IR 05000397/1998011
| ML17284A771 | |
| Person / Time | |
|---|---|
| Site: | Columbia  |
| Issue date: | 10/01/1998 |
| From: | NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV) |
| To: | |
| Shared Package | |
| ML17284A770 | List: |
| References | |
| 50-397-98-11, NUDOCS 9810060341 | |
| Download: ML17284A771 (36) | |
Text
i ENCLOSURE U.S. NUCLEAR REGULATORY COMMISSION
REGION IV
Docket No.:
License No.:
Report No.:
Licensee:
Facility:
Location:
Dates:
Team Leader:
Inspectors:
Accompanying Personnel:
Approved By:
50-397 NPF-21 50-397/98-11 Washington Public Power Supply System Washington Nuclear Project-2 Richland, Washington July 6 through August 28, 1998 C. J. Paulk, Senior Reactor Inspector, Maintenance Branch W. M. McNeill, Reactor Inspector, Maintenance Branch J. E. Whittemore, Senior Reactor inspector, Maintenance Branch M. Shlyamberg, P.E., Consultant NuEnergy, Inc.
Dr. Date A. Powers, Chief Maintenance Branch Attachment:
Supplemental Information 981006034i 98100i PDR ADOCK 05000397
-2-EXECUTIVE SUMMARY Washington Nuclear Project-2 NRC Inspection Report 50-397/98-11 This routine, core inspection was performed to assess the effectiveness of Washington Nuclear Project, Unit 2, personnel in identifying, correcting, and preventing problems that affect the safe operation of the plant.
~Oerattone Operations personnel were effective in the'identification and resolution of conditions adverse to quality (Section 2).
Proce Maintenance The gold card process was useful to identify human performance issues.
However, it was occasionally used to improperly include procedural violations and equipment issues.
This was due to a combination of factors, which included a lack of questioning attitude and minimal management involvement. The failure to initiate PERs, when required (two examples), circumvented the corrective action program and was identified as a noncited violation, pursuant to Section VII.B.1 of the NRC Enforcement Policy, of dure PPM *1.3.12 (Section 4).
Maintenance personnel were effective in the identification and resolution of conditions adverse to quality (Section 2).
The work control process was properly implemented with respect to the corrective action program (Section 3).
~En ineerin
~
Engineering personnel were not always effective in the resolution of conditions adverse to quality. The engineering personnel's performance was indicative of a lack of attention to detail. This was evidenced by a noncited violation of Criterion XVIto Appendix B of 10 CFR 50, pursuant to Section VII.B.1 of the NRC Enforcement Policy, for the lack of timely implementation of corrective actions for a condition adverse to quality. In addition, the licensee's actions associated with the turbine building service water system erosion and corrosion degradation; the potential for reactor core isolation cooling system bypass of primary containment; and the incomplete evaluations of water hammer events in the turbine building service water system indicated a lack of a questioning attitude (Section 2).
i-3-Re ort Details Summa of Plant Status The licensee was in the process of increasing power after the annual maintenance and refueling outage when the inspection began.
The unit achieved and maintained approximately 100 percent power for the duration of the inspection.
~Back round Aspects of the corrective action program in effect at Washington Nuclear Project, Unit 2, were reviewed in the operations, maintenance, and engineering functional areas, as discussed below. Since the corrective action program envelopes all technical disciplines and all of the systematic assessment of licensee performance (SALP) functional areas, this report format wilt deviate somewhat from the typical structure specified by NRC"Manual Chapter 0610,
"Inspection Reports."
Scope (40500)
The team analyzed the problems identified in the documents listed in the attachment to determine the licensee's effectiveness in performing:
Initial identification and characterization of problems; Elevation of problems to the proper level of management for resolution; Root-cause analyses; Disposition of any operability/reportability issues; Implementation of appropriate corrective actions; and Evaluation of repetitive conditions.
The documents reviewed included problem evaluation requests, work requests, work orders, goto cards, response to NRC information notices, operability evaluations, audits, and self assessments.
2.
aa Problem Evaluation Requests (PERs)
~Sco e
The team reviewed the 157 PERs listed in the attachment.
The sample was based, in part, on the risk significance of the system or components identified in the PER. The review evaluated the effectiveness of licensee personnel in identifying, correcting, and preventing the occurrence of problems that affect safe plant operations.
Also, the team reviewed root-cause analyses, as applicable, to assess their quality and effectivenes Observations and Findin s b.1 Operations-Related PERs The team did not identify any significant issues during a'review of 32 PERs that identified and reported problems related to the operations functional area.
Generally, the performance of operations-related identification of problems and implementation of corrective action was good.
b.2 Maintenance-Related PERs The licensee staff initiated PER 298-0107, on February 4, 1998, to address programmatic weaknesses in the PER corrective action process.
Five actions were identified as necessary to address the weaknesses, but the only significant progress was on the actions to provide additional training to appropriate personnel on the corrective action process and the root cause analysis process.
There was no indication of significant progress in the areas of clarifying criteria for identifying repeat problems, establishing review criteria for various steps of the process, or identifying process enhancements.
The team noted that several extensions had been granted to develop the required actions.
125 Vdc Electrical Distribution S stem The team noted, during the review of the vertical slice of 125 Vdc electrical distribution system maintenance activities, one PER for which the corrective action was not timely. The licensee staff wrote PER 296-0618, on August 7, 1996, because Appendix R-related emergency battery-operated lights failed during their annual 8-hour discharge tests.
Corrective actions were established to adjust the charging float voltage and replace the batteries.
On January 6, 1998, the licensee personnel started implementing the corrective actions to adjust the float voltage and replace the batteries.
These actions were not completed until August 25, 1998.
During the 2 years after establishment of the corrective action and its implementation, licensee personnel issued PERs 297-077, 297-0159, 297-0295, 297-0614, 297-0693, 297-0694, and 298-0274 as "noted" PERs for the additional test failures. The "noted" disposition was for trending purposes because the licensee had established, but had not performed the corrective actions for PER 296-0618.
Consequently, licensee management had at least seven opportunities to identify that the corrective actions had not been implemented.
The licensee staff revised the originally projected completion date of December 5, 1996, four times, in accordance with the procedure.
The most current projected completion date was August 1, 1998. Quality Department Surveillance Report 298-017, dated June 16, 1998, identified this lack of timeliness, but did not make a finding on the issue.
PER 298-0034 identified a programmatic weakness regarding the timely implementation
-5-of corrective actions for other issues, but did not identify PER 296-0618 as part of the scope of the problem.
Corrective actions for PER 298-0034, in part, were to communicate management expectations regarding extension approvals for corrective action due dates.
The expectation was that, after the first extension, subsequent extensions would require higher management approval.
However, this management expectation was not formalized into a procedural requirement.
Criterion XVIof Appendix B to 10 CFR Part 50 requires that conditions adverse to quality shall be promptly identified and corrected.
The team identified the failure to complete corrective actions until August 25, 1998, for a condition adverse to quality, which was identified in August 1996, as a violation.
Problem Evaluation Request 298-1121 was initiated on August 25, 1998, to address the untimely completion of corrective actions for Appendix R battery-powered light deficiencies.
The corrective action was to revise Procedure PPM *1.3.12A to require that the extension of corrective action plans be reviewed for safety-significance and that a written justification be provided, with a discussion of the safety-significance, for the extension request.
The team found the proposed corrective actions to be appropriate to correct the condition adverse to quality and, if properly implemented, to prevent recurrence.
The team considered that the issue of not controlling extensions to due dates which result in untimely performance of corrective actions to have been licensee-identified.
This non-repetitive, licensee identified and corrected violation is being treated as a non-cited violation (50-397/9811-01), consistent with Section VII.B.1 of the NRC Enforcement Policy.
b.3 Engineering-Related PERs Based on the sort of PERs by discipline, there were?16 PERs designated as engineering related PERs.
The team selected 62 PERs for detailed review based on their description.
These selected PERs were grouped for review as follows.= There were 5 PERs associated with the reactor core isolation cooling system; 10 PERs associated with the turbine building service water system; 10 PERs associated with information notices; 10 PERs associated with heating, ventilation and cooling systems; 10 PERs associated with the service water system; 4 PERs associated with the emergency diesel generator systems; and 13 PERs not associated with any specific system.
Reactor Core Isolation Coolin RCIC S stem The RCIC system is a safety-related system, which was credited as a backup system to the high pressure core spray system for the control rod drop accident.
Additionally, the
, RCIC system automatically initiates on a predetermined low level in the reactor vessel; hence, RCIC actuates during loss-of-coolant accidents and other events that lead to loss-of-level in the reactor vesse The team noted that Calculation NE-02-93-09, "RCIC Leak Through Oil Cooler Pipe Out RCIC-RV-33," was performed in response to PER 293-0302.
A licensee engineer identified that, on loss-of-dc power, primary containment is bypassed as the result of losing power to condensate Pump RCIC-P-4 (which is located in the RCIC pump room in the secondary containment).
The driving force assumed in the calculation was the static pressure of the suppression pool with primary containment pressurized.
The team found that the licensee's engineers did not identify, or consider, the possible
" scenario of the RCIC pump operating with a loss-of-level control. As such, in Calculation NE-02-93-09, the licensee's engineer calculated that the teak rate would only be 25 LPM (6.5 gpm), not the 61 LPM (16 gpm) that would be present with the pump running.
Problem Evaluation Request 297-0936, dated, November 20, 1997, was issued to document that RCIC condensate level control would not function during a station blackout event. The following description of this condition was based on the information provided in the subject PER and discussions with the licensee's engineers The RCIC condensate tank (a barometric condenser vacuum tank) automatic level control is not available during a station, blackout event (complete loss-of-ac power).
The barometric condenser is an atmospheric condenser that receives RCIC turbine drains and gland seal steam leak-off and main steam valve stem leak-off, which are condensed at atmospheric pressure using cooling spray from the discharge of the RCIC pump. This was the result of a modification to the control circuit that was made prior to the issuance of the plant's operating license, by General Electric Nuclear Engineering.
This modification added an ac relay powered from a non-Class 1E power panel.
This non-Class 1E, 120 Vac Power Panel E-PP-7AZ, is supplied from a safety-related bus, and is not shed during load sequencing.
During RCIC operation, approximately 61 LPM (16 gpm) of discharge flow are diverted to the barometric condenser to provide cooling spray for any gland seal leakage from the RCIC turbine. This cooling spray and gland seal leakage accumulate in the barometric condenser vacuum tank. Tank level is controlled by level Switch RCIC-LS-11, which starts and stops Condensate Pump RCIC-P-4 on high and low level, respectively.
Pump RCIC-P-4 normally returns vacuum tank coolant to the suction side of Pump RCIC-P-1 (main RCIC pump). Switch RCIC-LS-11 provides its signal to Pump RCIC-P-4 via Relay RCIC-RLY-K75, which is powered from the 120 Vac source discussed above.
During a postulated station blackout event, Relay RCIC-RLY-K75 remains de-energized.
Consequently, a high level signal from Switch RCIC-I S-11 would not start Pump RCIC-P-4 to reduce level in the barometric condenser vacuum tank. With an incoming flow of approximately 61 LPM (16 gpm), the tank would fillin approximately 10 minutes, activate Relief Valve RCIC-RV-F033, and result in an overflow to the RCIC room equipment drain. This overflow would first fillthe control rod drive pump room sump, due to the common connection of drain lines, and eventually flood the floor in
-7-rod drive pump room and the RCIC pump room. The licensee's calculation, based on a leak rate of only 25 LPM (6.5 gpm), determined that the approximate flood level in the RCIC and control rod drive pump rooms would be approximately 28 mm (1.1 in) after 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> of operation.
The team calculated that, on the basis of a leak rate of 61 LPM (16 gpm), the level of water in the RCIC and control rod drive pump rooms would be approximately 25 mm (1 in) in approximately 2.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br />.
The licensee's engineers evaluated, in Calculation ME-02-89-21, this condition to be acceptable and determined that there was no operability issue.
The engineers'ecision was based on the facts that, in the licensing basis, high pressure core spray, not RCIC, was the system credited for coping with the station blackout event.
However, the assumptions used for Calculation ME-02-89-21 did not consider Pump RCIC-P-1 to be operating and providing 61 LPM (16 gpm) into the barometric condenser.
Two corrective actions for the non-Class 1E components (e.g., Switch RCIC-LS-11 and Relay RCIC-RLY-K75) were identified by?he licensee's engineers.
The first corrective action was to revise the RCIC room temperature calculation to incorporate the additional heat load associated with this event (completed May 29, 1998). The second corrective action was to revise the station blackout event emergency operating procedure to direct the operator to manually operate Pump RCIC-P-4 based on the high and low level alarms (completed February 23, 1998).
Both corrective actions were completed prior to this inspection, but several months after the identification of the problem.
The team's review of the PER evaluation and corrective actions, as well as review of
'related documents and discussions with the licensee's engineers, identified that the engineers failed to appropriately disposition PER 297-0936.
The following related documents were reviewed: Washington Nuclear Project, Unit 2, Updated Safety Analysis Report, Sections 5.4.6 and 15.2.6; Calculation EQ-02-97-01; "Evaluation of Active Failure of RCIC-LS-11 in Seismic Event"; Component Summary Sheets for RCIC-LS-11 and RCIC-RLY-K75; Technical Merriorandum TM-2050, "Classification of Design Base Functions and Primary Containment Isolation Boundaries of the RCIC System"; and, the safety evaluation screening for the revision of the station blackout procedure.
The discussion below provides details of the team's evaluation.
The document, which set the basis for the qualification of the RCIC components and sub-components, was Technical Memorandum TM-2050. This document did not include the closed loop extension of primary containment for the RCIC system, the steam supply, and exhaust portions of the RCIC system, as well as the barometric condenser and its auxiliaries (see page 14 and Figure 1 of TM-2050). The guidance on electrical equipment qualification provided on page 13 of TM-2050 essentially limited the need to impose the environmental qualification requirements only to components which were required to perform either containment isolation, or high energy line break detection and isolation functions. Consequently, the limitswitch and relay were classified as non-
-8-environmental qualification-related.
The justification for qualification of the level switch (provided in CGER No. C92-0027) explicitly, and erroneously, stated that "[t]he RCIC steam supply and exhaust lines between RCIC-V-8 and RCIC-V-40 are not considered an extension of primary containment (Tech Memo 2050, Rev. 1)." The calculation, EQ-02-97-01, further stated that this switch only had passive, i.e., pressure boundary, safety-related requirements.
On the basis of the information, discussed above, the team found that some aspect of the restoration of the original classification (i.e., safety-related) for the RCIC system was limited to a passive, pressure boundary function only. The definition of the scope of the piping equipment for closed loop extension of primary containment for the RCIC system was only appropriate for a large break loss of coolant accident, since the large break loss of coolant accident would result in a prompt isolation of the containment isolation valves.
However, for other than large breaks, application of this definition neglected the fact that the entire RCIC system would be an extension of primary containment for an appreciable period of time. Additionally, the licensee's calculation of the room temperature for the event described in PER 297-0936 concluded that the final room temperature would be below the set point for of the high energy line break isolation system; therefore, the RCIC system would not isolate under those conditions.
Use of the nonsafety-related components and the nonsafety-related power source has a potential to bypass one of the three defense-in-depth boundaries (i.e., fuel cladding, reactor coolant pressure boundary, and primary containment).
Since the function of the nonsafety-related components or power sources cannot be credited for prevention or mitigation of the consequences of the accident, failure of the Relay RCIC-RLY-K75, or its power supply, or Switch RCIC-LS-11, would result in the loss of Pump RCIC-P-4 with all of the consequences described in the subject PER and the liftingof Relief Valve RCIC-RV-F033.
This scenario could occur not only during a station blackout event, but also during any design basis large or small break loss of coolant accident, or whenever the RCIC system was in operation and certain nonsafety-related components fail.
The licensee's engineers informed the team that they performed an analysis in accordance with 10 CFR Part 100 and General Design Criterion 19 for a rupture of the barometric condenser and that the results of this analysis were bounded by the design basis large break loss of coolant accident.
The team noted that the analyzed loss of the condenser pressure boundary was a postulated failure and, therefore, other failures (active or passive) were not required to be postulated.
On the other hand, loss of the pressure boundary due to a preexisting condition cannot be considered a postulated failure and, therefore, other failures (active or passive) must be considered.
Examples of failures that should be considered are failures affecting the integrity or the capability of the secondary containment.
Since no such analyses were performed by the licensee's engineers, the current configuration was considered to be an unanalyzed configuratio Both the engineering and operations department safety evaluation screenings for the change to Procedure PPM *5.6.1, "Station Blackout," failed to take into account that RCIC would be initiated during a loss-of-coolant accident and addressed only the station blackout event.
However, the team found even this limited review was inadequate.
The Final Safety Analysis Report, Section 15.2.6.5, "Loss of AC Power,
'adiological Consequences,"
states that "[w]hilethe consequence of this event does not result in fuel failures, it does result in the discharge of normal coolant activity to the suppression pool by means of SRV operation. 'Since this activity is contained in the primary containment, there will be no exposure to personnel."
Neither the PER, nor the safety evaluation screening, evaluated this apparent discrepancy'between the licensing basis and the condition described in the PER.
Additionally, in Calculation NE-02-93-10, "Post-LOCA Flooding and CRD/Condensate Pump Rooms," the dose rate was calculated for 25 mm (1 in) of suppression pool water on the floor of the RCIC pump room to be approximately 1.0 mR/hr. This was to be compared to a dose rate of 0.01 mR/hr during a loss-of-ac power without the relief valve lifting.
The team's review of the safety evaluation screening for the procedure revision identified that the documented justification of acceptability of the change was inadequate, since it was essentially limited to restatement of the scope of the revision (i.e., to add manual operation of Pump RCIC-P-4 in the station blackout procedure).
The team found the safety evaluation to be inadequate in that there was no consideration of the increase in exposure to personnel as a result of a station blackout event.
. The original design was based on the automatic control of the level of the barometric condenser.
Implementation of the second corrective action (i.e., revision of Procedure PPM *5.6.1) effectively resulted in the crediting of operator actions in place of automatic actions.
Crediting of operator actions in place of automatic actions was the subject of NRC Information Notice 97-78, which was reviewed and documented by licensee personnel in PER 297-0860.
(For additional details, see review of information notices below.) The revision of Procedure PPM * 5.6.1 was implemented at approximately the same time as the review of PER 297-0860, which failed to identify that operators'ctions were credited in place of automatic action.
During the followup week, the team was informed that personnel at three other plants had been contacted by the licensee's engineers.
It was stated that none of those plants had an isolating relay similar to Relay RCIC-RLY-K75 installed in the RCIC system at Washington Nuclear Project, Unit 2. However, all thr'ee plants had the same classification for the level switch. Specifically, the level switch was safety-related for pressure boundary protection, but was not Class 1E-qualified for electrical performance.
A licensee engineer initiated PER 298-1119 on August 25, 1998, to evaluate the analysis of the potential leak path from the suppression pool and primary containment through relief Valve RCIC-RV-33 in the event of a failure of the non-Class 1E level Switch RCIC-LS-11.
No resolution or corrective actions were provided to the team by the end of the inspectio The team found that the RCIC system may be outside of its design basis under certain conditions (e.g., loss-of-power to Relay RCIC-RLY-K75 or loss of Switch RCIC-LS-11, and the RCIC system in operation).
This was based on the bypassing of primary containment when the barometric condenser tank relief valve lifts during the operation of the RCIC system with a loss of the automatic condenser tank level control. As such, Region IV willfurther evaluate the possible generic implications of this design. *This issue will be tracked as an inspection followup item (50-397/9811-02).
Turbine Buildin Service Water All PERs reviewed for the turbine building service water (TSW) system provided a good documentation of the extent of the problems.
However, the team found that the corrective actions were not always broad.
For instance, when blockages of the thermal pressure relief valves were identified, the PERs (297-0371 and 297-0388) appropriately identified the remaining pressure relief valves, vents, drains, and instrument connections, as potential suspects for blockages.
However, the corrective actions were limited to only replacement of all pressure relief valves.
In addition, the evaluation of the heat exchangers affected by blockage of the pressure relief valves was limited to a visual vessel examination.
No acceptance criterion was used in this evaluation and personnel used to perform this evaluation did not,have suitable experience or training.
. That is, the system engineer who performed the visual inspection had no training or experience in nondestructive examination or stress analysis in order to determine the acceptability of the vessel.
Out of 10 PERs reviewed by the team, 5 were generated to address erosionicorrosion-related problems.
The information provided in these PERs and discussions with the TSW system engineer resulted in the following observations about the condition of this system.
Although this was an open water system, it did not receive effective chemical treatment.
The system suffered from erosion and corrosion,, which included through wall leakage attributed to presence of active Desulforibro sulfur-reducing bacteria.
Micro biologically-induced corrosion was also present.
Additionally, small diameter lines experienced blockages with corrosion products, silt, etc. The team found the lack of treatment was chronic. Based on the information that was provided in PERs, the team found that licensee management did not pay sufficient attention to this issue.
Also, the proposed modifications were either not implemented or poorly implemented (e.g., the system did not have a dedicated carbon corrosion inhibitor or a dispersant treatment).
Modification BDC 91-0299-OA specified and installed pumps to provide adequate chemical treatment.
The pumps, however, were under-designed.
This problem was scheduled for resolution during the 14" refueling outage, in 1999. The biocide treatment had been discontinued in September 1991.
Since that time, the system did not have an effective biocide treatment.
Modification BDC 94-0084-OA was installed to remedy this problem; however,.the pump was inducing significant pulsation in the plastic suction lines, to a point that management was concerned with potential suction pipe failures (PER 297-0627).
The problem again appeared to be inappropriate pump selection.
There was no firm date for the resolution of this proble Lack of effective chemical treatment for this system was synonymous with lack of effective preventive maintenance.
Additionally, the situation was further exacerbated by a lack of any predictive maintenance.
Based on the information provided by the system engineer, until recently, the corrosion coupon rack (which could have provided erosion and corrosion rate data) had its drain plugged, preventing the receipt of any data.
Only once in 1996 did the system receive an ultrasonic evaluation and the results of the ultrasonic evaluation indicated that the pitting corrosion was pervasive.
The team found that the TSW maintenance strategy was limited to after-the-fact replacement of leaking piping. There was, however, a major maintenance plan (MM1306), which was supposed to replace some of the problem portions of TSW piping.
During the 13" refueling outage, small bore (25 mm (1 in) and smaller) TSW piping in the main steam piping tunnel was replaced.
During the 14~ refueling outage, 17.4 cm (6 in) and smaller TSW piping in the reactor feedwater pump area was scheduled for replacement.
However, even after the scheduled replacements only, a very small fraction of the total TSW will have been replaced.
The team expressed a concern that the TSW system could experience a catastrophic loss of a pressure boundary due to a water hammer event, similar to the recent event experienced by the fire protection water system and discussed in NRC Inspection Report 50-397/98-16.
The team's concerns were based on known cases of through-wall corrosion and lack of predictive maintenance for the TSW system and a lack of knowledge about the (operational) susceptibility of the TSW system to water hammer.
The licensee provided the team with the following information about the TSW system.
The ruptured component within the fire protection system was a cast iron valve. AIIof the TSW system large bore components and piping were steel, which were not nearly as brittle as the cast iron. AIIof the known through water leaks in the TSW system were in relatively small piping. Allof the leaks were pinhole leaks, which happened in smaller stagnant branches, whereas all main branches have flow most of the time. In addition, the qualitative review of the areas adjacent to these leaks did not reveal any abnormal loss of wall thickness.
Although no formal program existed to monitor wall degradation, there were numerous valve replacements, including large bore valves.'n every case, the adjacent piping was evaluated (qualitatively) for an abnormal loss of wall thickness and the wall thickness.
was found acceptable.
Although this information somewhat alleviated the team's concern, the team found that the licensee's lack of programmatic approach to this potential problem was a shortcoming.
The team viewed the following indicators as evidence of a lack of programmatic approach:
absence of any system-based evaluations of a potential for operationally induced water hammer events and mitigative measures, if any; and, an absence of any documented monitoring providing quantitative observations about loss of wall thicknes During the exit, licensee management indicated that they would be establishing a task force to address all of the above-discussed TSW issues.
Information Notices 0 erationalEx erience Feedback Out of the 10 PERs selected for review, the team noted that the licensee's review of the 3 PERs discussed below were weak. The weaknesses included incomplete scoping of the issues, excessive reliance on existing barriers, and a lack of thoroughness.
The remaining 7 PER reviews were found to be adeq'uate.
PER 297-0624 IN 97-16
"Preconditionin of Plant SSCs Before IST or Technical S ecification Surveillances" The team found this PER evaluation to lack in thoroughness.
The subject information notice addressed two potential me'chanisms which could lead to preconditioning or "grooming." The first mechanism, identified in the information notice, was preconditioning due to inappropriate steps within the surveillance procedure.
The second mechanism was "grooming" due to the scheduling of preventive maintenance prior to the surveillances.
This PER extensively addressed the first mechanism.
However, with the
'xception of the response from the motor-operated valve group, this PER response failed to evaluate the effect of scheduling maintenance activities prior
.to scheduled surveillances.
The licensee personnel indicated, during the discussions with the team, that this issue was not evaluated.
However, their
. subsequent review (after the team's questions) of the maintenance and surveillance schedules determined that there was no inadvertent "grooming" associated with scheduling.
PER 297-0680 IN 91-50 Su lement
"Water Hammer Events Since 1991" The team found the evaluation performed in this PER to be incomplete in the scoping of the issues.
The disposition of this PER was limited to evaluation of the operationally observed water hammers and the type of water hammer events described in Generic I etter 96-06. The information provided in this PER did not provide any evidence that a systematic review of other systems, which operate infrequently and may experience conditions not observed during normal plant operation, was performed for the w'ater hammer susceptibility. The team's discussions with the licensee personnel confirmed that the review was narrow in scope.
Moreover, the team's concern about the TSW water hammer susceptibility, discussed above, was not addressed by the licensee in the PER. Also, the recent water hammer event, which led to the loss of pressure integrity of the fire protection system, might have been avoided if the review had been broader.
For example, the gages in the fire protection system were frequently replaced because they were sticking, reading higher than expected, or over-range Each. of these conditions was a potential result of a significant hydraulic pressure transient.
(Additional information on the fire protection system with respect to hydraulic pressure transients is contained in NRC Inspection Report 50-397/98-20.)
PER 297-0860 IN 97-78
"Creditin of 0 erator Actions in Place of Auto Actions and Modification of 0 erator Actions Includin Res onse Time" The team found that this PER evaluation excessively relied on existing barriers.
The licensing personnel based their evaluation on the existing programmatic
'arriers in place, which were supposed to preclude the crediting of operator, actions for automatic actions.
However, the licensee personnel did not provide any information to indicate that these barriers were previously in place and whether there were recorded breaches of these barriers (documented in the PER database, or the previously used problem and nonconformance data bases).
The team's review of a list of PERs identified at least one other potential example where these barriers were ineffective (PER 297-0936, discussed above).
c.
Conclusions The team identified examples of weak evaluations (e.g., potential RCIC bypass of primary containment, TSW erosion and corrosion degradation, susceptibility to water-hammer events, and crediting of operator actions in place of automatic actions),
and untimely determination and implementation of corrective actions (e.g., emergency battery-operated lights). While the PER program was generally implemented in a manner to comply with the requirements of 10 CFR Part 50, Appendix B, as demonstrated by the relatively low number of findings, a noncited violation pursuant to Section VII.B.1 of the NRC Enforcement Policy was identified for the failure to implement timely corrective actions.
An inspection followup item related to the evaluation of the potential bypass of primary containment under certain circumstances.
Operations and maintenance personnel were effective in the implementation of the PER program; however, engineering personnel were not always effective. The engineering personnel's performance was indicative of a lack of attention to detail and a lack of a questioning attitude.
3.
Work Orders and Work Requests a.
~Scc e
The team reviewed the 59 work orders and work requests listed in the attachment.
The selection of the documents was based, in part, on the risk significance of the system or component.
This review evaluated the licensee's effectiveness in identifying and correcting conditions that may affect the safe operation of the plan b.
Observations and Findin s The team reviewed more than 40 work orders that were related to the plant operations area.
The team did not identify any work orders that had not been successfully completed, or that should have resulted in additional corrective action. About half of the operational work orders that were reviewed had been initiated to perform surveillance, post-maintenance or post-modification testing, and then canceled, before the testing was completed.
In order to confirm that previously scheduled surveillance and testing were completed, the team found it necessary to conduct additional interviews and request additional documentation.
The licensee's practice of canceling work orders for testing and converting them to work requests, other work order tasks, or scheduled surveillances, made it difficultfor the team to followthe outcome of the testing. This difficultywas also experienced by the licensee's staff as evidenced by their inability to quickly find and provide the test results needed to evaluate about 20 selected previous determinations of operability on the basis of successful testing.
Licensee personnel acknowledged the observation of difficultyobtaining specific test data information from work control documents in the data base.
The team found that the licensee's work processing program included a low-level program called "fix-it-now." This program was of a low level, in that, little planning and documentation was associated with its activities. Work was accomplished on the basis of a work request and, in some cases, without the development of a work order. A nonquality procedure defined the program.
Fix-it-now activities included corrective work activities for mundane work, such as replacing light bulbs, adjusting manual valve packing, removing or installing vent or drain caps, etc.
It also included minor maintenance, such as repairing or replacing manual valve hand wheels; tightening of
'mechanical joints; adjusting pump packing; tightening bolts on brackets, hangers, and cabinets; or replacing missing bolts.
Overall, the team identified no problems where the licensee had not identified and corrected problems or trends in accordance with the corrective action program (i.e., no
'xamples were identified where a PER was required to address a problem and one was not written).
C.
Conclusions The work control process was properly implemented with respect to the requirements of the corrective action progra Gold Cards-15-
~Sco e
The licensee used the gold card process to report human performance issues, both beneficial (proper use) and counterproductive (misuse).
The team selected 248 gold cards from the list of 827 misuse gold. cards issued between January 1, 1997, and June 29, 1998, for review. The selection was made, in part, on the basis of the description provided in the list. The criterion for selection was based. on the premises that misuse gold cards were only to be issued for personnel performance issues.
The team examined the gold card process (and its implementation) to ensure that it had not been used to circumvent or bypass the corrective action program.
Observations and Findin s The team noted that the gold card process was not considered, by the licensee, to be part of the corrective action program.
The quality assurance organization, however, provided oversight and tracking for gold card activities. The process was implemented by a desk-top,instruction that was not controlled by the site procedure control process.
Prior to the inspection, the team obtained a list of gold cards initiated since February 1997.
During this review of recently initiated gold cards discussed below, the team identified cases where the initiation of a corrective action document (e.g., work request, technical evaluation, or problem evaluation request) was required, but not implemented.
Gold Card 6619 Gold Card 6619 was initiated to report a collision of the refueling bridge with a stationary hoist mounted beside the spent fuel pool. According to the gold card, on April 18, 1997, the bridge was being moved when the top of the bridge trolley impacted the top of a jib crane (i.e., new fuel handling crane) that was not positioned correctly. The gold card further stated that an instrument air line (integral to the fuel bridge) was damaged by the collision.
The air line was only slightly damaged due to minor bending and was quickly straightened out with no loss of function identified. Licensee representatives could not provide the team with any knowledge of what evolution wa in progress or why the bridge was being moved at the time of the collision. Personnel directly involved in the incident were no longer employed at the facility. Two licensee representatives stated in different interviews that the most likely evolution in progress at the time of the event was training, but this could'not be substantiate As a result of interviews with licensee personnel and review of Procedure PPM *2.14.1,
"Refueling Bridge Operation," Revisions 15 and 16 (Revision 15 was in effect when the event occurred), the team determined that requirements existed to align jib cranes to avoid collision before moving the fuel bridge. Two places in the procedure, Steps 4.10 and 5.17, specifically cautioned operators to assure adequate clearance between the fuel bridge and other components on the refueling floor during movement of the fuel bridge.
The collision of the fuel bridge with the jib crane occurred because personnel operating the bridge did not comply with the safety-related procedure that controlled fuel bridge operation.
Failure to report the procedure noncompliance, as required by the Procedure PPM *1.3.12, was identified as a violation. (See disposition below.)
Gold Card 14881 The team noted that Gold Card 14881 was issued on June 8, 1998, after it was discovered that a chemistry supervisor had signed off Procedure PPM '3.1.1, stating that all required tasks had been completed to support a mode change.
The team was informed, during an interview with the chemistry supervisor, that the procedure was signed off because of poor communications between himself and another employee.
The team found the actions to be more than a personnel performance issue.
This failure to initiate a PER, in accordance with Procedure PPM *1.3.12, was identified as a violation. (See disposition below.)
Audits Self Assessments and Surveillances During the review of audits and self assessments, the team noted that, in two reports, the'ssue of gold cards being used instead of PERs had been identified by the licensee's auditors. The reports were Audit 296-088, 'WNP-2 Corrective Action Program," dated January 31, 1997, and Surveillance Report 297-071, "Documentation of Procedure Noncompliance,"'dated December 5, 1997.
In Audit 296-088, the licensee's auditors reviewed PER 296-0834, which had identified an adverse trend in the improper use of gold cards.
The corrective actions from the PER were to provide training to all initiators and coordinators, not just to an individual department as was done in the PERs that were evaluated to identify the adverse trend.
In Surveillance Report 297-071, the auditors reviewed approximately 7000 gold cards,.
both proper use and misuse, issued between September 1996 and October 1997.
From that sample, they selected approximately 50 for additional review as possibly requiring PERs. That sample was reduced to 12 after further evaluation.
Licensee personnel then initiated PER 297-0883 to document that those 12 gold cards met the guidelines for PER initiation. The team noted, however, that, according to the final PER results, only 2 of the 12 gold cards required PERs to be generate Corrective Actions for the Failure to Initiate PERs During this inspection, the team selected 248 gold cards out of a total of 827 for review.
The team identified 2 gold cards that required PERs, and licensee personnel promptly initiated PERs for those 2 issues.
Neither of these 2 issues had been identified during the audit or surveillance by licensee personnel.
However, the team credited the licensee with identifying the generic issue of the improper use of gold cards, and determined that the licensee had implemented corrective actions for this issue.
The licensee initiated PER 298-1122 on August 25, 1998, to address the failure to write PERs for issues identified on gold cards.
On August 28, 1998, the team was provided with a copy of PER 298-1122, which was complete with resolutions and corrective actions.
From this PER, the team noted that the cause. for this violation was attributed to the failure of management to review or perform a rigorous review of the issued gold cards.
The team found that a lack of a formalized program, a lack of a questioning attitude, and a minimal management oversight contributed to the continuation of improperly using the gold card process and circumventing the corrective action program.
The corrective actions prescribed and taken were the additions of a signature by a manager or supervisor for the review of the gold card issue against PER initiation criteria; the initiation of a more rigorous review of all gold cards by quality programs personnel; and the issuance of an e-mail message to all gold card coordinators to reemphasize the expectation that all gold cards be reviewed by the originator's manager and/or supervisor against the PER initiation criteria.
The team found that the issue of not implementing Procedure PPM 1.3.12, when required, had been identified by the licensee and corrective actions had been implemented to correct this condition adverse to quality. The added actions prescribed in PER 298-1122 further strengthened the previous corrective actions.
The two issues discussed above that involved the licensee's failure to initiate PERs was a violation of Procedure PPPM*1.3.12. This non-repetitive, licensee identified and corrected violation is being treated as a non-cited violation (50-397/9811-03), consistent with Section VII.B.1 of the NRC Enforcement Policy.
Conclusions The gold card process was useful to identify human performance issues.
However, it was occasionally used to improperly include procedural violations and equipment issues.
This was due to a combination of factors, which included a lack of questioning attitude and minimal management involvement. The failure to initiate PERs, when required, circumvented the corrective action program and was, pursuant to Section VII.B.1 of the NRC Enforcement Policy, a non-cited violation of Procedure PPM *1.3.1.
Conclusions The corrective action program at Washington Nuclear Project, Unit 2, generally has been effective. Overall, its implementation has been improved, particularly with respect to the performance of the operations and maintenance organizations.
There were a few exceptions to good performance, which were identified as noncited violations, that indicated a weakness in the ability of the engineering organization to implement the corrective action program.
The evaluations of the NRC information notices was occasionally poor (e.g., incomplete scoping of the issues, excessive reliance on existing barriers, and a lack of thoroughness).
The responses to the PERs was occasionally poor (e.g., inadequate scope, minimal corrective actions, less than timely corrective actions).
6.
Exit Meeting Summary The team presented the inspection results to members of licensee management at the conclusion of the inspection on July 24, 1998.
ln addition, a followup exit was conducted via telephone on September 28, 1998. The licensee's management acknowledged the findings presented.
One manager stated that he did not think that a violation would have been identified for failure to implement the PER program if there was no gold card process.
The team asked the licensee's representatives whether any material examined during the inspection should be considered.
No proprietary information was identifie ATTACHMENT SUPPLEMENTAL INFORMATION PARTIALLIST OF PERSONS CONTACTED Licensee Personnel D. Atkinson, Quality Manager P. Bemis, Vice President, Nuclear Operations D. Coleman, Regulatory Affairs Manager F. Diya, System Engineering Manager D. Feldman, Assistant Operations Manager P. Insera, Licensing Manager D. Kerlee, Acting Quality Programs Manager S. Oxenford, Operations Manager G. Sanford, Maintenance Manager F. Schill, Licensing Engineer G. Smith, Plant General Manager C. Whitcomb, Assistant to the Plant General Manager NRC Personnel S. Boynton, Senior Resident Inspector INSPECTION PROCEDURE USED IP 40500 Effectiveness of Licensee Controls in Identifying, Resolving, and Preventing Problems
~Oened 50-397/9811-01 50-397/9811-02 50-397/9811-03 Closed 50-397/9811-01 50-397/9811-03 ITEMS OPENED AND CLOSED NCY Failure to take prompt corrective action for a condition adverse to quality.
IFI Adequacy of the design of the RCIC barometric condenser vacuum tank level control system.
NCV Failure to initiate PERs.
NCV Failure to take prompt corrective action for a condition adverse to quality.
NCV Failure to initiate PER Procedures-2-DOCUMENTS REVIEWED 297-0671 297-0676 297-0679 297-0680 297-0682 297-0693 297-0694 297-0696 297-0701 297-0714 297-0716 297-0730 297-0734 297-0746 297-0748 297-0779 297-0791 297-0800 297-0811 297-0836 297-0838 297-0860 297-0883 297-0936 297-0440 297-0444 297-0448 297-0463 297-0470 297-0473 297-0490 297-0491 297-0511 297-0551 297-0568 297-0579 297-0588 297-0590 297-0602 297-0605 297-0614 297-0624 297-0625 297-0626 297-0627 297-0641 297-0658 297-0661 MMP-DG1-B101, "Diesel Generator 1 Mechanical Inspection," Revision 3 NOS-30, "Control of Nonconformances and Corrective Actions," Revision 12 Operational Quality Assurance Program Description, Revision 11 OI-09, "Expectations for Supervisory and Peer Oversight," Revision K Ol-23, "Human Performance Improvement Program," Revision B PPM "1.3.12; "Problem Evaluation Request (PER)," Revision 25 PPM *1.3.12A, "Processing of Problem Evaluation Requests (PER)," Revision 7 PPM *1.3.58, "Conduct of Chemistry," Revision 7 PPM "1.3.60, "Verbal Communication Policy," Revision
PPM *1.5.4, "Control of Measuring and Test Equipment-Transfer Standards" PPM *2.14.1, "Refueling Bridge Operation," Revisions 15 and 16 PPM *3.1.6, "Startup Instrument Rack Valve Lineup," Revision 13 PPM *4.601.A4, "601.A4 Annunciator Panel Alarms," Revision 11 PPM "10.1.20, "Vacuum Cleaner Control," Revision 4 QAP-6, "Stop Work Authority," Revision 4 SWP-ASU-01, "Evaluation of Programs, Processes, and Suppliers," Revision
TSP-CRD-C101, "CRD Scram Testing With Auto Scram Timer," Revision 2 Problem Evaluation Re uests 296-0338 297-0208 296-0618
.297-0217 296-0637 297-0219 296-0674 297-0225 296-0806 297-0265 297-0018 297-0275 297-0025 297-0295 297-0027 297-0307 297-0043 297-0311 297-0077 297-0313 297-0091 297-0333 297-0102 297-0352 297-0104 297-0360 297-0105 297-0367 297-0115 297-0371 297-0124 297-0373 297-0130 297-0388 297-0131
'97-0412 297-0141 297-0421 297-0159 297-0426 297-0170 297-0432 297-0176 297-0437 297-0178 297-0438 29?-0184 297-0439 297-0955 297-0964 297-0983 297-0990 297-0997 297-1008 297-1011 297-1012 297-1028 297-1032 297-1048 298-0013 298-0034 298-0046 298-0078 298-0102 298-0107 298-0109 298-0110 298-0114 298-0123 298-0137 298-0147 298-0206
298-0220 298-0223 298-0239 298-0252 298-0265 298-0266 298-0268 298-0274 Work Orders DGR8 14 DYG5 03 DYG5 07 FJZ6 08 FMRO 05 GCL3 03 GCW7 05 GFP8 02 GFP8 03 GFX1 01 298-0276 298-0297 298-0309 298-0333 298-0370 298-0423 298-0425
, 298-0477 GHN301
'KD6
GKJ6 01 GLV3 03 GNL6 01 GNY8 01 GSN1 06 GTM9 07 GWC1 01 GWF3 01-3-298-0495 298-0529 298-0539 298-0542 298-0571
.
298-0598 298-0610 298-0615 GWW5 10 GZR6 08 HBL3 01 HBT4 01 HHY5 01 HHZO 04 HKNO 01 HMW7 03 HPT7 01 HRR3 08 298-0621 298-0633 298-0634 298-0645 298-0661 298-0672 298-0696 HSGS 05 HSH9 02 HSJ8 05 HSK606
'TM103 JBF2 04 JHK9 01 JLB3 02 JLM4 02 JMH7 11 298-0730 298-0753 298-0754 298-0771 298-0913 298-0922 JYT9 01 KFR8 02 KTG8 04 LLD5 01 LMF8 01 LRY1 04 LSSO 01 kkk 97001400 97002161
" 97003832 97004674 Gold Cards 4760 4877 5007 5121 5153 5154 5208 5210 5214 5215 5269 5276 5300, 5447 5487 5500 551?
5518 5519 5554 5556 5579 5654 5669 5715 5784 5963 6026
. 6158 6237 6366 6367 6408 6412 6484 98000343 98000437 98001107 6511 6593 6601 6612
~ 6619 6651 6693 6694 6695 6723 6872 6878 6881 6887 6908 6955 7092 7141 98001257 98001666 98001684 7215 7236 7507 7510 7518 7630 7637 7647 7683 7734 8000 8066 8096 8097 8100 8119 8166 8291 8316 8500 8549 8560 8562 8620 8839 9001 9002 9003 9004 9006 SOOT 9122 9180 9191 9196 9241 98002071 98004099 9314 9399 9401 9402 9403 9405 9407 9422 9423 9424 9429 9431 9432 9433 9434 9436 9504 9562
0
9571 9579 10210 10289 0290 0318 0335 0336 0339 0342 0344 0347 0351 0354 0355 0421 0615 0616 0700 0701 0703 0722 0723 0724 0725 0726 9582 9704
9708
9776
9778 9779
9780
9781 9782
9783
9784
9837
9838
9888 9983
10042
10139
10140
10141
.
10143
10144
10145
Self Assessments 10727 10728 10729 10730
.
10731, 10742 10743 10745 10749 10750 10766 10768 10769 10803 10835 10836 10837 10838 10839 10965 11172 11183 11202 11283-4-11299 11331 11426 11427 11428 11432 11444 11446 11458 11460 11464 11465 11467 11526 11528 11534 1'I545 11562 11596 11660 11667 11674 11676 11677 11678 11679 11738 11741 11780 11800 11803 11828 11830 11856 12003 12245 12280 12304 12370 12371 12734 12926 12928 12981 13019 13079 13082 13086 13096 13463 13589 13706 13935 13936 13941 13942
'13997 14038 14164 14324 14521 14626 14746 14881 15014 15053 15056 19942 297-065 PSO Audits 296-088 297-050 297-081 298-004 Quality Recommendations, December 8, 1997 December 22, 1997 WNP-2 Corrective Action Program, January 31, 1997 WNP-2 Corrective Action Audit, August 29, 1997 Joint UtilityManagement Audit (JUMA) WNP-2 Quality Assurance Program, December 15, 1997 WNP-2 Corrective Action Audit, March 10, 1998 Evaluation of Generic Communications Information Notices Information Notice 87-010, Supplement
Information Notice 91-085-R1 Information Notice 91-50, Supplement
Information Notice 97-14 Information Notice 97-16 Information Notice 97-27 Information Notice 97-38 Information Notice 97-41 Information Notice 97-78
Information Notice 97-90-5-Other Documents Calculation EQ-02-97-01, "Evaluation of Active Failure of RCIC-LS-11 in Seismic Event,"
Revision 0 Calculation NE-02-93-09, "RCIC Leak Through Oil Cooler Pipe Out RCIC-RV-33," Revision 0 Calculation NE-02-93-10, "Post-LOCA Flooding of RCIC and CRD/Condensate Pump Rooms,"
Revision 0 Calculation Modification Record 94-0628 for Calculation NE-02-93-10, Revision 0 CCER No. C97-0027, "Component CER Summary Sheets for RCIC-LS-11," Revision
CGER No. C97-0086, "Component CER Summary Sheets for RCIC-RLY-K22 and RCIC-RLY-K75,"Revision 0 Electrical Wiring Diagram EWD-6E-004, "Electrical Wiring Diagram Reactor Core Isolation Cooling System Pump RCIC-P-4 (E51-C004)," Revision 10 General Electric Elementary Diagram 807E173TC, "RCIC Sys," Revisions 16 and 17 NUREG 0892, WNP-2 Safety Evaluation Report, Section 5.4.3, "Reactor Water Cleanup System"
~ Regulatory Guide 1.26, Revision 3 Safety Evaluation for Reclassification of RWCU System, Transmittal NO. 14644, May 15, 1995 Safety Evaluation Screening for Revision of Station Blackout (SBO) Procedure PPM 5.6.1, February 23, 1998 Technical Memorandum TM-2050, "Classification of Design Base Functions and Primary Containment Isolation Boundaries of the RCIC System," Revision
WNP-'? FSAR, Sections:
3.6.1.15.2, Postulated Ruptures of Reactor Water Cleanup System 5.4.6, Reactor Core Isolation Cooling System 5.4.8, Reactor Water Cleanup System 7.6.1.3, Leak Detection System - Instrumentation and Control 15.2.6, Loss of AC Power
)