05000369/LER-2007-002

LER-2007-002,

Docket Number
Event date:
Report date:
3692007002R00 - NRC Website

BACKGROUND

Applicable Energy Industry Identification (EIIS) system and component codes are enclosed within brackets. McGuire unique system and component identifiers are contained within parentheses.

Containment Floor and Equipment (CFAE) sump discharge containment isolation valve [WD] (WL):

The Unit 1 CFAE sump discharge containment isolation valve, 1WL-65B, is part of the Liquid Waste System [WD] (WL). 1WL-65B is located on the discharge side of the CFAE sump pumps outside containment. 1WL-65B is normally closed and is remotely opened to allow pump out of the CFAE sump to the Liquid Waste system [WD] (WL) for processing. CFAE sump discharge inside Containment Isolation Valve (CIV), 1WL-64A, is normally open. Both valves are interlocked with the CFAE Sump pumps and Incore Instrumentation Sump pumps such that the pumps cannot be started unless the valves are opened, and the pumps will trip if any of these valves should close. The safety function performed by these valves is to close upon receipt of Phase "A" containment isolation signal (Engineered Safety Feature). Valves 1WL-65B and 1WL-64A are Grinnell Diaphragm valves (Duke Item # 05B-153) with Rotork electric actuators (model 11NA1-57).

Rotork actuators are vulnerable to slip the primary switch in the open direction when manually operated. For this slip to occur, the actuator is over rotated during manual operation until the over travel guide bar is against the stop on the actuator. A diaphragm valve stem backseat cannot reach the backseat of the valve bonnet therefore there is no hard stop for open travel. Because a diaphragm valve can continue past full open, once the over travel guide bar is against the actuator stop, the overload clutch will slip to prevent actuator damage. When this occurs, the open and closed limit switch settings are shifted. Consequently, the open limit actuates later in the open direction and the closed limit switch actuates earlier in the closed direction which results in leakage past the seat of the valve.

Essentially, the physical stroke length has not changed but the starting point from the open de-energized position has changed.

Containment Isolation Valve as-found Isolation Valve Leak Rate Test (LRT) are performed during end of cycle refueling outages. Individual leakage rates are not specifically part of the acceptance criteria of 10 CFR 50, Appendix J. Leakage rates exceeding individual limits only result in the containment being inoperable when leakage results in exceeding the overall acceptance criteria of 1.0 La (maximum allowable Type A leakage rate at pressure).

McGuire Technical Specification (TS) 3.6.1 - Containment:

TS specify that the containment shall be OPERABLE in MODES 1, 2, 3 and 4.

Maintaining the containment operable requires compliance with the visual examinations and leakage rate requirements of the Containment Leakage Rate Testing Program in accordance with TS Surveillance Requirement (SR) 3.6.1.1.

The TS 3.6.1.1 Bases states, "Failure to meet specific leakage limits for the air lock, secondary containment bypass leakage path, and purge valve with resilient seals (as specified in LCO 3.6.2 and LCO 3.6.3) does not invalidate the acceptability of the overall containment leakage determinations unless the specific leakage contribution to overall Type A, B, and C leakage causes one of these overall leakage limits to be exceeded.

As-left leakage prior to the first startup after performing a required Containment Leakage Rate Testing Program leakage test is required to be� 0.6 La for combined Type B and C leakage, and Type A leakage." At� 1.0 La the offsite dose consequences are bounded by the assumptions of the safety analysis.

TS 3.6.1, Condition A, require an inoperable containment to be returned to an operable status within one hour.

TS 3.6.1, Condition B, require the unit to be placed in Mode 3 within 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> and Mode 5 in 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> whenever the required action and associated completion time is not met.

McGuire Technical Specification (TS) 3.6.3 - Containment Isolation Valves:

TS 3.6.3 specify that each containment isolation valve shall be operable in Modes 1, 2, 3, and 4.

Per 3.6.3 Note 4: Enter applicable Conditions and Required Actions of LCO 3.6.1, "Containment," when isolation valve leakage results in exceeding the overall containment leakage rate acceptance criteria.

As per Condition A, if ONE containment isolation valve in a flow path that contains two isolation valves is inoperable, the affected penetration flow path shall be appropriately isolated within 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />.

As per Condition B, if TWO containment isolation valves in a flow path that contains two isolation valves are inoperable, the affected penetration flow path shall be appropriately isolated within 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br />.

If the required action and associated completion time of Condition A or B are not met, then TS 3.6.3, Condition F, states that the respective Unit must be in Mode 3 within 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> AND in Mode 5 within 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br />.

TS Surveillance Requirement 3.6.3.8 requires the combined leakage rate for all reactor building bypass leakage paths is .07 La when pressurized to Pa, 14.8 psig. If reactor building bypass leakage not within limit, TS 3.6.3.D requires leakage to be restored within limit in 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />. If the required action and associated completion time are not met, then TS 3.6.3, Condition F, states that the respective Unit must be in Mode 3 within 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> and in Mode 5 within 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br />.

McGuire TS 3.8.1 AC Source-Operating TS 3.8.1 specifies that two diesel generators (DGs), capable of supplying the onsite Essential Auxiliary Power Systems, shall be operable in Modes 1, 2, 3, and 4. With one DG inoperable, required action B.2 specifies that the required feature(s) supported by the inoperable DG shall be declared inoperable when its required redundant feature(s) is inoperable. The completion time for required action, B.2 is within 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> from discovering that one DG is inoperable concurrent with inoperability of redundant required feature(s).

EVENT DESCRIPTION

Note: The following timeline shows the sequence of pertinent activities that occurred between the last acceptable LRT of CIV 1WL-65B and the as-found LRT failure on March 22, 2007.

9/27/05 Local Leak Rate Testing performed and 1WL-65B meets acceptance criteria during 1EOC17 refuel outage.

10/03/05 Work order 562987 (PT 1WL-65B/Elec/Mech Inspection of Rotork Actuator) was performed on 1WL-65B. IP/0/A/3066/002D (Rotork Actuator Preventative Maintenance) directs the manual operation of 1WL-65B and it was cycled at least twice manually during the PM.

The requirement to LRT the valve following manual operation was waived.

10/14/05 Unit 1 entered startup Mode 4 at 1558 hours0.018 days <br />0.433 hours <br />0.00258 weeks <br />5.92819e-4 months <br /> 10/16/05 (WL Train B Valve Stroke Timing - Quarterly) was PT/1/A/4502/002 B performed on 1WL-65B.0 (.6)0 The stroke time was six tenths of a second different than previous recorded stroke times but within the acceptance criteria specified in the test procedure.

Unit 1 entered shutdown Mode 5 at 1904 hours0.022 days <br />0.529 hours <br />0.00315 weeks <br />7.24472e-4 months <br /> 3/10/07 3/22/07 Valve 1WL-65B failed Isolation Valve Leak Rate Test PT/1/A/4200/001 C Enclosure 13.40 (Test Sheet for Penetration M- 374) 4/20/07 It was determined that the failed LLRT on penetration M-374 resulted in exceeding the combined leakage rate for reactor building bypass leakage.

On March 22,0Containment Isolation Valve 1WL65B failed Isolation Valve 2007, Leak Rate Test PT/1/A/4200/001 C Enclosure 13.40 (Test Sheet for Penetration M-374).

Investigation revealed that the most probable cause of 1WL-65B failing its leak test was an open limit switch slip due to manual operation past the open limit. Although the exact slippage event could not be determined, a periodic valve stroke timing test performed on 10/16/05 indicated a change in the stroke time of 0.6 sec had occurred. The condition is assumed to have existed since the last successful leak rate test completed September 27, 2005. Subsequent calculations indicated the amount of slippage to be about 0.114 inches of stem travel in the closed direction which would have resulted in additional seat leakage.

The estimated leakage exceeded the combined leakage rate for reactor building bypass leakage paths (5.07 La; 9820 sccm) and the specific conditions of TS 3.6.3 D were applicable during Modes 1 through 4. The leakage rate also exceeded the combined leakage rate for penetrations and valves subject to type B and C test (5.60 La; 84,200 sccm) and Integrated Leak Rate Testing (ILRT) acceptance criteria (5.75 La; 105,000 sccm). TS 5.5.2 (Containment Leakage Rate Testing Program) specifies the acceptance criteria for leakage rates and TS 3.6.1 condition A applies whenever leakage exceeds the acceptance criteria for type'B and C test or Integrated Leak Rate Testing.

On October 14, 2005, Unit 1 entered Mode 4, which requires an operable containment system in accordance with TS Section 3.6.1 (Containment) and 3.6.3 (Containment Isolation Valves). Since the inoperability of 1WL-65B is being considered to have existed since the last successful leak test completed September 27, 2007, several TSs applied when Unit 1 entered Mode 4 and until the unit was placed in Mode 5 on March 10, 2007. The condition of the valve was not recognized and no TS required actions were taken to ensure remained in operation for a period longer than allowed by TS sections 3.6.1, 3.6.3, and 3.8.1. In addition to not meeting these required completion times, surveillance requirements 3.6.3.8 and 3.6.1.1 were not met. This condition represents an operation prohibited by Technical Specifications and is reportable per the requirements of 10 CFR 50.73 (a) (2) (i) (B).

Although individual leakage rates are not specifically part of the acceptance criteria of 10 CRF 50, Appendix J, combined leakage rates exceeding the acceptance criteria of 0.75 La result in the containment being inoperable. Therefore, Unit 1 was operated with an inoperable containment during cycle 18 and the condition is reportable per 10 CFR 50.73 (a) (2) (i) (B), "Any operation or condition which was prohibited by the plant's Technical Specifications.

For short periods of time during the cycle, the associated emergency power source was not available to the redundant CIV (1WL-64A). Unit 1 train "A" Diesel Generator (1A D/G) was removed from service 10 times for a total of 88.15 hours1.736111e-4 days <br />0.00417 hours <br />2.480159e-5 weeks <br />5.7075e-6 months <br /> during cycle 18, Mode 1 through 4. In addition to being reported as a condition prohibited by TS, this condition is also reportable per 10 CFR 50.73 (a) (2) (v) (C) as a condition that could have prevented the fulfillment of the safety function. This is because CIV 1WL-65B's normally open redundant feature, CIV 1WL-64A, would not have automatically closed assuming a loss of offsite power. Loss of offsite power is assumed concurrent with 1WL-65B's inoperability and inoperability of the 1A D/G due to planned maintenance.

CAUSAL FACTORS:

The most probable cause for LRT failure was manual operation of CIV 1WL-65B past the open limit causing the open limit switch to slip. The actuator was over rotated until the over travel guide bar was against the stop on the actuator. Once the over travel guide bar was against the actuator stop, the overload clutch slipped to prevent actuator damage. When this occurs, the open and closed limit switch settings are shifted. Consequently, the open limit actuates later in the open direction and the closed limit switch actuates earlier in the closed direction which results in leakage past the seat. Corrective actions following a previous occurrence (LER 370/2005-006) included a requirement to perform leak rate test following manual operation of containment isolation valves with Rotork operated diaphragm valves. The LRT was not performed due to human error.

The shift in valve stroke time was used to justify the extent of condition operability of similar Rotork operated diaphragm valves; however, corrective actions were not developed to identify future step changes in valve stroke time as indicators of potential failures of these valves.

The root cause was determined to be barriers that were put in place following a previous LRT failure of CIV 2WL-65B (LER 370/2005-006) were not robust enough to prevent human error from waiving a LRT requirement which resulted in the repeat condition on CIV 1WL-65B.

CORRECTIVE ACTIONS

Immediate:

1. 1WL-65B limit switches internal to the rotork actuator were adjusted followed by a successful Leak Rate Test.

Subsequent:

1. Actions were taken to evaluate the Unit 1 and 2 rotork operated diaphragm valves for extent of condition and no additional discrepancies were identified.

Planned:

1 Revise Appendix J program to perform Local Leak Rate Test (LRT) on all unit specific Rotork motor operated diaphragm valves prior to Mode 4 during refueling outages.

2 Perform Valve Stroke Timing on all unit specific Rotork motor operated diaphragm valves prior to Mode 4 during refueling outages.

3 Install positive controls (such as locks or tamper seals) to the hand/auto levers of all Rotork Actuated Diaphragm Valves after successful LRT and revise the appropriate administrative and technical procedures to control the locks and/or tamper seals.

SAFETY ANALYSIS

Based on a review of the potential consequences of excessive leakage through valve 1WL-65B, it is concluded that there is no change in the estimated Core Damage Frequency (CDF) or Large Early Release frequency (LERF) for McGuire.

The McGuire PRA categorizes containment isolation failures of less than 6 inches in diameter as small. Small containment isolation failures are not included in the LERF estimation because they have been shown to make a negligible contribution to early offsite health consequences. The WL line that 1WL-65B is in is a 4 inch line and would therefore be classified as a small containment isolation failure. The event of interest is not a complete failure of the valve but leakage beyond the allowable.

Consequently, there are no LERF implications for the excessive leakage experienced by the valve.

This pathway does not support any accident mitigation function and does not impact the core damage frequency calculation directly. Any potential contribution for diversion of containment sump inventory is judged to be negligible given that the redundant valve must fail, the sump pumps should not be running, and the small effective area of the failure would allow a very long time before the sump inventory could deplete. Based on this qualitative evaluation, there is a negligible impact on the CDF evaluation.

ADDITIONAL INFORMATION

Since 2WL-65B had a similar failure during 2E0017 (PIP M-05-1794), this failure is classified as a Recurring Event (LER 370/2005-06).

Additional actions have been identified during the root cause investigation that strengthens station programs, procedures and processes to ensure corrective actions minimize the potential of recurring events.