05000454/LER-2008-001

LER-2008-001, Technical Specification Non-Compliance of Containment Sump Monitor Due to Improper Installation During Oriainal construction

Docket Number
Event date:	03-28-2008
Report date:	06-13-2008
4542008001R00 - NRC Website

A. Plant Condition Prior to Event:

Event Date: March 28, 2008 Unit 1 was in Mode 6, Refueling No structures, systems, or components were inoperable at the start of this event that contributed to the initiation or mitigation of this event.

Background:

This report involves the Unit 1 Reactor Coolant System (RCS) [AB] leakage detection Instrumentation.

The Limiting Condition for Operations (LC[}) for Unit 1 Technical Specifications (TS) 3.4.15, "RCS Leakage Detection Instrumentation," requires one containment sump (RF) [IJ] monitor and one containment atmosphere particulate radioactivity monitor (PR) [IL] (i.e., 1PRO11A). The RF monitor consists of the containment floor drain sump flow monitor (i.e., 1RF008) and the reactor cavity sump flow monitor (i.e., 1RF010). TS 3.4.15 is applicable in Modes 1 through 4. If the required RF sump is inoperable, Condition A requires the performance of the RCS leak rate surveillance once per 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> and restoration to operable status within 30 days.

For the Unit lcontainment sump monitor to accurately quantify RCS leakage, the RCS leakage is collected and directed to the weir (measurement) box in each sump. This allows the detection of a RCS leak of one gallon per minute (gpm) within one hour.

B. Description of Event:

On March 28, 2008, with Unit 1 in a refueling outage, an Engineer (non-licensed) identified a potential operability issue with the 1RF008. He noticed that the sump's cover was flush with the containment floor and had two instrument tubetrack penetrations through its cover with each having an approximately five linear inch narrow gap on its perimeter. These gaps, open to the sump below, would allow RCS leakage to flow directly into the sump and bypass the collection system and weir box. The issue was placed into the Corrective Action Program and the condition was corrected by installing plates around each of the penetration gaps to block the flow path. This was completed prior to entering the mode of applicability for TS 3.4.15. In addition, an action was assigned to the Engineering Department to assess the status o past operability of the 1RF008.

On April 14, 2008, the past operability assessment concluded that the tubetrack gaps caused the 1RF008 to be inoperable dating back to installation. RCS leakage flowing through these gaps could, adversely impact the 1RF008 design requirement to detect a one gpm leak within an hour.

An investigation revealed that the original design and construction of 1RF008 in the 1976 timeframe should have had it installed three inches higher, such that the sump cover would be three inches above the floor elevation. This installation detail would have ensured, by being three inches higher than the �NRC FORM 366A (9-2007) PRINTED ON RECYCLED PAPER floor, that an RCS leak to the containment floor in the vicinity of the containment sump cover would be directed to the 1 RF008 leakage collection system and weir box and not allowed to bypass through the penetration openings in the cover.

The 1 RF008 was in an inoperable condition longer than allowed by TS 3.4.15. 10 CFR 50.73 (a) requires the reporting of events that occurred within three years from the time of discovery.

Consequently, since this condition existed since April 14, 2005, then it is reportable to the NRC in accordance with 10 CFR 50.73 (a)(2)(i)(B). The 60 day time clock for submitting a Licensee Event Report to the NRC began on April 14, 2008, when it was concluded the 1 RF008 was inoperable in the past.

The same installation error for 2RF008 occurred on Unit 2, however a GEMS level instrumentation modification installed in Spring 2004 changed the design such that all RCS leakage is measured regardless of the flow path taken to the sump.

C. Cause of the Event:

Due to the time frame the installation error was made (i.e., 1976), the investigation was unable to determine the cause for this installation error. The most probable cause was inadequate quality control oversight of the installation process.

Throughout Unit 1 service life there were several design changes to various aspects of 1 RF008, however the review of the 1 RF008 design during these modifications did not identify this installation error. These missed opportunities to identify this deficiency are attributable to human performance weaknesses such as inattention to detail and having too narrow a focus of the task.

In the 2004 on Unit 2 timeframe, during a design change review for the GEMS level instrumentation modification, it was identified that this three inch installation discrepancy existed. However, only the structural impact was reviewed and it was found to be acceptable. The adverse impact on operability of the 1 RF008 was not recognized.

D. Safety Analysis:

There was minimal safety significance to this condition. A RCS leak in Modes 1 through 3 and most of Mode 4 would flash to steam. Most of this steam would be condensed in the Reactor Containment Fan Cooler System [BK] and the condensation directed to the 1RF008 collection system as designed.

In Mode 4 RCS temperature ranges where RCS may not flash completely to steam then only a leak in the vicinity of the 1RF008 cover could have had any significant portion of its flow diverted through the open gaps. Since the identified bypass area is small in relation to intended design path area to the collection system, it is reasonable to assume that a much greater portion of the leakage would be correctly collected and directed to the weir box. If one conservatively assumes of 50 percent of the leakage flow was diverted through these gaps, then the 1RF008 would not detect a one gpm leak rate � until the actual leak was at two gpm. Based on engineering judgment, this is still considered small enough for Operators to take appropriate actions to place Unit 1 in a safe condition (i.e., Mode 5) The TS required 1PRO11A was capable of detecting a RCS leak of one gpm within an hour. During the brief periods of time when both 1RF008 and 1PRO11A were inoperable, Operators had several additional means to identify RCS unidentified leakage. These include the RCS Volume Control Tank level indicators, post accident containment sump [IP] level instruments, containment pressure and temperature indicators, and Pressurizer [AB] Level indicators.

E. Corrective Actions:

The 1RF008 penetration openings were blocked such that RCS leakage could not enter the sump via the sump cover.

The design drawings for the Unit 1 and Unit 2 RF008 will be revised to indicate the as-built installation.

The Engineering lessons learned from the missed opportunities in identifying this condition will be communicated to the Engineering staff.

F.Previous Occurrences:

Issue Report (IR) 113088 dated June 24, 2002, describes incorrect installation issue with the RF008 instrumentation.

IR 522160 dated August 21, 2006, describes the removal of the six inch lip at the bottom of the ECCS sump outer screen, which could result in online RCS leakage spilling into the ECCS sump instead of the RF sump.