05000254/LER-2012-003

LER-2012-003, Degraded Flood Protection Barrier

Docket Number Sequential Revmonth Day Year Year Month Day Year Quad Cities Nuclear Power Station Unit 2 05000265Number No.
Event date:	08-05-2012
Report date:	10-04-2012
Reporting criterion:	10 CFR 50.73(a)(2)(i)(B), Prohibited by Technical Specifications
2542012003R00 - NRC Website

PLANT AND SYSTEM IDENTIFICATION

General Electric - Boiling Water Reactor, 2957 Megawatts Thermal Rated Core Power Energy Industry Identification System (EllS) codes are identified in the text as [XX].

EVENT IDENTIFICATION

Degraded Flood Protection Barrier

A. CONDITION PRIOR TO EVENT

Unit: 1 Event Date: August 5, 2012 Event Time: 2350 hrs Reactor Mode: 1 Mode Name: Power Operation Power Level: 100%

B. DESCRIPTION OF EVENT

On August 5, 2012, station operating personnel initiated surveillance, Reactor Building [NG] Floor Drain [DRN] Sump Ball Valve [ISV] Leakage Testing. This surveillance verifies the leak tightness of the Reactor Building Floor Drain Sump Ball Valves for Unit 1 and Unit 2. The drain sump ball valves (one associated with each Emergency Core Cooling System corner rooms, four per unit) are normally closed to isolate the four-inch drain line between the ECCS corner room and associated reactor building basement sump. Each Unit has four ECCS corner rooms, the drain sump ball valve prevents propagation of water from the reactor building basement into the associated ECCS corner room during a postulated flooding condition (the drain valves also prevent water propagation between ECCS corner rooms should a leak initiate inside a corner room).

During the performance of the surveillance on Unit 1 on August 5, 2012, the drain sump ball valve associated with the 1B Core Spray [BM] corner room exhibited leakage (pencil sized stream). Operations declared the 1B Core Spray subsystem inoperable at 2350 hours0.0272 days <br />0.653 hours <br />0.00389 weeks <br />8.94175e-4 months <br /> on August 5, 2012, in accordance with procedure, "Control of In-Plant Flood Barriers and Watertight Submarine Doors." An engineering evaluation was completed supporting a temporary drain plug, which was installed in the 1B Core Spray corner room floor drain; the 1B Core Spray subsystem was then declared operable at 1615 hours0.0187 days <br />0.449 hours <br />0.00267 weeks <br />6.145075e-4 months <br /> on August 6, 2012. This was completed prior to performing the surveillance on the 2B Core Spray Room.

During the Unit 2 portion of the surveillance on August 6, 2012, the drain sump ball valve associated with the 2B Core Spray corner room exhibited similar leakage (pencil-sized stream). Operations declared the 2B Core Spray subsystem and the Unit 2 RCIC [BN] system (which shares the same ECCS corner room) inoperable at 1758 hours0.0203 days <br />0.488 hours <br />0.00291 weeks <br />6.68919e-4 months <br /> on August 6, 2012. In addition, given that the fire door between the 2B Core Spray corner room and 1A Core Spray corner room is not designed to be leak tight, the 1A Core Spray subsystem and Unit 1 RCIC system were concurrently declared inoperable. A temporary plug was installed in the 2B Core Spray corner room floor drain and the ECCS equipment associated with the 2B and 1A Core Spray corner rooms was declared operable at 1814 hours0.021 days <br />0.504 hours <br />0.003 weeks <br />6.90227e-4 months <br /> on August 6, 2012.

Surveillance, Reactor Building Floor Drain Sump Ball Valve Leakage Testing, was completed on August 9, 2012. The remaining drain sump ball valves on Unit 1 and Unit 2 tested successfully.

An apparent cause investigation was initiated to determine the causes of the failed leak tests of the 1B and 2B Core Spray Room Floor Drain Ball Valves.

Subsequent inspections on both failed valves (1B and 2B Core Spray Room Floor Drain Ball Valves) found that due to wear between the valve stem and associated actuator the valves were not completely closed when at the actuator travel stop, which prevented further hand-wheel closure. Since the same condition was found for each valve at the time of the surveillance (i.e., valve actuator stop prevented full closure), this evidence would indicate that the condition existed prior to the surveillance. Since the as-found leakage was the result of the valve actuator not properly indicating the valve position, the as-found leakage condition also likely existed prior to discovery, therefore this event is reportable (Units 1 and 2) per 10 CFR 50.73(a)(2)(i)(B), which requires the reporting of any operation or condition which was prohibited by the plant's Technical Specifications.

C. CAUSE OF EVENT

The drain sump ball valves were previously float-operated globe valves. These valves failed to provide adequate flood protection due to debris collecting on the seat and not allowing the valve to fully seat. A 1998 modification installed the manual ball valves on the floor drain lines from each ECCS room. The drain sump ball valves are four­ inch valves manufactured by Apollo Valves. An extension connects the valve to a hand wheel operator. The extension interfaces with the valve using a valve-to-actuator coupling. The hand wheel operator has a local indicator displaying whether the valve is open or closed.

Subsequent inspections discovered wear between the valve stem and associated valve actuator; therefore, the valves were not completely closed when the actuator was of its travel 'stop' position, which prevented further hand-wheel closure. The actuator stop set screw was out of adjustment due to service wear, which induced play in the actuator­ to-valve coupling such that the position indicator indicated full close even though the valves were slightly open (this was visually verified using a bore-scope). This condition resulted in the as-found leakage. The actuator gearing located outside the sump had no visible play or wear. The actuator stops were subsequently adjusted allowing additional turns of the hand wheel to achieve full valve closure. Full closure was visually verified by use of a bore­ scope.

The apparent cause of the 1B and 2B Core Spray Room Floor Drain Ball Valves to fail their leakage test was due to a failure to anticipate during design development that wear related degradation would occur at the valve-to-actuator coupling utilized in this unique configuration, allowing the valve to not be fully seated despite the actuator indicating full travel. A contributing cause to this event was wear-related degradation of the lubrication of the valve internals, which led to increased friction, contributing to the wear induced degradation of the valve-to-actuator coupling that was not anticipated during design development.

D. SAFETY ANALYSIS

System Design The design basis for ECCS corner room flood protection is discussed in Updated Final Safety Analysis Report (UFSAR) Section 3.4.1.2.2 (Protection of the Emergency Core Cooling System). The internal flood protection features ensure a single failure (flood) will not preclude safe shutdown. The ball valves prevent propagation of water from the reactor building basement into the associated corner room during a postulated flooding condition (the drain valves also prevent water propagation between ECCS corner rooms should a leak initiate inside a corner room). The drain sump ball valves will not mitigate flooding from potential overhead sources since there are open stairwells down into the corner rooms. The ball valves are normally closed, but may be opened temporarily to facilitate maintenance activities (such as system draining).

Safety Impact Reactor Building Floor Drain Sump Ball valve leakage tests were performed on both units on August 5 and 6, 2012.

The ball check valve that prevents backflow from the drain into a corner room was found to leak in the 1B Core Spray and the 2B Core Spray/RCIC rooms. This means that a flood in the Unit 1 Torus Room could also flood the 1B Core Spray Room, and a flood in the Unit 2 Torus Room could flood the 2B Core Spray/RCIC Room. In the case of flooding in the Unit 2 Torus Room and 2B Core Spray/RCIC Room, the 1A Core Spray/RCIC Room would also flood because the door between the 1A Core Spray Room and the 2B Core Spray Room is a fire door and not a flood barrier.

The safety impact is considered minimal given the following:

• The as-found condition would not have impacted safe shutdown capability. Given an internal flooding event on either Unit 1 or Unit 2, with no credit for operator action, at the time of discovery at least one division of low pressure and one high pressure ECCS system would have been available to support safe shutdown (i.e., for each unit, two (2) Residual Heat Removal (RHR) [BO] subsystems, one (1) Core Spray subsystem, and the High Pressure Coolant Injection [BJ] system were unaffected).

• The as-found drain sump valve leakage was small (pencil sized stream). Given the size of a typical ECCS corner room, significant flooding is not an immediate threat to ECCS equipment.

• Flooding in the reactor building basement would generate a control room alarm [LA] from high-level in the reactor building floor drain sump. The control room alarm would prompt an investigation of the area by operating personnel. It is reasonable to conclude the leakage would have been detected and mitigated as part of the expected operator response.

Risk Insights A risk assessment was completed based on the as-found conditions. Unavailability for the 1B Core Spray pump needs to be considered for the case of a Unit 1 reactor building basement flood. Similarly, unavailability for the 2B Core Spray pump, Unit 2 RCIC, 1A Core Spray pump, and Unit 1 RCIC needs to be considered for a Unit 2 reactor building basement flood. The reactor building basement flooding Core Damage Frequency (CDF) contribution combined with RCIC unavailability is 4.30E-08/yr (Core Spray unavailability does not impact this contribution). This results in a change in CDF of 6.2E-09/year which is considered negligible.

In conclusion, the overall safety significance and impact on risk of this event were minimal.

E. CORRECTIVE ACTIONS

Immediate:

1. The 1B Core Spray and 2B Core Spray (CS) room floor drain sump ball valves were adjusted allowing an additional 1.5 turns of the operator to achieve full closure. Full closure was visually verified by use of a bore­ scope. The subsequent leakage tests were successful.

2. Revised Reactor Building Sump Ball valve leak test PM frequency from 4 years to 2 years.

Follow-up:

1. A PM requirement will be developed for the drain sump ball valves of the 1(2) A RHR, 1(2) B RHR, 1(2) A CS, and 1(2) B CS, to perform periodic repair/replacements of the valve-actuator coupling and gearbox, and a PM will be developed to perform periodic repair/replacements of the valve internals.

F. PREVIOUS OCCURRENCES

The station events database, LERs, EPIX, and NPRDS were reviewed for similar events at Quad Cities Nuclear Power Station. This event was the 1B and 2B Core Spray Room Floor Drain Ball Valves failed their leakage test due wear related degradation that occurred at the valve-to-actuator coupling that allowed the valve to not be fully seated despite the actuator indicating full travel closed. Based on the cause of this event and associated corrective actions, the events listed below, although similar in topic, are not considered significant station experiences that would have directly contributed to preventing this event.

• Station Events Database — Issue Report 931415 (06/15/09), RHR A Room Floor Drain Ball Valve Will Not Fully Stroke. Operations performed a leak check on the 2A Residual Heat Removal (RHR) Room drain valve (2-4899­ 124) to verify the valve holds water when closed to perform its function as a flood barrier. The valve was cycled full open to verify the ability to pass and then closed. The linkage was found to bind during stroking the valve fully open, and required adjustment or lubrication. The valve operator (hand wheel gear box) was rebuilt, but the valve still did not open properly. The valve and actuator were ultimately replaced. Following replacement of the ball valve and actuator, the components were inspected and noticeable wear (mushrooming) within the valve coupling was visible. Slop, or play, was also noted in the valve coupling to the point where one and one-half turns of the hand wheel were noted before any movement occurred within the valve. This event is similar in that a valve actuator was not properly indicating the valve position, however in this event the valve position indication error occurred while in the open position and did not impact the leak tightness of the valve, hence this event did not result in a station investigation or LER. Therefore this event is not directly applicable and is not considered a significant station experience that would have directly contributed to preventing the event of this current LER.

• LERs - A review of LERs at Quad Cities Nuclear Power Station over the past 10 years did not identify any similar events that were associated with leakage due to valve actuators not properly indicating the valve position.

G. COMPONENT FAILURE DATA

The primary cause of the event was the valve actuator (stem extension) was not properly indicating the valve position.

The stem extension manufacturer is AVK-Carbo-Bond, Inc., Model No. 7'-3" extension. The valve manufacturer is Apollo, Part No. 88A-24A-21-TC. The valve operator manufacturer is Mastergear, Part No. 9804.

This event has been reported to ICES as Failure Report No. 300769.