IR 05000272/1998007
| ML18106A804 | |
| Person / Time | |
|---|---|
| Site: | Salem  |
| Issue date: | 08/04/1998 |
| From: | NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | |
| Shared Package | |
| ML18106A803 | List: |
| References | |
| 50-272-98-07, 50-272-98-7, 50-311-98-07, 50-311-98-7, NUDOCS 9808110180 | |
| Download: ML18106A804 (23) | |
Text
'.
U. S. NUCLEAR REGULATORY COMMISSION Docket Nos:
License Nos:
Report Nos:
Licensee:
Facility: *
Location:
Dates:
Inspectors:
Approved by:
9808110180 980804 PDR ADOCK 05000272 G
REGION I
50-272, 50-311 DPR-70, DPR-75 50-272/98-07, 50-311 /98-07 Public Service Electric and Gas Company Salem Nuclear Generating Station, Units 1.and 2 Hancocks Bridge, NJ.
June 29, 1998 - July 16, 1998 Aniello L. Della Greca, Sr. Reactor Engineer, EEB, DRS Jimi Yerokun, Sr. Reactor Engineer, SEB, DRS William H. Ruland, Chief Electrical Engineering Branch Division of Reactor Safety
- EXECUTIVE SUMMARY Salem Inspection Reports 50-272; 311 /98-07 June 29, 1998 - July 16, 1998 This inspection reviewed engineering activities related to previously identified issues. The report covers a two-week onsite inspection and two additional days of followup inspection at the NRC Region I offic Engineering Based on the their review of applicable engineering and other plant documents, the inspectors concluded the following:
The inspectors identified no areas of concern regarding the safety evaluation and conclusions in the UFSAR change that formalized the use of flow transmitters with variable damping in the steam lines. (E8.1)
Acceptable actions were taken to address the MOV design-basis setpoint configuration control, and to validate the RHR pump curves for the required flow and NPSH in the recirculation mode of operation. (E8.2 and E8.10)
The calculations and associated engineering evaluations for the component cooling flow acceptance criteria, for the verification of computer generated results and for establishing operator required time to complete the switchover of the ECCS pump suction from the RWST to the containment sump were comprehensive and technically sound. [E8.4, 6, and 11]
The original ( 1994) operability determination regarding the slow voltage buildup of the emergency diesel generator was inaccurate, but the subsequent (1997)
evaluation properly characterized the issue. [E8. 7]
The licensee had adequate measures in place to be able to comply with the requirements of TS section 3.5.3.b.2 regarding the ability to establish RHR flow into the RCS hot legs during a LOCA. [E8.8]
The proposed change to the surveillance procedure did not adequately address the TS requirement regarding periodic verification of the ability of CREACS to remove the assumed heat load. However, the procedure had not yet been revised, no violation of NRC requirements occurred. [EB.13]
The licensee-selected method for measuring. the primary containment average air temperature was representative of the overall containment atmosphere and the safety impact of using a potentially incorrect method would be minimal. The licensee acceptably satisfied the surveillance requirements of the Salem technical specifications and the intent of the TS bases. (E8.14)
The actions taken to address the Votes testing deficiencies and the missed testing of a containment electrical penetration breaker were acceptable. (EB.16 and 17)
ii
..
Report Details From June 29 to July 17, 1998, the NRC condu'cted an engineering inspection at the Salem Nuclear Generating Station. The objective of the inspection was to assess the effectiveness of the engineering functions in providing for the safe operation of the plan The assessment was accomplished through a review of the licensee's actions to address previously identified issues. During the inspection period, both Salem units remained at or near full powe Ill. Engineering ES Miscellaneous Engineering Issues (IP 92903)
E (Closed) Unresolved Item 50-272; 311 /96-07-07: Safety Evaluation Under 10 CFR 50.59 Design Change Package 1 ED-3328 was issued to prevent inadvertent safety *
injections due to spurious high steam flow signals during turbine stop valve closur This issue was addressed during the NRC review of Salem Restart Issue 11.38. At the time of the original review the inspectors identified four issues that required licensee clarification. Three of the four items were closed in a subsequent review, as documented in Inspection Report (IR) 50-272; 311 /97-08. The fourth issue, pertaining to the use of a time-delay function in the transmitter electronic board, remained open pending PSE&G's completion of UFSAR clarifications and the associated safety evaluation. The purpose of this inspection was to verify completion of these activitie Change Notice 97-08 addressed the addition of a new paragraph to section 10.3.2.3 of the UFSAR. This new paragraph formalized the use of transmitters with.
variable damping adjusted to minimize spurious transient high steam flow signal The safety evaluation associated with the change properly addressed the impact of the signal delay on the setpoints stated in the Technical Specification, had undergone peer review, and had been approved by SOR The original concern with this issue was that the addition of damping to the operation of the transmitter might delay the initiation of associated safety function The inspectors' review of this issue, during the current inspection, determined that margins as defined by 10 CFR 50.59 were already included in the determinations of the technical specification setpoints. Therefore, no further concerns and no violations of NRC requirements existed with this issue. In addition, the inspectors had no concern with the safety evaluation for the UFSAR revision and the conclusions contained therein. This item is close E (Closed) Unresolved Item 50-272; 311/96-11-10: Review Corrective Actions to AR 960607116 In September 1996, the NRC review of the Salem Unit 2 Motor Operated Valve (MOV) program, NRC Restart Issue 111.a.23, identified that past plant changes had been implemented without appropriate consideration given to their impact on the MOV design-basis setpoint documents. The purpose of this inspection was to review the actions taken by the licensee to ensure that future design changes were properly factored in the MOV progra Resolution of the configuration control issues was tracked by AR 96060711 Since its original issue, on June 7, 1996, this AR became a vehicle for addressing and tracking a variety of MOV-related issues. Currently, the AR included 34 actio requests (CRCAs), some of which were issued merely to transfer the closure responsibility from one engineer to another. Most of the issues in the AR were closed. A few remained open, including CRCA No. 29, related to the updating of MOV calculations, and CRCA No. 34 regarding revision of the design change package (DCP) specialty review checklis Following appropriate program changes, the NRC found the Salem Unit 2 MOV program acceptable, as noted in Inspection Report 50-272; 311 /97-03. As a result, NRC Restart Issue 111.a.23, Adequacy of Motor Operated Valve Program, was closed for Unit 2. The MOV program was reviewed again in March 1998 to address the Unit 1-specific program and again it was found acceptable. The AR issues, including the updating of MOV calculations, are program enhancements that are being properly tracked and resolved by the licensee in accordance with their corrective action progra CRCA No. 34, addressing the configuration control concern expressed by the NRC during the original review of the MOV program, was scheduled for closure by July 15, 1998, but the actions to address the concerns were still under review during the inspection. Following the inspection, on July 22, 1998, the licensee provided a copy of procedure NC.DE-AP.ZZ-007(0), Revision 10, used to address required specialty review of design or configuration change packages, both major and minor, permanent and temporary. The checklist associated with this procedure had been revised to clarify the conditions when MOV specialty review is require The inspectors' review of the proposed procedure changes found them acceptable and sufficient to address the NRC configuration control concerns. Regarding past configuration control, no specific deficiencies were previously identified by the NR Also, as stated above, the NRC found the MOV program was acceptabl Therefore, no violation of NRC requirement occurred with this issue. This item is close E (Closed) Inspection Follow-up Item 50-272/96-21-09: Complete Evaluation and Work On Unit 1 SJ Hot and Cold Leg Injection Valves DCP 2EC-3461, involving the design and installation of orifice plates in series with the Unit 2 cold and hot leg injection line throttle valves, was issued to remove the potential of safety injection (SI) pump runout damage. The actions to address the concerns related to this issue were evaluated and found acceptable during the NRC review of Salem Restart Issue 11.35, as documented in IR 50-272; 311 /96-20. The subject item was opened to followup on the closure of the DCP applicable to Unit The installation and testing of the design changes applicable to Unit 1 were reviewed during the NRC review of Restart Issue 11.35 for Unit 1. As documented in IR 50-272; 311 /97-21, the applicable DCP, 1 EC-3530, had been implemented. The inspectors' review of the post-modification testing, at that time, also determined that the flow orifices effectively mitigated cavitation in the cold and hot leg injection lines and provided the required runout protection for the charging and safety injection pumps. This item is close E (Closed) Unresolved Item 50-311/96-81-04: aasis for Component Cooling System Flow Balance Test Acceptance Criteria During a NRC inspection (Report 50-311 /96-81) the team identified a discrepancy in a flow balance test of the Component Cooling (CC) System. The concern was that the licensee did not have any documented calculations to support the acceptance criteria specified in the flow balance test procedure. The purpose of this inspection was to evaluate the licensee's resolution of this issu The licensee stated that, although they had not documented their bases in a calculation, the acceptance criteria were based on UFSAR design flows plus some margin to account for instrument error and pump degradation. Nonetheless, to resolve the NRC concern, the license prepared calculation, S-C-CC-MDC-1751, Component Cooling Flow Balance Acceptance Criteria, reflecting the acceptance criteria for component cooling flow to the safety loads for the CC system. The inspectors reviewed the calculation and the associated engineering evaluation (S-2-CC-MEE-1176, Revision 0, Component Cooling Safety Load Flow Requirements -
Unit 2), and found them to be comprehensive and technically sound. Appropriate considerations for instrument uncertainties and pump NPSH and degradation had been made in determining the quantity of flows from the CC pumps to their safety loads. The calculation had been reviewed and certified for design verification by the licensee in accordance with plant procedure NC.DE-AP.ZZ-0010-1. The calculation confirmed the adequacy of the acceptance criteri The inspectors found the licensee's actions to be appropriate. Based on the results of the calculation, the inspectors also concluded that no violations of NRC requirements had occurred with this issue. This item is close E (Closed) Unresolved Item 50-272; 311 /97-05-02: Electrical Penetration Backup Protection In a March 1997 inspection the NRC determined that, for 18 circuits, overload protection of electric containment penetration assemblies was not acceptable in that it did not meet the intent of Regulatory Guide (RG) 1.63. Specifically, the NRC determined that, for these circuits, the backup overcurrent sensors did not provide protection for high impedance faults that are less than six to seven times the circuit full load for more than approximately 70 seconds. PSE&G disagreed with the inspector position regarding conformance with the regulatory guide stating that RG 1.63, Revision 2, and earlier versions, required backup protection only at the maximum short circuit. PSE&G also stated that the NRC had previously accepted the current desig In a subsequent review (IR 50-272; 311 /97-08) the inspector reevaluated the technical issue and concluded that, for the circuits in question, it presented only a limited safety concern. Therefore, it was not a restart issue. However, the licensing aspect remained opened pending further review of the NRG-accepted design during the Salem licensing phas *
The NRC completed their review of the regulatory guidelines that were developed and used during the Salem licensing phase with the following conclusions:
For Unit 1, the NRG-accepted design did not explicitly require the installation of backup protective devices to assure protection of electric penetrations. Although not specifically documented, at that time, the NRC apparently believed that smaller sized conductors, located outside the penetration assembly would f.use open before the larger sized penetration *conductor and provide adequate protection. This was in conformance with the guidelines of Position C.1 of RG 1.63, Revision 0. RG 1.63, Revision 0, and IEEE 317-1972 are the licensing bases stated in the Salem FSAR of both unit For Unit 2, the NRG-accepted design explicitly required the installation of backup protection devices to assure protection of electric penetrations. Cables smaller than the penetration could not be used as fuses. This was documented in supplement 3 of the Salem Safety Evaluation Report (SEA). Consistent with the licensee's *
position, Supplement 4 of the SEA indicated that backup protection needed only be capable of interrupting the maximum available short circuit current. Supplements 3 and 4 to the SEA are also considered to be part of the Salem licensing base Therefore, the NRC agreed that the Salem current design was the accepted desig This item is close E (Closed) Unresolved Item 50-311197-06-02: Vendor Computer Software Controls While observing the control room area leakage testing, the inspectors noted that the leakage calculation had been performed by computer and that the computer printout of inputs and results showed the initial and final control room pressure to be reversed. In addition, the inspectors were unable to repeat the computer results using hand calculations and the formula specified in the test procedure. Thus, the inspectors questioned the licensee's controls placed on the software used for the leakage calculation and for software in general. The purpose of this inspection was to review the licensee's evaluation of the NRC concerns and the actions thereo The licensee evaluated the issue in performance improvement request (PIR)
No. 970415322. They determined that the inputs to the computer generated report were actual field-generated data. To verify the accuracy of the results, they prepared Engineering Evaluation (E.E.) No. S-C-SC-MEE-1212. This evaluation recreated the calculations performed by computer and confirmed that the computer results were accurate. The inspectors' review of the E.E. found it to be technically sound and properly focuse The inspectors also reviewed the discrepant data identified in the inspection repor They found that, although the initial and final pressure used as input in the leakage equation were. reversed, the results were not impacted.because the direction of the leakage was not defined. Regarding the accuracy of the results, an independent calculation performed by the inspectors identified negligible differences in the fifth decimal point that may be the result of the amount of significant digits used in the two calculation *
- Regarding the controls placed by the licensee on the use of software in general and in the leakage calculation, in particular, the inspectors determined that the requirements were specified in administrative procedure.NC.NA-AP.ZZ-0064(0). A summary review of this procedure identified no concerns regarding its requirements and application. In the specific application, the inspectors were unable to determine whether the reviewer, who was no longer employed by the licensee, had conducted
- a review of the program used in the spreadsheet to calculate the leakage or verified
.the results by independent method as required by 10 CFR 50, Appendix However, the licensee's subsequent verification and the inspectors' independent review showed that the results were correct. Therefore, no violation of NRC requirements occurred. This item is close ES. 7
{Closed) Unresolved Item 50-272: 311 /97-08-01: Operability Determination and Reportability of EOG Loadin In December 1994, during a mode operation test, the licensee observed that the emergency diesel generator {EOG) output circuit breaker had closed when the voltage was only 2200 V. The original licensee evaluation concluded that the slow voltage buildup was due to a voltage regulator anomaly and that the EOG was operable because the required voltage was available to the bus within the allowable time stated in the Technical Specification {TS). Salem TS surveillance requirement 4.8.1.1.2.a.2 specifies that the diesel generator shall be demonstrated operable by verifying that the voltage is between 3950 and 4580 V within 13 seconds after the start signa Subsequently, in February 1995, the licensee discovered that the voltage permissive-relay in the breaker close circuit allowed closure of the breaker when the voltage was only at 25 % of the nominal voltage. The licensee concluded again that the EDq was operable for the reasons stated in the original review, but initiated action to add a new relay that would not allow closure of the breaker until the EOG output voltage had reached the proper level (90% of the nominal voltage).
Because the event was in part due to a design deficiency that affected all EDGs of both Units, during the 1997 review of the issue, the inspectors questioned its reportability. The inspectors had several concerns, including adequacy of the first operability determination (OD) and the need to inform the industry of the deficient design. Reportability had also been questioned by the Site Operation Review Committee (SORC) during their review of safety evaluation No 97-168 associated with the above modification. The SORC open item, however, had been incorrectly closed.*
During the current review, the NRC inspectors determined that an engineering evaluation performed to address the NRC question {PIR 970717267) had concluded that the original 10 CFR 50.59 evaluation was faulty in that it had not addresse the fact that the EOG test did not duplicate design basis accident conditions. The licensee also concluded that the rate of rise of the EOG voltage during the test provided reasonable ai;;surance that the EDG would have been able to carry the design basis loads without stalling. Therefore, they recommended that Licensing issue a voluntary Licensee Event Report. This action was originally scheduled for completion on-August 18, 1997, but due date was deferred *several times and by the end of inspection no report had been issue Because the 1997 evaluation determined, by engineering judgement, that reasonable assurance existed that the diesel would be capable of performing its intended function, the inspectors concluded that no violation of NRC requirements had occurred regarding reportability and the licensee's decision to issue a voluntary report was appropriate. This item is close E8.8 (Closed) Unresolved Item 50-311/97-11-05: Adequacy of Abnormal Operating Procedure for Establishing RHR Flow into RCS Hot Legs during a LOCA in Mode 4 During a NRC inspection (Report 50-311 /97-11), inspectors identified a concern involving abnormal operating procedure (AOP) S2.0P-AB.LOCA-0001 (Q), Shutdown LOCA, Rev. _1. The procedure did not articulate specific steps for establishing hot leg recirculation through any flow path. Rather, the procedure relegated the responsibility for developing specific guidance to the Technical Support Center Engineering Staff. The issue arose during the inspectors' review. of the licensee's methods for satisfying the requirements of Technical Specification 3.5.3.b.2 for a RHR pump flow path capable, upon manual initiation, of discharging into two RCS hot legs. The issue was left unresolved pending verification of adequacy of the abnormal operating procedur The licensee had maintained that the AOP was consistent with the generic Westinghouse guidelines (Abnormal Response Guidelines, ARG-2, Shutdown LOCA, Revision 0, February 28, 1992) used for developing it. Nevertheless, they enhanced the AOP to include a two page set of instructions for establishing the hot leg recirculation flow path. The inspectors reviewed the Westinghouse guidelines and verified that the licensee's assertion had been accurate and that the AOP was consistent with the guidelines. The inspectors also reviewed the AOP enhancement and identified no discrepancy. The inspectors concluded that the licensee had, and continued to have, adequate measures in place to be able to comply with the requirements of TS 3.5.3.b.2; Therefore, no violation of NRC requirements had occurred. This item is close E (Closed) Inspector Followup Item 50 -33/97-11-06: Deletion of RHR Hot Leg Flow Path From Licensing Basis in Apparent Conflict with Technical Specification The NRC inspectors identified that the licensee had deleted the RHR hot leg flow path through valve RH26 from the Unit 2 design and licensing basis. It was not clear if. the deletion was in conflict with the technical specifications (TS) or no The issue was identified as an inspector follow up item so that the inspectors could *
further review and determine if a concern existed or no During the current review, the inspectors found that the deletion of the RHR hot leg flow path through valve RH26 from the EOP and FSAR was in accordance with a 10 CFR 50.59 Safety Evaluation. The licensee had determined that no TS change was necessary because the intent of TS 3.5.2.c.2 and 3.5.3.b.2 to have a flow path to two RHR hot legs was still met. Without RH26, the flow path was available via the intermediate safety injection pumps. The inspectors reviewed the licensee's determination that the change to the hot leg flow path did not require. a change to the technical specification and concurred with the licensee's determinatio Subsequently, however, PSE&G changed the TS to remove the hot leg path requirement of TS 3.5.2 in modes 1, 2 and 3. (NRC approved SEA dated September 11, 1997). Also, procedure AOP S2.0P-AB.LOCA-0001 (Q), Shutdown LOCA, was revised to provide the specific steps to take to establish hot leg recirculation when needed. The inspectors found the licensee's actions to be appropriate. This item is close E8.10 (Closed) Unresolved Item 50-311 /97-11-07: Unvalidated RHR Pumps NPSHr for Recirculation Mode NRC inspectors identified a concern with the. licensee's determination of the NPSH for the Unit 2 RHR pumps during Recirculation Mode of Operation. Specifically, the inspectors found that the RHR pump curves used did not show NPSH data beyond a pump flow of approximately 4,800 gpm and that the licensee, in their calculations, had assume an extrapolated value for NPSH required (NPSHr) of 24 feet for pump
- flow at 4,900 gpm. This value had not been validated by testin The inspectors had also questioned the licensee's use of "Containment Overpressure" in the NPSH available (NPSHa) calculations since the credit for
. containment pressure only applied to Unit 1. Licensee's calculation FSE/SS-PSE/PNJ-2017, "Salem Unit 1 and 2 ECCS Pump RHRS Recirculation," dated December 6, 1993, assumed the containment pressure to exist at the onset of a LOCA. *The assumption was inconsistent with the Salem 2 licensing basis as described in Section 6.1 and Appendix 3A of the UFSAR, and NRC Regulatory Guide 1. 1 as clarified in the NRC' s Standard Review Plan dated July 1981. This issue was left unresolved pending justification of the RHR pumps NPSH values used in calculation During this inspection, the inspectors reviewed the licensee's follow up actions. To resolve the questions associated with NPSHr, the licensee obtained a revised pump curve from the vendor_that showed the NPSHr for.flows up to 5,200 gpm. The
.
inspectors reviewed the pump curve and found that it appropriately reflected NPSHr data for flows above the 4,900 gpm used in the NPSH analysis. The inspectors also verified that the part numbers specified on the pump curve matched the licensee's inventory parts identification of the installed RHR pumps. For the NPSHa question, the licensee completed a revised calculation (.10 CFR 50.59 SE Cal #S-C-RHR-MDC-1711, Rev. 2) which did not take credit for containment overpressure. The calculated NPSHa was still more than the NPSHr for the recirculation mode of operation. For added conservatism, the containment sump level used for cold leg recirculation NPSH considerations was 80 feet (instead of the
81 '-7" to 81 '-9" previously submitted) to accommodate instrument uncertaintie The results showed that for cold leg recirculation at a flowrate of 4,900 gpm, the NPSHa would be 26. 7 ft versus a NPSHr of 22.8,ft.. For.hot leg recirculation at a flowrate of 4,980 gpm, the NPSHa would be 28.1 ft versus a NPSHr of 24.0 f Based on the above, the inspectors concluded that the licensee's actions to address this discreP.ancy had beer.i appropriate and no other concern was identifie Because the broader issues* associated with the RHR NPSH margin were identified as a violation of 10 CFR 50.59 (EEi 50-311/97-11-04)the inspectors also concluded that no additional violation existed. This item is close E8.11 (Closed) Unresolved Item 50-311 /97-11-09: Operator Ability to Perform Timely Switchover of ECCS Pumps Suction from RWST to Containment Sump Dudng a NRC inspection (Report 50-311 /97-11), inspectors identified a concern involving the ability of the operators to complete the switchover of Emergency Core Cooling System (ECCS) pumps suction from the Refueling Water Storage Tank (RWST) to the Containment Sump for the recirculation phase of a Small Break Loss of Coolant Accident (SBLOCA). Specifically, it was not clear if operators could complete the switchover successfully within an.appropriate, licensed time~ This issue was left unresolved pending completion of licensee actions to confirm the appropriate time and to demonstrate the ability of the operators to complete the switchover within that tim The inspectors had identified that the operator response times were more restrictive than originally assumed and licensed, and that the critical task times trained on in the simulator were incorrect. For the most limiting Large Break LOCA (LBLOCA)
case, the switchover (during the time frame between the RWST low and low-low level alarms) must be accomplished within 7.8 minutes; faster than originally licensed. Operators were judged on. completing the transfer in 11.8 minutes; and, they typically required approximately 9.5 minutes to complete all of the switchover steps. The inspectors considered the 11.8 minute acceptance criterion to be invalid, since it included an additional 1.8 minutes where forced recirculation flow to the core is stopped in a SBLOCA. It was also unclear whether operators could successfully perform the switchover even in the initial 10 minute period of a SBLOC The licensee subsequently determined that the appropriate time to complete the switchover was 9.5 minute during a LBLOCA and 11.2 minute during a SBLOC This determination was documented in an analysis titled, "10 CFR 50.59 SE, Update to EOP-LOCA-3 and FSAR: Revised RWST Draindown Evaluation."
Operators were trained and verified to be able to accomplish the task within the required time per commitment 97-0129, Train Operating Crew on The Revision to 2-EOP-LOCA-3, for New Response Time. The licensee notified the NRC in a letter dated June 13, 1997, of completion of efforts involving changes to procedures and operator training and validation to address this issu *
During the current inspection, the inspectors reviewed the new analysis, and the operator training documentation and found the licensee's actions to be complete.
and appropriate. Despite the discrepancy in estimated required times for accomplishing the switchover, operator response time was always within the worst case required time. Because this issue was also identified to be a violation of 10 CFR 50.59 (EEi 97-l82-01013), as also documented in section E8.18 of this report, no additional violation of NRC requirements was assessed. This item is close E8.12 (Closed) Unresolved Item 50-311/97-11-10: Use of Undocumented and Unlicensed
"NOSI" Code for Unit 2 Specific Design During a NRC inspection (Report 50-311 /97-11), the inspectors identified a concern involving what appeared to be the licensee's use of an Unlicensed code ("NOSI") for Salem Unit 2 specific Small Break Loss of Coolant Accident (SBLOCA) analysi They noted that, for long-term cooling, a licensee's March 1996 safety evaluation stated that "... the 10 CFR 50.46 analysis of record for Salem is based on limiting plant performance characteristics documented in Westinghouse letter NSAL-95-001. " This letter indicated that ECCS flow may need to be interrupted during switchover of ECCS pumps suction from the RWST to the containment SL!mp. A generic analysis using the "NOSI" code, which took credit for the downcomer flow during the swapover from RWST to Containment Sump mode of a SBLOCA, had been developed to study the effects of short periods of no ECCS flow during the switchover. However, the "NOSI" code was not licensed for use at Salem Unit The inspectors were concerned that the licensee was incorporating generic analyses into the Salem Unit 2 design basis using the undocumented (and unlicensed) "NOSI" code as the basis for a 1.8 minute SBLOCA period of no pumped flow to the cor During the current review, the inspectors determined that a subsequent evaluation of the issue by the licensee had found that the Unit 2 swapover analysis did not include the use of the "NOSI" code. The licensee also found that, although the Salem 2 specific analysis was based on Westinghouse letter NSAL-95-001, no credit had been taken for the effects covered by the "NOSI" code. This was later documented in a Westinghouse letter, dated July 16, 1998, to PSE&G. In this letter, Westinghouse specifically confirmed that the "NOSI" code had not been used for the Salem 2 specific analysis. Also, a new drain down analysis, performed by
- Westinghouse on May 20, 1997, clearly stated the exclusion of the use of the
"NOSI" code. The "NOSI" code reference had been deleted from the UFSA Based on their review of the Westinghouse analysis and the confirmation letter, the inspectors concluded that the licensee's actions were appropriate and no violation
'
of NRC requirements existed with this issue. This item is close E8.13 (Updated) Inspection Followup Item 50-311/97-16-03: Conformance of Calculations and Procedures with TS 4. 7.6.1.d(5)
Technical Specification (TS) surveillance requir.ement 4. 7.6.1.d(5} requires that every 18 months the licensee verify that each control room emergency air conditioning system (CREACS) train has the capability to remove the assumed heat
- load. During the Unit 2 power ascension testing, in July 1997, the NRC observed that this requirement was not being addressed by the surveillance testing. The inspectors also observed that some of the assumptions in heat load calculation S-C-CAV-MDC-1569 were not correct. Furthermore, because of ventilation system changes, the heat generated in control room panels 1 (2)RP3 and 1 (2)RP4 was now being removed by CREACS. The TS requirements were to be addressed in a future revision of surveillance procedures S1.RA-ST.CAV-0004(0) and S2.RA-ST.CAV-0003(Q), for Units 1 and 2,* respectively. The scope of this inspection verified the status of the licensee proposed changes to address the identified issue Revision of the current surveillance procedures was proposed in performance improvement request (PIR), No. 980129233, initiated on February 5, 1998. The PIR stated that the procedures were insufficient to verify the TS requirement and recommended several actions to achieve compliance. Proposed actions included measurement of air and chilled water flow, development of criteria for chiller coil fouling, and visual inspection of dampers and chilled water valves.. In the subsequent PIR evaluation, the assigned engineer disagreed with the PIR recommendations and stated that compliance could be achieved by recording the control room temperature during the monthly operation of the CREACS. The evaluation stated that a temperature of less than 85 ° F would demonstrate the ability of the CREACS trains to remove the assumed heat load. The evaluation had been closed without appropriate review and approva On April 17, 1998, the licensee issued another PIR, No. 980415202, to address procedural improprieties committed during closure of the first PIR. The subsequent evaluation addressed the previous procedural improprieties, but did not change the resolution method for the issue. A corrective action item was issued to revise the above procedures and include a step that would record the control room temperature during the performance of the surveillance test. The procedure itself had not been revised, but the evaluation had received the required approva The inspectors disagreed with the resolution of the issue for the following reasons:
( 1) measuring the control room temperature addressed only the ability of CREA CS to maintain the stated temperature at the current heat loading and cooling condition; (2) no specific basis had been provided for the acceptance criterion of 85 ° F; (3) the temperature measurement did not verify the capability of the system to remove the assumed heat load, as required by the TS; and (4) the temperature measurement did not consider system conditions such as outside air temperature, water temperature, chilled water flow, and chiller efficienc The inspectors concluded that the proposed change to the surveillance procedure did not adequately address the TS requirement regarding periodic verification of the ability of CREACS to remove the assumed load. However, because the procedure had not yet been revised, the inspectors also concluded that no violation of NRC requirements had occurred. This issue remains an inspector followup iterri pending completion of PSE&G revision of the surveillance procedure and subsequent NRC revie In conjunction with this review, the inspectors observed that, although heat load calculation S-C-CAV-MDC-1569 had been reissued in October 1997, the revision had not addressed minor discrepancies identified previously by the NRG. Therefore, the exact value of the heat loads on CREACS was not known. The licensee initiated action to revise the calculation prior to the issuance of the surveillance procedur E8.14 (Closed) Unresolved Item 50-272; 311 /97-020-01: Containment Air Temperature
Averaging The containment average air temperature is an input to the plant accident analysi Licensee Event Report (LER) No.95-004, dated May 18, 1995*, informed the NRG that the calculation of the primary containment average temperature had not been done in strict compliance with the surveillance requirements specified in Section 4.6. 1.5 of the Salem, Units 1 and 2, technical specifications. A NRG inspection of this issue, completed on November 13, 1997, determined that, since 1995, the licensee had used three methods for calculating the average temperature and that the current method, the arithmetic average of ten thermocouples, was the least conservative of the three. At that time, the inspectors also determined that the licensee had conducted a survey of the containment temperature and was evaluating the results. The purpose of this inspection was to review the results of the licensee's evaluation and the resulting corrective action,_
The engineering evaluation performed by the licensee, following the November 1997 inspection, is contained in E.E. No. S-C-C.V.-MEE-1280, Revision 0, dated January 29, 1998, "Salem Unit 1 and 2 Containment Temperature Monitoring Design Basis." This document did not address the results of the temperature survey. Instead, as indicated by the title, it had addressed only the design basi * The licensee had, however, submitted to Westinghouse a copy of an earlier evaluation, S2-CBV-MEE-1115, Revision 1, "Salem Unit 2 Containment Temperature Monitoring Design Basis." In a letter, dated June 24, 1998, Westinghouse stated that: "... strictly from the perspective of Containment Integrity [group]... it is reasonable to use an arithmetic average of 10 thermocouples (T/C)inside containment in order to determine the bulk average containment temperature."
Because the licensee had not addressed the results of the survey, the inspectors evaluated three aspects of the issue, as described belo Design/Licensing Basis The current revision of the Salem Units 1 and 2 TS Section 3.6.1.5 states: "Primary containment average air temperature shall not exceed 1 20 ° F." The corresponding TS surveillance requirement and TS bases, Section 4.6.1.5, state, respectively:
"Verify containment average air temperature is within the limit at least once per twenty four hours." and "... an average [containment air temperature] is calculated using measurements taken at locations within containment selected to provide a representative sample of the overall containment atmosphere." As evident in the words of the TS surveillance requirement, the verification method is left to the discretion of the licensee. Furthermore, no guidance is provided in the TS bases of
how and when the measurements should be taken such that they are representative of the overall containment atmosphere and provide assurance that the TS limit is not exceede An earlier revision of the same TS surveillance section required: "The primary containment average air temperature shall be the arithmetical average of the temperature at any 5 of the 10 following locations.... " The ten locations were described by elevation and geographical direction. In this case, although the method of averaging was specified, the selection of the points was again left to the *
discretion of the license Based on the above, the inspectors concluded that the current method used by the licensee met the requirement of the TS and that the measurement locations met the intent of the TS bases. The inspectors also concluded that no violation of NRC requirements existe Technical Adequacy of the Averaging Method The determination.of an overall containment average air temperature is very difficult to obtain because it would require an impractical amount of thermocouples scattered throughout the containment. During the construction phase of the plant ten temperature detectors were installed for this purpose. The areas selected were considered to be "representative of the overall containment atmosphere" because the areas also included the majority bf the heat sinks within the containment. Both heat sinks and air contribute to the definition of the accident profile. Heat sinks are major contributors because of the large amount of heat energy stored within the Because some of the temperature detectors are in the lower region of the containment, they read a lower temperature and tend to lower the calculated arithmetic average. Nonetheless they do represent the temperature of the air in which they are located and the also represent the temperature of the heat sinks contained in the are As stated previously, during August and September 1997, while the plant was operating, the licensee performed a survey of the upper containment temperatur For each of the 14 locations monitored, the licensee had developed temperature versus time graphs and identified the maximum and average temperature for the period. The inspectors calculated that the arithmetic average of the average temperatures was 118 ° F and that the average of the maximum temperatures was approximately 127°F. Because the peak temperatures did not occur at the same time or same day, this latter temperature did not represent the average of any given day, only the average of the highest temperatures in the period. In addition, because the peak temperatures were, in the most part, of short duration, the
. Inspectors believed that a detailed evaluation of the graphs would show that the upper containment average for the hottest day of the monitor period was several degrees lower than the above calculated average maximu *
To further evaluate the significance of the survey data, the inspectors compared, for three days in September (for which data from the ten TCs could be retrieved), the average temperature recorded in the upper containment with the temperature recorded by the ten TCs. The comparison showed that, for the same day, the arithmetic average of the three TCs above the operating floor was approximately the same as that of the TCs in the upper containment, whereas the average of the ten thermocouples for each of the three days was 11 to 1 2 ° F lower than the ave.rage temperature in the upper containment for that da The above readings suggest that there may be a discrepancy between the ten-sensor arithmetic average method used by the licensee and the 14-thermocouple average in the upper containment. However, limited heat sinks exist above the operating floor. Therefore, an average that was based on the average air temperature in the upper containment would be unnecessarily conservativ Significance of Concern To evaluate the significance of an assumed 10°F error in the initial temperature, the inspectors discussed this possibility with Westinghouse. Telephone conversations conducted between the inspectors, licensee engineering and supervisory personnel, and a Westinghouse representative, determined that: ( 1) accident analyses by Westinghouse had conservatively assumed the containment air, heat sinks, and external air temperatures to be all at 120°F; (2) the design basis LOCA and a main steam line break (MSLB) would cause the highest temperature and pressure in the primary containment; (3) the composite peak accident temperature for the Salem primary containment was 350 ° F;. (4) Westinghouse had previously evaluated higher initial containment temperature for other plants; (5) based on the analyses performed for two other four-loop plants, an increase in initial containment average temperature of 10°F would result in a corresponding containment peak pressure increase of approximately 0.3-0.4 psig, for both the LOCA arid the MSLB; (6) for a LOCA, the expected peak temperature increase was less than 1.6°F; (7) for a MSLB, the expected peak temperature increase was less than 6.8°F; and (8) in either case, the peak temperature had a duration of less than two minute Based on the above, the inspectors concluded, as had Westinghouse, that an error of 10°F in initial average primary containment air temperature would have minimal effect on the containment integrity. The inspectors also concluded that an increase.
of the peak Salem.temperature from 350 to 357°F for less than two minutes would
- have minimal impact on the environmental qualification of the inside primary containment equipment, when thermal lag and physical locations of the equipment (mostly in the lower elevations) are also considere Conclusions *
The inspectors concluded that the licensee-selected method for measuring the primary containment average air temperature was representative of the overall containment atmosphere* and that the safety impact of using a potentially incorrect method would be minimal. In addition, the licensee acceptably satisfied the surveillance requirements of the Salem technical specifications and the intent of the TS bases. Therefore, no violation of NRC requirements occurred during and after the various surveillance procedure changes. This item is close ES.15 (Closed) Inspection Followup Item 50-272; 311 /97-021-03: Guidelines for Instrument Loop Design Loading The environmental qualification testing of the Rosemount transmitters assumed a maximum power dissipation by the transmitters of 0.36 Watts. The inspectors'
review of transmitter loops identified no violations of these transmitter design limitations. The licensee's ins~ructions for instrument loop design, however, did not include guidance in this are During the current review, the inspectors determined that the licensee had revised the Rosemount environmental qualification maintenance and surveillance information sheet to require a review of technical standard DE-TS.ZZ-1022(0) for transmitter power supply and loop loading restrictions and had revised the technical standard to reflect the transmitter loop design requirements. The inspectors also determined that the licensee had also revised the technical standard for the Salem and Hope Creek setpoint calculations, SC.DE-TS.ZZ-1904(0) and DE-TS.ZZ-1001 (0),
respectively. This item is close ES.16 (Closed) Violation 50-272: 311 /97-021-04: Inadequate Evaluation of Votes Test Results During a review of the closure package of Restart Issue 11.24, the inspectors identified two examples where PSE&G had failed to properly document and evaluate the results of VOTES testing and to assure that the test requirements had been properly satisfie In their response to the Notice of Violation, dated April 3, 1998, the licensee attributed the violation to personnel error and inattention to detail by technicians, during the performance and verification of calculations associated with the VOTES test procedur The licensee reviewed the errors and determine that no physical changes to the valves were required. The licensee also reviewed a sample of Salem Unit 2 and
- Hope Creek valves. They identified no errors. To address the personnel error issues, they discussed the NRC findings with Salem and Hope Creek MOV techn.icians and stressed the importance of accurate completion of procedure requirements. Lastly, to avoid future errors, they clarified applicable section of the test procedure. The inspectors reviewed the results of the licensee's valve survey and the changes to the test procedure and found them acceptable. This item is close ES.17 (Closed) Violation 50-311 /97-021-05: Failure to Include Containment Penetration Breaker in TS-Required Surveillance Program While reviewing the Salem Engineering Backlog, the inspectors determined that the licensee had failed to demonstrate operability of a molded-case circuit breaker used as a containment penetration conductor overcurrent device. The circuit breaker located in Panel 97-2 was not in the licensee testing program and had not undergone inspection or preventive maintenance in the pas In response to the Notice of Violation, letter dated April 3, 1998, PSE&G attributed the violation to inadequate configuration control and personnel error. The circuit breakers were in vendor-supplied panels, but the vendor.documentation did not accurately reflect the panels configuration. In addition, for Unit 2, the licensee believed that, in 1992, one of the breakers had been inadvertently deleted from the maintenance test procedur To address the issue, the licensee tested the circuit breakers that had not been previously tested, corrected the procedure, SC.MD-ST.ZZ-0004(0), to include the missed breakers, and creat.ed recurring surveillance tasks to ensure that the third breaker would be periodically tested in accordance with current TS requirement The licensee also revised applicable panel drawing and design calculations to reflect the correct plant configuration and document the breaker-required testing. Lastly, the licensee initiated an action item to be completed by July 30, 1998, to review a sample of vendor-supplied panels to ensure that similar conditions do not exist elsewher The inspectors verified that the breaker had been tested, that the procedure had been revised to include the missed breakers, and that recurring tasks had been created to include the breakers in the plant surveillance program. In addition, the inspectors reviewed the design change package and associated safety evaluation prepared to reflect the panels as-built conditions. Lastly, the inspectors verified records of the revised calculations. Based on the above review, the inspectors concluded that acceptable actions had been taken by the licensee to address the issue and the root cause of the violation. This item is close E8.18 (Closed) EEi 50-311/97-182-01013and EEi 50-311/97-182-01023:10 CFR 50.59 and 10 CFR 50. 72 (and 73) Violations Associated with ECCS Suction Switchover On October 8, 1997, a Notice of Violation (NOV) was transmitted to PSE&G concerning three violations of NRC requirements. Two of the violations were:
(1) 10 CFR 50.59 - for implementation of a change to the facility that involved an unreviewed safety question without NRC approval, and (2) 10 CFR 50. 7 2 and 10 CFR 50. 73 - for not reporting conditions outside the plant design basis. In PSE&G's response (dated November 11, 1997) to the NOV, the licensee concurred with the violations and discussed their corrective actions. The 10 CFR 50.59 violation occurred when the licensee approved changes to emergency operating procedures 2-EOP-LOCA-3, Transfer to Cold Leg Recirculation, and to UFSAR Section 6.3 in its July 10, 1994 Safety Evaluation. These changes, involving the RWST drain down and operator response time, increased the probability of occurrence of a malfunction of equipment important to safety and thus introduced unreviewed safety questions. The 10 CFR 50. 72 and 50. 73 violation occurred when as of April 17, 1997, the licensee failed to report the condition which had been identified on November 1, 199 During the current inspection, the inspectors reviewed the licensee's actions and LERs 50-272/97-009-00and 50-272/97-009-01 that were subsequently issue The LERs, the closure of which was documented in IR 50-272; 311/97-12, were accurate in their description of the issues. The review also determined that RWST drain down time had been re-evaluated (May 20, 1997). The inspectors identified no concerns with this re-evaluation. Procedure 2-EOP-LOCA-3 had also been revised (May 30, 1997) to support the new RWST drain down time and the.
operation crews had been trained accordingly. The root cause investigation, completed in May 1997, had identified some deficiencies in the 10 CFR 50.59 Safety Evaluation Program implementation. The actions prescribed by the root cause investigation to correct the identified deficiencies had been completed and a process for periodic self assessment of the program established in accordance with plant procedures. Based on the above review, the inspectors concluded that the licensee's actions were appropriate and these items (EEi 50-311/97-182-01013and 50-272/97-182-01023)are close IV. Plant Support FB Miscellaneous Fire Protection Issues (IP 92903)
F (Closed) Unresolved Item No. 50-272 & 50-311/97-09-06: Appendix R Exemptions This unresolved item questioned the technical basis for granted NRC exemptions from 10 CFR Part 50, Appendix R detection and suppression requirements in those plant areas where electrical raceway fire barrier systems (ERFBS) were provide Based on the issuance of the severity level IV violation (reference NRC letter dated October 8, 1997, enforcement action 02014) for the failure to install ERFBS consistent with the tested ERFBS configurations and PSE&G's commitment and Cable Raceway Fire Wrap Resolution Plan to address the indeterminate status of ERFBS, as discussed in PSE&G letters LR-N97320 and LR-N97357, this item is administratively closed. NRC concerns regarding licensee resolution of ERFBS discrepancies continue to be tracked via enforcement action 0201 F (Closed) Enforcement Item No. 50-272 & 50-311 /97-257-03014: Compensatory Fire Watches This item resulted in a severity level IV violation following an enforcement conference held with the licensee on July 10, 1997, for the licensee's failure to maintain compensatory measure fire watch patrols for inoperable ERFBS. Based on the adequacy of the licensee's corrective actions as discussed in PSE&G letter LR-N97320, dated May 16, 1997, and the NRC review of this issue in Janu.ary 1998, as documented in inspection report 97-21, section F8.8, this item is administratively close V. Management Meetings X1 Exit Meeting Summary The inspectors presented the inspection results to members of licensee management at the conclusion of the inspection on July 16, 1998. The licensee acknowledged the findings presente The inspectors asked the licensee whether any material reviewed during the inspection should be considered as proprietary information. No proprietary information was identifie *'
PARTIAL LIST OF PERSONS CONTACTED Public Service Electric and Gas Company C. Fricker D. Garchow S. Mannon D. McHugh M. Moncourtois G. Nagy J. O'Connor D. R. Powell R. Smith C. Smyth A. Spivak E. H. Villar J. Zudans Quality Assurance Director, Design Engineering Mechanical Design Engineering Supervisor, Electrical Engineering System Engineering *
r Salem System Engineering Manager Electrical Engineering Director, Licensing Salem Environmental Qualification Engineer Manager Salem Licensing Manager System Engineering Senior Licensing Engineer Manager Mechanical Engineering U. S. Nuclear Regulatory Commission J. Linville S. Morris Chief, Division of Reactor Projects, Branch 3 Salem Senior Resident Inspector
- INSPECTION PROCEDURES USED IP 92903: Followup - Engineering IP 64704: Fire Protection Program ITEMS OPENED, CLOSED, AND DISCUSSED Discussed 50-311/97-16-03 IFI Conformance of Calculations and Procedures with TS 4.7.6.1.d(5)
Closed 50-272; 311 /96-07-07 URI Safety Evaluation Under 10 CFR 50.5 ; 311/96-11-10 URI Review Corrective Actions to AR 9606071'1 /96-21-09 IFI Complete Evaluation and Work On Unit 1 SJ Hot and Cold Leg Injection Valve /96-81-04 URI Basis for Component Cooling System Flow Balance Test Acceptance Criteri ; 311 /97-05-02 URI Electrical Penetration Backup Protection 50-311 /97-06-02 URI Vendor Computer Software Control ; 311/97-08-01 URI Operability Determination and Reportability of EOG Loadin /97-11-05 URI Adequacy of Abnormal Operating Procedure for Establishing RHR Flow into RCS Hot Legs during a LOCA in Mode /97-11-06 IFI Deletion of RHR Hot Leg Flow Path From Licensing Basis in Apparent Conflict with Technical Specificatio /97-*11-07 URI Unvalidated RHR Pumps NPSHr for Recirculation Mod /97-11-09 URI Operator Ability to Perform Timely Switchover of ECCS Pumps Suction from RWST to Containment Sum ~311/97-11-10 URI Use of Undocumented and Unlicensed "NOSI" Code for Unit 2 Specific Desig ; 311 /97-20-01 URI Containment Air Temperature Averagin ; 311/97-21-03 IFI Guidelines for Instrument Loop Design Loadin ; 311 /97-021-04 50-311 /97-021-05 50-311/97-182-01013 50-311/97-182-01023 50-272; 311 /97-09-06 VIO Inadequate Evaluation of Votes Test Result VIO Failure to Include Containment Penetration Breaker in TS-Required Surveillance Progra EEi 10 CFR 50.59 Violation Associated with ECCS Suction Switchove EEi 1 Q CFR 50.72 (and 73) Violation Associated with ECCS
- Suction Switchove URI Appendix R Exemptions 50-272; 311 /97-257-03014 EEi. Compensatory Fire Watches
ADFCS ANSI AR ASME CCHX CFCU CFR CJP COTSS CR DCP ECCS l&C ISi ISLT MCR MT N/A NDE NPS NOP NRC NSRB OTSC PMT PT
.PSE&G RCP RCS RHR RPM RT SER SGFP SI SNSS SORC SRO SSFI SW TRB TRIS TS UFSAR UT LIST OF ACRONYMS USED Advanced Digital Feed.water System American National Standards Institute Action Request American Society of Mechanical Engineers Component Cooling Heat Exchanger Containment Fan Coil Unit Code of Federal Regulations Code Job Package *
Commercial Off The Shelf Software Condition Report Design Change Package Emergency Core Cooling System Instrumentation and Controls lnservice Inspection lnservice Leak Testing Modifications Concerns and Resolution Magnetic particle Examination Not Applicable Non-destructive examination *
Nominal Pipe Size Normal Operating Pressure Nuclear Regulatory Commission Nuclear Safety Review Board On-The-Spot Change Post-Maintenance Testing Dye Penetrant Examination Public Service Electric and Gas Reactor Coolant Pump Reactor Coolant System Residual Heat Removal Revolutions per minute Radiographic examination Safety Evaluation Report Steam Generator Feedpump Safety Injection Senior Nuclear Shift Supervisor Station Operations Review Committee Senior Reactor Operator
- Safety System Functional Inspection Service Water Test Review Board Tagging Request Inquiry System Technical Specification Updated Final Safety Analyses Report Ultrasonic Examination