to shift or to conduct i briefings to new operations crews on the status of the testing activit No administrative controls existed to ensure that the spray valve stayed in the position desired for test. Normal controls that could have been used were out-of-service tags, information tags, or direct control of the equipment by the personnel performing the testing or maintenance activity. In this case, the pressurizer spray valve was requested to be repositioned, but no administrative means of ensuring that it would remain ja its desired position were utilized An administrative means of controlling bie position of the spray valve hand switch in this case was particularly important in view of the several plant and maintenance activities that were occurring during the execution of DB-MI-0305 The method employed by station personnel to place the spray valve into the hand and clased position was not rigorous. The test procedure prerequisite section required that five control room component positions be verified and signed off before start of the test. The normal method for the use of sign-offs as an event-free tool was to accomplish one step, sign it off, and then proceed to the next step. However, in this case, a control room operator verified that all five steps were completed, and then informed the test leader that the conditions were satisfied. The test leader then signed off that all five steps had been completed at the same time, thereby not fully utilizing the event-free too The procedure stated in its purpose section that the test be performed with the l RCS pressure less than 30 psig or as approved by the shift supervisor (Section 1 4) and that the procedure be performed in mode 5 or 6 or in a mode approvw by the shift supervisor (Section 1.5). There was no evidence (e.g., log
l i
[1
!
,
l l entry, procedure step sign-off, or interview results) that a shift supervisor had l granted permission for the test to be conducted above 30 psig. In fact, the test l was commenced with the pressure above 30 psig, and the pressurizer spray valve inadvertently cycled with the pressure at about 250 psig.
.
Criterion XI of Appendix B to 10 CFR Part 50 states, in part, that test procedures shall
! include provisions for assuring that all prerequisites for the given test have been met. A prerequisite for the performance of DB-MI-03052," Channel Calibration of 58A-ISP/TRC02A2/3A4 RCS Pressure and Temperature to RPS Channel 2 and
- Pressurizer Power Relief Valve," was that pressurizer spray valve RC-2 hand switch be l placed into the "close" position. Contrary to this, on May 6,1999, test procedures l pertaining to the Channel Calibration of 58A-ISPTTRC02A2/3A4 RCS Pressure and Temperature to RPS Channel 2 and Pressurizer Power Relief Valve did not assure that the prerequisites for the performance of the test were met such that pressurizer spray l valve RC-2 would remain in the "close" position during the conduct of the test. This )
l Severity Level IV violation is being treated as a Non-Cited Violation, consistent with
{
,
Appendix C of the NRC Enforcement Policy. This violation is in the licensee's corrective j l action program as CR 1999-0839 (NCV 50-346/99004-01).
The licensee's initial response to this event was to evaluate the feasibility to utilize an l information tagging scheme during testing activities to ensure that an inadvertent mis-positioning of plant equipment would not occur in the future. Additionally, plans were to l l evaluate the procedure for this test to ensure the prerequisites are located in the j appropriate location.
' '
l Conclusions The failure to ensure the pressurizer spray valve RC-2 hand switch was in the "close" position as specified in test instructions resulted in the unexpected cycling of the valve and a 10 psig drop in RCS pressure during the test. The effect on plant operations was minimal due to the rapid identification and resolution of the pressure decrease by control room operators. This was a Non-Cited Violation of test control requirements.
i M1.2 Pressurizer Sorav Valve Breaker inadvertentiv Opened Durina EDG Outaag Inspection Scope (62707)
l The inspectors reviewed the circumstances surrounding an event where normally closed j breakers BF-1260 (providing power to pressurizer spray valve RC-2), and YF-104
- (providing power to low voltage switch gear room (LVSGR) ventilation damper HV-5314)
I were found open on April 15,159 Observations and Findinas An overhaul of emergency diesel generator (EDG) 2 was ongoing during the week of April 12,1999. Station electricians, mechanics, instrumentation and control, contractors, operators, supervision, and oversight personnel all performed work during the overhaul within the relatively small confines of the room. Essential 480 volt Motor Control Center (MCC) F12B and essential 120 voit MCC YF1 are both located next to each other near to EDG 2. A narrow passage between the EDG and MCCs F12B/YF1 requked that personnel transiting this passage be careful so that the MCCs breakers
L
.
e would not be inadvertently operated. This is especially important because F128/YF1 breakers provide power to EDG and LVSGR ventilation dampers (support equipment for risk significant systems), an auxiliary feedwater valve (auxiliary feedwater is a risk significant system), and the pressurizer spray valv On April 15, with the EDG outage still ongoing, an equipment operator performing his routine rounds in the EDG 2 room found normally closed breakers YF-104 and BF-1260 both open. Operators shut the breakers to restore them to the desired configuratio The inspectors subsequently determined that the breakers could have been open for up to 7 hours8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br />. The effect of having BF-1260 open was that the pressurizer spray valve would not have opened to control RCS pressure in the event of a transient. Additionally, the PORV was not available for RCS pressure control, because it had been isolated with its upstream block valve in order to address PORV leakage. However, station risk assessment personnel subsequently determined that the spray valve and PORV not available did not contribute significantly to core damage frequency. The effect of YF-104 open was that LVSGR 2 ventilation was unavailable, which served to provide cooling for safety-related invertors, voltage regulators, battery chargers, batteries and switchgear that were in the room. Station risk assessment personnel determined that the risk model combination of the EDG 2 out of service with the LVSGR ventilation system unavailable caused core damage frequency to substantially increas The licensee fact finding effort did not determine the actual root cause of the breakers being open. The apparent root cause was that a person who had been in the EDG 2 room performing maintenance-related activities accidentally and unknowingly brushed up against YF-104 and BF-1260 while attempting to transit between F12B/YF1 and the EDG. This explanation was supported by the fact that: (1) YF-104 and BF-1260 are at
,
about hip level, (2) they are in close proximity to each other, and (3) they are both
. located very close to the EDG. The inspectors reviewed the factors involved which included an inspection of the proximity of the EDG to the breakers and determined that the licensee's apparent cause determination was reasonabl Related to the fact finding effort was a question as to why control room operators did not discover that breaker BF-1260 had been opened. When BF-1260 opened, the closed indicating light for RC-2 was de-powered. In response, control room operators stated that the light was very difficult to see because of excessive glare that was caused when the control room ceiling tiles were removed. The inspectors evaluated the ability of control room operators to observe that the RC-2 closed indicating light had extinguished during the conduct of their routine duties and determined that it would have been very difficult for a control room operator to have noticed that the closed light was out without performing a detailed walkdown of the pane Station personnel were in process of determining a method to physically prevent breakers on the panels from being accidentally opened during high activity levels in the room. Measures that were being considered were to: (1) provide a barrier directly in front of F12B and YF1, (2) post a continuous watch over the panel during high activity l levels, and (3) add an item to the maintenance pre-job brief checklist to discuss any
'
potential effects of maintenance activities on nearby plant equipment (item 3 has been cornpleted).
. Conclusions The pressurizer spray valve and low voltage switchgear room 2 ventilation systems were made unavailable for up to 7 hours8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br /> when two breakers were inadvertently opened during an emergency diesel generator (EDG) 2 maintenance outage. The licensee determined that this event caused core damage frequency to substantially increase during this period. The most likely cause was inattention-to-detail by workers who transited a relatively narrow passage between the breakers and EDG 2 during the EDG 2 maintenance outage and inadvertently contacted the two breakers and caused them to ope M1.3 Surveillance Activities (61726)
The following surveillance activities were reviewed during the inspection period:
-
DB-SP-03357 RCS Water Inventory Balance
-
DB-OP-03007 Shutdown Margin The execution of the activities was in conformance with procedural requirements. The procedures provided sound methodology to measure the parameter desired and presented acceptance criteria that was consistent with TS requirement M2 Maintenance and Material Condition of Facilities and Equipment M2.1 Plannina and Execution of Mid-Cycle Outaae The inspectors evaluated the planning, execution, and control of mid-cycle outage by conducting interviews, attending shift briefs, observing coordination activities and monitoring the status of the plant. The following major activities took place during the outage:
Reactor coolant pump 2-1 upper thrust bearing repaired
+
Containment air cooler 1-1 motor replaced
-
Rupture disk to pressurizer coae safety discharge kne installed to prevent pressurizer code safety leakage into the containment atmosphere
Leaking pressurizer code safety valves replaced
-
Leaking power-operated relief valve replaced
Letdown cooler rupture disks replaced
-
Extent of condition in containment for having wrong fasteners on pressure retaining components performed
+
Leaking auxiliary feedwater steam admiscion valve repaired
-
Main steam safety valves SP1784, SP17B6, SP1787 replacement
-
Turbine bypass valves repaired
-
Valves RC-2 and MU-1 A repaired
+
Corroded yokes on several RCS pressure boundary valves replaced
+
Deluge permissive modifications to the main / auxiliary transformers and the startup transformers
-
EDG generator exhaust modifications
.. .
.
-
.
. .
. '
i i'
f The inspectors determined that plant management prepared for and executed the maintenance outage in an organized and deliberate manner. Plant material condition
.
was improved during the outage which should increase plant reliabilit lil. Enaineerina E2 Engineering Support of Facilities and Equipment E2.1 Inadeauate Desian Basis Calculations for Emeraency Diesel Generator (EDG) i Ventilation System
- Inspection Scope (37551)
The inspectors reviewed the circumstances surrounding errors in EDG room ventilation system calculation Observations and Findinas During the loss of offsite power event that occurred as a result of a tornado that hit the site on June 24,1998, EDG Room 1 air temperatures exceeded the EDG operating procedure limit of 120 degrees and EDG 1 was declared inoperable. USAR Section 9.4.2.1.2.3 stated that the EDG room temperature would not exceed i 120 degrees with 95 degrees outside temperature and USAR Table 7.2-3 stated that the
- calculated maximum worst case temperature for the EDG rooms was 120 degrees. An
- obvious factor that caused the elevated room temperature was a failed hydromotor for the EDG 1 room ventilation system recirculation damper Even though the EDG 1 room temperature exceeded 120 degrees and EDG 1 was administratively declared i
inoperable, the EDG remained functional and supplied power to its essential busses l
throughout the event.
! One of the follow-up items from the event was to review the EDG ventilation system design. Bechtel Calculation 25.8, Revision 0 provided a basis for sizing the EDG l ventilation systems so that the ventilation systems would be in compliance with design requirements imposed by USAR Section 9.4.2.1.2.3 and Table 7.2-3. During this review, Bechtel Calculation 25.8 was determined erroneous in that: (1) it used an l incorrect density factor in calculating the volumetric flow rate requirement, (2) it did not l include the ventilation fan motor heat load, (3) the engine heat load was calculated I based on a different engine vendor (which was much lower than the heat load from the installed EDG), and (4) there were differences between the assumed ventilation air l l
intake configuration and the as-built ventilation air intake configuration. The corrected j l calculations indicated that the EDG room temperature could reach 133 degrees with the !
EDG running at 110 percent load at 95 degrees ambient temperatur i As a result of the findings, the licensee conducted an extent of condition review for potential errors in other ventilation calculations, initiated a new calculation for EDG room temperature, and planned to update the USAR with revised EDG room temperature limits. Additionally, modifications were initiated to: (1) redirect the air flow from the EDG generator ventilation exhaust, (2) provide more airflow through one of the EDG control cabinet enclosures, and (3) change the room ventilation controllers to a more robust design. As of the end of the inspection period, the licensee had completed the EDG
.
! generator ventilation modification and the EDG control cabinet modification. The results of the post maintenance testing indicated that substantialimprovements were made in the temperature profile for the most limiting EDG control cabinet component Additionally, the licensee performed a review of EDG component temperature '
certifications to evaluate any imminent failure concerns at high temperature condition During the review, EDG differential relays were determined to be the limiting components in that the manufacturer did not certify that they would function for a 40-year life span if they were subjected to temperatures above 131 degrees. The differential relays were not subject to significant temperature excursions above 130 degrees for an appreciable amount of time. The fact that the EDGs continued to function demonstrated that the differential relays did not degrade to the extent that the operation of the EDG was impaire In addition to the corrective actions mentioned above, the licensee planned to establish a new EDG room temperature operability limit based on ambient outside temperature and to revise the USAR description to more accurately describe the as-built design basis of the room ventilation syste CriMrion ill of Appendix B to 10 CFR Part 50, states, in part, that measures shall be established to assure that applicable regulatory requirements and the design basis as specified in the license application, for those structures, systems, and components to which this appendix applies are correctly translated into specifications. drawings, procedures, and instructions. Contrary to this, the design basis was not correctly l translated for the EDG room ventilation system in that it was not correctly sized such l that the EDG room temperature would not exceed 120 degrees at 95 degrees outside temperature. This Severity Level IV violation is being treated as a Non-Cited Violation, consistent with Appendix C of the NRC Enforcement Policy. This violation is in the licensee's corrective action program as PCAOR 98-1646 (NCV 50-346/99004-02). Conclusions An original design calculation error led to a condition where the EDG room ventilation i system was sized such that it was incapable of meeting its design basis of maintaining EDG room temperature below 120 degrees with ambient outside air temperature of 95 degree E7 Quality Assurance in Engineering Activities E7,1 Mode Restraint Ljst a, inspection Scope (71707)
!
l
'
The inspectors reviewed the documentation associated with station personnel's
! disposition of about 41 items that were on the mode restraint list prior to operations personnel initiating plant startup activities from the mid-cycle maintenance outag l Observations and Findinas Of the 41 items reviewed, the inspectors determined that 16 of the items did not have sufficient documentation to determine independently whether the condition had been
i
'
l
'
l acceptably resolved. The inspectors presented these questions to station personnel who provided followup information that enabled the inspectors to determine that all but one of the 41 items were adequately resolved. The remaining item, documented by CR 1999-0719, presented a concern that a 4-inch steam generator blow-down line had unjacketed NUKON insulation installed on the line. This discovery was made as part of an extent of condition review that had previously determined that NUKON insulation had been installed elsewhere in containment without having been evaluated for ECCS sump blockage during a loss-of-coolant accident (LOCA). The inspectors previously dispositioned the original issue as a Non-Cited Violation of Design Contro The justification for this condition provided in the CR was that the probability of a high energy line break occurring in the vicinity of unjacketed NUKON insulation located near the ECCS emergency sump was small. The inspectors questioned engineering personnel on the validity of this justification. Engineering personnel provided the clarification that due to the large size of the ECCS emergency sump screen, much less than 50 percent of the ECCS emergency sump would be clogged by the insulation. The USAR stated that the ECCS emergency sump was designed to perform its design function with up to 50 percent of the screen blocked. Additionally, the licensee indicated that the NUKON insulation was not susceptible to agglomeration and would not tend to collect on the ECCS emergency sump screen. The inspectors considered the new information and determined that the ECCS emergency sump function would not be jeopardized in the event of a LOCA in the vicinity of the unjacketed NUKON insulatio Subsequent to the conversation with the licensee, the licensee was planning to update l the documentation of the disposition of CR 1999-0719 with the justifications that were i discussed with the inspecto l l Conclusions The licensee resolved over 41 mode restraint items prior to startup from the mid-cycle outage in a comprehensive manner. However, the licensee's documentation for some of these items did not provide appropriate justification for issue resolution. The licensee i provided acceptable verbal justification for these conditions and indicated that an update j
'
to the documented resolution would be performe E8 Miscellaneous Engineering issues (37551) l
'
E (Closed) Unresolved item 50-346/96014-05 (DRP): Technical Specification Interpretation Report (TSIR) For Shield Building Integrity. This involved an event on November 26,1996, where a Howout panel shear bolt had been inadvertently stepped on and broken. The licensee determined that the blowout panel would fait during a design basis LOCA event, and declared the blowout panel inoperable. The licensee entered TS 3.6.5.2, Shield Building Integrity, which allowed up to 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> to return the blowout panel to an operable status before requiring a plant shutdown. The blowout panel was repaired in approximately 11 hours1.273148e-4 days <br />0.00306 hours <br />1.818783e-5 weeks <br />4.1855e-6 months <br />.
- The inspectors questioned the licensee's decision to not declare both trains of EVS l inoperable, because the EVS negative pressure boundary would be degraded during a LOCA event to the extent that neither of the EVS trains would be able to draw down the shield building negative pressure boundary pressure in the time assumed in the USA If both trains of EVS are inoperable, TS 3.0.3 would apply which allows the condition to
l
,
c
.
,
exist for only 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> before initiation of plant shutdown that would reach hot standby conditions within the next 6 hour6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> The licensee justified the action with TSIR 89-0007 which stated that the only TS that applied for the situation was TS 3.6.5.2 (Shield Building). Supporting the licensee's TSIR was NRC Inspection Report 50-346/91001 that stated the licensee's position was '
acceptabl After further review, the NRC technical staff determined that the licensee's TSIR was not I i
entirely correct. In a Task Interface Agreement memorandum dated June 11,1998, the Office of Nuclear Reactor Regulation (NRR) staff concluded that the EVS must be declared inoperable if it could not perform its design function, and the appropriate TS q should then be entered. After being informed of this decision, operations personnel :
generated Standing Order 98-014, which required that operations personnel defer to I licensing personnel concerning any future uses of TSIR 89-0007. Licensing personnel I subsequently submitted License Amendment Request (LAR) 98-0010 to relocate the surveillance requirement for the EVS draw-down test from TS 3.6.5.1 (EVS) to TS 3.6.5.2 (Shield Building Integrity). This provision is similar to the Combustion j Engineering Owners Group Standard TSs. The failure of the licensee to invoke !
TS 3.0.3 in this situation constitutes a violation of minor significance and is not subject to formal enforcement actio V. Manancment Meetinas X1 Exit Meeting Summary The inspectors presented the inspection results to members of licensee management at the conclusion of the inspection on May 13,1999. The licensee acknowledged the findings presented. The inspectors asked the licensee whether any materials examined curing the inspection should be considered proprietary. No proprietary information was identified.
l
v
.
!
.
PARTIAL LIST OF PERSONS CONTACTED !
. :
Licensee l l
G. G. Campbell, Vice President Nuclear J. H. Lash, General Manager, Plant Operations l R. E. Donnellon, Director, Engineering and Services i S. P. Moffitt, Director, Nuclear Support Services l J. L. Freels, Manager, Regulatory Affairs l J. L. Michaelis. Manager, Maintenance l M. C. Beier, Manager, Quality Assessment !
C. A. Price, Manager, Business Services G. A. Skeel, Manager, Security F. L. Swanger, Manager, Design Basis Engineering P. R. Hess, Manager, Supply S. Garchow, Manager, Nuclear Training D. H. Lockwood, Supervisor, Compliance !
D. C. Geisen, Supervisor, Plant Engineering J. J. Johnson, Supervisor, independent Safety Engineering Group j R. B. Coad, Jr., Superintendent, Radiation Protection J. E. Reddington, Superintendent, Mechanical Services M. J. Roder, Superintendent, E/C G. M. Wolf, Engineer, Regulatory Affairs D. W. Schreiner, Senior Maintenance Advisor R. M. Cook, Compliance Engineer NRC T. J. Kozak, Chief, Division of Reactor Projects Branch 4 K. S. Zellers. Resident inspector, Davis-Besse
,
INSPECTION PROCEDURES USED l
IP 37551: Onsite Engineering IP 61726: Surveillance Observations IP 62707: Maintenance Observation IP 71707: Plant Operations l lP 92700: Onsite Follow-up of Written Reports of Nonroutine Events at Power Reactor Facilities
20
r-
,
',
ITEMS OPENED, CLOSED, AND DISCUSSED Opened j 50-346/99004-01(DRP) NCV ' Prerequisites for a test not assured during performance of l test resulting in inadvertent 10 psig reduction in RCS pressur /99004-02(DRP) NCV inadeouate design basis calculations for emergency diesel
! generator ventilation system.
Closed l
50-346/96014-05(DRP) URI TSIR for Shield Building Integrity
- 50-346/98012-01(DRP)' URI Entry into 10 CFR 50.54(x)
i 50-346/99004-01(DRP) NCV Prerequisites for a test not assured during performance of l test resulting in inadvertent 10 psig reduction in RCS j pressur /99004-02(DRP) NCV Inadequate design basis calculations for emergency diesel generator ventilation system.
! LIST OF ACRONYMS AND INITIALISMS USED CFR Code of Federal Regulations CR Condition Report EDG Emergency Diesel Generator ESF Engineered Safety Feature
,
EVS Emergency Ventilation System ICS Integrated Control System l IFl Inspection Followup Item l lR inspect'on Report
! LOCA Loss-of-Coolant Accident MCC Motor Control Center MDFP Motor Driven Feed Pump NCV Non-Cited Violation NRC Nuclear Regulatory Commission i
PCAQR Potential Condition Adverse to Quality Report !
l PDR Public Document Room
! PORV Power Operated Relief Valve l l RCP Reactor Coolant Pump l l
RCS Reactor Coolant System l RPS Reactor Protection System ;
.SRO Senior Reactor Operator 1 TI Temporary Instruction TS Technical Specification ;
TSIR Technical Specification interpretation Report !
USAR Updated Safety Analysis Report l VIO Violation
21