IR 05000272/1996080
| ML18102A732 | |
| Person / Time | |
|---|---|
| Site: | Salem  |
| Issue date: | 12/31/1996 |
| From: | NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | |
| Shared Package | |
| ML18102A731 | List: |
| References | |
| 50-272-96-80, 50-311-96-80, NUDOCS 9701090185 | |
| Download: ML18102A732 (80) | |
Text
Docket Nos:
License Nos:
Report Nos:
Licensee:
Facility:
Location:
Dates:
Inspectors:
Approved by:
U.S. NUCLEAR REGULATORY COMMISSION 50-272, 50-311 DPR-70, DPR-75
REGION I
50-272/96-80, 50-311 /96-80 Public Service Electric & Gas Company Salem Nuclear Generating Station, Units 1 & 2 Hancocks Bridge, New Jersey 08038 May 6 - May 17, 1996 October 21 - 25, 1996 E. Kelly, Branch Chief, Systems Engineering Branch, DRS P. Swetland, Sr. Resident Inspector, Millstone, DRP F. Bower, Reactor Engineer, Systems Engineering Branch, DRS B. McDermott, Resident Inspector - Susquehanna, DRP M. Shlyamberg, NuEnergy, Inc. - Contractor James T. Wiggins, Director Division of Reactor Safety, Region I 9701090185 961231 PDR ADOCK 05000272 G
EXECUTIVE SUMMARY Salem Nuclear Generating Station, Units 1 &. 2 NRC Inspection Report 50-272/96-80, 50-311 /96-80
- As of May 1996, the design (via the Configuration Baseline Documents) was not effectively validated in all aspects, and the resolutions for some closed deficiencies were inadequate. The licensing and design bases of the systems reviewed by the team (fuel handling ventilation, emergency control air, and service water/containment fan cooling) were not well understood nor were they accurately articulated in the UFSAR and CB *
The fuel handling ventilation (FHV) system design and licensing bases were poorly understood and fragmented prior to June 1996. Proper system operatior ':1*.l*l..S
'".*. "*. ***
,..,,.
questionable, but all of the May 1996 team's findings were subsequently addressed, including a reanalysis of the fuel handling accident submitted to the NRC on June 4, 1996. Corrective actions associated with the FHV system and fuel handling activities were described in letters dated May 29 and 31, 1996, prior to resuming fuel movement on June 10, 199 *
Accident analyses in UFSAR Chapter 15 did not, in all cases, match other internal UFSAR sections, nor did they match certain design specifications. In one instance, the operability of the containments for both Salem units was impacted due to operation for extended periods with containment fan cooler (CFCU) fan flow/heat transfer outside of the plants' design bases. The licensee subsequently addressed these issues, and reported them in several Licensee Event Reports (LERs 96-15 and 20) issued in August and September 199 *
Testing appeared weak and ineffective in a number of instances, particularly for the
. compressed air (CA) system. Subsequent to the May 1996 team observations, component adjustments, repairs, reanalysis of the load calculation and retesting restored the design basis capacity of the emergency compressor *
An FSAR Project conducted in July-August 1996 was well-managed and represented an effective effort to validate of the Salem Updated Final Safety Analysis Report (UFSAR). The project had strong elements of oversight and independent confirmatio *
In the FSAR Project, over 400 input assumptions for UFSAR Chapter 15 safety analyses were validated by reference to calculations, test procedures and other design documents. With the exception of 9 instances involving incomplete source documents and some relatively minor discrepancies, the Chapter 15 reviews demonstrated a high degree of consistency and accuracy, providing assurance of operation within the bounds of safety analyses. The Chapter 15 reviews were complemented by a review of all 162 Unit 2 Technical Specification License Amendments issued through May 1996, with similar result ii
... _/
Executive Summary
In response to NRC concerns about the resolution of previously identified design deficiencies(DEF's) and technical justifications for operability, the UFSAR Project evaluated 1, 752 DEF's and over 300 Engineering Evaluations. The DEF screening resulted in detailed.review of 500 deficiencies associated with design or licensing bases questions for the 16*safety analyses systems referenced in UFSAR Chapter 15. Reviews of the Engineering Evaluations were later expanded to cover all important Salem Unit 2 systems. The licensee concluded that corrective action processes were adequately addressing such deficiencies based upon: (a) the relatively large sample and few (seven AR's) problems found; as well as (b) the significance of the problems, which were primc;irily related to documentation; and (c)
the fact that these issues related to process concerns more reflective of the 1 988-93 timeframe, rather than post-199 iii
.. *
'
TABLE OF CONTENTS PAGE EXECUTIVE SUMMARY ii E1 Conduct of Engineering * *......... *....................... *....
E1.1 Special Team Inspection - Background and Scope........ *.. *....
E2 Engineering Support of Facilities and Equipment.......................
E2. 1 Service Water System....................................
E Containment Cooling. *.............................. *....
E Fuel Handling Area Ventilation.......... * *..................
E Control Air System................. *...................
E7 Quality Assurance in Engineering Activities............ *........ *...
E7. 1 Configuration Baseline Documents (CBDs)....... *.............
E7.2 UFSAR Changes. *....................... *......... *...
24-E7.3 Commitment Verification Project............................
E Salem FSAR Project............................. *.......
X1 Exit Meeting Summary.......................... *.............
iv
E1 Conduct of Engineering E1.1 Special Team Inspection - Background and Scope E2 Based upon recent industry experience, an NRC inspection was initiated, in May 1996, to examine the fidelity between the Salem as-built plant configuration, design, and the current licensing bases in the Updated Final Safety Analysis Report (UFSAR). Because of the Salem Unit 1 plant status, Unit 2 systems were initially considered based upon risk significance, operational history and design change activities. Included in the system-level reviews were open deficiencies and problem reports, operability determinations, associated engineering memoranda and licensing change notice The team "screened" ten systems during the first week of the May 1996 inspection to better understand the licensee's programs for maintaining configuration control and implementing corrective action for identified deficiencies. Three systems -- fuel handling ventilation, control air, and service water/containment cooling -- were then selected for more in-depth review during the second week based upon modification history and open design/licensing bases issues. A limited "vertical slice" evaluation of design was performed utilizing interviews with cognizant engineers, review of associated Configuration Baseline Documents (CBDs) and other design information such as calculations, and comparison of this information against the current licensing bases.
Engineering Support of Facilities and Equipment E Service Water System* Inspection Scope The team reviewed the documents listed below and walked down portions of the Salem service water (SW) system to determine the fidelity between its design basis, current licensing basis, and as-installed configuratio *
UFSAR, Sections 6.2.2, 9.2, 9.4.4, 15. *
Technical Specification, Sections 3/4.3. 7.4, 3/4. *
Configuration Baseline Documentation for Service Water System; DE-CB.SW-0047 (0), Revision 6, dated 6/28/95
Service Water System Unit 2, Initial Readiness Review, dated 9/18/95
Service Water System Operational Performance Inspection, dated 12/5/94
Discrepancy Evaluation Forms (DEF): DES-93-00106, DES-93-00121, DES-90-01434
Calculation S-C-SW-MDC-1350, Revisions 0 and *
10 CFR 50.59s for Modifications 1 EC-3316 & 2EC-3274 (Throttled Manual Isolation Valves for the CCHX)
Configuration Baseline Document Maintenance Procedure; NC.DE-AP.ZZ-0021 (Q), Revision 4
2 Observations and Findings The service water system performs both safety and nonsafety-related functions. Its safety-related function is to supply cooling water from the Delaware River to the safety-related heat loads that must be cooled during postulated design basis accident conditions. The system is comprised of two flow loops, which are normally cross connected, that are supplied by a total of six pumps. The licensee considers the system operable with up to two of the six pumps out of servic The SW pumps' ability to deliver the expected design basis flow is dependent upon the pumps having adequate net positive suction head (NPSH). For proper pump operation, the NPSH available must be greater than the NPSH required. The NPSH available is dependent on the river water level, river water temperature and, to some extent, atmospheric pressur NPSH Available and Single Failure Vulnerabilities The licensee identified several equipment and operational problems between 1990 and 1994, where the SW system would not perform its intended safety function due to inadequate NPSH or single failure *
In August 1990, PSE&G identified that the SW pumps would not be capable of providing the required flow due to inadequate NPSH at the design basis low level and high temperature river conditions. (DES-90-01434)
In November 1993, PSE&G identified that the single failures of either an air-operated Component Cooling Water (CCW) heat exchanger (HX) flow control valve or a temperature control valve would divert significant cooling flow from safety-related equipment and cause a loss of NPSH for the SW pump *
In June 1994, PSE&G identified the potential for SW pumps to cavitate (due to pump run out) during the recirculation phase of a LOCA when only two SW pumps are aligned to both CCW HXs. This condition was beyond previously analyzed conditions and could have affected the ability of the SW system to perform its design basis function. Licensee Event Report (LER)
50-272/95-025 documented this proble Corrective actions to resolve the SW pump NPSH issues have been implemented, including installation of new SW pumps which require less NPSH at the required design flow, and use of manual valves to limit excessive SW flow through either CCW HX in the event of a single failure. A modification is also planned to install orifice plates in the CCW HX lines to reduce the operational burden associated with maintaining the correct throttle positions of the manual valves. Based on review of the new pump performance curves (and related updates to UFSAR Table 6.3-13 and Section 9.2), the inspector considered these corrective actions to be acceptable.
' \\
Design/Licensing Document Inconsistencies In May 1996, the inspectors identified several apparent discrepancies between the UFSAR, the CBD, and the plant configuration. These issues, which have since been resolved by PSE&G, as of October 1996, included:
UFSAR Section 9.2.1.2 (and pending changes) stated that the service water flow through the No. 11, 12 and 21 CCW HXs is maintained at 10,000 gpm by means of a control valve. The CBD indicated that flow to the CCW HX has been limited to a maximum of 8000 gpm by means of throttled manual valves (for pump run out protection in case of a single failure).
- The UFSAR has been updated to reflect the actual plant condition, which is that flow is controlled based on CCW HX outlet temperature and the nominal flow rating of the HX is 10,000 gpm. The UFSAR has also been updated to indicate that the minimum HX flow of 8,000 gpm is necessary to accomplish the intended safety function as design basis river condition In May 1996, Table 6.3-13 of the UFSAR (and pending changes) indicated that the SW pumps had sufficient NPSH available at the design basis river water conditions. The UFSAR table was incorrect regarding the original SW pumps, but has since been updated to reflect the new pumps, with annotations to address the remaining original pumps on Salem Unit 1.
The SW flow rates and heat loads provided in SW CBD Table T-5 did not agree with the SW flow rates and heat loads provided in UFSAR Table 9.2-In response to this issue, the licensee has reconstituted the basis for the flow and heat load values in Table 9.2-1 using the original design documents and updated the UFSAR. The licensee stated that no clear basis could be found for the values originally listed in Table 9.2-1. The bases documents listed in the CBD were used to reconstitute the UFSAR Table, and therefore the two documents are now consisten System Design Temperature The maximum design temperature for the SW system components was calculated by PSE&G to occur downstream of the containment fan coil units (CFCUs) following a large break LOCA. The inspector noted the calculated value of 160°F was based on an assumption that the CFCUs were at their design basis fouling factor limit of 0.002. The inspector determined that this assumption may not be conservative since the maximum heat transferred to the SW system will result from a clean CFCU (i.e., a fouling factor approaching 0.00).
The inspector questioned the effects of a higher SW system design temperature on previously analyzed stresses for piping, supports, and containment penetrations. To address these questions, PSE&G initiated Condition Report (CR) 96-513197 to evaluate what fouling factor assumptions would be appropriat
..
PSE&G determined that using a more conservative (lower) fouling factor for determining the highest SW temperature downstream of the CFCUs was appropriate, and resulted in an increase in the maximum temperature from 160°F of 195°F. PSE&G reached the following conclusions regarding the impact of this temperature increase on the SW system:
With the SW system in a limiting pressure alignment, an adequate positive pressure margin exists in the CFCU outlet pipe to avoid flashing at 195° The revised vapor binding assessment was to be included in the next revision of the SW system design basis calculation (S-C-SW-MDC-1350).
- The formal basis for the CFCU zero fouling temperature was included in the *
CFCU heat transfer basis calculation (SC-CBV-MDC-1637, Revision 0).
PSE&G determined that no field changes were necessary. Stress calculations and associated supports were reviewed for a CFCU discharge piping temperature of 195°F during the design basis accident. PSE&G determined that the piping and supports meet applicable ASME Code allowable stresses for this faulted conditio The inspector concluded that PSE&G incorporated the more conservative CFCU fouling factor assumption into the design basis temperature for the SW system piping and associated supports. This conservatism provides additional margin in the design by accounting for circumstances which are unlikely to occur simultaneously, and which would serve to limit the duration of post-accident containment peak temperature Service Water System Conclusions Replacement of the* SW pumps, implementation of fixed flow restrictions for the component cooling water heat exchanger lines, and changes to the emergency operating procedures corrected the problem of inadequate net positive suction head and the single failure valve vulnerabilities. As discussed in Section E7.4 of this report, the Service Water System also received considerable review for additional design and licensing basis issues since the May 1996 NRC team inspectio PSE&G did not initially address Service Water System design basis deficiencies, identified between 1990 and 1993, in accordance with NRC requirements for corrective action. These deficiencies, reported in LER 50-272/95-25, caused the plant to be outside its design and licensing bases (prior to 1995) and is a violation of 1 O CFR 50 Appendix B, Criterion XVI, Corrective Action. This violation meets the criteria of Section Vll.B.2 of the NRC Enforcement Policy. Therefore, discretion is being exercised and the violation is not being cited.
E Containment Cooling Inspection Scope The inspector reviewed the documents listed below to assess the consistency between the design basis, current licensing basis, and plant configuration for the containment fan coil units (CFCUs). The CFCU's are cooled by the service water system, and designed to recirculate and cool the containment atmosphere in the event of a design basis acciden * *.
UFSAR, Sections 6.2.2, 9.2, 9.4.4, 15. *
Technical Specification, Sections 3/4.3.7.4, 3/4. *
Configu,r.ation Baseline Documentation for Containment Building Ventilation System; DE-CB.CBV-0026 (0), Revision 2, dated 12/7 /92
Discrepancy Evaluation Forms (DEF): DES-93-00106, DES-93-00121, DES-90-01434
CFCU Motor Cooler Reduced Water Flow Evaluation, Revision 2, dated January 1994, by Westinghouse Electric Corporation
Calculation S-C-SW-MDC-1350, Revisions 0 and *
10 CFR 50.59s for Modifications 1 EC-3316 & 2EC-3274 (Throttled Manual Isolation Valves for the CCHX)
Configuration Baseline Document Maintenance Procedure; NC.DE-AP.ZZ-0021 (0), Revision 4 Observations and Findings Each Salem containment has five CFCUs that are designed to remove the heat from equipment during normal plant operation, and to limit the peak containment pressure following an accident to within design basis pressure. The Containment Spray system is redundant to (and can be used in conjunction with) the CFCUs to accomplish the safety function. The design capacities of these systems are such that adequate heat removal can be achieved with all five CFCUs, both CS pumps, or the combination of three CFCUs and one CS pump. The minimum heat removal capability required for the three CFCUs in the Salem accident analysis is 25 million BTU/hou Reduced CFCU Heat Removal Capability During the inspector's review of the SW system, two deficiencies were examined that affected the heat removal capacity of the CFCUs. Both deficiencies were originally identified by PSE&G in 1993, and one remained open as of the May 1996 inspection. Specifically, these deficiencies involved reduced air/steam flow through the CFCUs during accident conditions and discrepancies in the assumptions (in the UFSAR) for CFCU heat removal capability.
.,.
Discrepancy evaluation form (DEF) DES-93-00106, dated April 1993, identified that the calculated air flow through each CFCU during accident conditions would be 7,000 cubic feet per minut'e (cfm) less than design value of 47,000 cfm assumed in the containment accident analysis. PSE&G closed the DEF in April 1994 based on a Westinghouse containment analysis "Fuel Upgrade and Margin Recovery Program:
LOCA Containment Integrity Analysis," WCAP 13839. Based on this analysis, PSE&G determined that even with the reduced air flows, the containment pressure under the worst case conditions would be below the design limit. However, the inspector found that PSE&G's resolution had previously relied on an analysis that was not part of the Salem licensing basi A second DEF (DES-93-00121) was issued in May 1993 to address a different, but related, discrer~~Y with the CFCUs. This DEF;.identified that the heat removal rate for the CFCUs listed in UFSAR Section 6.2.2.2.1 (81 million BTU/HR) was lower than the value of heat removal assumed in Chapter 15 of the UFSAR (83 million BTU/hr). During the May 1996 inspection, this DEF was still under review by PSE& Reduced Air Flow PSE&G accepted a Westinghouse analysis (WCAP 13839) for resolution of the April 1993 air flow design basis discrepancy (DES-93-00106) without performing an evaluation to determine whether this change to the facility constituted an unreviewed safety question. The inspector determined this discrepancy had been neither properly addressed as a condition adverse to quality, nor properly evaluated as a design chang The CFCU air flow discrepancy reflected the differences between the air flow from a single CFCU fan and the calculated air/steam flow that can be achieved with multiple CFCUs discharging to a common duct. A later calculation justified the reduced air flow by reducing the fouling assumption and a lower air flow rate for design basis conditions of 39,000 cfm. PSE&G justified reducing the design basis fouling factor from 0.002 to 0.0015 using past CFCU performance test results (excluding their performance when the SW chlorination system was inoperable).
DES-93-00106 was subsequently resolved on October 8, 1996, when the Station Operations Review Committee approved a 50.59 Safety Evaluation for calculation S-C-CBV-MDC-1637, Revision CFCU Heat Transfer In May 1993, PSE&G assigned the resolution of the CFCU heat capacity discrepancy (DES-93-00121) a long term priority based on the expectation of minimal impact, and because the containment spray system provides redundant protection. At that time, PSE&G concluded that no operability determination was necessary.
- DES-93-00121 was administratively closed in July 1996, based on the technical issues being incorporated into Condition Report (CR} 96-0515126 generated during the May *1996 NRC inspection. The licensee's corrective action for this CR included changing the UFSAR to reflect the current licensing basis information for the CFCUs (heat removal of 83.6 million BTU/hr, 40,000 cfm air/steam flow, and 90°F SW temperature}. The heat capacity issue was resolved, along with the air flow issue, on October 8, 1996, through approval of the 50.59 Safety Evaluation for calculation S-C-CBV-MDC-1637, Revision Test Acceptance Criteria The inspector identified that PSE&G's CFCU testing had been previously inadequate because the test acceptance. t::ri:t:~ria specified @n air flow of 4 7,000 cfm, rather
- than the expected 40,000 cfm. Test results from the first quarter of 1993 showed CFCU fouling was significantly greater than assumed in the containment accident analysis. PSE&G's evaluation of the test results (at that time} concluded that the CFCUs were operable based on low river water temperature at the time of the test discovery. However, this evaluation was inadequate because it did not address the possible operability impact during previous operation when the river water temperature was significantly highe In response to the inspector's observations regarding the CFCU test, and the inadequate resolution of the heat capacity issues, PSE&G issued CR 96-051 512 The licensee reevaluated the test results using the correct heat removal value and an integrated approach to assessment of the system's ability to meet its design basis. PSE&G attributed the excessive CFCU fouling to a period of approximately nine months when the SW chlorination system was inoperable. Therefore, during the first quarter of 1993 when both Salem units operated in Mode 1, CFCU heat transfer capability was determined to be outside of design basis requirements and was reported to the NRC in LER 50-272/96-01 CFCU Component Design Temperature
- "CFCU Motor Cooler R~duced Water Flow Evaluation," dated January 1994, qualified the CFCU motors for service at 271°F. UFSAR, Chapter 6.2.2, states that other elements of CFCU system (coils, HEPA filters} are qualified to 271°F, whereas the duct gaskets "are suitable for temperatures up to 300°F." The inspector noted that these temperature qualifications were less than the maximum containment accident temperature of 351.3°F listed in UFSAR, Chapter 15.4.8.2.4 for a main steam line break (MSLB). This discrepancy raised a generic question regarding the design and qualification temperature for safety-related equipment inside containmen Discrepancies between the peak containment temperatures in the Salem design and licensing basis were reported to the NRC in LER 50-272/95-016-01. PSE&G concluded that the limiting temperature of 351.3°F (MSLB) was acceptable based on evaluations of the EQ program equipment, the containment liner, the containment spray piping supports, the containment hatches, and other steel structures. PSE&G submitted license change request (LCR) S96-06, dated June 18, 1996, to add the MSLB peak temperature to TS 5. *
Corrective actions included a change (LCR S96-13) in the containment pressure and temperature data listed in TS to make it consistent with the UFSAR. However, the inspector determined that the basis for LCR was not complete because PSE&G failed to evaluate the effects of a higher peak containment temperature on equipment not covered by the EQ program; examples included CFCU duct gaskets and HEPA filters. Although the need for this analysis had been previously recognized by PSE&G, the evaluation was still in draft form and had been overlooked during preparation of the LCR. On October 25, 1996, PSE&G initiated a CR to evaluate the impact of the higher design temperatures on safety-related equipment inside containment. PSE&G subsequently completed an analysis of mechanical components *,.,,side containment, and summarized their conclusions in a letter submitted to the NRC on December 6, 199 Results of PSE&G's CFCU "Vertical Slice" In addition to the findings discussed above, two new issues were identified by PSE&G engineering personnel during their review of the CFCUs as part of the Salem FSAR Project (Section E. 7.4). The first issue resulted from a startup-testing related modification in 1976 to prevent water hammer which added a time delay to the isolation of nonessential SW loads. As a result of this delay, the design basis SW flow to the CFCUs in certain single failure scenarios could not be achieved in the TS-required time of 45 seconds. The second issue resulted from evaluation of Westinghouse NSAL-96-003 for applicability to Salem. PSE&G determined that as SW is restored to the CFCUs under LOOP/LOCA conditions, water trapped in the CFCUs would be subject to significant thermal expansion. The thermal expansion could result in pressurization of the CFCU components and SW piping beyond design limits and might also affect the ability of the SW side valves reope These issues were reported to the NRC in LER 50-272/96-020 on September 18, 1996, and PSE&G attributed these conditions to the following:
PSE&G failed to update the plant design basis or consider the impact on other analyses and design inputs after implementation of the 1976 design chang *
The potential for heat up of the CFCU during fan coast-down was not recognized until review of Westinghouse NSAL-96-00 *
The ESF response time testing procedures did not address SW valve timing as it affects the supply of cooling water to the CFCU To resolve the valve timing issue, PSE&G is proposing a TS change (License Change Request LCR S96-13). The potential for CFCU over-pressurization was resolved through installation of a pressure relief modification and, as part of the corrective action process, the potential for similar situations in other equipment remains under evaluation by PSE&G.
Resolution of CFCU Capability Issues The 50.59 Safety Evaluatidn for Calculation S-C-CBV-MDC-1637, Revision 0, approved on October 8, 1996, brought the current licensing basis up to dat PSE&G has committed to testing the CFCUs once per cycle in response to Generic Letter 89-13. A PSE&G restart action item captures the need to establish new test criteria, however PSE&G plans to clean all the CFCUs each refueling outage until sufficient test data is collected to justify a reduced frequency. Based on review of the existing CFCU test data, the inspector considered this action to be reasonable, provided the chlorination system is reliabl Containment Cooling Conclusions E.Ci)l
'*
Design basis discrepancies identified by PSE&G in 1 993 were not properly addressed as significant conditions adverse to quality. The issues identified during this inspection show that, prior to May of 1996, PSE&G failed to: (1) address the operability impact for past operation; (2) resolve the issues in a timely manner; and (3) properly evaluate certain technical resolutions that changed the design of the facility. Although not adequately addressed betwe.en 1990 and 1994, PSE&G has since reported these problems (LERs 50-272/95-016-01 and 50-272/96-015) and is taking appropriate actions to correct them. This problem is attributed to discrepancies associated with activities prior to 1995, and was addressed by broad licensee programs and actions. These issues represent examples of a violation of 10 CFR 50 Appendix B, Criterion XVI, Corrective Action. The violation meets the criteria of Section Vll.B.2 of the NRC Enforcement Policy. Therefore, discretion is being exercised and the violation is not being cite PSE&G's resolution of the design basis inconsistencies regarding CFCU air flow and heat removal capability were ultimately acceptable. Changes to the design basis calculations needed to resolve these issues were properly evaluated in 1996 as changes to the facility under 10 CFR 50.59. PSE&G's actions to clean the CFCUs each refueling outage, until performance testing results justify a lesser frequency, is considered a reasonable approach since 1993 problems with reliability of the SW chlorination system had a significant impact on CFCU foulin Fuel Handling Area Ventilation Scope The Fuel Handling J.\\rea Ventilation (FHV) system ensures that all radioactive material released from an irradiated fuel assembly will be filtered through the HEPA filters and charcoal absorber prior to discharge to the atmosphere (TS Bases 4.9.12). The FHV was designed to exhaust the spent fuel pool area at 60 air changes an hour with a system operating flow rate of 19,490 cfm +/- 10% Because of the potential for radioactive releases from the spent fuel, defective fuel cladding or a fuel handling mishap, the building is maintained at a slight negative pressure (greater than or equal to 1 /8 inch water gauge). The total capacity of the FHV system is designed to maintain the building between 60 ° F and 105 ° F, and the
'I
expected design conditions are expected to be less than 70% relative humidity (FSAR Section 9.4). The inspectors reviewed following documents and walked down the Salem Unit 2 Fuel Handling Ventilation system to sample the fidelity between design basis, the current licensing basis, and the existing plant configuration:
Configuration Baseline Documentation (CBD) for Fuel Handling Area Ventilation System DE-CB.FHV-0021 (Q)
Unit 2 Technical Specification 3/4.9.12, Fuel Handling Area Ventilation System
Facility Operating License, License Number DPR-75, Salem Nuclear Generating Station Unit No.2, Docket No. 50-311
Safety Evaluation Report for Dockets 50-272. ~.nd 50-311, October 11, 1974; Section 15.0, Accident Analyses and Section 15.1, Ventilation Systems
Safety Evaluation Report for Dockets 50-272 and 50-311, Section 15.2.2, Fuel Handling Accident
UFSAR Section 9.4.3, Fuel Handling Area Ventilation
UFSAR Appendix 3A, Public Service Electric & Gas Positions on USNRC Regulatory Guides
UFSAR Section 15.4.6, *Fuel Handling Accident
UFSAR Section 11.1.3, Fuel Handling Sources
Amendment No. 131 to Unit 2 License
On-Site Safety Review Group (SRG), Review of the Salem HVAC Systems Maintenance and Equipment, SRGC 87-084, Revision 1, dated August 21, 1987 Observations and Findings The inspectors encountered difficulty in performing a comparison of the design and licensing basis described in the Technical Specifications, UFSAR, and CBD due to the number and extent of existing and planned change notices to various documents. The following were a sample of discrepancies noted by the inspectors *
{in May 1996) that were not identified as part of recent system readiness reviews, or other design activities prior to May 199 Auto-Start Surveillance During a review of the Technical Specifications (TS), the inspectors noted that TS Surveillance Requirement 4.9.12.d.2 states that "At least once per 18 months by verifying that on a high radiation test signal, the system automatically starts (unless already operating) and directs its exhaust flow through the HEPA filters and charcoal absorber banks." During discussions with PSE&G personnel, the inspectors were informed that the Unit 2 Fuel Handling Ventilation system did not have an auto-start feature. The inspector questioned whether an operability determination had been performed.
'*
PSE&G personnel provided a copy of condition report (CR) 00951005119 which documented that QA personnel had previously identified this concern on October *6, 1995, while reviewing a draft of LER 272/95-24. The CR's recommended corrective actions included initiating an emergency license change request (LCR) and reporting the issue in a supplement to LER 272/95-24. The need for an emergency TS change was documented in memo LR-195601.
Single Failure Vulnerability The inspectors' review of the UFSAR and the CBD found that the FHV system is required to be able to sustain the failure of an active component without losing the
.capability of maintaining a negative pressure and fiitered effluent (UFSAR Section 9.4.3.3). An open item (number 10) identified during thc COD development
- -~.,,
process highlighted that the charcoal filter train may be subject to isolation due to a single failure of the air-operated (fail open) inlet or outlet damper, and that an unmonitored release from the FHB could result. This concern was originally identified on DEF 01301, dated June 19, 199 The DEF was closed on October 20, 1995, based on Design Engineering's conclusion that the dampers were not active components. The inspector noted that the original target date to resolve the discrepancy was December 15, 1991, but the DEF was assigned a low priority because the FHV system "does not contribute to the core damage frequency and therefore was not modeled in the PRA." During discussions with engineering personnel, those individuals initially reiterated that the dampers were passive' components that were verified to cycle at least once each 18 months. The FHV CBD, Section 2.3.1.H, also indicated that the two 50% capacity exhaust fans met Regulatory Guide (RG) 1.52 and IEEE 279 _with respect to single failure consideration The inspector noted that there are other "active" components in the system (e.g., a fire damper in the common exhaust plenum, a single nonsafety-related 100%
capacity supply fan c:ind the charcoal filter deluge system) similarly affected, and therefore in question with respect to single failure. Further, the normally open inlet/outlet dampers on the non-emergency filter train (without charcoal), which the licensee intends to reposition closed via a design change, could similarly fail open creating a bypass leakage pat As part of the team's evaluation of this issue, an unrelated operability determination (OD)95-002, dated May 27, 1995, for the control room ventilation system was also reviewed. Attached to this OD was a memorandum documenting a March 22, 1974, internal PSE&G meeting addressing the design basis and licensing aspects of various Salem HVAC systems, with special emphasis on the application of single failure criteria. The minutes of the meeting stated:
'.
"... although the Salem design does not conform to single failure criteria, it was contended that inherent to any HVAC system exists the ability to make short-term temporary repairs to insure operability, thereby reducing the importance of single failu.re criteria.... The AEC has reviewed the Salem FSAR, and with exception of the Control Room HVAC System, has not raised any serious questions with regard to Salem HVAC licensability.*. any change in the HVAC design basis at this time, could not be afforded without sacrificing some input on existing schedules. Limited engineering and design manpower and additional field work support the basis for this conclusion. If it is determined at a later date that the existing design basis is unacceptable from a licensing viewpoint... sufficient flexibility exists in the present HVAC design to make field modifications at a future unit outage."
As stated in the conclusions of this section, PSE&G subsequently committed to compensatory measures until such time as the single failure requirements for the FHV system were resolved. Determination of the systems' original licensing and design basis with respect to single failure has been forwarded to the NRC's Office of Nuclear Reactor Regulator (NRR), and is therefore unresolved (URI 50-311/
96-80-01).
Charcoal Heat Load The inspectors reviewed UFSAR Section 9.4.3.3 and noted that it stated that an analysis was made of the heat loading on the filters, assuming that all of the available radioiodine released (using NRC Safety Guide No. 25 assumptions) is adsorbed on the filters. Per the UFSAR, the resultant heat loading was determined to be "approximately 1 BTU per hour." This statement is reiterated in Section 7.0 of the FHV system CBD. However, CBD open item #13 noted that an analysis to support the statement was not availabl DEF-90-01304 was initiated on.June 19, 1990, to document this issue; the DEF was recently closed on September 11, 1995, based on the results of a new Calculation S-C-FHV-MDC-1471 that showed a heat generation rate of less than 2 Btu/hr for the "conservative" case and less than 1 Btu/hr for the "expected" cas The calculation noted that although the conservative case exceeds 1 Btu/hr, the statement in the UFSAR had been substantiated because ( 1 ) "a fuel handling accident is not likely to break all the fuel rods in a damaged assembly"; and (2)
"much of the gamma radiation will not be absorbed as heat within the charcoal."
The inspector noted that the "expected case" did not appear to agree with the UFSAR Section 15.4 analysis for a fuel handling event, and that no safety evaluation (per 10 CFR 50.59) was performed to more formally add the reanalyzed design to the licensing basis (PSE&G merely closed the DEF). The design and licensing basis event (described in the UFSAR and the NRC's SER) assumed that all the fuel rods are damaged, just as assumed in the conservative case more recently analyze *
Concerning the conservative case the inspector noted that: ( 1 ) the analyzed heat load was 2 Btu/hr versus the 1 Btu/hr described in the UFSAR; (2) activity was considered at 1 68 hours7.87037e-4 days <br />0.0189 hours <br />1.124339e-4 weeks <br />2.5874e-5 months <br /> after shutdown versus the 1 00 hours0 days <br />0 hours <br />0 weeks <br />0 months <br /> specified in Section 15.4; and (3) the activity values used in the calculations were taken from a table of fuel handling sources from.UFSAR Section 11. 1.3 that do not agree with the values in similar tables in Section 15.4 (discrepancy was described in detail above).
Aggravating this was the inconsistency with rerack Amendment Number 131.
Further, it was not clear which table would match the current collection of fuel in the pool, and planned for the next core reloa PSE&G performed new calculations that verified that the heat load from iodi adsorption was greater than the 1 BTU/hr described in the FSAR. The revised heat loading is well below the ignition point of the charcoal, and the licensee plans to revise the UFSAR using 10 CFR 50.5 Maintenance of FHB Ambient Temperature The inspectors reviewed CBD Section 4.2 and UFSAR Section 9.4.3.1, and noted both state that part of the FHV design and licensing bases is for the system to maintain the building between 60°F and 105°F. The SER (Section 9.8) recognized that the ventilation systems were designed to maintain a safe level of temperature and cleanliness in the rooms or compartments served. During the onsite period, (prior to June 1996), licensee personnel were unable to provide the basis for this design basis temperature range, but stated that it was most likely related to personal comfor CBD Section 3.0 stated that the FHB local hot water unit heater capacities were inadequate. Further, the System Index Database in the SRR report listed Design Change Package (DCP) 2EE00044, identified as necessary for restart to modify the heating coil piping and instrumentation to prevent freeze-up, and classified this item as a repetitive failure. Discussions with licensee personnel indicated that the FHB was believed to have dropped below the minimum design and licensing basis temperature of 60°F during the last winter; however, the building temperatures were not being monitore The licensee later provided two documents that indicated that the FHB had dropped below its design basis minimum temperature. Problem Report (PR) 00950216116 indicated that a heating coil was ruptured (apparently due to freezing) and the FHB dropped below the design basis minimum. Condition Report (CR) 00960207103 indicates that the FHV was outside its design basis due to ice formation on the FHB shipping bay floor. More recently this condition was documented in a 10 CFR 50.59 evaluation, dated November 9, 1995, approving a change to the FHV operating and ala.rm response procedures. These procedures previously instructed the operators to manually control the heating water, or to shutdown the FHV supply fan (and thu_s make the system inoperable) when the supply duct temperature switch dropped below 40°F. The November 1995 change allowed the operators to continue to operate the system utilizing administrative actions to monitor building
\\
'
temperature. During a walkdown of the FHV supply system, the inspector noted that the heating coil thermostat (2TA6255C) was set at 58°F versus the FSAR required value of 60°F. The licensee initiated AR 00960529205 to investigate this issu The inspector also questioned FHV system/building temperatures at the higher end (viz. 105°F, as documented in both the CBD and the UFSAR). Specifically, the alarm setting for the supply air duct is currently set for 120°F, based upon a table referenced in the FHV CBD. The inspectors reviewed set point calculation SC-FHV001-01, "2 Fuel Handling Building DP Alarm and Control" associated with maintaining the FHB at a negative pressure of 1 /8 inch water gage. During this review, the inspectors noted that the process design inputs and assumptions were based on a FHB temperature range of 60°F to 120°F. The inspectors noted that the November 1995 safety evaluation had not considered this set point calculatio Operability Determination Operability Determination (00)95-114, dated October 23, 1995, described an evaluation of high vibration levels on the FHB exhaust fans. The OD concluded that the number 21 fan was operable and that the number 22 fan was operable but degraded. The OD noted that the bearings were not in "imminent danger of failure;" however, the vibration on the 22 fan was indicative of bearing wear. The OD noted that the bearing should be replaced at the next maintenance opportunity and that, until bearing replacement could be accomplished, weekly monitoring was sufficient to identify negative trends and initiate corrective action. Monthly vibration monitoring was recommended for the 21 fan. Discussions with PSE&G personnel indicated that this OD was initially needed because replacement *bearings were not on han *
The 21 fan was reported to be slightly above the high vibration alarm level, and the 22 fan is well above (1.6 versus 1.207 inches per second) the high critical alarm level. The inspector noted the following concerns with the OD: ( 1) no bearing replacement, expiration, or re-evaluation date was specified; (2) no clear basis was identified for the established vibration limits; (3) the justification for exceeding the established limits and supporting the assertion that bearing failure was not imminent was not clearly articulated; and (4) a vibration limit to preclude imminent failure was not clearly identified. The inspectors observed that the operability evaluation could have been improved if these issues had been initially addressed. The licensee subsequently repaired the bearings prior to commencement of fuel handling activities on June 10, 199 Offsite Dose Calculations During the review of LER 95-24, the operability/reportability determination associated with (AR) 00960506228, licensing memo LR-195601, and discussions with licensee personnel, the licensee's initial position (and explanation) was that the FHV system was not credited for the fuel handling design' basis accident (OBA).
~. *.' -. ; :
During the May 1996 onsite inspection period, the licensee* could not provide the original offsite dose calculations (referenced in the NRC's original 1974 SER and more recently the May 1994 SER associated with Unit 2 Amendment 131 ) to support this statement. The NRC's SER indicates that charcoal filtration was assumed in the licensing review of the fuel handling OBA. Subsequent, discussions with licensee personnel indicated that the offsite dose calculations performed to support the fuel pool reracking in 1994 did credit the FHV syste The reracking was approved by license Amendment 131 on May 4, 1 994. The inspector noted that this information (e.g., offsite radiological consequences) was planned to be updated into the licensing and design basis documentation such as the UFSAR during the next scheduled update in June 1996 (Change Notice 95-25).
The source terms upon which the original SER and Amendment No. 131 are based (page 17, Section 2.8, Table 1) were inconsistent, and did not appear to agree with the current design and licensing basi By letters, dated May 29 and June 4, 1996, PSE&G provided information (including revised calculations) concerning the spent fuel pool re-rack amendment dose assessment for the fuel handling accident. Charcoal filtering efficiency was credited in this analysis. Additionally, PSE&G identified two potentially non-conservative assumptions associated with the fuel handling accident analysis. These were related to iodine removal of spent fuel pool decontamination factor (500) and charcoal filter efficiencies (99%) that were inconsistent with the guidance in Safety Guide (S.G.) 2 % Bypass flow The inspector reviewed DEF DES-91-00628, originally initiated on September 26, 1991, which identified that leakage around the charcoal and HEPA filters exceeded the TS requirement for less than 1 % bypass. This DEF was inappropriately closed in January 1992, based on a statement that the bypass leakage was determined at a higher than expected differential pressure, and that if the expected differential pressure was used, the bypass would be acceptable. No calculation to support this statement was provided at that tim The Salem Technical Specification Surveillance Improvement Program (TSSIP)
identified concerns related to the adequacy of current test procedures to verify this surveillance requirement, as documented in action request (AR) 0096050622 This AR identified that using a common particulate test medium, DOP, (as required by TS 4.9.12) in lieu of halogenated hydrocarbon refrigerant gas, could misrepresent total bypass flo NRC Generic Letter (GL) 83-13 addresses surveillance requirements for HEPA and charcoal absorber units, specifically testing for bypass flow with halogenated hydrocarbon refrigerant. The bases for TS 4.9.12 state that GL 83-13 should be used as procedural guidelines for surveillance testing. The licensee subsequently tested the fitters in accordance with revised procedure 2-IOP-1 * *
E *
FHV Conclusions The inspectors concluded that the licensing and design bases for the Salem FHV system were not well understood nor well integrated prior to June 1996. The lack of clarity and detail with respect to these bases may be a contributing cause of this apparent lack of understanding. Prominent examples included: (1) the single failure design considerations for filter train dampers.(and other equipment); (2) the safety role of charcoal filters (bypass flow, heat load, accident analysis assumptions);
(3) the significance of temperature and pressure control in the building, and the relative merits of an automatic start feature; and (4) uncertainties (and unavailability) of design basis accident calculation The lack of clarity was evidenced by conflicts between the TS, UFSAR, and CBD, as well as internal inconsistencies within the documents themselves. These conflicts appeared to be the result, in part, of weaknesses in maintaining these documents. Efforts to maintain the documents were hampered by the apparent unavailability of records and weak technical resolutions of some issues. The team concluded that fuel handling activities should not commence until these issues were fully and effectively addresse Subsequent to the May 1996 inspection, PSE&G addressed these issues as discussed in a letters to the NRC dated May 29, 1996 and May 31, 1996.
Additional inspections independently verified the actions associated with PSE&G's commitments. A reanalysis of the fuel handling accident was submitted on the Salem docket by letter dated June 4, 1996. Following several conference calls between the NRC and PSE&G, fuel handling activities were resumed on June 10, 199 FHV system design and testing discrepancies which existed prior to 1996 were satisfactorily addressed by PSE&G prior to commencement of fuel handling activities on June 10, 1996. The problems were described in LERs 95-24 and 96-05, but represented licensed activities prior to 1995 and have been addressed by broad licensee programs and actions. These issues represent examples of a violation of 10 CFR 50 Appendix B, Criterion XVI, Corrective Action. The violation meets the criteria of Section Vll.B.2 of the NRC Enforcement Policy. Therefore, discretion is being exercised and the violation is not being cite Control Air System Scope The inspector's review of the design and licensing bases for the safety-related Control Air (CA) system, included comparison of selected portions of the documents listed below, discussions with the system manager, and a brief system walkdow The inspector focused on the capability of the system to perform as described in the Salem UFSAR Section 9.3~ 1 design evaluation. Available documents were reviewed to ascertain how selected aspects of the design basis described in the UFSAR compare with the system's Configuration Baseline Document, calculations, installed equipment, test procedures and test results. The licensee's assessments of the system;s status and design bases were also reviewed for insight into past performance, potential problem areas, and effectiveness of corrective action *
Updated UFSAR Section 9.3.1; and UFSAR Questions 5.44 and NRC Safety Evaluation Report (SER), Sections 9.0 and 9.10 Salem IPE Report, Section 3.4.21 DE-CB.CA-0014(0), Revision 3, "Salem CBD Control and Station Air Systems" and outstanding change notice Control Air/Station Air System Initial Readiness Review Report, dated September 18, 199 Control Air System Load Study S-C-CA-MDC-0462, dated November 11, 1990 PSE&G Response To NRC Generic Letter 88-14, "Instrument Air System Problems Affecting Safety-Related Equipment," March 31, 1989 PSE&G Response To NRC Generic Letter 88-14, Supplemental Information, June 25, 1990 S-C-CA-MEE-0433-01, Comparison Of Power Operated Relief Valve (PORV)
Air Usage Requirements Versus Control Air System Containment Accumulator Capacity, December 30, 199 *
S2.0P-PT.CA-0001 (Q), Revision 5, Emergency Control Air Operability
S2.0P-PT.CA-0002(Q), Revision 1, Emergency Control Air Compressor Flow Capacity Check
Salem System Program Assessment Pilot Report, September 12, 1995
Control Air System Functional Review, March 6, 1989 Observations and Findings System Description The compressed air system is shared by both Salem units and is comprised. of a nonsafety-related Service Air (SA) system and a safety-related Control Air (CA)
system. The CA portion has two parallel headers that serve safety-related load The air supply for the two CA headers is normally provided by the nonsafety-related SA compressors; however, each CA header is automatically backed up by an Emergency Control Air Compressor (ECAC). Safety-related air operated equipment in each Unit is supplied from a local Redundant Air Panel that ensures a reliable air supply from the remaining CA header in case of a single failure. During normal operations, equipment in each Unit is preferentially supplied from one of the CA headers as determined by it's redundant air panel. Unit 1 equipment is preferentially aligned to the to the 'B' CA header, whereas Unit 2 equipment is preferentially aligned to the 'A' CA header. This design comports with the SER statement that any single component failure will not result in the loss of system function. There are no existing Technical Specification requirements for the system. UFSAR Section 9.3.1 states that each ECAC has a capacity of 500 standard cubic feet per minute (scfm) at 110 pounds per square inch gage (psig), and can provide the safety-related control air requirements for both units. Each ECAC is powered from a safety-related bus that is backed up by an emergency diesel generato.*
Design Capability The licensee performed Control Air System Load Study S-C-CA-MDC-0462, Revision 0, dated November 11, 1990, in accordance with commitments to Generic *
Letter 88-14, and in response to an independent System Functional Review (SFR)
performed in 1989. The calculation states that the worst case loading of a single ECAC occurs on the safety-related portion of the 'B' CA header. The calculation
_shows a maximum continuous load of 489 scfm (including an assumed leakage of 50 scfm), and provides an analytical basis that supports the 500 scfm ECAC capacity listed in the UFSAR. Thus, there exists an 11 scfm (2%) margin for compressor degradatio Tested Capability A preoperational test of the CA system performed in September 1979, showed that No. 2 ECAC compressor capacity was 493 scfm at 100 psig. The test procedure indicated that a deficiency report was generated based on the capacity being less than the acceptance criterion of 534 scfm; however, the licensee was not able to retrieve documentation for the resolution of this findin In May 1989, in accordance with a PSE&G commitment to Generic Letter 88-14, the licensee performed a capacity test of the No. 1 ECAC. The test aligned the ECAC to half of the air users it would supply under design basis conditions, and checked the load and unload cycles of the compressor. The licensee concluded that the compressor was loaded (i.e., working continuously to maintain pressure) 45%
of the time; however, the inspector's review of the same test results identified an error and found that the compressor was loaded 71 % of the time. These results indicate the capacity of a single ECAC was overestimated by PSE&G (particularly its ability to supply both Salem units per design). The 1989 test.procedure contained no explicit acceptance criteria, and no formal evaluation of the results was apparen The 1989 test was inadequate. A normally open manual valve was used as the boundary between the safety and nonsafety-related portions of the air header,
- . rather than the 3-inch check valve CA920 which is required to perform this isolation under design basis conditions. PSE&G stated that the CA920 check valves are periodically inspected but that no leak test had ever been performed. Based on the size and hard seat design of this valve, the inspector concluded that zero leakage would be unlikel The licensee also performed temporary surveillance procedure OP-TEMP-9002-2 in June 1990, to test the ECACs' flow capacity. Problems with the test equipment resulted in inconclusive results for the No. 1 ECAC; the No. 2 ECAC showed a capacity of 495 scfm (below the design basis value of 500 scfm). The licensee accepted the Unit 2 test by averaging data taken before the test instrument was driven to,..full scale by moisture in the system. The No. 1 ECAC data were accepted based on similarity with initial No. 2 ECAC data, and the results of the aforementioned 1989 test. The inspector concluded that these test results did not demonstrate the capability of the ECAC compressor *,
Periodic Testing and Maintenance PSE&G performs internal ECAC inspections every two years and an overhaul every five years. The post maintenance testing for this work did not apparently verify the compressors' capacity; this activity alone would invalidate previous test result Surveillance procedure, S2-0P-PY.CA-0001 (Q), Revision 5, is performed monthly to verify operability of the No. 2 ECAC and its associated control air dryers. The surveillance verifies an automatic start of the compressor on low header pressure and development of pressure at the inlet of the air dryer. The inspector identified i:hat the acceptance criterion for automatic start between 82 - 88 psig was inconsistent with the UFSAR description of a start below 85 psig. The licensee was not able tQ identify a detailed basis for the 85 psig set point listed in the UFSA As of October 1996, PSE&G had not yet determined a basis for the ECAC automatic start pressure setpoint. The equipment is currently set at a nominal value of 85 psig and PSE&G has initiated a CR to evaluate the appropriate setpoint, but does not consider this a short term issue. The inspector also noted that the monthly ECAC operability test only verified its low pressure automatic start;. the monthly test does not provide assurance regarding the ability to supply compressed air or to maintain the integrity of the air syste By design, the ECACs have three automatic start signals. The UFSAR states that during plant operation, the ECACs can be tested to verify automatic starting and operability. Monthly surveillance procedure S2.0P-PT.CA-0001 (Q) verifies an automatic start on loss of header pressure and the Mode 3 blackout accident load test S1.OP-PT.CA-0001 verifies the safeguards actuation start signal. However, the third automatic start, for loss of all three SA compressors, is not tested. During review of the 1 989 CA System Functional Review, the inspector noted that this test inadequacy was identified (as part of the self-assessment) but had not been formally tracked nor earmarked for resolutio Design Temperature Issues The inspector found two inconsistencies within Configuration Baseline Document (CBDl Section 5.5.1. The ECAC design specification requires an 85°F maximum cooling water supply for the compressor's coolers. Although the normal chilled water supply meets this criterion, the UFSAR, CBD, and procedures allow alignment of the coolers to the service water system. This would typically be done to support chilled water system maintenance. According to UFSAR Section 9.2.1.2, the maximum service water temperature is 90°F. Secondly, the inspector found that the specification requires a maximum air inlet temperature of 100°F, but that the room's ventilation system can maintain temperature below 110°F under design basis conditions. Therefore, the inspector questioned whether the single ECAC operating during the systems' design basis scenario could trip on high compressed air temperatur '*
A function of the station air system discussed in the UFSAR is to provide forced air cooling for hot piping containment penetrations. UFSAR Section 3.8.1.6.5, UFSAR Question 5.44, and CBD Section 5.10 are consistent with respect to the system being required to maintain concrete temperatures adjacent to hot pipe penetrations below 150°F for structural.reasons. Although the air flow is balanced in accordance with a Westinghouse system manual, there was no basis in the design for the flow settings. This issue was being addressed as part of Salem Restart Issue T-2, "Reliability of Control Air System," separate from this inspectio June - August 1995 Events Problem Report 950612236 identified apparent problems maintaining 'B' CA header air pressur~ fo!lt).wing a switchyard transient on June 7, 1995, that caused loss of the SA compressors. The ECAC apparently did not adequately load or maintain header pressure during the transient. During followup of this event, the licensee identified three changes to the CA system that had been implemented prior to the event; test valves (or "tap offs") had been previously installed in the CA heade Problem Report 950829400, initiated August 29, 1,995, documented discovery that the No. 2 ECAC ('A' CA header) was unable to supply half of its safety-related loads (viz. one of the two Salem units) when aligned to support maintenanc Subsequent investigations found the intercooler and aftercooler drain traps blocked open by internal interference, and that a manual crank on the compressor's unloader valve was not fully retracted. After these conditions were corrected, the compressor appeared to be loaded 65% of the time. A loading of less than 50%,
with the given alignment (one Salem unit), would have been a general indication the system was functional. The licensee concluded the unexpected additional demand was due to numerous system air leaks. The inspector considered this finding an indication that the 'A' header was seriously degraded between June and August 199 Self Assessment The inspector found that the conclusions of the licensee's 1989 System Functional Review (TR-881 64-01 ) regarding the adequacy of routine testing, and questions regarding capacity of the compressors and their ability to supply the safety-related loads, agreed with the findings of the NRC team. The inspector also noted that only 6 of 40 recommendations documented in the 1989 report had been tracked by the licensee. The SFR report further highlighted aspects of the ECAC, including "... a general understanding regarding the high redundancy of the system... a perceived low level of importance... and that the capacities of the compressors and their ability to supply the safety-related portion... is in question." Included in the 6 tracked items were two recommendations regarding the capability of the system to perform its design function that apparently had not been resolved by the license *A more recent independent licensee assessment of the air system performed in September 1995, found that "... in general, the health of the SA/CA system is poor... " as reflected by the dependency on temporary air compressors when making alignment change *
Resolution of Deficiencies (1)
Design capability was addressed by PSE&G as of October 1996, by reviewing the CA loading assumptions in calculation S-C-CA-MDC-1 639 to ensure it is current with all system design changes. After reconsideration of all CA system changes made over the past three years in the loading calculation, PSE&G determined that the predicted system requirements surpass the required design basis compressor capability by approximately 9%.
(2)
Tested capaL.ility - PSE&G resolved the ECAC performance testing issues through revision of ECAC capacity test procedure S1 (2).0P-PT.CA-0002(0),
Revision 6, to provirle. better evaluation. Qf test results. In addition, allowances in the design basis calculation were made to account for CA920 check valve leakage, and a test procedure was developed to provide periodic verification of the valve's condition. The inspector noted that the revised ECAC procedure tests both the compressor's capability and the CA system performance. The inspector concluded that PSE&G's actions adequately resolved the CA system testing deficiencies identified during the May 1996 inspectio (3)
(4)
Periodic Testing and Maintenance - The monthly surveillance procedure S1(2).0P-PT.CA-0001 (0) was being revised to indicate an acceptable start pressure above the 85 psig value described in the UFSAR. PSE&G has not resolved the issue of periodic capability testing, but is considering either monthly or quarterly testing in conjunction with the existing monthly surveillance. The CA capacity test procedure S1 (2).0P~PT.CA-0002(0) was revised to include a test of the automatic start circuit associated with loss of all SA compressors. This capacity test for the ECAC's.is now designated as the post-maintenance testing for the two and five-year maintenance activitie The ECACs were successfully tested following completion of the five-year maintenance overhauls performed during the extended Salem outage. The No. 1 ECAC achieved 515 scfm at 110 psig and the No. 2 ECAC achieved 511 scfm at 110 psig. These results demonstrated that the compressors meet their design basis capacity of 500 scfm at 110 psi Design Basis Discrepancies - Computer evaluations of the ECAC' s determined that the machines were not limited by either the increase in Service Water temperature to 90°F, or the 105°F OBA inlet air temperature of the mechanical penetration. During review of this discrepancy and evaluation of recent test results, PSE&G determined that the limiting factor was a trip set point for high discharge air temperature. The set point was increased from 300 to 340°F through a design change, necessitating modifications to the pipe supports.
. '
The original system description for the containment penetration cooling air flow control valve adjustments (to maintain the surrounding concrete below 150°F) were located by PSE&G following the May 1996 inspection. This information was captured in calculation S-C-PC-MDC-1657 and as a result of this reconstitution, PSE&G determined that adjustments to the flow control valves for the penetration coolers are necessary. The inspector considered this action appropriate however, the valve settings are based on calculation only and no efforts are planned to confirm the penetrations are being maintained below 150°F. This issue was being evaluated separately as part of NRC Restart Item T-Degraded conditions or components were identified by PSE&G during the
- recent testing of the ECJ\\Cs. Fourteen work documents were completed for the No. 1 ECAC and twelve work documents were completed for the No. 2 ECAC in order to adjust, repair, or replace degraded components. PSE&G also recognized that testing the compressors at their design values identified deficiencies that would not be evident during a normal 20 minute loaded run; examples included lubrication oil pressure degradation and the need to revise'
the high air temperature set point. The operability surveillance test procedure (monthly) is currently under review to establish a method and frequency for periodic verification of actual system performance based on the lessons learned from the recent system testin Conclusions The inspector's review of design and licensing basis documentation, surveillance test procedures, and past test results during the May 1996 team inspection led to the following conclusions:
PSE&G's past performance testing, in the aggregate, had never confirmed the capability of the emergency control air compressors to perform their intended safety function described in the UFSAR. The test scope was not adequate, and PSE&G's review of the test results was ineffectiv *
Although testing was planned during the extended Salem outage, PSE&G had not previously identified the inadequacies in the past test results or the existing test methodolog *
The limited design margin and past functional failures in the Summer of 1995, coupled with the poor operational and maintenance history, led the inspector to conclude that it is unlikely the ECAC system had been able to perform its design basis function prior to October 199 *
The lack of Technical Specification requirements for operability and surveillance testing of the ECAC system contributed to a general misunderstanding of the system's importance to safety, and consequently a poor maintenance and operational histor E7 E *
Corrective actions taken by PSE&G during the extended Salem outage addressed the testing inadequacies and evaluation errors identified by the NRC in May 199 The ECAC component adjustments, repairs, and replacements made by PSE&G during their subsequent corrective actions have restored the design basis capacity of the ECAC compressors, and enhanced testing has adequately demonstrated this capacity. PSE&G's update of the system's load calculation and repair work on components assure the integrity of the air system being supplie PSE&G's failure to implement corrective actions for the degraded condition of the CA system is a violation of 10 CFR 50, Appendix 8, Criterion XVI, Corrective Action. The violation meets the criteria of Section Vll.8.2 of the NRC Enforcement Policy. Therefore, discretion is being exercised and the violation is not being cite Quality Assurance in Engineering Activities Configuration Baseline Documents (CBDs)
PSE&G started development of the configuration baseline documents (CBD) for selected plant systems/structures in 1988/1989. The stated purpose of the CBD project was to capture design basis information, with a goal to gather design basis information (for each of the selected systems/structures) in one comprehensive document for use in engineering, installation/modification, operating and training activities. Major objectives of the project were to (1) eliminate/minimize use of invalid or suspect design basis data as design input to future engineering project development and analysis; and (2) improve confidence that safety margins are not exceeded due to design changes. The project scope included assimilation of design basis information into a useable form, verification of accuracy and technical correctness, and indexing of references, and the identification of missing, incomplete or incorrect data. Full system walkdowns and as-built verification were not included in the scope of the CBD project. The CBDs are "controlled" document PSE&G originally developed 49 CBDs for safety-related systems/structures. The CBDs reviewed by the inspectors were found to be comprehensive in scope and contained detailed information. The inspectors' limited review of the CBDs, design-related activities and discussions with the licensee nonetheless resulted in the following observations:
CBDs were never effectiv~ly validated. Currently there is a project underway to validate a Umited number of CBDs for both the Hope Creek and Salem plant *
Review of the configuration baseline document "maintenance" procedure raised a question regarding independent (ANSI N45.2.11) verification. The change notices generated as a part of design change packages are independently verified; however, this verification is not specifically focused upon the CBD change but rather the design package as a whole. The inspectors were concerned -- although no such instances were identified --
that the CBD's could or may have been used for design input in the context of Standard ANSI N.45.2.1 E *
- Although, a CBD revision is required by procedure when the number of change notices exceed 1 5 or when an outstanding change notice is older than 2 years, the Fuel Handling Ventilation CBD had 17 outstanding change notices, of which seven were older than 2 year The licensee subsequently instituted a "freeze" on the use of CBD's pending further activitie UFSAR Changes In August 1995, PSE&G was cited (50-35"4/95-10-01) for failing to incorporate changes into the Hope Creek UFSAR as required by 10 CFR 50.71 (e). PSE&G concluded that this violation also applied to Sq_lern,. As a result, a thorough and detailed root cause analysis was completed in October 1995. This analysis noted that a relatively significant backlog of change notices for Salem and Hope Creek has existed for several years, and that the significance of this backlog was not elevated to PSE&G management's attention until 1994. The analysis reported that in November 1994, the total number of backlogged CNs having the potential for exceeding the 10 CFR 50. 71 (e) requirements was estimated to be 145. Of these, approximately 50% were believed to require UFSAR changes for Salem and Hope Creek. The root cause analysis found a breakdown in the procedural processes required for updating the UFSAR. The breakdown appeared to be caused by a lack of ownership and management oversight of the change process. Corrective actions included updating the program procedures, closing the backlog of identified change notices with an interim UFSAR update in December 1995, and providing training on licensing basis complianc More recently, additional actions have been undertaken to review the adequacy of the UFSAR update process. PSE&G Licensing staff sampled license amendment SERs and completed subsequent to* May 1996, a review of all license amendment The system readiness review program, although not focused or intended to search for licensing basis update concerns, identified approximately 80 items that could result in UFSAR changes. Further, the Onsite Safety Review (OSR) group performed an independent assessment of the UFSAR update process that was focused on the adequacy of the process, although not intended to address whether the plants were being operated consistent with their UFSAR commitments. This assessment concluded that the UFSAR update program was adequate to support Salem restart and the continued operation of Hope Creek. It was noted that some programmatic improvements were still needed, and that UFSAR discrepancies continue to be identifie During the May 1996 team inspection, PSE&G Licensing initiated an action plan to perform a vertical slice evaluation on a sample of systems to, in part, identify weaknesses with the UFSAR and its update program. An overview of the FSAR project is discussed in Section E7.4 of this repor *
E7.3 Commitment Verification Project In December 1995, PSE&G undertook a review to verify that commitments previously made to the NRC were in fact completed. This verification did not evaluate the adequacy or effectiveness of the actions taken. Commitments were viewed as a written statement identifying future action or a completed action in a docketed response to a NRC regulation, finding, or recommendatio One hundred percent of commitments closed from 1990 to 1995 and 25% from the previous five years were reviewed by PSE&G. Of the 2,453 commitments reviewed, the licensee found that 7 had never been implemented, 7 were implemented but inadvertently changed, and 31 were not properly implemente One of the 45 commitments requires a design change to inst?l!-ci.1rbs in the....
entrances of the diesel generator room E Salem FSAR Project Inspection Scope At a meeting held on May 23, 1996, the NRC presented examples, such as those described in this report, wherein the licensing and design bases were not well understood by PSE&G, and that further verification of the Updated Final Safety Analysis Report (UFSAR) was needed. At that time, work was in.Progress or completed by PSE&G which included as-built configuration walkdowns, Maintenance Rule implementation preparations, the Technical Specification Surveillance Improvement Program, and an EOP Upgrade Progra To establish reasonable assurance that Unit 2 would be operated within its licensing basis on restart, PSE&G committed to conduct a Salem FSAR Project to integrate and more clearly focus the full scope of ongoing licensee efforts that related to licensing basis conformance. The Plan also included substantial additional engineering review activities not previously planned. The FSAR Project Plan was presented to the NRC on June 11, 1996, and was the subject of a Public Meeting held on July 2, 1996. From the June 11, 1996, meeting input, the NRC raised 17 questions regarding the scope of the FSAR Project. These were answered in a letter to the NRC, dated August 23, 1996, providing clarification on the following aspects of the project:
Safety Analysis (FSAR Chapter 15) review and matrix preparation
UFSAR "Macro" Review of 47 plant systems
Vertical slice review of 7 systems
Deficiency Evaluation Form (DEF) Closure Review
Review of Engineering Evaluation Closure and related JCO's
Safety Evaluation Review
Process Evaluations
. '
Initial "vertical slices," performed as a pilot effort, examined the Emergency Diesel Generators, Auxiliary Feedwater, Service Water, and 115 Volt AC Distribution systems. These were performed to establish an approach and procedure, and were followed by full review on seven systems including Fuel Pool Cooling, Safety Injection, Reactor Protection System, Auxiliary Building Ventilation, Fuel Handling Building Ventilation, Containment Building Ventilation, and miscellaneous Ventilation Systems (Intake Structure, EDGs, Switchgear & Penetrations).
For the 47 principal plant systems, "Macro" reviews selected primary UFSAR attributes and parameters, and verified a sample of the attributes for confirmation of implementation in drawings, procedures, design change packages (DCPs),
configuration baseline documentation, change notices and license change request These sampled UFSAR attributes were also compared to the Technic!'Jt Specifications and NRC Safety Evaluation Reports (SERs) for consistenc Problems identified during the July-September 1996 conduct of the FSAR Project were documented as Action Requests (ARs). Approximately 190 ARs were written, of which 20 were designated to be resolved prior to plant restart. Four of the issues.identified during the course of the FSAR Project were reported to the NRC as Licensee Event Reports (LERs):
LER 272/96-005-04, Tech Spec Surveillance Implementation deficiencie *
LER 272/96-015-00, CFCU heat removal capability related to biofouling
LER 272/96-018-00, NonSafety-Related RWST piping
LER 272/96-020-00, Containment Fan Coil Units, 45-60 second timing issue The final results and conclusions of the FSAR Project were presented to the PSE&G Management Review Committee (MRC) on September 24, 1996. The licensee concluded that the FSAR project, when combined with related tasks, provided reasonable assurance that Salem Unit 2 would operate within its licensing and design basis. The MRC accepted this reasonable assurance conclusion and the FSAR Project closure documentation packag Regionally-based NRC inspectors followed licensee progress in implementation of the FSAR Program Plan using NRC Inspection Procedure 40501 as guidance. The inspectors conducted several oversight visits throughout the June-September 1996 time period and performed a final onsite inspection in October 1996. The NRC will develop its view on the adequacy of this effort prior to restart, in connection with the Inspection Manual Chapter 0350 proces Deficiency Evaluation Review The inspector assessed the implementation of the licensee's Deficiency Evaluation Form (DEF) review process in July - August 1996. The process was assessed to verify that DEFs which may have insufficient closure documentation would be identified, and subsequently incorporated into the licensee's corrective action progra *
X1
The licensee reviewed previously closed DEFs to verify that the DEF closure was consistent with the plants design/licensing bases. The process selected approximately half of Salem's DEFs for review. The DEFs were selected based on the safety significance of the systems. A total of 59 DEFs were identified as having potential discrepancies with the licensing/design basis. The licensee initiated 7 Action Reports (ARs) to further evaluate these issues. The issues ranged in significance from an absence of accessible documentation on the design basis of nonsafety-related diesel auxiliaries to a potential seismic design inadequacy in refueling water storage tank purification system. Most of the 59 DEFs were potential NRC Regulatory Guide 1.97 qualification issues that had previously been assigned low priority. The inspector selected a random sample of apprmdmately 50 DEFs and found that the licensee had properly evaluated all of the DEFs reviewe Conclusions The inspector concluded that the scope and implementation of the licensee's program for reviewing DEFs were appropriate. The DEFs with insufficient closure documentation had been properly identified and incorporated into the corrective action program. Regarding the comprehensiveness of the overall FSAR project, an NRC safety system functional inspection (SSFI) is being conducted in November-December 1996. The SSFI team is intended to evaluate the component cooling water (CCW) system, and to independently assess the effectiveness of the FSAR project and the licensee's overall activities related to the Salem Unit 2 design and licensing bases. The SSFI team will report o*n its findings and conclusions separate from this inspectio Exit Meeting Summary The team leader, and DRS Division Director and other NRC representatives discussed the preliminary findings (as of May 1966) with licensee management on May 23, 1996. Slides from PSE&G's presentation are attache As discussed elsewhere in this report, several other meetings and telephone conferences have occurred since the ~.
,_
'<
ENCLOSURE *2 UNITED STATES NUCLEAR REGULATORY COMMISSION
REGION I
475 ALLENDALE ROAD KING OF PRUSSIA, PENNSYLVANIA 19406-1415 May 8, 1996 MEMORANDUM TO:
Thomas-T. Martin, Regional Administrator FROM:
..f0 ri.... James T. Wiggins, Director ~7~./7L71/
-_....- Division of Reactor Safety ~
~~
'--
SUBJECT:
SPECIAL TEAM INSPECTION CHARTER FOR REVIEW OF SALEM LICENSING BASIS Because of recent industry experience, a Special Team Inspection (STI} will be conducted to determine whether there are significant licensing basis conformance issues at the Salem station. The team will review the licensee' efforts to identify deviations from licensing basis requirements, with special consideration given to the adequacy and timeliness of proposed licensee corrective actions for any deficiencies foun Attachment 1 represents the charter for the Special Team Inspectfon and details the scope of the inspection. The inspection shall be conducted as a regional initiative during the period of May 6-17, 1996, and will be managed by Gene Kelly of DR A proposed charter was discussed during the Salem Assessment Panel at their April 19, 1996 meetin Docket Nos. 50-272, 50-311 Attachments:
I. Special Team Inspection Charter STI Membership cc w/Attchts:
J. Stolz, NRR W. Dean, OEDO L. Olshan, PD 1-2, NRR W. Kane, ORA, RI R. Cooper, DRP, RI R. Gallo, DRP, RI S. Barber, DRP, RI D. Screnci, PAO, RI D. Chawaga, SLO, RI
ATTACHMENT l
- CHARTER -
SALEM STI 96-80 (May 6-17, 1996) The team's principal objectives are to determine:
(1) if licensing basis conformance issues exist; (2) if so, how significant are those issues; and (3) are the licensee's activities sufficiently scoped and appropriately timed to react to those issue.
The team will determine and assess the:
(1) scope; (2) approach; (3)
status; and (4) results thus far of licensee efforts to evaluate and improve their conformance to the current Salem station licensing basi Emphasis will be placed upon the licensee's efforts to identify and resolve discrepancies in configuration baseline documents (CBDs),
particularly as they relate to the Updated Final Safety Analysis Report (UFSAR) and the Salem licensing basi.
The team will perform a very limited safety system functional inspection (SSFI) of systems selected based on their risk significance, and their maintenance, modification and material condition histor On a sampling basis, the team will compare some details of the actual plant configuration with the current Salem licensing basis, and will evaluate the licensee's appreciation of the difference.
During the first week of the inspection, the team will survey available licensee information for portions of the following ten systems:
Control Area Ventilation
Fuel Handling Ventilation
Turbine Bypass Steam
AM SAC
Station Air and Control Air
Reactor Vessel Level Instrumentation
125 VDC & Batteries
Containment Building Ventilation
4 kV Electric (Vital Buses) Depending upon the findings {namely extent and significance) from the initial survey reviews, three systems will be selected for more detailed evaluation. Those evaluations will include the following: Determine if licensing.basis conformance issues exist and assess their significance. The team will examine relevant documents including engineering evaluations, open JCOs and operability evaluations, LERs written on the system since 1994, and UFSAR Change Notices and License Change Requests.
\\
Attachment 1 2 Evaluate the effectiveness of the design basis program, including configuration baseline documents, associated discrepancy evaluation fonns, and the licensee's current efforts to reconstitute and maintain the design and licensing basis. Also, interview system managers, and evaluate the findings from recent licensee system re.adiness reviews, particularly Level 1 problem report Compare the results of CBDs and system readiness reviews against the. licensing basis requirements for operations, testing and maintenance published in the Technical Specifications (TS),
Updated Final Safety Analysis Report (UFSAR), NRC Safety Evaluations and other fonnal licensing conunitments.
.... ~*....
ATTACHMENT 2 SALEM STI ADMINISTRATIVE REQUIREMENTS STI TEAM COMPOSITION:
Gene Kelly, Chief, Systems Branch, Division of Reactor Safety Fred Bower, Reactor Engineer, Division of Reactor Safety Brian McDermott, Resident Inspector, Susquehanna Paul Swetland, Senior Resident Inspector, Millstone M. Shlyamberg, Contractor PLANNED OATES:
Onsite (2 Weeks):.
Team Debrief to Licensee:
Management Meeting:
LOGISTICS:
May 6-17, 1996 May 17, 1996 May 24, 1996 The team will be based in Building TB2 outside the protected area, in a conference room {OSR Workroom) at (609) 339-105 The PSE&G licensing contact is David Dotson at {609) 339-1282. A draft inspection report or detailed summary will be ready for NRC management review on May 17, 1996.
Presentation to USNRC..
- * *Design/Licensing Basis:* **.
.
.
.
Control Air System
May 23,1996
- .
.
...
, *.' -
fTI
- z p
(./)
c:
- c IT1 w
.
.
Conclusions 1 CA System is not in Tech Specs, not
,credited in Safety Analysis, or
- Chapter 15 UFSAR Accident Analysi Unable to use.data obtained from earlier testing to adequately access that
~
.
'.
~
system meets design basi *
1 Ad.ditional testing will be performed to provide assurance that system meets design basi *
\\..!
System Design I Licensing Basis Control Air (CA) *System is not contained in Tech Spec. I\\
1 UFSAR describes system in* Sections 3.8-31, 7.3'-2, 7.6-6, and 9.3~ *
..,
.
.
.
NRC SE:R that the CA System.d,~s,ign was. satisfactor >> S.ER.~c~eptance basis was that any.sjngle c.omponent failure will not result in loss of system functio..
...
,.*
System Design I Licensing Basis
- *
.
~..... -..: ~;: ~ *=. -~"=*:**.. *:*.:*:=****=* :--:-- r *->~:.""-==:*:*.*.: *==.:**.:*.*:.*::.. ::-:.:::*~-... :"t'~~tt:.. ::::~**=.. *=:: *!......... :.
=
.
___
_
- !;.**
- .....
~'""'...,.,
- *
- >
- ~
~
1 CA System is not_credited in Safety. ':*'
Anal,ysis, system assumed r.ot availabl CA.System in not credited in,
- :::. :.*
Chapter 15 Accident Analysi I>> Basi.s:, P$E&G conctuded that CA: System is l)Ot required for safe shutdown during accident condition *,
~.*
"'.;.
"'
System Design /Licensing Basis 1 In hyp:othetical event that all,.CA-*lost.:
( d-eemed not credible),* au *cc~ntrol '.valves fail to preferred position If affected valves need modulation to go to Hot Standby or Cold Shutdown, use of local handwheels and portable nitrogen bottles is permitte >>Operations procedure 81 (2).0P-AB.CA-0001 (Q) addresses thi~.
.
Commitments General Design Criteria 1 PSE&G committed to Atomic *industrial Forum & Atomic Energy Commission *
versions as... stated in UFSAR.. : !";*
- ... ;
1 1 OCFR50' Appendix A Criterion met, with* exceptions.noted in CBD and.
.UFSAR Section 3~ 1.3. *
- >>* R*eference:. UFSAR Appendix 3A:*. '~PSE&G Positio.n~-on Reg Guides"m
..,,
...
I-I
L
--
Commitments As licensed Regulatory Documents 1 1 OCFRS0.49 for equipment in harsh environmen Reg Guide**1 :22 1 Reg-Guide 1.29
'
.
1 Reg Guide* 1.53 Reg Guide 1. 75 t
'
.
r
.
Commitments As licensed Regulatory Documents 1 FSAR Questiqn 9.41: "_Operational *or ;Je$t Data to Verify the Functional Reliab1ili.ty.. of ithe
- .
...
.
.
. '
- .. -.,
.
ECAC's".
-
l I
\\
..
"'
1 F$AR Question 9.42: "Listing of Critical seNiCe Equipment Supplied by CA".
.
.
1 FSAR Question 7.4: Submit Analysis to show results of Sudden Loss. of ~CA".
1 FSAR Question 9.45:""Describe *Methods to Bri.ng Reactor to Cold SD".
- /'...
.
-
I
- System Design
- ".* *:'... : ~~-. :~~.f!jo :\\:=' "~ ** ~:: ':"...... :*.~ :*.*.,:-:*.*.,.*. _: ~*:** *~*: ** :**:* :~ ::: :: : :":. ;:* *. !~ ':*
-:~......
~*~ :: **.: !::: *. ~.. *:. *.:...* *..
.: :... : *.,.....,,...
' *
~
1 The Control Air System has two parallel and redundant header Each header provides air for b.oth:* *
Safety Related and *Non-Safety Related load..
,..
1 Normal supply iS from Station Air Compressor' * f
' *
System Design.
1 Emergen.cy Cont~ol Air Compressor~s.'.
supply a backup source. of air to Safety Related Equipment. *
Each.~ Stati.on Air Compressor ha.*s '. ::.
capacity to supply Station Air & Control Air Systems for both Units.*
.
,.
I.
I
. '
'
'*
f
.*
.*. ~'
.
System Design 1 Each Emergency Control Air Compressor: has capacity to satisfy*
require.d flow of all Safety Related :air users for both Units*.
1 T)h.ere*are (3) Station Air Compressor's, (2) Emergency *Contro,I Air Compressors, and. (1.) Station Blackoµt Co.mpresso :...
- System Timeline Highlight~
. **
~ * ;.: *, *..... :.- ~... :~: ** ::-! :**:,,r:. "**.!..;:*.:. ;.;:.... ;.. : *::*... ; ~, ~y*:.. ::.; *.;:.,:\\., *:*=-. ** :*.'": ** **
- _,, > **" ~;:~ :~.-::..:: ** * : :.
- ": **.: : ** "'"" ~.. ::,
- * ***
1 Emergency Control Air Compressor capacity test performe /79 (#2), 4/90 (#1 /2), 6/90 (#2).
1 CA System load test p.erforme /89 (#1 ), 4/90 (#1/2); 7/92 (#1 ).
1 CA920 ma*intenance performed 3/92, 8/92, 3/93, 11 /93, 10/9 ~ I
- ..
.
~ I
-,,,
System Timeline Highlights 1 Automatic start of Emergency Control Air-Compre,ssor on loss of (3) Station Air Compressors* tested-8/76, 5/79.:
Monthly Periodic Test performed
~~Emergency, Control Air Compressor:*.;_ *
Operability".;
- -. *
1 SEC functional testing performed evei'y refueling includes Emergency Control Air Com-presser * I
- *
Results Testing and Validation of Design
'l; **
- * *.: *: ~ ***,.: >'!.:;."'
';" *
... *~ !:.,:!*._:::_..',*{* *. :-i:* ** ** * ** ***:::*,. *,.;, ~ ;.-.;:-..~*: **.* '-i:':(.:*."'***:. "'"' *' **** ***,,,,,...
- !~.
- ~."
- *.,*. *
o:' Jo 1. Test DCP* 1 SX-2286 obtained :CA ;
.
-***
System field dat This information was used to develop computer model "Integrated Air Manageme:nt* Program", this model.. was Verified:& Validate '
'
1 PSE&G*calculations were developed to *
document result ~ I
. *
- -
.Results Testing and Validation of Design
'
. **
~
.
...... *": ::;!.. * '*:.::**, :*;:.** :*!i..... *.~*::.* -::<:*.. :.*;:.:.:YO.":;.;~*... :. *:::.'.:'.::::::::::-.* *:: ':* :*:*:*:::"*:"'..*:,:".. ___..... ::.: *.. :::.:.**~.:*it.'...*'*~*-:.*~}:.... *.;:: :.. !. \\
o
..
< '"
~
"'
..
- ** "
"'
- """".,. *.,,*.
- :.....
"
.... '* * ~*..
',...........
""
'*
1 * Load, study concluded that worst case system load was-*489 Standard Cubic Feet: per Minute (SCFM).
Using vendor supplied data~* PSE&G calculated that Emergency Control Air Compressor capac-ityJs 525 SCF.,
- ...
. _,
-
i 1 Capacity and system load field tests
- I were performe *
Results Testing and Validation of De:sign 1 Results of testing concluded that system design *basis functions were validated and testing was considered adequate at the time. (1989.-1992)
1 Self.identified i'n Au!gust 1995 areas of::
weakness~ in,CA Syste*.Had, plan in place to perform additior11al testing prior to Mode 4 to reconfirm.*
assumptions. (AR #950829400)
... '......
Assessment of Design Capability
- lt.,)....
- ;,* :l',i ""'
' ~ ** '
1 * Detailed load study and flow series model provided adequate design review of system load Earlier testing did not adequately provide assurance that system met intended desig. t
- *.
- Assessment of Design Capability 1 PSE&G.:*will perform additio11al a*cti.ons prior to Restart to provide assurc;inpe
.
.
.
.
.. '
,j that CA system performance meets intended design capabilit * *
1 Self identified concerns with CA system necessitated delay of Final System Readiness Review and presentation of USNRC Restart Issue (T2), "CA System Reliability", until.July..
..... '. -~
..
- Actions
'.
~=
- ~. *. **:,* *. ':.- :-.
.'~::;,..:*~:;."'*:.:... ;...: :.*.:,:;.;.,... **:.:: *. :*.;: ~- :';;. :~:.-.,~**.'*' :*;-.,.-:;:;. *:=: :*: ;. ~::.;: ;. *. *: ':'!......... i;-:;::.. ~~:.* :'. *:~;.,.... :* ~ *.. ~,f~ ** : * * * *** *
- * *
-~ *
"
"
- **
...
- ~.*.-.. J ** ~
ii!
- *
.. *****.-. ; * * *
~
"'
1 Test CA920 check.valve and cluantify leakage. (Not required for lr1-Service Testing Program).. : ;...
1 Test Emergency Control.Air, Compressor capac.ity. (Self: Identified)
1 Evaluate CA *.system loading assumptions and testin... '.. ;
i ~
.
- '
- '
.
..
.,
. - ~
I
.
'
Actions
"
"" ~*
- *
, *..,_... *
- :* *" : *:*.. ~ "!.... ~: :*.
'¥::.-~:.:,... : *-.*.-:* :' *:* ~.:.:.::***..'.. :-1:. : :*"'.;~: *:-::.-:".:.:"*: :': :*:: ::.: :-.::* :: '*') :*:,"=:!" :.: "'..... :.*: "" ** :.::.' : -': -~; ~::: ** ;;(... * * ** *:: ** ". :*
- . : :. ; ~...
- *
-.,,. ;,~ ~
"
'* *
'
.. * '*: ""'" * **,"'....
- "'**:*....... * ~
- .
- -.** *
- . * * *,,.
- nO '
..
1 Assessment of all CA System calculations and load studies will be complete.....
\\
1 Test all (3) Emergency Control Air Compressor auto start functions:.....
..
.
.
'
.
,,.
>> Los.s. of :all Station Air Compressors auto st~rt recent,ly verifie " ~
.
....-.. --
- '
Conclusions
1 CA System is not in Tech Specs, not credited: in *Safety Analysis, or Chapter :15 UFSAR Accident A,nalysi Unable to *use data. obtained from earlier testing to adequately-access that system meets design basis..
..*... ' *.. -'::*:>
Additional testing Will be performed t'O provide assurance that system meets design *basi ~
.. : --...
-
-
---
~
- NRC Licensin /Design Basis Review, CFCU's
- Based on the recent NRC audit of Salem licensing and design basis adherence, a number of related issues were identified which may call CFCU capability into question:
- Fan Flow Capability
- FSAR is inconsistent in it's documentation of CFCU input into the Accident Analysi Design does not account for affects of zero-fouling
- Temperature input to CFCU motor/cooler low flow analysis
- -
CFCU Capability, Conclusions 1) The current CFCU design and material condition supports the Accident Analysis. The primary concern is updates of design and licensing basis documentation as plant configuration and analysis techniques evolve Basis:
-
Current analysis of CFCU performance (including reduced air flow considerations) shows in excess of 83.6 Mbtu/hr capabilit The CFCU' s were cleaned this outage. Test results show that a clean CFCU has significant margin over Accident Analysis requirements and also shows that this capability lasts in excess of 18*month ) Presently AR-960515126 is assessing past 89-13 testing to assess if past testing results and material conditions supported the accident analysi *----------------------------------.
- CFCU Capability, Conclusions (cont'd)
3) CFCU licensing/design documentation, and plant configuration are not always consisten ) Most of the CFCU and SW items discussed in the audit were.
previously self identified or were addressed in existing analysi Basis:
-
Low Air Flow was identified in a restart required DEF (93-121) which will force a reconciliation of the issu Accident Analysis inputs ( CFCU required performance) were identified in a restart required DEF (93-121) which will force a FSAR chang The Zero Fouling question was not raised prior to the audi F was considered in the assessment of CFCU moto:* co'oler performance at reduced fl.ow. Documentation could haye been better.
. *.
...
- CFCU Design Basis Functions
- Remove heat from containment during normal plant operations
- In conjuction with Containment Spray, CFCU' s maintain containment temperature and pressure within design limits during LOCA and Main Steam line break accidents. Primarily achieved by steam condensatione
- System Desig!! and Accident Analysis Evolution
- CFCU design and it's relationship to the Accident Analysis has evolved since initial plant licensin Replacement of Westinghouse coils with Marlo Coils.(1982-1985)
- Service water design temperature change from 85 F to 90 F. (1991-1992)
- Development of new design tools for assessment of CFCU design capability.(1982, 1992)
- Changes of input into the Accident Analysis.(1990, 1991, 1992, 1996)
- -
- '
- Adherence to 4licensing and Design Basis
- There are instances where Licensing and Design Basis documentation has not been updated:
-
Westinghouse coils still referenced in the FSAR and the Containment Ventilation C}3 *
-
Approved computer codes used to determine coil capacity have not been updated in the FSA Fouling factors used in the design of replacement coils were updated in the Service Water CBD but not in the FSA FSAR Section 6.2 was not consistently updated as inputs to the Accident Analysis change CFCU air flow rates referenced in the FSAR were not revised based on testing and calculation result I
.,
CFCU Low Air Flow (40,000 scfm.vs. 47,000 scfm)
- In 1991 GL 89-13 testing was instituted. As a result, potential low air flow aud other issues were identifie of three CFCU' s tested just prior to RF0-9 failed. DR# STD-91-030( w/50.59) was initiated and dispositioned as acceptable based on current river water temperature.2/1 /91
-
The 2 remaining CFCU' s failed testing during shutdown RF0- The fallowing were generated to assess operation after the first CFCU failur >> JPO:PSE-9-042 (3/6/1991)
>> EE:S-1-CBV-MEE-0538 (3/6/1991)
>> LER:91-005-00 (4/6/1991)
- I I
-- -*-
- --*---
CFCU Low Air Flow (cont'd)
- The response, while it did address immediate operability and reportability, was weak* in the following areas:
-
The inconsistencies in the FSAR and CBD still remained unchanged
-
Longterm tracking ofthe airflow concern (DEF's93-106, and 93-121) received low priority. DEF 93-106 was closed out based on the Margin Recovery Project which has not been approved by the the NRC (a requirement based on the scope of the revision to the acci~ent analysis).93-121 is still open and is coded restart required for Unit The LER and associated JPO looked only at the intervening time between original discovery (the first CFCU test failure) and RF0-9 shut down. (A.R.-960515126 to validate reporting requirements)
- CFCU Input Paramlers to the Accident Analysis
Currently the Accident Analysis assumes 65Mbtu/hr/unit CFCU capability in the Main Steam line break analysis, and 83.6 Mbtu/hr/unit for the LOCA analysis. FSAR Section 6.2 states that the input to the accident analysis is 8 lMbtu/hr/uni *
This was identified along with the low air flow issue as part of DEF 93-12 *
As stated previously, this issue received low priority based on the assessment that the* reduction in CFCU capacity was minimal. No action was taken to correct the FSA.,
- Zero Fouling
- Since the advent of 89-13, cleaning of CFCU coils has been taking place. As a result, there is the potential that exit piping could exceed current design temperature. A.R.-
960513197 has been initiated to address the following:
-
Impact on system analysis relative to potential flashing in the outlet pipin Impact on SW piping and hanger qualification The Corrective Action Program will force an 'Operability and reportability determinatio I
CFCU Motor/Cooler Low Flow Analysis
- It was noted that the CFCU motor/cooler analysis performed by Westinghouse used the 271 F peak temperature associated with LOCA.vs. the 351 peak associated with the Main Steam line break.*
-
Westinghouse considered using 351 F but selected 271 F based on the short duration (approx 160 sec) that the temperature peaks above the 271 F value during the Main Steam line break even Westinghouse is providing a clarification on the basis for the 271F selection which will be incorporated into Salem's design
- documentatio,.
.
,
I
- - - -
- - - - - - -------------------
CFCU Capability, Actions
- Reassess LER/JPO/EE to determine if Justification for Past Operation scope met requirement..
- Correct inconsistencies identified in the FSAR and Containment Ventilation CB * Complete the assessment of impact related to the "Zero Fouling" questio * Update CFCU motor/cooler low flow analysis to clarify the basis for selection of 271.
,
.
--~~----~~
.
.
~
.
.,
'
PRESENTATION TO USN RC
-.
DESIGN/LICENSING BASES UNIT 2 FUEL HANDLING BUILDING HVAC SYSTEM
May 23~ 1996
\\
.*
INTRODUCTION
Current design consistent with current.
licensing/design bases (with exception of T.S. exhaust fan auto-start req'mt.)
- System design* ffmits offsite dose to well below 1OCFR100 lhlrits (also well below *
more stringent limits of NUREG 0800)
Control room dose < GDC-19 limits
Recent procedure changes increase defense-in-depth by ensuring system aligned to charcoal filter and operating when fuel or load being handled
Key issues self identified
UFSAR clarifications required
.
l SYSTEM SAFETY FUNCTION As originally licensed
Limit offsite dose to small fraction of 10CFR100 limits
Fuel handling accident Since licensed
Limit offsite dose to small fraction of
10CFR100 limits
Fuel pool rerack
Limit control room personnel dose to
<GDC-19 limits
Based on revised calculation in conjunction with control room HVAC redesign
The system supports the accident analysis
- '
)
'
'\\
LICENSING BASIS Original SER & Amendments *
System design considerations
Seismic I for applicable portions
Backup 1 E power
Fire detection
Load handling
Section 15.4 accident analysis
NRC dose assessment demonstrated releases are a small fraction of 1OCFR100 limits
License Amendment 88, Salem Unit 2
Incorporated clarification to basis for testing - referenced R.G. 1.52 Rev 2 (ANSI N510-1975)
.
).
LICENSING BASIS (can't)
License Amendment 151/131 for Salem Units 1 & 2
Most recent pool rerack project
Maintained original licensing basis+.
under 1OCFR100
Filtered release assumed
Safety Analysis Report
Exhaust system seismic Class I/SR
100% capacity, redundant HEPA filter trains to maintain bld negative pressure
Automatically diverts to HEPA/
charcoal filter on high radiation
.,, '
'
,*,
LICENSING BASIS (can't)
- * 1 OOo/o capacity HEPA/charcoal filter train, non-redundant as described in FSAR Section 9.11.1 & Figure 9.11-1 and response to Question 133 regarding exception to R.G.-1.52
UFSAR Section 9.4.3.3 requin?s...
!.,
- ..
clarification
Control room exposure limits - <5 Rem whole body or equivalent per
Technical Specifications
< -.125" H20 building pressure
Charcoal filter tested for > 99°/o efficiency @ 19,490 cfm
High rad statt (unless running), shift to HEPA/Charcoal filters. (Mod in.
progress to install prior to restart)
.,.
'
...
---~---------
i
LICENSING BASIS (can't)
Reg. Guides
R.G. 1.52 (R2) - ESF Atmospheric Cleanup System
Exception to redundant charcoal filter train
R.G. 1.25 - Assumptions Used for Evaluating the Potential Radiological Consequences of a Fuel Handling Accident in the Fuel Handling and Storage Facility for Boiling and Pressurized Water Reactors.
-~
......
.....,
. *"'.
..
LICENSING BASIS (con't)
R.G. 1.13 - Spent Fuel Storage Facility Design Basis
Controlled building leakage during refueling operations
Early detection of fuel damage by rad monitors
Filtration system to limit release of radioactive iodine and other radioactive materials
Local and control room alarms on hi rad
Hi rad actuation of filtration system o
Exception to high rad actuation of filtration system (FSAR 9.1.2.1)
Must assume all rods in one bundle fail
Leakage assumptions per.25
r
~-
.,_
'
.
..
l
..
~
Design Bases ("Safety Significant)
- . Two exhaust fans (only one required to perform safety function)
Maintains bldg at negative pressure
- within Tech. Spec. limits
Area rad monitors shift exhaust to HEPA/charcoal train on high ra *
Exhaust fans 1 E powere *
Continuously monitored exhaus *
Single HEPA/charcoal filter train
Accident rad. exposures w~ll below limits (see table)
Assists fuel pool cooling upon loss of normal cooling
- . Temporary by 50.59 until SFP cooling upgraded
-
.....
' j" i
Design Bases (Non-Safety Significant)
1 supply, 2 exhaust fans
~ 20,000 cfm flow through building
- . Normal temperatures - 60°F to 105°F
Min/max temperatures - 40°F to 120°F.
Supply fan non-1 E powered and non-
..
.
se1sm1c
Exhaust through HEPA filter train
-
-
"-.
I
- . ~
Conclusions
Current design consistent with current licensing/design _bases (with exception of T.S. exhaust fan auto-start req'mt.)
- System design limits offsite dose to well below 1OCFR100 limits (also well below more stringent limits of NUREG 0800)
Control room dose < GDC-19 limits
Recent procedure changes increase defense-in-depth by ensuring system aligned to charcoal filter and operating when fuel or load being handled
Key issues self-identified
UFSAR clarifications required
.,
~
' \\.
"_t
~
"l
- , '.
FUEL HANDLING ACCIDENT ANALYSIS 10CFR100 Limit NUREGBOO Guidelines Rerack Analysis (SER 151/131)
Original Analysis (Salem SER)
Current Decayed Fuel TWO HOUR LOW POPULATION EXCLUSION AREA ZONE Thyroid (rem)
300
13
<10-5 W. Body (rem)
6
1
<10-3
Thyroid (rem)
300 75 <10-6 W. Body (rem)
6
<1
<1