05000282/LER-2011-002
| Prairie Island Nuclear Generating Plant Unit 1 | |
| Event date: | 07-01-2011 |
|---|---|
| Report date: | 08-23-2011 |
| Reporting criterion: | 10 CFR 50.73(a)(2)(iv)(A), System Actuation |
| 2822011002R00 - NRC Website | |
was manually tripped from approximately 100% steady-state power. The manual reactor trip was in response to CV-31183, the right (south) steam turbinel Right Stop Valve (RSV)2 failing closed as the result of an Electro-Hydraulic (EH) oil leak located at the RSV. As designed, 11 and 12 Auxiliary Feed Water 4 (AFW) pumps auto started on steam generators low water level.
The PINGP Unit 1 turbine-generator consists of a steam turbine that drives an electrical generators.
High-pressure steam enters the turbine through two turbine stop valves and four governing valves.
One turbine main steam stop and two main steam governing valves form a single assembly. An electro-hydraulic (EH) actuator controls each turbine stop valve so that it is either in the wide-open or closed position.. The stop valve has a hydraulic cylinder equipped with a test port blank flange sealed with an O-ring.
Timeline:
July, 1997 Actuator was rebuilt October, 2009 Unit 1 Control Valve and Stop Valve Actuators replaced during 1R26 (Unit 1 refueling outage, Cycle 26) December, 2009 Noted small oil leakage from Right Stop Valve (RSV) CV-31183 actuator and scheduled repair for 1R28 April, 2011 Unit 1 operated without any significant issues involving the RSV for the operating cycle (October, 2009 — April, 2011) April —June, 2011 1R27 - Performed Left Stop Valve work, no work done with or to the
RSV
June 11, 2011, approximately 00:38 CDT Unit 1 in Mode 1, end of 1R27 July 1, 2011, approximately 15:52 CDT RSV failed shut resulting in a Unit 1 manual Reactor trip Post-trip investigation determined that there was an EH system leak at the RSV. The leak was determined to have originated from the back side of the Actuator. The test port flange 0-Ring seal was found failed. A new 0-Ring was installed and Unit 1 entered Mode 1 at 17:03 CDT on July 2, 2011.
1 EIIS System Code: TA 2 EIIS Component Identifier: ISV 3 EIIS System Code:
TG 4 EIIS System Code:
BA 5 EIIS Component Identifier: SD 6 EIIS System Code: TB 7 EllS Component Identifier: SEAL The last known rebuild of the RSV actuator was in 1997 by J. M Grimstad Inc. This actuator remained in the warehouse for approximately twelve years until it was installed during 1R26 in October, 2009. It was in service for less than two years before the 0-ring failed.
On December 8, 2009, (soon after the 1R26 outage) a Work Request (WR) was initiated due to oil leakage from the RSV actuator. This WR was converted to Work Order (WO) and scheduled for 1R28 (along with other major valve assembly work). It was not recognized that the slight leakage could have been a precursor to the RSV 0-ring failure. Unit 1 had operated without any significant issues involving the RSV for the operating cycle (October, 2009 — April, 2011). No work was scheduled during 1R27 involving the known actuator leakage. Unit 1 was placed in Mode 1 on June 11, 2011 at 00:38 CDT. The RSV 0-ring failed requiring a manual Reactor trip at 15:52 CDT on July 1, 2011, Prior to disassembly, the as found torque values were recorded. Values of 50, 200, 250 and 375 in- lbs were recorded. The lowest values were recorded on the side the 0-ring failed on. The 0-ring condition was inspected and there was a crack all the way through the cross section. Additionally, the 0-ring had a split along the outer circumference that propagated approximately 90 degrees of the circumference. The 0-ring also showed signs that a bursting phenomena occurred. The 0-ring was found to be pliable and capable of being tightly bent 180 degrees about itself without showing any signs of distress.
Since the pressure force for this 0-ring is on the inside radius, a careful inspection of the inside radius was performed. No evidence of nicks, cuts, abrasion or foreign material were observed that indicate the initiating point of the failure was from the inside surface of the 0-ring. From the inspection, failure initiation appears to be from the outside surface.
The inspection identified two primary issues. First, the 0-ring was flattened somewhat. There was not enough flattening to classify it as a true compression set however, this flattening was an abnormal indication. This flattening is most likely age induced deformation of the 0-ring. Second, the 0-ring shows slight signs of extrusion on the low pressure edge. Extrusion occurs when an 0-ring is squeezed into the gap between the metal surfaces. The observed extrusion is not as pronounced as in the industry literature. However, a slight extrusion is apparent. This can be caused by excessive clearances, excessive pressure, improper sizing, etc. Inspecting the origin of the failure indicates the initiation point for the 0-ring failure is at the observed extrusion location. Extrusion of this type caused a stress concentration and appears to ultimately have caused the leakage with CV-31183.
The manual Reactor trip is reportable under 10 CFR 50.73(a)(2)(iv)(A) since there was a manual actuation of the Reactor Protection8 system (RPS). During the transient, Steam Generator water level went low (as expected) and the auxiliary feedwater system (AFW) actuated (as designed). The autostart of the AFW system is also reportable under 10 CFR 50.73(a)(2)(iv)(A).
6 EIIS System Code:
JC
NRC FOR
-2010) M 366A (10 LICENSEE EVENT REPORT (LER)
CONTINUATION SHEET
U.S. NUCLEAR REGULATORY COMMISSION 05000 282 YEAR 2011 - 002
NO
- 00
SAFETY SIGNIFICANCE
This event did not challenge nuclear safety as all plant systems responded as designed. Therefore, this event does not represent a safety system functional failure for Unit 1. However, the 0-ring failure resulted in the RSV failing closed and required the operators to manually trip the Reactor. This initiating event challenged reliable operation and any unplanned reactor trip is a significant event. The plant systems responded as designed so there were no radiological, environmental, or industrial impacts associated with this event. Therefore, this event did not affect the health and safety of the public.
CAUSE
The causal evaluation determined that right main turbine stop valve failed closed due to excessive oil leakage from a failed 0-ring. An age related compression set and slight extrusion near the outside contact surface resulted in the failed 0-ring.
The compression set caused the 0-ring to behave differently inside the groove. The slight extrusion condition causes stress concentration at that location. Combined with the slightly less torque values found on the failure location side, the 0-ring failed and ultimately burst causing excessive leakage and the RSV to fail closed.
A contributing cause was that actuators do not have serial numbers to allow for tracking. Additionally, no tracking mechanism exists for spare turbine control actuators resulting in the installation of an actuator that had been last rebuilt in 1997.
CORRECTIVE ACTION
A review of the past actuator replacements will be conducted to determine if proactive 0-ring replacement is warranted. WR(s) will be initiated as necessary to perform replacements.
Preventive Maintenance (PM) procedures will be revised to ensure that the 0-rings on the Stop Valves and Control Valves are replaced when the actuators are replaced. This approach makes the contributing cause involving serial number tracking irrelevant.
The proper cap screw torque value(s) will be determined and evaluated.
PREVIOUS SIMILAR EVENTS
A LER search was conducted and no similar events involving age-related failures of 0-rings were identified in the last three years at PINGP.