IR 05000259/1995049
| ML18038B458 | |
| Person / Time | |
|---|---|
| Site: | Browns Ferry  |
| Issue date: | 10/11/1995 |
| From: | Moorman J, Peebles T NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| To: | |
| Shared Package | |
| ML18038B456 | List: |
| References | |
| 50-259-95-49, 50-260-95-49, 50-296-95-49, NUDOCS 9510180136 | |
| Download: ML18038B458 (14) | |
Text
~llREgy
~O
++*++
UNITED STATES NUCLEAR REGULATORY COMMISSlON
REGION II
101 MARIETTASTREET, N.W., SUITE 2900 ATIANTA,GEORGIA 303234199 Report Nos.:
50-259/95-49, 50-260/95-49, and 50-296/95-49 Licensee:
Tennessee Valley Authority 6N 38A Lookout Place 1101 Market Street Chattanooga, TN 37402-2801 Docket Nos.:
50-259, 50-260 and 50-296 License Nos.:
DPR-33, DPR-52, and DPR-68 Facility Name:
Browns Ferry Nuclear Power Station Units 1, 2, and
Inspection Conducted:
September 18-22, 1995 Inspector:
Ja es H. Moorman, III rd./D.GS Date Signed Accompanying P
sonnel:
Jeffrey Shackelford, NRR Joel Munday, Resident Inspector, Browns Ferry Approved by:
Thomas A.
eebles, Chief Operations Branch Division of Reactor Safety D
e igned SUMMARY Scope:
This was a special, announced inspection in the area of facility control over activities which had been identified as significant to overall plant risk as
'eported in TVA's recent multi-unit probabilistic safety assessment submittal.
The purpose of this inspection was to evaluate the facility's performance in operations, engineering, and maintenance of selected risk-related components and activities.
Results:
The inspectors found that the facility's performance in operations, training, engineering, and maintenance was adequate with respect to items identified as important to risk by the multi-unit probabilistic safety assessment.
Within the areas inspected, there were no adverse findings related to multi-unit operation.
9510%80l36 95iOiO PDR ADQCK 05000259
PDR Enclosure
~ '
ll il~
REPORT DETAILS 1.0 Persons Contacted Faci 1 ity Empl oyees
- T. Abney, Nuclear Assurance and Licensing
- C. Crane, Assistant Plant Manager
"H. Crisler, Outage
- T. Dexter, Training Manager
- H. Fecht, Corporate Nuclear Assurance and Licensing
- R. Gilbert, Operations D. Hill, Operations Training Manager
- J. Johnson, Site guality Manager
- H. Jones, Senior Engineer E. Kirby, System Engineer
- W. Jones, Nuclear Assurance and'icensing
- R. Machon, Site Vice President
- R. HcHahon, Probabilistic Safety Assessment Corporate Engineer
- T. Overlid, Corporate Nuclear Assurance and Licensing
- P. Salas, Site Licensing Manager
- T. Shriver, Nuclear Assurance and Licensing T. Simmons,,
System Engineer D. Smith, System Engineer
- S. Wetzel, Licensing
- J. White, Outage
- L. Will,iams, Engineering and Maintenance Manager J.
Woodward, System Engineer Other.facility employees contacted included engineers, mechanics, technicians, operators, and office personnel.
NRC Representatives
- L. Wert; Senior Resident Inspector
- Attended Exit Interview 2.0 Background The Browns Ferry site consists of three General Electric BWR-4 reactors with Hark I containments.
Units I and 2 share some systems and crosstie capability exists for other systems.
Crosstie capability also exists for some systems between Unit 2 and Unit 3.
Due to the number of shared systems between Units 2 and 3, the NRC requested that TVA determine how the return to operation of Unit 3 would affect the core damage frequency (CDF) of Unit 2.
This analysis was conducted and the results were documented in the Browns Ferry Multi-Unit Probabilistic Safety Assessment (PSA).
The PSA was transmitted to the NRC on April 14, 1995.
The multi-unit PSA assumes the most limiting case for Unit 2 operational risk is with all three units operating.
With three units operating, it is assumed that any unit will be unable to provide support to another unit Enclosure
Cl 4I i
Report Details through shared and/or crosstied systems.
Given this assumption, an estimate of the Unit 2 CDF was provided.
The results indicated that the CDF for Unit 2 increased by a factor of four.
As presented in the multi-unit PSA, scenarios initiated by loss of offsite power contributed
percent to the CDF and scenarios initiated by internal flooding contributed 22 percent.
The inspectors conducted an in-depth evaluation of the top sequences with these initiators.
An evaluation was performed in the areas of operator performance, training, and procedures associated with the important human actions.
Additionally, engineering and technical evaluations were made in the areas of equipment reliability and availability of the important plant equipment used in the mitigation of the important scenarios.
The following sections provide a brief overview of the functional sequences which were investigated.
The findings and observations associated with the important actions and equipment failures are documented in the report sections associated with important operator actions and equipment reliability issues.
3.0 Review of Dominant Accident Sequences 3. 1 Turbine Building Flooding
~
~
~
The top accident sequence reported in the multi-unit analysis was that of a turbine building flood coupled with independent failures of the residual heat removal (RHR) pumps.
The flood was assumed to be of sufficient magnitude and severity so as to,fail the balance of plant equipment necessary to maintain the condenser as a viable heat sink.
This loss of the condenser, in conjunction with the RHR failures, would presumably lead to eventual core damage.
The inspectors reviewed the licensee's practices and procedures as they related to the detection and mitigation of potential turbine building flooding events.
It was determined that the licensee's flooding analysis had not explicitly specified the potential flood sources which might cause this event.
The inspectors performed a general turbine building walkdown in an attempt to identify those systems whose failure could lead to a
turbine building flood.
The condenser circulating water (CCW),
raw.
cooling water (RCW) and condensate/feed system were identified as those systems which could provide large volumes of water in the turbine building.
However, it was not apparent that these sources would provide a
sufficient volume of water to fail the balance of plant (BOP) equipment as assumed in the PSA study due to the extremely large volume of the turbine building.
The licensee acknowledged that no formal volumetric analysis had been performed to substantiate the assumption that a flood of this magnitude could occur and incapacitate the BOP systems.
Enclosure
Cl
il~
iJ Report Details
The i'nspectors determined that while no separate abnormal operating instruction (AOI) existed for turbine building flooding, the losses of the individual systems and components as a result of a postulated flood would be covered by the individual system AOIs.
The AOIs generally provided operators with adequate guidance for event mitigation.
. In addition to an assessment of procedural adequacy, the inspectors reviewed the material condition and inspection programs associated with the piping systems which could potentially result in turbine building floods.
It was determined that the CCW piping adjacent to the main condenser was subjected to periodic visual inspections-during each waterbox cleaning.
These cleaning activities occurred during every refueling activity and also during any power reduction which resulted in the unit being at or below 70 percent power.
No potential piping degradation had been noted which would indicate that accelerated corrosion existed in the system.
The RCW system is supplied by CCW and is a much lower volume system.
However, flooding events in the RCW pit could also lead to losses of the BOP systems due to the dependencies involved.
The RCW piping is inspected any time the system is opened for maintenance.
The licensee acknowledged that some microbiologically induced corrosion has been noted in isolated
~
sections of the RCW system, and the affected piping has been replaced.
Additionally, the suction bellows for the 2A RCW pump failed during service approximately 18 months ago and was replaced.
The licensee has inspected all of the other RCW pump suction and discharge bellows.
It was determined that some of the RCW suction and discharge bellows had 'been sprayed with corkmastic piping insulation.
The bellows are of an elastomer construction and are not designed to be coated or treated with any type of material.
The licensee instituted a program tp replace all of the affected bellows, and only the ID suction bel.lows remained to be replaced.
[The necessary parts were on order and awaiting shipment.]
The inspectors noted that the suction bellows for the raw service water (RSW)
pumps had also been coated with corkmastic.
The RSW pumps are collocated with the RCW pumps and would also pose a flooding hazard in the RCW pits.
However, the RSW system is comprised of much smaller diameter piping and the system's volume is substantially lower than the RCW or CCW systems.
The RSW piping is a much smaller diameter and the system volume is substantially lower than the RCW or CCW systems.
The licensee acknowledged this deficiency and indicated that actions would be taken to resolve the issue.
Additionally, it was noted that while not credited in the analysis, adequate indications would be available to the plant operating staff to detect turbine building flood events.
Further, for the worst case CCW pipe breaks, isolation of the flood source would be possible by securing the CCW pumps and/or shutting the pump discharge valves which are located outside the turbine building.
3.2 Total Loss of Offsite Power (LOSP) with Emergency Diesel Generator (EDG)
Failures Loss of offsite power sequences comprised approximately 39 percent of the total-CDF in the multi-unit analysis.
The inspectors investigated the Enclosure
0 il~
Report Details licensee's programs, procedures, and assumptions'egarding these scenarios.
It was found that the licensee had implemented adequate procedural controls and had instituted proper maintenance programs associated with the major components which contributed to these postulated scenarios.
Analytical errors in sequences numbers three and seven, were identified.
Even though these particular sequences did not actually lead to core damage, they were assumed to do so. [Electrical power to the RHR system was recovered in these sequences such that core damage would not occur.]
The licensee acknowledged these errors and instituted an investigation to determine the extent of the error.
Additionally, for certain other.LOSP sequences, the analysis assumed that offsite power was recovered, but that the operators would not take the necessary actions to recover the condenser as a heat sink when concurrent RHR system failures were experienced.
Thus, these scenarios would also lead to core damage when an obvious success path existed.
The.licensee's training program had incorporated the necessary guidance
.to recover the condenser as a heat sink under these circumstances.
4.0 System Availability
~
~
~
~
~
4.1 RHR System Availability The inspectors reviewed the licensee's practices and maintenance history associated with the RHR system.
Particular emphasis was placed on the inter-unit crossties between Units 2 and 3.
The inspectors did not identify any maintenance or unavailability deficiencies associated with the RHR system.
It was determined that the major system components were being maintained in accordance with the required maintenance and testing programs, and that no significant maintenance or availability problems existed.
The Unit 3 RHR system was in a turnover status at the time of the inspection, and final system status was pending the results of operational, performance testing.
It was noted however, that even though the individual components necessary for inter-unit crosstie had been tested individually, no integrated flow testing of this capability had been performed subsequent to pre-operational testing.
The ability of the units to share RHR systems was inferred from the individual component/system testing programs.
The inspectors concluded that the licensee had satisfactorily addressed the inter-unit capabilities of the RHR systems.
4.2 Residual Heat Removal Service Water (RHRSW) System Availability The inspectors reviewed the licensee's practices and maintenance history associated with the RHRSW system.
Particular emphasis was placed on the inter-unit crossties between Units 2 and 3.
During the inspection, the B2 RHRSW pump was being replaced by the spare pump and removed for periodic overhaul.
The results of a recent Unit 3 heat exchanger inspection showed the 3C and 3D RHR heat exchangers to be severely fouled on the SW side.
The li'censee attributed this to the extended layup of the system and instituted a flushing program to clean the heat exchangers.
Additionally, Enclosure
0
Report Details 4.3 4.4 an extensive piping replacement program had replaced a considerable amount of RHRSW piping in preparation for the Unit 3 startup.
Subsequent to the piping replacement and heat exchanger flushing, the licensee had collected heat exchanger performance data to use as a baseline for trending future performance of the Unit 3 heat exchangers.
The inspectors determined that the licensee had implemented adequate maintenance practices with respect to the RHRSW system.
Major system components were maintained and tested in accordance with the applicable guide1ines and regulations.
However, the RHRSW crosstie lines which provide the path for standby coolant injection to the RHR system had a portion of piping with a volume of stagnate water.
The licensee indicated that this portion of piping would be flushed out during performance of the RHRSW/Emergency Equipment Cooling Water valve operability tests.
The inspectors determined that the licensee's controls related to maintaining the availabil.ity of the RHRSW system were adequate.
RCW System Availability The inspectors noted a poor maintenance practice with respect to the RCW system.
The licensee had coated the bellows associated with the RCW pump suction and discharges.
as well as the suction bellows of the RSW pumps with corkmastic insulating material.
[See paragraph
.3. 1 Turbine building flooding.]
As noted earlier, these bellows are of an elastomer construction and the vendor recommends that no types of coatings be applied.
Certain coating materials can actually degrade the elastomer as well as obscure the physical condition of the bellows making periodic inspections difficult to accomplish.
The licensee had previously identified this discrepancy and had taken steps to replace the affected RCW system bellows.
As of the date of the inspection only one RCW suction bellows (pump ID) had a coated 'bellows installed and a work request had been initiated to remedy the defect.
However, as also discussed earlier, the RSW pump suction bellows exhibited the same deficiency, and this problem had not been documented by the licensee.
The licensee acknowledged that the RSW pump suction bellows also needed to be replaced and indicated that the appropriate corrective actions would be initiated.
Control Air System Availability The inspectors reviewed the licensee's maintenance practices and history associated with the Control Air System.
Particular emphasis was placed on the recent modification installed to prevent a fault in Units'
or 2 from affecting operation of Unit 3.
A Unit 2-3 isolation valve was installed.
In conjunction with excess flow check valves already in the system, the capability now exists to isolate a fault in one unit from the non-affected unit.
These valves are not tested to ensure operability and there are no current plans to test them.
The quality of air in the system is monitor ed frequently.
Any reduction in air quality would provide reason to test any valve or component in the system whose operation was suspect.
Procedures used to direct operator actions in a loss of control air event appeared adequate for event mitigation.
Enclosure
~
i
~
~
Report Details 4.5 Vital 250 Vdc Battery Board Availability The inspectors reviewed. the licensee's practices and maintenance history associated with Battery Boards 2 and 3.
Recent modifications have increased battery capacity as well as reduced loads.
Si.nce the modification, there have been no adverse trends in maintenance.
Procedures for aligning alternate chargers to the battery boards were adequate.
The inspectors noted minor discrepancies between the procedure and breaker labeling.
The licensee initiated corrective action for the discrepancies.
5.0 General Observations 6.0 The licensee's multi-unit PSA contained analytical errors and conservative modelling assumptions of a nature that made it difficult to assess whether valid qualitative insights could be obtained from the analysis.
Several of the.top (i.e., highest frequency)
sequences were erroneously assumed to lead to core damage when in fact the sequences had been recovered prior to plant damage.
Additionally, several important success paths had not been credited in the analysis.
Thus, many sequences which were assumed to lead to core damage may have been recoverable (or reduced in frequency) if these potential success paths had been analyzed and/or credited.
In particular, the operator actions to realign the high, pressure injection source suction to the condensate storage. tank, vent the drywell, and restore the condenser as a heat sink subsequent to the recovery of offsite power were not analyzed even though these actions are specified by the plant operating procedures.
Additionally, no credit was taken for operator detection and isolation of potential flooding sources in the turbine building flood scenarios even though adequate instrumentation and equipment exists to accomplish these activities.
Interviews with plant operators indicated that they would indeed pursue these potential success paths in an actual event. [All of the aforementioned success paths are proceduralized and are part of the operator training and requalification programs.]
Thus, while all of the observed analytical errors and modelling issues were of a conservative nature, the net effect may have been to elevate the importance of some scenarios relative to others such that true importance measures and ranking would be difficult.
Therefore, it is possible that while the analysis may have provided a conservative estimate of the upper bound of the CDF during multi-unit operations, its use would be of limited value for prioritizing important operator actions, systems, components, or maintenance activities.
Exit Interview At the conclusion of the site visit, the inspectors met with representatives of the plant staff listed in paragraph one to discuss the results of the inspection.
The licensee did not identify as proprietary any material provided to, or reviewed by the inspectors.
The licensee did not express any dissenting comments.
Enclosure
II I
O~
II>