05000412/LER-2005-001
| Docket Number07 19 2005 2005 -0001 -001 11 18 2005 | |
| Event date: | 07-19-2005 |
|---|---|
| Report date: | 11-18-2005 |
| Reporting criterion: | 10 CFR 50.73(a)(2)(i)(B), Prohibited by Technical Specifications |
| 4122005001R01 - NRC Website | |
PLANT AND SYSTEM IDENTIFICATION
Westinghouse-Pressurized Water Reactor {PWR} Containment Isolation System {JM} Main Steam System {SB}
CONDITIONS PRIOR TO OCCURRENCE
Unit 2: Mode 1 at 100 percent power There were no systems, structures, or components that were inoperable at the start of the event that contributed to the event other than as described below.
DESCRIPTION OF EVENT
On July 14, 2005, one of the two Main Steam System isolation valves (2MSS-S0V105C) on the steam supply line from 21C steam generator leading to the steam turbine-driven auxiliary feedwater pump (TDAFWP) unexpectedly opened at Beaver Valley Power Station (BVPS) Unit No. 2. The control room operators verified that no start signal to the TDAFWP existed or was necessary and that the redundant steam isolation valve (2MSS-S0V105F) on that steam line remained closed. 2MSS-S0V105C was declared inoperable. The TDAFWP remained operable via steam lines supplied from steam generators 21A and 21B.
During follow-up investigation actions on July 19, 2005, it was determined that solenoid operated valve (SOV) 2MSS-S0V1050, located in the Main Steam Valve Room, inappropriately opened due to a shorted coil in the SOV. This caused an associated control circuit relay (3-MSSCTX) located in the Emergency Switchgear Room to fail when it opened at an amperage beyond its design current interruption rating. This relay failed by melting its contacts in the solenoid operated valve (SOV) coil circuit. The molten metal from this relay's contact melted through the relay's outer clear plastic dust cover and fell on the relay located below it (3D-SSRAB). The molten material also melted a small hole through the outside plastic relay dust cover of 3D-SSRAB, which could potentially cause this relay to fail. A failure of relay 3D-SSRAB could inhibit the control signals going to containment isolation valves 2SSR-SOV-128A1/129A1/130A1 (sample valves) and 2PAS-S0V105A1 (post accident sample valve), potentially preventing them from automatically closing during a Containment Isolation Phase A (CIA) signal.
Upon learning of this information on July 19, 2005, the control room operators conservatively declared the four containment isolation valves affected by relay 3D-SSRAB inoperable, deactivating these valves closed in accordance with BVPS Unit 2 Technical Specification 3.6.3.1, Action a. This action to declare these four containment isolation DESCRIPTION OF EVENT (Continued) valves inoperable was conservatively taken even though the initial visual inspection indicated that relay 3D-SSRAB may still have adequately functioned.
Relay 3D-SSRAB was replaced. Subsequent bench testing of the original relay 3D-SSRAB indicated that this relay would not pick up when its coil was energized with 125 VDC and would not have functioned to transfer the CIA closure signal from the protection system to the four applicable containment isolation valves between the time when relay 3-MSSCTX failed on July 14, 2005 and when these four valves were deactivated closed on July 19, 2005.
REPORTABILITY
BVPS Unit 2 Technical Specification 3.6.3.1 requires that the each containment isolation valve be operable. When it was first discovered on July 19, 2005 that Relay 3D-SSRAB was potentially degraded, the four containment isolation valves potentially adversely affected by Relay 3D-SSRAB were de-energized closed in accordance with Technical Specification 3.6.3.1 Action a. Subsequent testing showed Relay 3D-SSRAB was not operable due to the molten metal that dripped from relay 3-MSSCTX on July 14, 2005.
Action a requires "With one or more penetration flow paths with one containment isolation valve inoperable, isolate the affected penetration flow path within 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> by use of at least one closed and deactivated automatic valve." Since this was unknown on July 14, 2005 that Relay 3D-SSRAB failed making containment isolation valves 2SSR-S0V 128A1/129A1/130A1 and 2PAS-S0V105A1 inoperable, the containment penetrations associated with these four valves were not isolated with a deactivated closed valve within four hours as required by Technical Specification Action a. This was a condition prohibited by plant Technical Specification, and is reportable pursuant to 10 CFR 50.73(a)(2)(i)(B). A Licensee Event Report is required to be issued within 60 days of July 19, 2005, when the degraded Relay 3D-SSRAB condition was discovered.
The failure of the solenoid coil on the steam isolation valve which initiated this event was reviewed for potential 10 CFR Part 21 reportability. As described below, this coil failure was attributed to a manufacturing defect due to a lack of required insulation on a SOV coil lead connection. 2MSS-S0V105A, B and C are isolation valves on the three steam lines leading to the steam-driven auxiliary feedwater pump. A coil failure on this SOV conservatively causes the SOV to fail open, allowing steam to travel to the safety related turbine driven auxiliary feedwater pump. However, these steam isolation valves also perform a containment isolation function on the outside of containment for these steam lines, which requires these valves to close upon a containment isolation signal. Each line has a second SOV in series (2MSS-S0V105D, E and F), which can also be used for an REPORTABILITY (Continued) additional outside containment isolation capability. Each of the three steam lines are directly connected to its associated steam generator, which is a closed system. Thus, the primary barrier credited for the containment isolation function on these steam lines penetrating containment, per General Design Criteria 57, is the closed system on the inside of containment provided by the steam generator tubes, which would be unaffected by any coil failure on the outside steam isolation valves. Thus, this coil failure mechanism in this BVPS application is not a significant safety hazard and not reportable pursuant to 10 CFR Part 21 since the containment isolation function would not be completely lost by a coil failure on these steam isolation valves.
A failure of a coil in other similar BVPS SOV applications was evaluated and it was determined that a SOV coil failure in these applications would not lead to a loss of a required safety function. Thus, this coil deficiency is not reportable pursuant to 10 CFR Part 21 since it does not involve a substantial safety hazard for BVPS applications.
CAUSE OF EVENT
The direct cause of this event was that the failure of 3D-SSRAB was a passive failure from unexpected collateral damage at a location remote from the SOV failure. This relay failure presented no immediately recognizable indications. This resulted in not meeting the 4 hour4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> requirement of Tech. Spec. 3.6.3.1 Action a. Operator actions at the time of discovery of the potential failure of 3D-SSRAB were appropriate and timely.
The root cause of this event was that there was no process requirement that would prompt personnel who identified the failure of 2MSS-S0V105C to conduct an immediate investigation of the remainder of the circuit to look for failure of other components in the circuit in the four hour requirement of the Technical Specification. The only way that the collateral damage to relay 3D-SSRAB may have potentially been found within the four hour requirement of Technical Specification 3.6.3.1 Action a would have been a visual inspection.
EVENT ANALYSIS
Relay 3-MSSCTX failed due to 2MSS-S0V105C failing. The initial laboratory analysis determined that the coil of 2MSS-S0V105C failed due to a poorly crimped connection in the SOV coil made by the manufacturer, which led to the connection heating. This heating ultimately led to insulation failure, resulting in the coil shorting. Subsequent laboratory analysis found that this crimped connection was uninsulated, leading to the connection shorting to the coil wire. Field measurements of the SOV coil resistance indicated that the resistance of the failed coil was approximately 4.5 ohms compared with the approximate 90 102 ohms for a good coil. With this greatly reduced resistance, the current in the circuit to EVENT ANALYSIS (Continued) resistance of the failed coil was approximately 4.5 ohms compared with the approximate 90 102 ohms for a good coil. With this greatly reduced resistance, the current in the circuit to 2MSS-S0V105C was much higher than normal. When the SOV opened, its Position Switch PS2 opened, deenergizing relay 3-MSSCTX, which opened contacts 113-114 to break the SOV circuit which was carrying a current estimated at approximately 10 amps due to the reduced coil resistance. This 10 amps was well beyond the contact's break rating of 5 amps, but remained below its overcurrent device setting. This melted contacts 113-114 of relay 3-MSSCTX. The molten material from contacts of the 3-MSSCTX relay melted through its relay outer plastic dust cover and fell on to the 3D-SSRAB relay below it, melting a small hole through the outer plastic dust cover of 3D-SSRAB. The material that melted its way into relay 3D-SSRAB came to rest in contact with the moving and fixed portions of the relay contact block. The material cooled, adhering to these surfaces such that the relay contacts would not move when 125 volts DC was applied. Since relay 3D-SSRAB remained in its normal arrangement and was not called upon to operate from July 14, 2005 until it was found damaged on July 19, 2005, there was no indication that relay 3D-SSRAB would not operate.
The initial failure analysis performed by Beta Laboratory determined that the failure of 2MSS-S0V105C was due to a poor internal crimped connection that attaches one of the two external leads to the coil windings which shorted to the coil windings through the fiber tape.
The report indicated that the SOV was in good condition with the exception of localized heating in the area of the crimped connection. The coil in 2MSS-S0V105C failed after only 6 months of service. At that time, the most likely reason for the failure was a poor electrical connection inside the crimped connection that led to overheating at the connection and eventually shorting to the coil windings.
Subsequent to the coil failure for 2MSS-S0V105C on July 14, 2005, 2MSS-S0V105A unexpectedly failed open on September 1, 2005. This failure of 2MSS-S0V105A did not involve any other collateral relay failures. A post-failure analysis indicated that this failure on 2MSS-S0V105A was similar to the failure of 2MSS-S0V105C on July 14, 2005. The lead wire of the coil had penetrated the fiber tape barrier on the outside of the coil and shorted to the coil. This penetration occurred at the location where the start coil wire and its stranded lead wire were crimped together on each of the failed coils. The crimp connections were uninsulated on both failed coils. The bare metal crimp of the lead wire penetrated the fiber tape barrier and coil wire polyimide insulation. This caused a majority of the coil windings to be bypassed. The cause of the failure of these two SOVs was inadequate insulation during manufacturing of the solenoid coil leads crimp connections. It is believed that the thermal expansion that the coil experienced due to heat from its proximity to the Main Steam Line and being continuously energized contributed to the puncture of the fiber tape and coil wire insulation, eventually allowing the crimp connector to short to the coil windings. These two EVENT ANALYSIS (Continued) premature coil failures involved Target Rock replacement coils manufactured by Rotech Electronics, Inc. in lot number 9105 manufactured in 1991. Though detailed failure analysis of coil failures prior to 2005 at BVPS is not available, a review of past BVPS operating data shows that eight coils from lot number 9105 have failed, six of which occurred within 6 months of installation (which includes the two failures in 2005). These coils failures are premature since the coils have a 21 year qualified life. Only Target Rock replacement coils manufactured by Rotech in lot number 9105 and used in the steam supply isolation to the BVPS turbine driven auxiliary feedwater pump application have failed prematurely at BVPS.
SAFETY IMPLICATIONS
The four containment isolation valves (CIVs) adversely affected by the 2MSS-S0V105C controlling relay (3-MSSCTX) melting event on July 14, 2005 were:
- 2PAS-S0V105A1, inside CIV for post accident containment atmosphere sample, is normally closed and fails closed.
- 2SSR-S0V128A1, inside CIV for primary coolant hot leg sample, is normally closed and fails closed.
- 2SSR-SOV129A1, inside CIV for containment sump and Residual Heat Removal System liquid sample, is normally closed and fails closed.
- 2SSR-S0V130A1, inside CIV for pressurizer relief tank gas sample, is normally open and fails closed. It is normally isolated upstream by 2SSR-S0V125A, 125B, 140, 141, and 2SSR-A0V119A, all of which are normally closed and fail closed.
2SSR-S0V128A1, 2SSR-S0V129A1 and 2PAS-S0V105A1 are normally closed and remained closed from July 14 through July 19, 2005. Thus, these valves would have already been closed in their safety position if a condition causing a CIA condition would have occurred. However, 2SSR-S0V130A1 would not have been closed since it is normally open. However, there are other valves in this sample line which would have been closed preventing any direct path out of containment. In addition, 2SSR-S0V130A1 remained capable of remote manual operation from the control room and could have been manually operated closed as directed by the Emergency Operating Procedures.
The plant risk associated with the BVPS Unit 2 event that occurred on July 14, 2005, due to the failure of controlling relay (3-MSSCTX) and the subsequent failures of the containment isolation valves is considered to be very low. This is based on the very low risk impact of having 1 out of 3 main steam lines to the TDAFWP being isolated and the negligible risk impact from the affected containment isolation valves. These affected containment isolation valves are currently screened from the PRA containment isolation model due to their very low risk of contributing to a radioactive release.
Based on the above, the safety significance of the event that occurred on July 14, 2005, was very low.
CORRECTIVE ACTIONS
1. All coils manufactured by Rotech Electronics, Inc. in lot number 9105 at BVPS will be evaluated for replacement.
2. Provisions will be incorporated into appropriate operating procedures to require operating personnel to visually inspect the seal-in relays associated with the 2MSS-S0V105 series valves failing open.
3. An operating experience report has been issued on the subject providing the details of the event and component failures. An additional operating experience report has been issued on this subject describing the additional evaluation findings on this coil defect.
Completion of the above and other corrective actions are being tracked through the BVPS corrective action program.
PREVIOUS SIMILAR EVENTS
A review found one potential prior BVPS Unit No. 1 and one potential BVPS Unit No. 2 Licensee Event Report within the last five years involving either relay or containment isolation valve discrepancies.
BVPS Unit 1 LER 2000-007, "Technical Specification Non-Compliance Due to Misinterpretation of Containment Isolation Valve Requirements for GDC 57 Penetrations.
BVPS Unit 2 LER 2001-003, "Condition Inadvertently Exceeds Technical Specification Allowed Outage Time.
BVPS Unit 1 LER 2000-007 is not similar to this LER because it involved the human error of not complying with the containment isolation valve Technical Specification as written and misapplied previous NRC correspondence. BVPS Unit 2 LER 2001-003 is not similar to this LER because it involved a human error in not declaring an undervoltage relay inoperable due to not recognizing the importance of unexpected status light indications which signaled a failed solid state input/ protection channel.
COMMITMENTS
There are no new commitments made by FirstEnergy Nuclear Operating Company (FENOC) for BVPS Unit No. 2 in this document.