IR 05000272/1991026

Safety Insp Repts 50-272/91-26,50-311/91-26 & 50-354/91-19 on 910911-1022.No Violations Noted.Major Areas Inpsected: Operations,Radiological Controls,Maint & Surveillance Testing,Ep,Security & Engineering Technical Support

ML18096A372
Person / Time
Site:	Salem, Hope Creek
Issue date:	11/22/1991
From:	Jason White NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I)
To:
Shared Package
ML18096A371	List: IR 05000272/1991026 ML18096A370
References
50-272-91-26, 50-311-91-26, 50-354-91-19, NUDOCS 9112110041
Download: ML18096A372 (46)

Text

Report No License No Licensee:

F aci Ii ties:

Dates:

Inspectors:

Approved:

U. S. NUCLEAR REGULATORY COMMISSION

REGION I

50-272/91-26 50-311/91-26 50-354/91-19 DPR-70 DPR-75 NPF-57 Public Service Electric and Gas Company P.O. Box 236 Hancocks Bridge, New *Jersey 08038 Salem Nuclear Generating ~tation, Units 1 and 2 Hope Creek Nuclear Generating Station September 11 - October 22, 1991 T. P. Johnson, Senior Resident Inspector S. M. Pindale, Resident Inspector S. T. Barr, Resident Inspector H. K. Lathrop, Resident Inspector R. L. Nimitz, Senior Radiation Specialist J. C. Stone, Salem Project Manager, NRR S. Dembek, Hope Creek Project Manager, NRR R. J. Urban, Project Engineer*

/A/

- -

~*

L,N. R. W ~Chief

~Reactor Projects Section 2A Areas Inspected: Routine resident safety inspection of the following areas: operations, radiological controls, maintenance and surveillance testing, emergency preparedness, security, engineering technical support, safety assessment/quality verification, and licensee event reports and open item followu Results: The inspectors identified two non-cited violations for Hope Creek. In addition, the inspectors identified four unresolved items, three for Salem and one for Hope Creek. An executive summary follows.

9112110041 911126 PDR ADOCK 05000272 G

PDR

OPERATIONS EXECUTIVE SUMMARY Salem Inspection Reports 50-272/91-26; 50-311/91-26 Hope Creek Inspection Report 50-354/91-19 September 11 - October 22, 1991 Salem: The Salem uriits were operated in a safe manner. Operator response to events demonstrated excellent procedure compliance and communication and effective supervisor oversight. Radiation monitoring and engineered safety feature system actuations were reported, and licensee actions were appropriate. A Unit 1 forced outage and several Unit 2 power reductions demonstrated a conservative and safe approach to unit operations. Also, licensee outage planning activities for these power reductions and the forced outage were effective. Poor control of fire protection troubleshooting activities resulted in the loss of safety related ventilation systems due to charcoal deluge system actuation. The licensee convened a Significant Event Response Team to investigate the event, and the implementation of its recommended corrective actions is unresolve Hope Creek: The unit was operated in a safe manner. A detailed walkdown of the Standby Liquid Control (SLC) System determined that the system was properly lined up and operable.

. A programmatic weakness was noted in the control of procedure changes (unresolved);

procedures themselves were well written and usable. An unplanned Technical Specification 3.0.3 entry was made when both of the primary containment instrument gas compressors were simultaneously out of service, rendering both trains of the main steam isolation valve sealing system inoperable. Licensee actions were appropriate. The coordination associated with the restart of a reactor water cleanup pump was effective and demonstrated safe plant operation. The licensee is addressing the confusing wording of the technical specification on the safety auxiliaries cooling syste RADIOLOGICAL CONTROLS Salem: Periodic inspector observation of station workers and Radiation Protection personnel implementation of radiological controls and protection program requirements did not identify any deficiencies. The licensee appropriately responded to a heat exhaustion even * Continuing followup regarding onsite storage of ammonia resulted in an unresolved item remammg ope Hope Creek: The inspector, through periodic observation of station workers and Radiation Protection personnel implementing radiological controls and protection program requirements, noted several positive attributes. However, two observations (an unposted radioactive material storage area and improper personnel dosimetry) resulted in NRC identified non-cited violations. The "B" floor drain sample tank was inadvertently released to the Delaware River due to personnel error. The radiological consequences of this release were insignificant and the licensee's followup was appropriate.

MAINTENANCE/SURVEILLANCE Salem: Several pressurizer power operated relief valve (PORV) failures resulted from multiple causes, including insufficient vendor and procedure guidance, inadequate administrative controls, and historical lack of aggressive management attention. Also, the.

safety classification of the PORV is unresolved. Event followup investigation continue The licensee response to a slow main steam isolation valv~ closure time was appropriat Two safety injection pump testing operi items were close Hope Creek: Minor-post maintenance material deficiencies associated with the standby liquid control system were identified in routine observations. A pressure perturbation caused by post accident sampling system valve stroking resulted in a half-scram and a full recirculation runback to 60% power. Appropriate corrective actions were take EMERGENCY PREPAREDNESS No noteworthy findings were identifie SECURITY No noteworthy findings were identifie ENGINEERING/TECHNICAL SUPPORT

• Salem: Review of the management of engineering work activities determined that they were being performed in accordance with applicable procedures and were being properly prioritized and executed. The licensee responded appropriately to a _lOCFR 21 report regarding one of their auxiliary feedwater pumps. An open item regarding the main steam vent valve panels was close Hope Creek: Review of the management of engineering work activities determined that they were being performed in accordance with applicable procedures and were being properly prioritized and executed. A spurious start of a standby filtration, recirculation, ventilation system fan was due to the failure to modify a system operating procedure to drain instrument lines. This was another example of a programmatic weakness in the licensee's procedure change control process, as noted previously in the SLC system walkdown. Good safety perspective was noted in the licensee's decision to not implement a design change while at power, due to the scram risk. The core spray pump motor seismic design was reviewed and

• concluded to be acceptabl SAFETY ASSESSMENT/QUALITY VERIFICATION Salem: Although the Station Operation Review Committee properly addressed causal factors regarding pressurizer power operated relief valve failures, the initial root cause determination was not completely characterized. After discussions with management, the root cause was determined to be unknown at this time. Station management displayed a conservative approach to unit operations.

.

-

Hope Creek: Self-assessment initiatives by onsite Quality Assurance and Safety Review were performed well and were effective in ass_essing performance, determining root causes, and recommending corrective actions. An engineered safety feature actuation notification

• was withdrawn following a licensee review of reporting requirements. -The licensee's interpretation was acceptabl Common: Salem and Hope Creek General Manager "state of _the station" meetings were effective in communicating the licensee's self-assessment of station performance to the workers.* A review of the 10CFR50.59 Safety Evaluation Program identified specific deficien_cies at both Salem and Hope Creek; these were appropriately responded to by the licensee. However, one evaluation, concerning the Salem BF-19 valves, is unresolve Licensee controls for items suspended in the unit's spent fuel pools were appropriat SUMMARY OF OPERATIONS Salem Units 1 and 2 Unit 1 operated at power during this period except for a forced outage (September 16-25, 1991) to repair an electrohydraulic control system lea Unit 2 operated at or near full power except for power reductions on September 22, 1991, and October 18-20, 1991, due to steam generator chemistry problems and again on October 11, 1991. due to ventilation system Technical Specification complianc.2 Hope Creek The unit operated at or near full power during the period, except for an unplanned power reduction during surveillance testing on September 23, 199.

OPERATIONS Inspection Activities The inspectors verified that the facilities were operated safely and in conformance with regulatory requirements. Public Service Electric and Gas (PSE&G) Company management control was evaluated by direct observation of activities, tours of the facilities, interviews and discussions with personnel, independent verification of safety system status and Technica Specification compliance, and review of facility records. These inspection activities were conducted in accordance with NRC inspection procedures 60705, 71707, 71710, and 9370 The inspectors performed normal and back-shift inspections, -including 21 hours2.430556e-4 days <br />0.00583 hours <br />3.472222e-5 weeks <br />7.9905e-6 months <br /> of deep back-shift inspection.2 Inspection Findings and Significant Plant Events 2.2.1 Salem Unit 1 Forced Outage At 7:45 a.m. on Monday, September 16, 1991, the licensee initiated a shutdown of Salem Unit 1 due to an unisolable electrohydraulic control (EHC) system fluid leak. The small leak had been observed for a few days and was being monitored. The licensee had previously decided to shut the unit down beginning Tuesday evening (September 17). However, on Sunday (September 15), the EHC leak increased to approximately 35 liters per hour. Station management subsequently made the decision to immediately shut down the uni The unit initially proceeded to Hot Standby (Mode 3). However, a body to bonnet leak was observed on one of the two pressurizer spray valves (lPS 1). This required entry into Cold Shutdown (Mode 5) for valve repair. The leaks on the EHC system and lPSl were repaired, and other corrective and preventive maintenance activities were complete *

The unit was restarted on September 25. Main steam isolation valve testing (see Section 4.3.1.D) delayed power ascension testing. The unit was synchronized to the grid on September 27, 199 The inspector observed the unit shutdown, the forced outage maintenance activities, and unit restart activities. Operator procedure compliance and communication were noted as excellent. Shift supervisor oversight was effective in providing the necessary command and control of operational activities. The inspector witnessed the reactor startup and criticality, which was achieved at 5: 16 p.m. on September 25. The inspector also concluded that licensee station and operations management demonstrated an excellent safety perspective related to their decision to immediately shut the unit down due to an increased EHC leak rate to perform equipment repair Unit 2 Power Reductions Due to Main Condenser Leak At 12:10 a.m. on September 22, 1991, operators on Unit 2 entered abnormal operating procedures (AOP) due to a main condenser hotwell high conductivity alarm. Apparently, a previously plugged condenser tube failed, resulting in river water intrusion into the condenser hotwe!l. Steam generator conductivity peaked at 30 micromhos per centimeter. Per the AOPs, unit power was reduced to 20%. Both the 23A and 23B circulators were removed from service and inspected. The failed condenser plug was replaced. The unit remained at 48 % power for 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> to recover from accumulated axial flux difference (AFD) penalty minutes, as required by Technical Specification The inspector reviewed licensee actions and confirmed proper AOP implementatio Licensee compliance with Technical Specification requirements regarding the AFD was verified to be correc Subsequently on October 18, 1991, management decided to begin a Unit 2 power reduction from 100% in order to remove chloride ions from the steam generators. The turbine generator was taken off-line, and a chemical hideout recovery evolution was performed. The licensee initiated this hideout recovery evolution because of a vendor calculation that cone! uded the chloride crevice concentration was such that the steam generator tubes were subject to an accelerated denting over time. During this power reduction, the unit remained in Mode 2 at 0% power. Chloride ion concentration increased from 1 ppb to 110 pp After a 12 hour1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> soak, chloride concentration was increased to about 15 % and maximum blowdown was initiated on all steam generators to remove these chloride ions from solutio Final chloride concentration on all steam generators was within their chemistry goal ( < 20 ppb). Upon completion of these evolutions, the unit was synchronized with the grid at 8: 12 p.m. on October 20, 1991. The unit was subsequently returned to full powe The inspector reviewed licensee power reduction and outage plans and associated implementing procedures. This included core reactivity management calculations and plan The inspector also attended selected pre-outage meetings and discussed these plans with operations, outage, chemistry, reactor engineering, and management personnel. The inspector conclude_d that licensee plans were appropriate and appeared proactive. Also, the

licensee demonstrated a conservative approach to safe unit operations and attention to secondary chemistr Unit 2 Power Reduction Due to Loss of Ventilation Charcoal Filters On the morning of October 10, 1991, Nuclear Site Fire Protection Department (NFP)

personnel were performing surveillance procedure MlO-SST-045-2, "Functional and

. Circuitry Test of Thermister Heat Detectors for Iodine Removal and Containment Pressure Relief Charcoal Filter When tested, however, the No. 21 Iodine Removal Unit (IRU) fire protection deluge valve failed to operate as required. The Operating Shift was informed, the IRU deluge system declared inoperable, and the appropriate Technical Specification (TS 3. 7.10. 2. b.1) was followed. NFP initiated a work request, and troubleshooting of the affected system was undertaken in accordance with procedure SC.IC-GP.ZZ-0006(Q),

11 Salem Generating Station Control Equipment - Troubleshooting."

Troubleshooting continued into the morning of October 11, when a relay in the IRU deluge system circuit was identified as the potential cause of the affected valve not trippirig. In order to test that relay, the contact that supplies power to the relay was jumpered, and the appropriat~ breaker was closed. When the breaker was closed at 11:42 a.m. on October 11, the fire protection deluge system was actuated for the charcoal filters of the following systems: the auxiliary building ventilation system, the. control room emergency air conditioning system, the containment pressure relief system, and the fuel handling building ventilation system. The charcoal filters for the iodine removal system were not affected because_ the suspected relay was indeed* faulty, and the deluge valve for the IRU s was inoperable. The charcoal beds of the other four systems were severely wetted, and the systems were subsequently declared inoperabl The declaration of inoperability of the auxiliary building ventilation and control room emergency air conditioning systems at 12:04 p.m. placed the unit in two concurrent 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> Technical Specification Action Statements (TSAS 3. 7. 7 and TSAS 3. 7.6, respectively). The containment pressure relief and the fuel handling building ventilation systems do not have any associated TSASs unless fuel movement or activities are in progres Once in the TSASs, the Salem Operations, Technical, Maintenance, and Chemistry/Radiation Protection Departments began to take the necessary measures required to replace the affected charcoal beds and retest the ventilation systems to verify operability.. Concurrently, Salem Station management and PSE&G Nuclear Licensing began to prepare a request for NRC Region I to approve a TS waiver of compliance. The purpose of the waiver was to extend the 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> TSAS to complete testing of replaced filtration system The I icensee performed charcoal bed replacements throughout the evening and completed the work early on the morning of October 12; followed by operability testing of the auxiliary building ventilation a*nd control room emergency air conditioning systems. Uncertain that the testing would be satisfactorily completed prior to the expiration of the 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> TSAS, PSE&G submitted their waiver request to Region I for approval. PSE&G discussions with the NRC were continuing when plant management decided to begin a slow plant shutdown while the waiver was being deliberated. The TSAS was to expire at 12:04 p.m. on October

• 12, and the shutdown was initiated at 11:26 a.m. Shortly after noon, all required testing of the two affected ventilation systems was satisfactorily completed, the systems declared operable, and PSE&G withdrew their request for the TS waiver of compliance. The shutdown was halted at 12:36 p.m. with plant power reduced to just less than 90 percent power. Subsequently, the plant was returned to full powe * *

The resident inspector w~s present at the site when the inadvertent deluge occurred on October 11, and tot.ired the auxiliary building to assess the effects of the event. The resident noted that the licensee had taken effective and timely action to isolate the affected portions of the fire protection system, clean up the areas that had been sprayed; and post the required fire watches to compensate for the inoperable deluge system. The inspector also followed the licensee's preparation of the TS waiver of compliance as well as the plant management's waiver approval process. Further, the inspector witnessed the completion of the charcoal replacement and operability testin * *

Through observations and discussions with licensee personnel, the inspector concluded that the event was a result of poor maintenance control-and personnel error resulting from a lack of proper training. The NFP technicians involved in troubleshooting the deluge system inadequately implemented procedure. SC.IC-GP.ZZ-0006(Q); the individuals did not maintain an accurate account of the performance of troubleshooting activities as required by the procedure and did not properly account for thejumpers that were placed in the actuation circuit. Further, the individuals were not sufficiently knowledgeable of the circuits they altered and failed to realize that one of the relays that was jumpered supplied power to actuation relays other than those for the IRUs. Consequently, the deluge system was actuated when the power supply breaker was close Poor maintenance control was demonstrated in that the operating shift was not properly advised when the troubleshooting plan changed and was unaware of the possible implications of the maintenance work being performed. Additionally, the adequacy of the indications in the control room for actuations of fire protection system may require further evaluation by the license The General Manager - Salem Operations convened a Significant Event Response Team (SERT) to investigate and provide an independent review of the event and its causes. A draft copy of the SERT's report was reviewed by the inspector prior to the end of this report period. The inspector found the report to be thorough; the report identified all of the inspector's concerns and contained several short term and long term corrective action recommendations. Recommended corrective actions included additional training for the Site Services Department and an upgrade of the work standards for the Site Services Department, and changes to the Fire Protection System actuation system and its indications in the control room The inspector also noted that, despite the weaknesses that led to it, the licensee's response and actions were very appropriate. The work on the affected ventilation systems and their retests, as well as the preparation of the TS waiver request, were conducted in a timely and conservative manner and displayed good work and cooperation by the various departments and personnel at the Salem Station. Pending the acceptance and implementation of the

SERT's corrective actions, however, PSE&G's followup of this event will remain an unresolved item (UNR 50-272/311;91-26,.Ql). Unit 1 Main Steam Isolation During Startup On September 23, 1991, during reactor heatup in Mode 4 (Hot Shutdown), a main steamline isolation (MSI) occurred at Unit 1 due to a spurious high steamline flow coincident with a low steamline pressure (600 psig) or a low average temperature (543 degrees F) signal. The residual heat removal system was in service for decay heat removal. When the unit is in Mode 4, the low steamline pressure and low average temperature bistables are in a tripped condition. The high steamline flow logic requires one out of two channels to trip per steam generator (SG) in two of the four SGs to satisfy the trip coincidenc Prior to the event, one high steamline flow channel for the No. 14 SG was in the tripped condition for functional surveillance testing. Both high steamline flow channels of the N SG subsequently actuated spuriously, thereby satisfying the MSI trip coincidence. Only the four main steamline drain valves closed as a result of the actuation; the main steam isolation valves (MSIVs) and MSIV bypass valves were already clo.sed. The licensee reported this event to the NRC per 10 CFR 50. 72 reporting requirements. Similar MSis have occurred previously while the unit was in Mode 4 (June 3, 1990 and August 12, 1990).

The licensee had evaluated the previous Mode 4 MSis, and attributed the cause of the actuations to be condensation buildup in the main steam lines. The licensee is continuing an assessment of the existing main steamline flow measurement design (sensing lines).

The inspector reviewed the licensee's current plans to resolve this issue and verified that further evaluation is planned for consideration of main steam flow measuring modificatio The inspector concluded that the licensee's response to this event was appropriate and had no further questions at this tim Radiation Monitor Engineered Safety Feature (ESF) Actuation The following ESF actuation occurred and was reported by the licensee during the period:

Unit Radiation Monitor Date

2Rl2B October 10, 1991 1:38 Systems responded as designed, causing a contaj.nment ventilation isolation. As stated in previous LERs and management meetings, licensee actions include short term and long term equipment upgrades. The inspector reviewed licensee actions regarding this event. The licensee intends to submit an LER for this event. No unacceptable conditions were noted..

2.2.2 Hope Creek Engineered Safety Feature (ESF) System Walkdown The inspector independently verified the operability of the Standby Liquid Control System (SLC) by performing a walkdown of all accessible portions of the system to confirm that the system component lineup and procedures matched plant drawings and the as-built configuration and to identify equipment conditions which could degrade performanc System operating, surveillance and maintenance procedures were reviewed for technical accuracy, ease of performance, and completeness. This inspection was conducted in accordance with NRC inspection procedure 7171 The physical walkdown of the system was performed on October 17, 1991, one week after

_maintenance had, been performed on the "B" SLC subsystem. The inspector determined that the system was properly aligned for Mode 1 operation, matching the positions required by the computer-generated (TRIS) line-up sheets and the piping and instrumentation diagram (P&ID). Required locking devices and pipe end caps were properly installe Overall material condition appeared good. However, there were a number of discrepancies which indicated that the post-maintenance recovery had not been completely effective. For example, the inspector noted pieces of debris in the "B" pump drip pan (tape, wire), on top of two electrical junction boxes on the north wall facing the storage tank (tape, paper scraps),

and under the pump suction piping attached to the storage tank (insulation, wire, tape).

Additionally, the protective cover for the "B" pump crankcase breather was missing, and packing leaks were noted on valves BH-V063 and BH-V06 *

• The inspector reviewed 16 SLC operating, surveillance, and maintenance procedures to assess their completeness, technical accuracy, and ease of performance. In general, the inspector found the procedures to be dear and concise, yet adequately detailed so less experienced personnel should be able to correctly perform the intended evolution. Pictorial aids in the SLC pump overhaul procedure (MD-CM.BH-001) were clear and detaile Although some minor editorial discrepancies were noted, no errors with any safety significance were foun The inspector noted that an on-the-:spot change (OTSC) to procedure HC.OP-IS.BH-0001,

"Standby Liquid Control Pump AP208-Inservice Test," apparently had not been incorporated in a number of similarly affected procedures. When questioned on this issue, operations management indicated that the procedure writers group would incorporate the OTSC in companion procedures with alphabetically related components within two weeks. OTSC N l lA to HC.OP-IS.BH-0001 (SLC Pump AP208 Inservice Test) was issued on September 23, 1991; as of October 22, 1991, a similar change to companion procedure HC.OP-IS.BH-0002 (SLC Pump BP208 Inservice Test) had not been made. The same was true for several other SLC procedures, including the SLC system operating procedure, OP-SO.BH-001. The inspector reviewed the administrative procedure governing procedure development and change, NC.NA-AP.ZZ-0032, and determined that it did not adequately address incorporation of changes into companion or related procedures. Based on the above, the inspector concluded that a programmatic weakness existed in that there was apparently no

.

-

process to assure that necessary changes would be incorporated in a:n procedures affected by an OTSC for a specific procedure. This issue is an unresolved item (UNR 50-354/91-19-01).

The results of the reviews and walkdown were discussed with licensee management, who -

_ immediately initiated action to correct the material deficiencies noted by the inspector and investigate the procedure issues. These efforts were ongoing when the report period close Operations management will document their corrective actions and results when complete Service Water Loop Outage On October 10, 1991, at 3:15 a.m., the licensee removed the "i3" service water (SW) pump from service to repair its discharge valve and strainer. In order to properly isolate the pump, the _entire "B" SW loop was tagged. This included both the "B" and "D" SW pumps and placed the licensee in the following Technical Specification (TS) Action Statements (TSAS):

TS 3. 7.1.1.a.2, Safety Auxiliaries Cooling System (SACS),

TS 3. 7.1.2.a.3, SW,

-TS 3. 7.3.a, Control Room Emergency Filtration System, TS 3.6.2.2.a, Suppression Pool Spray, and TS 3.6.2.3.a, Suppression Pool Cooling The-most limiting of these TSASs was 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />. TS 3.7.1.1.a.2 requires that SACS be realigned for the affected emergency diesel generator (EDG) coolers. However, the licensee was unable to realign the "B" EDG due to a leaking SACS isolation valve. Therefore, the

"B" EDG was declared inoperable and TS 3.8.1.1.b was entered. This was also a 72 hour8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> TSA During the morning control room tour, on October 16, 1991, the inspector questioned the licensee's interpretation of these TSASs. The inspector determined that TS 3.7.1.1. (regarding SACS) is somewhat confusing due to multiple commas and "and/or" usage. As a result of this wording, one could conclude that rio inability to realign both EDGs to the other SA CS loop implies a 12 hour1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> TSAS. However, this is inconsistent with TS 3. 8. regarding the in operability of a single EDG. The inspector discussed this item with plant management, NRC headquarters, and regional personnel and concluded that a 72 hour8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> TSAS was appropriate in this case. However, the licensee agreed to submit a TS amendment change to improve and clarify the wording of TS 3. 7.1.1.a.2. In addition, the licensee stated that it was their intention to replace the affected SW and SACS valves with a modified "soft seat" butterfly valve design in order to improve system reliabilit Reactor Water Cleanup (RWCU) Pump Start On October 11, 1991, the inspector witnessed the starting of a RWCU pump in the reactor building. The evolution was performed in_ accordance with system operating procedures, and supervision was present in the field. Communications, both face-to-face and between the field and the control room, were proper. Radiation protection technician coverage was

-*

effective (see Section 3.2.2.A). Overall, the evolution demonstrated the licensee's ability to safely and effectively operate the uni * Open Item Followup (Closed) Unresolved Item (50-*J54/90-01-02). Control of Systems Alignments During Mairitenance. A post scram review following the January 6, 1990, reactor scram and turbine trip on high moisture separator level determined that one of three normal drain level controllers was out of service due to an isolated air supply valve. Apparently, post maintenance restoration activities failed to open the air supply valve. The licensee committed to reviewing this event and restoration issues with maintenance and operations personnel (see Licen*see Event Report 90-01), No further problems were noted with system restorations during recent outages (forced, refueling and scheduled system).

The licensee has reviewed and updated selected system alignment checklists (TRIS). The inspector verified implementation_ of these corrective actions and discussed them with the appropriate licensee personnel. The inspector had no further questions regarding this issue, and the unresolved item is considered close Unplanned Entry Into Technical Specification (TS) 3. At 9:15 a.m., on September 6, 1991, the "A" primary containment instrument gas (PCIG)

air compressor was out of service for maintenance (filter replacement) when the running "B" PCIG air compressor tripped. Since PCIG is required to support operability of the main steam isolation valve sealing system (MSIVSS), the licensee declared both MSIVSSs inoperable, a condition not addressed in TS 3.6.1.4; consequently, TS 3.0.3 was entere The "B" PCIG compressor could not be.immediately restarted. However, the "A" PCIG compressor was returned to operable status and started 18. minutes later, and TS 3.6. Action Statement for one MSIYSS inoperable was satisfie The inspector reviewed this event with.operations and licensing personnel and concluded that *

the licensee's actions were appropriate and in conformance with the applicable TS Action Statement.

RADIOLOGICAL CONTROLS Inspection Activities PSE&G 's conformance with the radiological protection program was verified on a periodic basis by the NRC resident staff. These inspection activities were conducted in accordance with NRC inspection procedures 71707 and 93702. In addition, a regional specialist inspector reviewed selected activities in accordance with NRC inspection procedure 83750.

• 9 Inspection Findings and Significant Plant Events 3.2.1 Salem Heat Exhaustion Event in Unit 1 Containment During Outage Oi1 September 19, 1991, while Salem Unit 1 wasin Hot Standby (Mode 3) during the forced maintenance outage, workers entered the containment structure in order to prepare a pressurizer spray valve (lPSl) for repair work. At the time, the unit had been shut down for less than three days, and the containment environment was still warm, as evidenced by a wet bulb globe temperature of 94 F degrees in the vicinity of the pressurizer. As a result of the high temperatures, one of the workers began to suffer from symptoms of heat exhaustion and had to leave the containment structur The NRC resident inspector learned of the event shortly after it occurred and investigated the licensee's control of the work being performed inside containment. The inspector determined that plant management was aware of the one employee that had been overcome by the heat and was evaluating the matter when contacted by the NRC. The licensee subsequently decided to halt work inside the containment structure until the plant reached Cold Shutdown (Mode 5).

The affected worker was taken to the site medical facility, examined, and sent home. He missed no work time as a result of the event. As part of the investigation, the resident inspector verified that the licensee's procedures for working in a stressful environment had been in place at the time of the event. The inspector determined that the Salem Radiation Protection group had properly implemented PSE&G Nuclear Department Safety Directive M 10-ISD-O 15, 11 Occupational Heat Stress, 11 prior to and throughout the work that was being performed and that the plant management's decision to halt work subsequent to the one employee being overcome was pruden Update on Effect of Onsite Ammonia Storage on Control Room Habitability (Open) Unresolved Item (50-272/91-25-02). During inspections conducted in August and September 1991, NRC inspectors identified a concern with the control of ammonia at the Salem Station, especially with respect to control room habitability. As a result of that concern, the Salem Chemistry group revised the procedure for the receipt of ammonium hydroxide to minimize the possibility of inadvertent exposure of personnel to ammonia vapor. The revision of Salem Chemistry procedure SC.CH-AD.ZZ-0470(Z), "Receipt of Bulk Chemical Tank Trucks, 11 included the following changes:

Provision of sufficient workers required for deliveries as specified by the Chemistry Supervisor, Receipt of ammonium hydroxide deliveries only during backshifts, Notification of Safety and Fire Protection of ammonium hydroxide deliveries,

Provision of diagrams for the roping off of Unit 1 and Unit 2 Turbine Buildings, Having the Control Room ventilation system placed in "Accident-Inside Air" and certain Turbine Building supply and exhaust fans shut off, Requirement for ammonium hydroxide delivery trucks to have a pump, and Requirement for all east side Turbine Building doors to be closed during ammonium hydroxide deliverie The first ammonia (ammonium hydroxide) delivery after the revision of this procedure occurred on September 18, 1991, and the resident inspector was present to witness the implementation of the new procedure. The inspector noted that the ammonia hydroxide transfer was conducted in a controlled manner, the procedure was adhered to, and the procedure revisions adequately prevented the inadvertent exposure of personnel to ammonia vapor Also as a result of the identified NRC concerns, PSE&G had committed to performing a preliminary safety assessment of onsite ammonia storage and its potential effect on control room habitability. This preliminary assessment was conducted by PSE&G Nuclear Engineering and Plant Betterment and provided to the resident inspector on September 19, 1991. The evaluation concluded that if a postulated accident occurred at the Salem ammonia storage vessel, habitability of the Unit 1 or Unit 2 control rooms would not be affected. The inspector reviewed the licensee's report, the approach taken, and assumptions made, and identified no deficiencies in PSE&G's bases for arriving at this conclusion. The inspector concluded that PSE&G was making good progress in the resolution of this unresolved ite Also discussed as part of this inspection was PSE&G's commitment to provide, within 30 days of Inspection Report 50-272 and 311/91-25, a summary of other toxic chemicals stored on site and a discussion of planned actions PSE&G would take relative to their impact on control room habitability. After further discussion with the NRC specialist inspector and the Region I Office, it was agreed that PSE&G will supply this list to the NRC prior to November 29, 1991. This item remains open pending the submission of this list, the completion of a final 10 CFR 50.59 safety evaluation for control room habitability relative to ammonia storage, and any required Final Safety Analysis Report revision, as discussed in NRC Inspection Report 50-272 and 311/91-2.2.2 Hope Creek Routine Observations The inspector's reviews of the radiological controls program included the following aspects:

Radioactive and contaminated material control,

.

Posting barricading and access control (as appropriate) to Radiation, High Radiation, and Airborne Radioactivity Areas, Adherence to radiation work permits,

..

Provision and proper use of dosimetry devices, and Radiological controls performance and initiative Within the scope of these reviews, the following positive attributes were noted:

The station exhibited a clean and orderly appearance. The licensee was performing extensive painting and_ cleaning effort.

The licensee assigned a Radiation Protection Supervisor to track and monitor corrective action on leaks (e.g. valve leaks). Licensee personnel indicated only about a dozen leaks were present. Catch bags and drip bags were used to minimize spread of contamination. Station contaminated area was noted as being 7%~

The licensee initiated a number of efforts to reduce personnel exposure, including a *

controlled shutdown during the last outage to minimize crud bursts, use of video cameras for collection and observation of routine activities in station Radiation Areas, and modification of operator round schedules to take, advantage of normal routine power reductions resulting in less operator exposur The licensee added two additional channels to the radiation monitoring system to *

monitor radiation dose rates inside and outside the traversing incore probe detector roo The licensee was closely monitoring radiological occurrence reports. Monthly summaries were provided to station manager.

Technician coverage during a Reactor Water Cleanup Pump start was appropriate in ensuring that proper radiation protection measures were take *

The following non-cited violations were identified and reviewed during this inspection period:

Unposted Radioactive Material Storage Area At 2:00 p.m., on September 11, 1991, the inspector noted that the south access point to the refueling floor (elevation 201 feet) was not posted as a Radioactive Material Storage Are The licensee's radiation protection procedure HCRP-TI.ZZ-0204 (Q), "Posting of Signs and Barriers," Revision 0, specifies in section 5.1.8, that each room or area with radioactive material in excess of 10 times the amounts specified in 10 CPR 20 Appendix C shall be posted as "Caution Radioactive Materials". The refueling floor contained quantities of radioactive material which were estimated to be about 100 times the amounts specified in 10 CFR 20 Appendix C value The licensee immediately re-posted the access point. The licensee postulated that during painting the posting was removed: The licensee's Senior Radiation Protection Supervisor*

reviewed the posting procedure with all appropriate radiation protection personnel on September 12, 1991. In addition, all other access points to the refueling floor were verified

• to be properly posted. The licensee also reduced the area of responsibility of radiation protection supervisors to allow for more in-depth review of smaller areas of the facilit The inspector concluded that the lack of posting of the south access door was a violation of Technical Specification 6.1 L However, since this matter had minor safety significance and was corrected by the licensee (including actions to prevent recurrence), in accordance with 10 CPR 2, Appendix C,Section V.A, a Notice of Violation will not be issue Worker Inside RCA Without Proper Dosimetry At about 5:30 p.m., on September 12, 1991, the inspector observed a worker inside the radiological control area (RCA) (102 foot elevation service radwaste hallway) without an integrating pocket dosimeter. The individual was wearing his personnel dosimetry (thermoluminescent dosimeter). The licensee's radiation protection procedure NC.NA-AP.ZZ-0024, Revision 1, "Radiation Protection Program," specifies in section 3.9 and Attachment 1 that it was the responsibility of each individual to wear a self reading pocket dosimeter or ALNOR (an integrating alarming dosimeter) as required by radiation protectio The individual immediately left the RCA, was counseled, and a radiological occurrence report was generated. The licensee also performed a dose assessment and concluded that the individual had been in the RCA for only 15 minutes and had not entered any significant radiation dose rate areas (greater than 0.2 mR/hr). Operations personnel were informed of the incident and were reminded of the need of greater attention to detai The inspector concluded that failure to wear an ALNOR was a violation of Technical Specification 6.11. However, since this matter had minor safety significance and was immediately corrected in accordance with 10 CPR 2, Appendix C,Section V.A, a Notice of Violation will not be issue Inadvertent Release of the "B" Floor Drain Waste Sample Tank (FDWST) to River On October 11, 1991, radwaste operators had recirculated and sampled the "A" PDWST for release to the Delaware River. At 6:56 p.m., operators lined up and commenced pumping the "A" tank to the river. Approximately 30 minutes later, the operator noted that the "A" tank level had not decreased, but the "B" PDWST was decreasing, indicating that the tank contents were being released. The release was terminated and shift management informe The licensee calculated that about 5,200 gallons had been discharged. A sample was obtained from the "B" PDWS The licensee's investigation determined that an incorrect valve lineup had been made due to personnel error. During the performance of the valve lineup the operator inadvertently selected the outlet valve for the "B" tank instead of the "A". The incorrect selection was not noted by the supervisor during his verification of the lineup. Chemistry management discussed this incident with the resident staff and Region I. No notifications were made, as the Emergency Classification Guideline (7F) action level (any liquid release that exceeds Technical Specification limits for 15 or more minutes) was not reached; sample results frorn

the "B" tank showed it to be almost identical to the "A" tank, which had been approved for release and were within Technical Specification limit FDWST Permit N Date A

910291-L 10/11/91 B

910292-L 10/11/91 B

910294-L 10/12/91 Chemical Results microCurie per milliliter mRem 1.43E-03 2.04E-03 1.43E-03 1.17E-03 1.43E-03 1.65E-03 Comment as approved for discharge*

as original! y discharged after recirculation before completion of discharge Additionally, no liquid effluent high radiation alarms were received during the discharg The inspectors noted that the radiological significance of this event was minimal. The licensee's actions taken in investigating the incident were appropriate. Implementation of corrective action, including procedural enhancements and counseling of the operators involved, was ongoing when the inspection period ende.

MAINTENANCE/SURVEILLANCE TESTING Maintenance Inspection Activity The inspectors observed selected maintenance activities on safety-related equipment to ascertain that these activities were conducted ill accordance with approved procedures, Technical Specifications, and appropriate industrial codes and standards. These inspections were conducted in accordance with NRC inspection procedure 6270 Portions of the following activities were observed by the inspector:

Salem 1 Sa_] em 1 &2 Salem 2 Salem 2 Work Request (WR)/Order (WO) or Procedure Description Troubleshooting and Main Steam Isolation Valves vanous Various Various Various slow closure Power Operated Relief Valves 2A emergency diesel generator preventive and corrective maintenance Charcoal Bed Replacements

Hope Creek Various Hope Creek 910926142

Standby liquid control "B" loop maintenance Diesel fuel oil piping cathodic protection anode replacement The maintenance activities inspected were effective with respect to meeting the safety objectives of the maintenance progra.2 Surveillance Testing Inspection Activity The inspectors performed detailed technical procedure reviews, witnessed in-progress surveillance testing, and reviewed completed surveillance packages. The inspectors verified that the surveillance tests were performed in accordance with Technical Specifications, approved procedures, and NRC regulations. These inspection activities were conducted in accordance with NRC inspection procedure 6172 The following surveillance tests were reviewed, with portions witnessed by the inspector:

Salem 1 Salem 1 Salem 1&2 Salem 1

. Hope Creek Hope Creek Procedure N Test SP(0)4.0.5P-AF(l3)

13 auxiliary feedwater pump inservice test SP(0)4. 7. Main Steam Isolation Valves timing test SP(0)4.0.5. V-MISCl Power Operated Relief Valves testing SP(0)4.0.5P-BA(l 1&12)

Boric Acid Transfer Pump QP-ST.GK-001 *

"B" Control Room Emergency Filtration Monthly Surveillance OP-ST.KJ-003

"C" Emergency Diesel Generator Monthly Surveillance The surveillance testing activities inspected were effective with respect to meeting the safety objectives of the surveillance testing program.

15 Inspection Findings 4.3.1 Salem Pressurizer Power Operated Relief Valves (PORV)

Overview On September 20, 1991, PSE&G discovered that both Unit 1 PORVs were inoperable, apparently for some period of plant operations between April - September, 1991. _

The inspector determined that the technical significance of this PORV inoperability is minor, since:

(1)

PORVs are functionally tested and made operable before placing the low temperature overpressure protection function in service; (2)

Per safety analyses, the primary and credited means of providing overpressure protection in Modes 1~3, the code safety valves, were available and operable during operations; (3)

Although the PORVs are used and referenced in three emergency operating

_ procedures (EOPs), there are alternate methods addressed in the EOPs if the PORVs -

are unavailable; and

-

(4)

The PORV failure mode would not result in an inadvertent blowdown or loss of coolant acciden However, the event indicated important management and programmatic issues regarding maintenance, planning, training and corrective action programs'. The inspector n~viewed the licensee's actions and response to this event and conducted an independent root cause evaluatio Seguence of Events and Licensee Response On September 20, 1991, the licensee notified the NRC via the Emergency Notification System to report that both PORVs on Salem Unit 1 failed to open during testing due to loose bolts on the cover for the air actuator diaphragm. The PORVs (lPRl and 1PR2) are reverse-acting air to open, spring to close, three inch relief valves manufactured by Copes-V ulcan. The PORV s function to relieve reactor coolant system (RCS) pressure during pressure transients to minimize the undesirable opening of the three spring loaded pressurizer code safety valves_while in Modes 1-3. The PORVs are also used during reactor cooldown and shutdown operations to prevent RCS overpressurization. The pressurizer overpressure protect.ion system (POPS) protects the RCS by-lowering the PORV lift setpoints when the

• RCS temperature is at or less than 312 degrees F, as required by Technical Specification (TS) 3.4. Air leakage from the actuator cover flange was discovered by the licensee when the unit was in Mode 4 (Hot Shutdown) and preparing to enter Mode 5 (Cold Shutdown).. Functional testing of the PORVs per TS 3/4.4.9.3 was in progress prior to reaching 312 degrees F reactor coolant temperature and placing POPS in service. The air leakage was sufficient to preclude the valves from opening upon demand. Unit 1 had been shutdown on September 16, 1991, due to an electrohydraulic control system oil leak. On September 22, 1991, the lPRl and 1PR2 PORV diaphragm cover bolts were tightened, and both PORVs were satisfactorily stroke tested and returned to an operable status. The unit then continued its cooldown to less than 312 degrees The licensee determined that the lPRl and 1PR2 actuator diaphragms had most recently been replaced during the last Unit 1 refueling outage on March 21, t991 and April 12, 1991, respectively. The work was performed by contractor maintenance personnel.. The licensee's event investigation reviewed several related issues, such as the possibility of improper bolt torque techniques and improper diaphragm material (relative to existing environmental conditions) that may have exhibited creep due to high ambient temperature. The licensee had experienced similar type *air operated valve failures in the past. However, in each case, only one of the two PORVs was affected. Control and oversight of contractor maintenance personnel, procedure adequacy, QA/QC involvement and training were also reviewed by the license The Unit 1 PORVs were successfully tested both after the actuator maintenance during the last refueling outage (Spring 1991) and prior to placing POPS in service during the last refueling outage recovery. In addition, the licensee has responded to Generic Letter (GL) 90-06 (response dated December 21, 1990) regarding improved reliability of PORV s, PORV block valves, and POP The Unit 2 PORV diaphragms were last replaced on April 7, 1990. In addition, the Unit 2 PORVs were last tested with satisfactory results in the POPS mode on May 15, 1991, during a scheduled mini-outage. PORV testing is currently required once per 18 months when the unit is in Mode 3 or 4. In addition, following the Unit 1 event on September 20, 1991, both Unit 2 PORVs were tested satisfactorily on October 3, 1991, to assure continued PORV.

operabilit On October 18, 1991, plant operators attempted to stroke Unit 1 PORV, lPRl, after a control room annunciator alarmed (POPS Channel I Auxiliary Air Pressure Low). However, the valve did not move off of its closed limit. Subsequent investigation identified that there were two eye bolts installed on the operator diaphragm cover (for lifting the actuator during maintenance). The other ten fasteners were the appropriate bolts. Air was leaking excessively at the eye bolt locations. Additionally, it was identified that several.of the remaining 10 bolts had loosened from the 20 ft-lb torque applied on September 22, 199 *

The eye bolts were replaced with the standard fasteners, the diaphragm was replaced, and the cover was torqued using a modified incremental torquing process on October 19, 199 Then, 1 PR2 was stroke tested while personnel were stationed at the valve. The valve stroke successfully, however, there was slight air leakage observed at the actuator cover. There were no eye bolts installed on 1PR The licensee stated that some actuator* covers have eye bolts installed at two opp<'>site positions to assist in actuator maintenance (removal). The contractor technicians who performed the September 22, 1991, maintenance to lPRl stated that the eye bolts had been in place and were reused per standard practice. At the end of the inspection period, the licensee was developing a PORV inspection and testing program to monitor PORV performance and long term effectiveness of completed maintenance. The licensee is also conducting a review to determine the appropriateness of the use of eye bolts for the purpose of fastening the actuator cover flang Technical Specifications and Safety Sienificance Unit 1 TS 3.4.3 (TS 3.4.5 for Unit 2) requires PORVs and block valves to be operable in Modes 1-3. TS 3.4.9.3 (TS 3.4.10.3 for Unit 2) requires the POPS relief valves to be operable when the RCS temperature is less than or equal to 312 degrees F (lower end of Mode 4). It appears that the failure occurre.d after startup from refueling on April 29, 1991, and before the discovery on September 20, 1991. If the PORVs are inoperable in Modes 1-3, the block valves must be closed in accordance with the TS Action requirement. Per TSs and the FSAR accident analyses, the PORVs are not credited to function (open) in order to limit RCS pressure during design basis transients. The three code safety valves (1PR3,4,5)

provide that function, and remained operable during previous unit operation. However, the PORVs are used and referenced in three emergency operating procedures (EOPs) which address postulated accidents that are beyond the plant design basis. These associated events are steam generator tube rupture, natural circulation cooldown and loss of secondary heat sink. In each of these cases, the inspector verified that there are alternate methods addressed in the EOPs if the PORVs do not functio During a normal unit shutdown, the PORV lift setpoints are adjusted for the POPS mode of operation and the PORVs are functionally tested prior to placing POPS in service. These actions are performed in Mode 4 when RCS temperature is greater than 312 degrees Therefore, at the time of the testing, neither TS 3.4:3 (Modes 1-3) nor TS 3.4.9.3 (Mode 4 -

less than 312 degrees F) apply for PORV operability. Consequently, upon discovery of the degraded PORV condition on September 20, 1991, no specific TS Action requirements were applicabl The inspector determined through document review and discussions with licensee personnel that the failure mode for the PORVs would not result in a PORV opening. That is, in the event of a catastrophic loss of air actuator integrity, the PORVs would spring close. Also,

the PORV block valves (1PR6 and 1PR7) were operable and capable of closing in order to isolate a faulty PORV. The block valves are tested every 92 days per TS At the end of the in.spection period, 1PR2 was inoperable due to valve seat leakage; its

.

associat~ block valve was closed per Unit 1 TS 3.4.3. Also, 2PR1 was likewise inoperable due to valve seat leakage; its associated block valve was closed per Unit 2 TS 3. NRC Review The inspector conducted a detailed independent root cause investigation of this event, and concluded that there were multiple contributing factors which resulted in this even Maintenance and Planning A revised maintenance procedure (November 1990) was not used during PORV maintenance (Spring 1991), because the maintenance planner was apparently unaware ofthe new procedure when the work order was generated. The inspector identified that there was no formal mechanism to ensure that recurring task (RT) work orders reference new implementing procedures when issued by the Procedure Upgrade Project (PUP). RTs are those activities that are performed on some specified frequency, such as the PORV diaphragm preventive maintenance activity (once every refueling).

PORV diaphragm replacement is typically performed by contractor Instrumenta,tion and Control (I&C) personnel, although PSE&G technician's are also qualified to perform the same valve actuator troubleshooting and repair activities. The inspector interviewed both contractor and licensee technicians and. reviewed related maintenance procedures and the vendor manual. The inspector found that the maintenance procedure previously used for maintenance, "M23A-Torquing Guidelines," contained only general guidance related to the PORV actuator diaphragm work. Likewise, the Copes-Vulcan vendor manual contained no*

specific guidance on torquing the diaphragm cover and did not specify torque values. Based upon the interviews with the I&C technicians, the inspector concluded that both the contractor and licensee technicians were knowledgeable of the existing documented PORV actuator maintenance requirements (with the exception of the new specific PORV maintenance procedure).

The inspector reviewed maintenance procedure No. SC.IC-PM.RC-OOOl(Q), "Pressurizer PORV Operator Maintenance Procedure," a new procedure developed by the PUP and approved for use on November 30, 1990. The procedure provides specific instructions for PORV operator removal, reassembly, maintenance and calibration, including a specific torque value -and process. The specific procedure was developed due to work practice weaknesses relative to the lack of documented specific torque values, previously identified (1989) by the licensee's Quality Assurance organization.

• Training The inspector reviewed the licensee's process for contractor qualification/certification. For_

the contractors involved with the PORV work, no deficiencies were identified. Contractor personnel are not normally included in the licensee's technical training programs. The inspector also reviewed training and qualification requirements for PSE&G technicians and found only minor weaknesses relative to detailed instructions on the torquing process. The inspector found that the licensee had previously recognized the noted deficiencies and had developed a Maintenance Continuing Training Module related to torquing procedures for mechanical fasteners. The Training Department had planned to begin the training in November 1991 and expects to provide the training to both PSE&G and contractor I&C technicians. The inspector reviewed the lesson plan and found it to be adequate to implement the stated objective Material The diaphragm material is an elastomer whose geometry can change under loaded conditions, and is subject to a creep phenomenon (loss of thickness or geometry change) if tightened uneven! y. The inspector ascertained that the diaphragm was originally purchased to a ambient temperature specification of 120 degrees F. The licensee stated that actual ambient temperature near the PORVs is in excess of 160 degrees F. The diaphragm material (Buna-N) is certified for 120 degrees F. The licensee contacted a Copes-Vulcan representative, who stated that no deterioration of the-Buna-N diaphragm should occur due to exposure t degrees F for a period of 18 months. However, there is no certified testing to confirm acceptable performance at the higher temperatures. Licensee inspection of the diaphragms that were removed on September 22, * 1991 did not identify any significant adverse deterioration due to high temperature exposure. The creep phenomenon appears to be a function of ambient temperature, diaphragm material properties, ~d uniformity of torquin A Copes-Vulcan representative informed the licensee on October 2, 1991 that a specific torquing process with recommended torque values will maximize the service life of the diaphragm. The letter also stated that uneven compressive loading causes excessive creep in overloaded areas of the diaphragm and a subsequent reduction of sealing capability. The vendor also stated that the actuator cover design was previously modified from a 12-bolt pattern to a 24-bolt pattern, allowing for a more uniform loading distribution.. The licensee had known about the vendor design change and is currently considering a station design change to allow the modified bolt patter Corrective Actions for Previous Deficiencies The inspector conducted a review of completed maintenance work orders relative to both Unit I and 2 PORYs. The inspector found that since December 1989, Unit 1 lPRl experienced air leaks on four other occasions and the bolts were tightened each time. Unit 1 I PR2 experienced air leaks on three other occasions since March 1987. Three air leaks were

identified for the Unit 2 PORVs since April 1987. The licensee's QA organization audit identified that since 1978, there were over 20 occurrences of loose bolts and/or diaphragm degradation identified. The inspector found that several action~ had been initiated by the licensee to resolve the problems, such as vendor interface, new procedure development, and diaphragm material/application investigation. However, the lack of discemable improvement and progress in this area indicates a lack of management involvement and/or aggressivenes The licensee's response to GL 90-06, dated December 21,1990, states that Salem Units 1 and 2 commit to include PORVs and block valves within the scope of a QA program per 10 CPR 50, Appendix B. The response also states that the PORVs and block valves are included in the plant operational QA List. Inspector review identified that the PORV.and actuator cfosure spring are classified as safety related. However, the diaphragm is non-safety related (NSR). The licensee stated that the diaphragm is NSR because the PORVs are designed to fail to the closed position on loss of air supply and that no provision is necessary to ensure activation of the valves should the air supply fail, since the valves are classified as inactive (FSAR Section 5.5.13). The inspector concluded that additional information is needed to determine whether 1) the PORV actuator diaphragm should be considered as safety related, and 2) whether the PORVs are inactive components since they are designed to open in the POPS mode of operation. The licensee initiated an effort to re-evaluate the safety classification of the diaphragm. Pending review of tpis evaluation, this is an Unresolved Item (UNR 50-272/311;91-26-02).

Licensee Corrective Actions The licensee implemented several corrective actions for this event, including actions that were previously initiated. These actions are listed as follows:

1)

Replaced diaphragm on lPRl and 1PR2, torqued actuator cover per maintenance procedure SC.IC-PM.RC-OOOl(Q), and successfully retested both PORVs on September 22, 1991 ;

2)

Successfully stroke tested Unit? PORVs (2PR1 and 2PR2) on October 3, 1991; 3)

Replaced diaphragm on lPRl torqued actuator cover using enhanced torquing process and successfully stroke tested 1PR2 (slight air leakage noted) on October 19, 1991; 4)

Initiated the development of an interim PORV testing and inspection program to verify operability for Unit 1 and Unit 2 PORVs; 5)

Conducted event investigations by both the QA organization and System Engineering; 6)

Continued previously initiated efforts to resolve diaphragm material property concerns as related to creep;

7)

Initiated efforts by System Engineering to certify diaphragm for high temperature;

. 8)

Began assessing a modification by System Engineering to change the existing 12-bolt actuator cover to a 24-bolt cover; 9)

Developed specific detailed training lesson plans on July 26, 1991; current plans are to train all contractor and licensee I&C technicians beginning November 1991;

10)

Temporarily implemented a process to ensure that PUP notifies the appropriate maintenance planner when new procedures are developed; permanent administrative procedure changes are currently being evaluated; and 11)

Initiated evaluations to confirm the proper safety classification for the PORV actuator diaphragm and to determine the appropriateness of use of eye bolts to fasten the actuator cove The inspector verified that the above actions are either completed or are currently in progress. The inspector reviewed the QA and System Engineering evaluations and concluded that both evaluations were complete and comprehensive. Each evaluation provided appropriate recommendations for management review.

Conclusions The inspector concluded that the safety significance of the PORV inoperability is minor, since (1) the pressurizer code safety valves, not the PORVs, are the credited means per the FSAR to provide RCS overpressure protection while in Modes 1-3; (2) EOPs reference alternate methods to complete actions if the PORVs are inoperable; (3) the PORVs fail to the closed position, thus preventing an inadvertent blowdown or loss of coolant accident; and 4)

the PORVs are functionally tested and verified to be operable before placing POPS into servic However, the PORVs do provide an important function, both in the POPS mode and when directed by EOPs for postulated events that are outside the design basis of the plant. This event indicated important management and programmatic issues regarding maintenance, planning and corrective action programs as indicated belo The inspector noted that several conditions must exist for the proper sealing of the PORV air actuator cover. These include proper diaphragm material for existing environmental conditions and a uniform loading on the cover and bolts. There were multiple causes that contributed to this event. The inspector concluded that the following root causes resulted in the PORVs being inoperable:

I )

Vendor and procedure guidance did not provide sufficiently detailed guidance and strict controls necessary for PORV diaphragm maintenance activities;

2)

Lack of aggressive management attention and/or oversight allowed degraded conditions to exist for an extended period of time without being effectively resolved (i.e. recurring PORV failures); and 3)

Administrative controls were inadequate to ensure the appropriate maintenance planning personnel became aware of the new PORV maintenance implementing procedur The inspector concluded that although many corrective actions had been initiated, aggressive management attention is necessary to fully and properly resolve this issu Steam Driven Auxiliary Feedwater (AFW) Pump Rooms During surveillance testing of the 13 AFW pump, the inspector noted that operations personnel throttled and closed the steam line drain trap isolation valves. The inspector questioned operators and system engineering personnel regarding this practice. Their response was that during steam driven AFW pump operation, the steam line drains blow steam into the room because the lines are not piped directly to a room floor drain. As a result, operator access to the room is impaired. The licensee has initiated design changes to modify these drain lines. This is being done under the Salem revitalization project. The inspector verified that these design changes are planned for the upcoming scheduled unit refueling outage Open Item Followup (Closed) Violation (50-272 and 311/90-12-01 and 03). Failure to verify and establish safety injection (SI) and charging pump flow rates per Technical Specification (TS) 3/4.5.2. The licensee responded to the violation and related issues in a letter dated July 10, 1990 and in LER 90-14. These documents were previously reviewed and found to be acceptabl Corrective actions are delineated in these documents. NRC Inspectiori 50-272 and 311/90-13 also reviewed these issues related to emergency core cooling system pumps. Based on the above reviews, these items are close (Closed) Unresolved Item (50-272 and 311/90-81-15). Non-code repair to ASME class 3 service water piping without prior NRC approval. *

The Integrated Performance Assessment Team (IPAT) reviewed the licensee activities concerning a welded patch repair to a through wall leak of the Unit 1 No. 11 Service Water*

(SW) header. The team noted that the temporary repair was not in accordance with ASME Section XI Article IWA-4000, "Repair Procedures." It was also noted that the temporary repair was made without prior NRC approval, as required by 10 CPR 50.55a(g)(5)(iii).

In response to this inspection finding, on June 6, 1990, PSE&G submitted a formal request for an ASME Section XI relief for this repair. On September 3, 1990, PSE&G received a

letter dated August 29, 1990, from NRR stating that NRR reviewed the PSE&G submittal and determined that the temporary repair was acceptable for the remainder of the current fuel cycle. NRR also stated that a repair that meets ASME code requirements must be completed during the ninth refueling outage (February 1991). During the ninth refueling outage, the resident inspector confirmed that the code repair had been completed and in accordance with ASME code requirement Additionally, the NRC issued Generic Letter (GL) 90-05, "Guidance for Performing Temporary Non-Code Repair of ASME Code Class 1,2,3, Piping," on June 15, 199 PSE&G revised Nuclear Administrative Procedure (NAP)-20, "Nonconfonnance Program," *

on November 8, 1990, to incorporate the guidance provided in GL 90-05 guidelines and be only implemented after receiving approval of the repair package from the NRC; The inspector had no further question Main Steam Isolation Valve (MSIV) Testing During the Unit 1 startup from its forced outage (see Section 2.2.1.A), the licensee performed procedure SP(0)4. 7.1.5, "Main Steam Isolation Valve Emergency Close Time Response, 11 in accordance with Integrated Operating Procedure 3, "Hot Standby to Minimum Loa In procedure SP(0)4.7. l.5, the test of the fast closure ability of the four MSIVs is described, and the acceptance criteria that the valves close in less than five seconds is incorporate On September 25, 1991, 11MS167 was the first MSIV to be tested, and it took 5.23 seconds to close. Technical Specification allows an eight second time. However, the licensee requires a five second closure time. Initial troubleshooting did not identify a readily apparent cause for the slow response time of the valve, but the Salem Instrument and Control group adjusted the bypass valve on the hydraulic actuator of the MSIV, and an acceptable closure time of 4.95 seconds was achieved. The next MSIV tested was the 12MS167, which also took longer than the allowed five seconds to close. Troubleshooting on this valve was again difficult, but the slow closure time was eventually attributed to a faulty Westinghouse Air Brake Company (WABCO) quick release valve on the 12MS171 vent valve air operato The WABCO valve provides a quick exhaust path for the MSl 71 air operator, allowing the MS171 to open more quickly, which in turn provides a vent path for the steam holding the MSIV open, causing the MSIV to fast-close. The faulty WABCO valve was replaced with the one spare valve PSE&G had in stock, and the 12MS167 was subsequently satisfactorily retested. The 13 and 14MS167s were both tested satisfactorily following the retest of 12MS16 The Salem Technical and Maintenance Departments later investigated the cause of the WABCO valve failure and determi.ned the cause to be a faulty diaphragm in the valve which prevented it from properly exhausting the air from the MS171 operator. The licensee believed that *less than optimum performance of the WABCO valve on the 11MS171 may have contributed to the slow response of the 11MS167. PSE&G did not have any previous

I L

...

.

.

maintenance or test history data on any of the MS169 cir MS171 vent valve WABCO quick release valves. The WABCO valves are not tracked in the Salem Managed Maintenance Information *system (MMIS), nor do they appear in any system_ piping and instrumentation diagrams. The only place the licensee could find the WABCO valves documented was in a note on the part drawings for the. MS169 and MS171 vcilves. To better monitor the performance of the WABCO valves and to be prepared for any future failures, PSE&G has planned and begun to implement vari~us corrective actions. A change in the surveillance test procedure for the MS 169 and 171 valves was instituted to better replicate the conditions of the MSIV fast closure tests, with additional attention paid to the WABCO valves. New, spare WABCO valves were acquired to*be on hand in the event of future failures, and PSE&G plans to conservatively replace all Unit 2 WABCO valves during the MSIV overhauls planned for that unit's refueling outage in January - March 1992. The licensee also considering adding the WABCO valves to MMIS so they may be better tracked in the futur The resident inspector was monitoring the licensee's progress in the Unit 1 restart when the problems with the MSIV testing occurred. Due to previous problems encountered with the

• Salem MSIVs (see NRC Inspection Report 50~272 and 311/90-20) and the extensive overhaul performed on the Unit 1 MSIVs during that unit's last refueling outage (see NRC Inspection Report 50-272 and 311/91-05), the inspector closely monitored PSE&G's investigation and resolution of the problems encountered during the Unit 1 startup, including observation of the testing of the MSIVs and subsequent troubleshooting, inspection of the failed WABCO valves

.and the WABCO valves installed on the other MS169 and 171 valves, and discussion of the problem with members of the Salem Operations, Technical, and Maintenance Department The 'inspector concluded that PSE&G should have had a better history and knowledge of these WABCO valves and their potential impact on MSIV performance, but once the problem was identified, it was investigated and corrected properly and positive steps were taken to help prevent and deal with it,' should it reoccur. The inspector had no further questions at the end of the report perio.3.2 Hope Creek Unplanned Power Reduction Due to Surveillance Testing On September 23, 1991, at 9:38 p.m., a reactor protection system (RPS) half scram (RPS Channel A 1) was received concurrent with a full runback of both reactor recirculation motor-generator sets due to several low reactor water level signals (Level 3, + 12.5 inches). The plant responded as designed, and reactor power was stabilized at about 60%. No engineered safety feature (ESF) actuations occurred. At the time of the event, operations personnel were performing *an in service surveillance test on the post. accident sampling system (PASS)

vaives required for primary containment isolation. After determining the low water level signals to be spurious, the RPS half scram was reset and reactor power was returned to 100% on September 24. *

The licensee determined that the apparent root cause of the half scram and recirculation runback was a pressure perturbation caused by operation of the PASS No. 15 jet pump sample isolation valves during the performance of HC.OP-IS.R~-101, "Post Accident

. Sampling System Valves Inservice_Test." The PASS sample line taps off the same instrument line that provides the high side pressure signal to the fuel zone level transmitter LT-N085A. The low pressure side of this tr:ansmitter shares a common instrument line with a number of safety related level and pressure transmitters. A few days prior to this event, the PASS panel had been used for training. The licensee concluded that panel operation could have resulted in a partial drainage of the sample lines such that a water hammer occurred when the sample isolation valves were. stroked open on September 23, 1991. The resultant pressure perturbation *propagated through the pressure diaphragm of LT-N085A and was sensed as about a 0 inch reactor water level by a number of transmitters on the associated instrument rack, triggering the observed half scram and recirculation runbac The inspector discussed this event and subsequent investigation with licensee operating and technical personnel. Operator actions in responding to, controlling, and recovering from the transient were appropriate and timely. In order to better understand the pressure phenomenon, the inspector reviewed a number of previous incident reports and licensee event reports (LERs). LERs 86-61, 87-30~ and 91-04, were particularly noteworthy in that they described actuations due to the propagation of a pressure pulse generated by several different means through an instrument rack. For example, in July 1987, the High Pressure Coolant Injection (HPCI) system was initiated when a technician valved in a reactor water level transmitter. It was also noted that the Rosemount transmitters used at Hope Creek have very quick response times so that even very short duration or minor pressure spikes could have resulted iri the generation of a trip signal. Previous corrective actions included enhariced technician training, procedural improvements, and a number of design changes to reduce the affects of pressure surges during surveillance testing. To address these latest events, the licensee stated that enhancements would be made to the PASS panel operating procedure and that in the future, all training on the PASS panel would be conducted with the planning and scheduiing group and the operating shift. The inspector concluded that these measures

appeared appropriat.

EMERGENCY PREPAREDNESS The inspector reviewed PSE&G's conformance with 10 CFR 50.47 regarding implementation of the emergency plan and procedures. In addition, licensee event notifications and reporting requirements per 10 CPR 50. 72 and 73 were reviewe *

No noteworthy findings were identified.

--- --

-

- ----------

26 SECURITY PSE&G' s conformance with the security program was verified ~n a periodic basis, including the adequacy of staffing, entry control, alarm stations, and physical boundaries. These inspection activities were conducted in accordance with NRC inspection procedure 7170 No noteworthy findings were identifie.

ENGINEERING/TECHNICAL SUPPORT Salem CFR 21 Reportability Concerning Auxiliary Feedwater (AFW) Pumps On September 19, 1991, Ingersoll-Rand (1-R) Company issued a 10 CPR 21 (Part 21) report regarding a failed AFW pump at another facility due to broken cast iron diffuser pieces. The apparent causal factors were corrosion and erosion/cavitation damage due to accumulated hours at low flow conditions. I-R recommended the following actions: (1) inspect any affected AFW pumps for possible internal damage; (2) replace damaged components with martinsitic or austenitic steel internals; and (3) conduct inservice testing at sufficient flow rates ( > 25 % design flow).

The inspector informed PSE&G of this Part 21 report, and the licensee's review determined the following:

Of the six AFW pumps at Salem (three per unit), only the number 22 pump is an 1-R pump. The remaining pumps are Byron-Jackson. *The number 22 'pump was installed in 1985 due to an inservice bearing failur The number 22 AFW pump is tested at about 50% design flow (approxima:t~ly 200 gallons per minute).

A spare I-R pump rotating assetnbly is onsit *The licensee contacted the vendor and reviewed the Part 21 and related reports. The licensee's corrective actions as related to the vendors recommendations are: (1) inspect the number 22 AFW pump during the 1993 refueling outage (the date is based on historical testing at 50% flow and no performance problems); (2) modify the spare rotating element by

~ending it to the vendor for material upgrades; and (3) continue inservice testing at higher than the minimum 25 % flow requiremen The inspector revi~wed the licensee's response to this issue and discussed it with licensee engmeers. The inspector concluded that the licensee's actions were appropriate.

27 Lack of Vent Openings for Main Steam Vent Valve ControlPanels (Closed) Unresolved Item (50-272 and 311/90-19-01). This item concerned the lack of vent holes on the cabinets which contain two of the four Unit 2 MSi69 solenoid operated valves and the resulting uncertainty regarding the environmental qualification of these valves. The MS169 valves are important because they provide for the fast closure of the units' main steam isolation valves (MSIVs), the MS167s. The vent path that was supposed to be provided in the. cabinets containing the MS 169s is designed to prevent the cabinet from collapsing and damaging the valves upon a steam line rupture in the room containing the cabinet This condition was identified by PSE&G and considered as a condition outside the design basis of the plant. A Design Change Package (DCP) was developed and issued, and the licensee remedied the condition by placing the missing vents in the appropriate cabinet PSE&G iss1,1ed a Licensee Event Report (LER 50-311/90-25) to document the deficient condition and, in order to assess the significance of the cabinets not having the vents, performed an evaluation of the structural adequacy of the MSIV vent valve instrument panels if subj~cted to a main steam line break (MSLB) pressure transient. For that accident, the postulated peak pressure reached in the area of the affected cabinets is 5. 8 psid, reached in less than 1.0 second. The licensee's evaluation included laboratory tests which subjected similar cabinets to a pressure of 8.65 psid achieved in 0.25 seconds. In this case, the maximum deflection of the cabinet cover was 0. 75 inches, with rio damage to the cabinet internals note The inspector reviewed and compared the LER, the DCP, and the safety evaluation which PSE&G prepared in. response to this deficient condition and inspected and confirmed the presence of the vents in the cabinets that were the subject of the DCP. The inspector determined the content and conclusions of the above mentioned documents to be acceptable and, in light of the data contained in the safety evaluation, concluded the safety significance of the missing vents to be negligible. PSE&G responded to this item in an acceptable manner, and this item is close.2 Hope Creek Core Spray Pump Motors Seismic Design An unresolved item (50-354/88-80-02) pertaining to the structural and seismic adequacy of weight arrangements added to the "B" and "D" core spray pumps to reduce vibration was closed in NRC Inspection 50-354/91-12. The licensee prepared evaluation H-1-BE-NEE-0506 for the seismic evaluation of the installed vibration damper and absorber, and evaluation H-1-SEE-0525 and H-1-BE-SDC-0739 for the decoupling effect on the vibration absorbe *

The Office of Nuclear Regulatory Regulation performed a review of the above evaluations (TAC No. 80461). The staff concluded that the addition of the damper or absorber has no adverse affect on motor performance and both the damper and the absorber are structurally adequate to withstand the seismic load during a Safe Shutdown Earthquake event. An earlier staff concern regarding the effect on the pump/motor frequency due to coupling between the pump/motor was also appropriately addressed by the licensee. The staff prepared a safety evaluation, which concluded that the modified motor/pump configurations are acceptabl Engineered Safety Feature (ESF) Actuation - "F" Filtration, Recirculation and Ventilation System (FRVS) Fan Start On September 14, 1991, operations personnel discovered that the "F" FRVS recirculation fan was running. After verifying that the fan had started spuriously, the fan was stopped. The licensee's investigation determined the cause to be a collection of moisture in the "C" FRVS fan low flow switch which provides an auto start signal to the standby fan (in this case, "F").

There have been five previous spurious auto starts of standby FRVS fans since 1987 (see LERs 87-16, 87-33, 90-06, 90-23 and 90-34), one of which in December 1990 was attributed to circumstances similar to this event (see LER 90-34 and NRC Inspection Report 91-01, Section 9.1).

Licensee corrective actions for the December 1990 event included expediting a design change to modify the auto start logic and revising the surveillance test procedures to include draining the flow switch instrument lines after FRVS testing. At the time of the September 1991 auto start, the licensee had modified the surveillance procedures as committed to in LER 90-3 However, in this most recent event, the "C" FRVS recirculation fan had been run earlier (the normal reactor building ventilation was out of service) using the system operating procedure, which had not been revised to include the draining of instrument lines. The inspector concluded that the failure to identify and modify the operating procedure was a second example of the programmatic weakness discussed in Section 2.2.2.A of this inspection report

• and that the licensee's corrective action in this area, as stated in LER 91-18, would not adequately address that weakness. Licensee technical personnel indicated they would determine the status of related procedure change incorporation and consider enhancements to the administrative procedure controlling procedure development and changes, NC.NA-AP. ZZ-003 Also in LER 91-18, the licensee changed their commitment to implement the design change to the auto start logic until either the unit was shutdown or no later than the fourth refueling outage. The inspector discussed this change with engineering personnel who stated that during a review of the feasibility of implementing the design change at power, risks of inadvertently removing safety related equipment from service and/or tripping the unit were identified. Since the safety significance of spurious FRVS starts was minimal, the licensee decided not to implement the design change while at power. The inspector reviewed the licensee's assessment and noted that the licensee exhibited a good safety perspective in their review and conclusions.

29 SAFETY ASSESSMENT/QUALITY VERIFICATION Salem Station Operation Review Committee (SORC) Meeting The inspector attended a SORC Meeting on October 18, 1991. A draft licensee event report (LER) was reviewed, Unit 1 LER No. 91-26, which addressed the failure of both pressurizer power operated relief valves (PORVs). The inspector verified that the SORC member composition and quorum requirements as specified in Technical Specification 6.5.1 were satisfied.

. The inspector noted that the SORC members appropriately reviewed the failure and properly evaluated the causal factors of the PORV failures as documented in the draft LER. The draft LER identified the root cause of the event as equipment failure. The SORC concurred with that conclusion following a discussion. However, the inspector concluded (see Section 4.3.1.A) that other multiple root causes resulted in this equipment failure which, in turn, resulted in rendering the PORVs inoperable. The inspector's concern regarding the definition of root cause was communicated to the SORC Chairman. After these discussions, the SORC reviewed the root causes stated in the LER. The root cause was restated as

"unknown" at this tim Unit Operations During the inspection period, station management demonstrated a conservative and safe approach to unit operation. Examples included:

Unit 1 shutdown to repair an electrohydraulic control system leak, Unit 2 power reduction due to river water intrusion into the steam generators, and Unit 2 power reduction to lower the chloride crevice concentration from the steam generator.2 Hope Creek Self-Assessment Initiatives The inspector reviewed two self-assessment initiatives regarding current Hope Creek performance. These included an independent performance based surveillance by station quality assurance (QA) of personnel errors and, a comprehensive scram review by a team led by the onsite safety review group (SRG). The SRG team's final report recommendations and conclusions were reviewed in NRC Inspection 50-354/91-23.

During the current inspection period, the inspector met with the SRG team leader to discuss the process utilized. The QA review concluded that personnel errors continue to be a major root cause of events. However, a 20% decrease was noted from the period August 1, 1990 to April 1,. 1991 (as compared to the previous SALP cycle). The QA review also noted the event reports were not being performed in a timely manne Overall, the inspector concluded that these two efforts were thorough, well performed, and appeared to be effective in their conclusions and recommendation Reactor Protection _System (RPS) Half Scram Reportability On September 23, 1991, the unit received a half scram (RPS Channel Al trip) and full recirculation system runback on low reactor water level (Level 3 - + 12.5 inches) due to surveillance testing on the Post Accident Sampling System (PASS). (For a full discussion of this event see Section 4.3.2.A of this inspection report.) The event was duly reported to the NRC Operations Officer and resident inspector (event no. 21889) per lOCFR 50.72(b)(2)(ii).

Subsequent to the event, the resident staff discussed with licensee personnel the licensee's perceived need to report half scrams as engineered safety feature (ESF) actuations. The licensee responded that their action was consistent with the emergency classification guideline (ECG) 18L, but would request their licensing group to determine if their classificatiqn of a half scram as an* ESF actuation was unduly conservative. Station licensing, in letter NLR-191533 dated September 26, 1991, to operation's management, stated that the trip of a single RPS channel (resulting in a half scram) did not, by design, activate a full reactor scram (i.e.,

no rod movement). Therefore, no ESF actuation occurred, and no report was require This position was derived from a review of NUREG-1022, "Licensee Event Report System,"

and the stated definition of "actuation" of multi-channel ESF actuation systems on page 1 The licensee subsequently withdrew the four hour non-emergency report on September 30, *

1991. The resident staff reviewed the licensee's interpretation of NUREG-1022, discussed the issue with the NRR project manager, and concluded that the licensee's interpretation was acceptabl.3 Common General Manager (GM) State of the Station Meetings The GMs for both the Hope Creek and Salem Stations conducted a series of "State of the Station" meetings with their respective employees. The purpose of these meetings was to present to station workers and supervisory personnel an assessment of performance during the first half of 1991. The GM, Salem conducted meetings on August 30, 1991, and the GM, Hope Creek conducted meetings on September 20, 199 The GM's addressed the following items:

Current performance levels and performance indicators,

Corporate vision, mission, and goals, Current weaknesses and strengths,

Technical issues,

_

Status of regulatory and industry evaluations and assessments, Current human performance issues and problems, and A long term message for continued hard work, attention to detail, commitment for higher expectations, and to fight complacenc The inspector attended one meeting, reviewed the viewgraphs and discussecI the meetings with each GM and selected station personnel. The meetings appeared to be well received by employees, including an interactive question and answer period. The inspector concluded that these "State of the Station" Meetings appeared to be an effective tool for communicating assessment of station performanc CFR 50.59 Program Inspection During the period of September 10 through September 12, 1991, the NRR Project Managers for Salem and Hope Creek conducted an inspection of PSE&G's 10 CPR 50.59 progra The inspection involved the review of the administrative procedure for implementing. the program and a sample of completed 50.59 Reviews and Safety Evaluations for design change packages, temporary modifications, deficiency reports, final safety analysis changes and procedures. This inspection was conducted in accordance with NRC Inspection Procedure 3770 Program Reviews Administrative Procedure NC.NA-AP.ZZ-0059(Q), Revision 0, "10 CPR 50.59 Reviews and Safety Evaluations" (NAP-59), is the document that provides the guidance for conducting a 10 CFR 50.59 (hereafter referred to as 50.59) Reviews and Safety Evaluations. A 50.59 Review determines if 50.59 is applicable to the proposal at hand. This is determined by answering the questions: Does the proposal change the facility as described in the Safety Analysis Report (SAR); Does the proposal change procedures as described in the SAR; or Does the proposal involve a test or experiment not described in the SAR? The bases for the answers must be provided. If the answers to the three questions are "no", 50.59 does not apply and further evaluation is not required. If the answer to any of the above questions is

"yes", then a 50.59 Safety Evaluation must be performed to determine if an uilreviewed safety question is involve For all 50.59 Reviews, NAP-59 requires a peer review and management approval to assure that the 50.59 Review was properly performe A 50.59 Safety Evaluation consists of answering the following questions and providing the bases for the answers:

L 1..

May the proposal:

(a)

Increase the probability of an accident previously evaluated in 'the SAR?

(b)

Increase the consequences of an accident previously evaluated in the SAR?

(c)

Increase the probability of occurrence of a malfunction of equipment important to safety previously evaluated in the SAR?

(d)

Increase the consequences ofa malfunction of equipment important to safety previously evaluated in the SAR?

(e)

Create the possibility of an accident of a different type than previously evaluated in the SAR?

-

(t)

Create the possibility of a malfunction of a different type than any previously evaluated in the SAR? Does Jhe proposal reduce the margin of safety as defined in the basis for any Technical Specification?

If the answers to the above questions are no, then an unreviewed safety question is not involved, and the proposal can be implemented. If the answer to any of the above questions is yes, NRC approval before implementation is required. For all proposals that receive a 50.59 Safety Evaluation, a peer review, management approval, and review and approval by the Station Operations Review Committee (SORC).is require.

.

For both a 50.59 Review and a 50.59 Safety Evaluation, theJicensee must also determine whether or not the proposal involves a Technical Specification change. In either case, if the answer is yes, NRC approval is required before implementation of the proposa The application of NAP-59 provides the mechanism for limiting-proposals that require further review by SORC. However, SORC reviews all design change packages, temporary modifications, and any proposal that requires a 50.59 Safety Evaluatio For review and approval of procedures, PSE&G uses Station Qualified Reviewers (SQR) to provide independent review of procedures. The SQRs also determine if a 50.59 Safety Evaluation is required for the procedure in question. The process defined in NAP-59 is used for documenting the 50.59 Review and the Safety Evaluation, if required. If a 50.59 Safety Evaluation is required, SORC approval is required before the procedure.can be implemente NSAC-125, "Guidelines For 10 CFR 50.59 Safety Evaluations," has been incorporated into the guidai1ce provided in NAP-59. It was noted that NAP-59 is inconsistent regarding the

• information to be included in the description of the change. Page 4 of 19 uses the word

"shall". to indicate that the applicable design, operation, and regulatory requirements are to be included; Page 1 of 18 of Exhibit 1 uses the word "should".regarding the same information. The licensee agreed to examine this in future revisions of NAP-5 Salem A sample of completed 50.59 Reviews and Safety Evaluations were examined by the project manager (see Attachment 1). The 50.59 Reviews and Safety Evaluations were prepared in accordance with the guidelines provided in NAP-59, with the exception of one, as discussed below. The information provided in the reviews and safety evaluations was complete and provided the rationale that supported the conclusion, i.e., either 50.59 is not applicable or an unreviewed safety question was not involved. The project manager had no questions concerning the 50.59 Reviews and Safety Evaluations except for the feedwater regulating valve BF-19 packing adjustment while at power, discussed belo Salem 1 and 2 BF-19 Packing Adjustment Issue On December 20, 1990, the Station Operations Review Committee (SORC) approved revision 3 to the Salem Generating Station Technical Department Safety Evaluation titled

"Packing Adjustment of BF-19 Valves" (S-C-F300-MSE-0706-3) dated November 21, 199 The BF-19 valves are the feedwater regulating valves at both Salem 1 and 2 and are listed as containment isolation valves in the Technical Specifications, Section 3.6.3 and Table 3.6-Technical Specification Surveillance Requirement 4.6.3. l requires a cycling test and

• verification of isolation time after maintenance, repair, or replaceme~t work is performed on the valve or its associated actuator, control, or power circuit. These valves are also in the inservice test (IST) program. The ASME Code,Section XI, Article IWV-3200, states that valves that undergo maintenance should be tested prior to return to service to demonstrate that the performance parameters are within the acceptable limits. Article IWV-3200 lists stem packing adjustments as maintenance that could affect valve performanc The packing adjustment procedure that was approved by SORC allows the packing to be adjusted with the units at power and maintaining the valves "operable" by performing a

• partial stroke test to show the valves are not binding. The original issue of the packing adjustment procedure is dated September 25, 1987. A 50.59 Safety Evaluation, dated June 12, 1987, had been prepared to support the original issue of the packing adjustment procedure. Subsequent revisions appear to have utilized the June 12, 1987, 50.59 Safety Evaluatio A review of the documentation concerning this issue revealed that on June 9, *1987, valve 23BF-19 had a packing leak, and a request was made to the NRC staff for an exemption from the IST program requirement that considered tightening the packing holddown nuts as a maintenance activity. Such exemption would permit the packing adjustment without

requiring a timed stroke test following the packing adjustment. The internal PSE&G memo states that the NRC staff approved the reques.

There are three additional internal PSE&G memoranda, dated June 16, 1987, August 14, 1987, and October 28, 1987 and two letters, dated July 1, 1987 and August 19, 1987, to the NRC that address the adjustment of the packing of specific BF-19 valves. All of the memoranda and letters document telephone conversations between PSE&G and the NRC or the valve packing vendor. The October 28, 1987, memorandum documents a telephone conversation between PSE&G and the NRC staff in which the safety evaluation that analyzed the acceptability of torquing the BF-19 valve packing without subsequently performing a timed, full stroke test of the valve is discussed. The memo notes that the NRC staff concurred in this approach as long as it could be demonstrated that the final torque value would.not affect valve closure time. The memo's final statement is that because it was demonstrated on two occasions that a torque of 13 ft-lbs did not affect valve closure time, it was acceptable to torque the packing without performing a timed, full. stroke tes From June 8, 1987, to December 15, 1990, the BF-19 valves in Salem Unit 2 had their packing adjusted a total of sixteen times. The final torque values ranged from a low of 12 ft-lbs to a high of 21 ft-lbs. In the same period, only one BF-19 valve at Salem Unit l had its packing adjusted. Of the 17 occasions of packing adjustment, eight resulted in a final torque of greater than 13 ft-lbs, five resulted in a final torque of 13 ft-lbs or less and four did not list the final torque value. On all occasions a partial stroke was done to assure the valve was not binding. However, only stroke test data taken in 1987 is referenced in Revision 3 of the packing adjustment procedure (S-C-F300-MSE-0706-3).

• The NRR Project Manager noted that the 10 CFR 50.59 safety evaluation, upon which the BF-19 packing adjustment activities were based, did not consider any torque values greater than 13 ft-lbs relative to the need for a timed stroke test following the adjustment. The project manager further noted that while a one-time exemption may have been approved by NRC for a specific instance, there was no basis for the licensee to assume the exemption applied for any occurrence of packing adjustment. It appeared to the project manager that the licensee had misinterpreted or misunderstood the limited extent and nature of NRC acceptance. Accordingly, the licensee elected to terminate use of the procedure for packing adjustment of the BF-19 valves until a license amendment request has been approved by the NRC. This item is considered unresolved pending the licensee's resolution of this matter, and NRC's determination that other valves were not similarly affected by the licensee's misunderstanding (UNR 50-272/311;91-26-03).

Hope Creek The inspector reviewed the 50.59 summary reports contained in the May, June, and July 1991 Monthly Operating Reports. Paragraph (b )(2) of 10 CFR 50.59 requires that the report include a summary of the Safety Evaluation. The 50.59 Safety Evaluation provides the bases

for the determination that the change, test, or experiment does not involve an unreviewed safety question. The inspector determined that the licensee's Monthly Operating Reports did not contain an adeqm1te summary of the Safety Evaluatio Th~ licensee agreed to address this concern in future Monthly Operating Reports. Subsequently, the August 1991 Monthly Operating Report was reviewed by the inspector and was found to adequately address the inspector's concern The inspector reviewed four licensee 50.59 Reviews and Safety Evaluations for conformance with the regulation. The documents reviewed were: Temporary Modification Requests (TMRs)91-023 and 91-029 and Design Change Packages (DCPs) 4EC-3012 and HHC-0204/24. The inspector determined that the four documents adequately addressed the requirements of 1OCFR50.59. The inspector agreed with the licensee's conclusions regarding 50.59 applicability and regarding the unreviewed safety question determinatio Additionally, the inspector reviewed the four 50.59 documents to determine if they conformed with NAP-59. The following comments were noted:

NAP-59 requires that the description o~ the change include a discussion of the applicabl design, operation and regulatory requirements that relate to the proposal. Exhibit 1 of NAP-*

59 gives specific examples of what should be discussed. The four documents the inspec_tor reviewed did not adequately include this discussion in the description section, Additionally, NAP-59 requires that the 50.59 Review and Safety Evaluation contain sufficient detail to allow a reviewer to independently arrive at the same conclusions. Contrary to this requirement, page 7 of 10 of TMR 91-023 did not contain sufficient information for the inspector to.arrive at the same conclusion. The inspector found acceptable the licensee's conclusion only after the licensee provided additional informatio The licensee agreed to address these concerns in future 50.59 evaluations. The inspector has *

no further questions at this tim * Spent Fuel Pool Item Survey The inspector surveyed items fo the spent fuel pools (SFPs) at Salem units and at Hope Creek. In addition, programs and controls for storage of items in the SFPs were also reviewed. The objective was to assess the risk associated with the storage of heavy items suspended in the SFP The SFPs at both Salem units contained very few items other than spent fuel assemblie Although an inventory list did not exist; a recently written procedure covering underwater storage of highly radioactive material required a list to be prepared by the end of 199 Hope Creek's SFP contained a variety of items that were recorded on an inventory lis However, Hope Creek did not have a formal procedure covering the storage of items in the

SFP. After further review, the inspector noted that the licensee. had previously identified this issue and a procedure was being writte The inspector concluded that there were no heavy objects suspended in the SFPs at either *

Salem unit or Hope Cree.

LICENSEE EVENT REPORTS (LER), PERIODIC AND SPECIAL REPORTS, AND OPEN ITEM FOLLOWUP 9. l LERs and. Reports PSE&G submitted the following licensee event reports, and special and periodic reports which were reviewed for accuracy and evaluation adequac Salem and Hope Creek Monthly Operating Reports for September 199 Salem and Hope Creek Safeguards Event Log No discrepancies were note Salem LERs Unit 1 LER 91-28 concerned a steam generator blowdown isolation valve failure on August 23, 1991. The event was reviewed in NRC Inspection 50-272/91-23. No inadequacies were noted relative to this LE LER 91-29 conc.erned a radiation monitoring system (RMS) sample pump failure on September 7, 1991. The pump was replaced and will be changed out every three months pending a new design as part. of the RMS upgrade project. No inadequacies were noted relative to this LE LER 91-30 (See Section 4.3. l.A)

!,,ER 91-31 (See Section 2.2.1.D)

Unit 2 LER 90-20, Revision 1, updated commitments associated with radiation monitoring system

• hardware upgrades including uninterruptible power supply installation and channel replacements. No inadequacies were noted relative to this LE LER 90-28, Revision 1, updated the root causes and corrective actions for non-conservative intermediate range monitor trip setpoints that occurred during a restart from a 1990 refueling

• outage. Unresolved item 272 and 311190-19-05 was opened pending an LER revision. No inadequacies were noted relative to the LER. Based on the licensee's corrective actions in the LER, the unresolved item is considered close * LER 91-12 concerned a 2A safeguards *equipment control (SEC) system actuation due to.

personnel error during surveillance testing. The event was reviewed in NRC Inspection 50-272/91-23. No inadequacies were noted relative to this LE LER 91-13 concerned four radiation monitoring system (RMS) actuations due to equipment design and/or component failure. These events were reviewed during NRC Inspection 50-272/91-23. Corrective actions included returning each RMS channel to service after repair, modifications scheduled for an uninterruptible power supply, and longer term equipment upgrades/replacements. No inadequacies were noted relative to this LER. _

Hope Creek LER LER 91-18 (See Section 7.2.B)

9.2 Open Items The following previous inspection items were followed up during this inspection and are tabulated below for cross reference purpose Report Section

. 272&311/90-19-_01 272&3ll/90-12-01, 03 272&311/90-19-05 272&311/90-81-15 272/91-25-02 Hope Creek 354/90-01-02 7..3..1 4.3..2..2. Closed Closed

' Closed Closed Open Closed*

1 EXIT INTERVIEWS/MEETINGS 1 Resident Exit Meeting The inspectors met with Mr. C. Vondra and Mr. J. Hagan and other PSE&G personnel periodically and at the end of the inspection report period to summarize the scope and findings of their inspection activitie Based on Region I review.and discussions with PSE&G, it was determined that this report does not contain information subject to 10 CFR 2 restriction.2 Specialist Entrance and Exit Meetings Date(s)

9/9-13/91 Subject Radiological Control Inspection Report N &311/91-25 Reporting..

Inspector Nimitz

• ATTACHMENT 1 Persons contacted during 50.59. Inspection at*Salem Nuclear Re~ulatory Commission

• J. Stone, Salem Project Manager, NRR

• S. Barr, Salem Resident Inspector

• S. Pindale, Salem Resident Inspector Public Service Electric and Gas Company

• E.H. Villar, Station Licensing Engineer

• John Hodson; Offsite Safety Review Engineer

• Scott Gillespie, Principal Safety Review Engineer, PWR

• J.M. Eggers, Licensing & Regulation

• V.J. Polizzi, Operations Manager

• M. Shedlock, Maintenance Manager

• T. Cell mer, Radiation Protection & Chemical Manager

• D. Lyons, Technical Engineer

• W.R. Schultz, Manager - Station Quality Assurance

• B.A. Preston, Manager - Salem Projects D. McCollum, Senior Project Engineer M. Alpaugh, Lead Engineer S. Pace, Lead Engineer F. Wiltsee, Engineer W. Centinaro, Engineer R. Kirk, Lead Engineer J. Wiedemann, Senior Staff Engineer OTHER

• E. Krufka, Lead Engineer, Atlantic Electric Persons contacted during 50.59 Inspection at Hope Creek Nuclear Reeulatory Commission

• S. Dembek, Hope Creek Project Manager, NRR

• K. Lathrop, Hope Creek Res.ident Inspector

• Public Service Electric and Gas Company

• D. Smith, *Station Licensing Engineer

• R. Beckwith, Station Licensing Engineer

• B. Hall, Technical Manager - Hope Creek

• R. Hovey, Operations Manager - Hope Creek

• Attended Exit on September 12, 1991 Following is a list of 50.59 Reviews and Safety Evaluations, except for procedures, that were examined during this inspection: Deficiency Report SMD-91-34 7, Loop 13 T-Cold RTD Replacement at Powe.

S-C-F300-MSE-0706-3, Packing Adjustment of BF-19 Valve.

1 EC-3065, Pkg 1, Narrow Range RTD Conax Connection Replacemen.

FSAR Change 91-04, Replacement of #12 Service Water Pum.

TMR 91-029, Temporary Power to Fuel Handling Buildin.

TMR 91-031, Mechanically Gagging BIT Relief Valve, 2SJ1.

Deficiency Report SMD 91-354, Repair of 10 inch service water lin.

Deficiency Report SMD 91-339, Repair of operator on valve 1ST90.

Temporary Modification 91-078, Repair of body to bonnet leak, valve 13RD8.

Temporary Modification 91-080, Repair of valve body leak, valve 21BS4 Following is a list of 50.59 reviews and safety evaluations for procedures that were examined: lIC-4.1.072, Rev. 5, 1R44A Containment High Range Channel Calibration Procedure. 50.59 Revie.

S 1.IC-FT.RCP-0033, -0034, -0035(Q), Rev. 0, Functional tests of steam generator narrow range water level transmitters. 50.59 Revie.

S2.0P-PT.SW-0021(Q), Rev. 1, No. 21 Containment Fan Coil Unit Heat Transfer Performance Data Collection. 50.59 Review. SC.MD-PM.CH-0003(Q), Rev. 1, Chiller Condenser Recirculating and Chilled Water Pump, Pump Internal Inspectio ~59 Revie.

2IC-2.10.202, Rev. 4,. 2FT-947 #22 Residual Heat Exchanger Outlet Flow for Safety Injectio ~59 Revie.

SC.MD-GP.ZZ.0022(Q), Rev. 1, Bolt Torquing and Bolting Sequence Ouidelines. 50.59 Revie.

Sl.OP-SO.CVC-0018(Q), Rev. 0, 11 Evaporator Distillate Ion Exch~ge Resin Removal. 50.59 Revie.

• Sl.OP-PT.SW-002l(Q), Rev. 1, Containment Fan Cooler Unit testing. 50.59 Revie.

Sl and S2.0P-ST.FHV-0001(Q), Rev. 0, Surveillance test of HEPA and charcoal filters. 50.59 Revie.

SC.MD-CM.CBY-OOOl(Q), Rev. 0, Containment Fan Cooler Units maintenance. 50.59 Revie.

SC.MD-PM.ZZ-0042(Q), Rev. 0, Inspection of Mission Duo-check valves in service water system. 50.59 Revie.

SC.MD-PM.OG-0019(Q), Rev. 0, Diesel preventive maintenance. 50.59.

Revie.

SC-MD-IS.4KY-0001(Q), Rev, 0, Inspections and testing of GE Magne-Blast circuit breakers. 50. 59 Revie.

Sl.OP-AB.ZZ-0002(Q), Rev. 0, Internal Flooding of Power Plant Building.59 Review and Evaluatio.

Sl.OP-PT.SW-0002(Q), Rev, 0, Flushing of service water system dead leg.59 Review and Evaluation.