05000313/LER-2005-003
| Docket Numbersequential Revmonth Day Year Year Month Day Yearnumber No. 05000 | |
| Event date: | 12-26-2005 |
|---|---|
| Report date: | 02-24-2006 |
| Reporting criterion: | 10 CFR 50.73(a)(2)(iv)(A), System Actuation |
| 3132005003R00 - NRC Website | |
A. Plant Status At the time of this event, Arkansas Nuclear One, Unit 1 (ANO-1) was operating in Mode 1 at approximately 95 percent power.
B. Event Description
At 1047 CST, on December 26, 2005, ANO-1 experienced an automatic actuation of the Reactor Protection System (RPS)[JC] due to a Main Turbine [TA] trip in response to low turbine bearing lube oil [TD] pressure. The low lube oil pressure condition was the consequence of a failure of oil ejector discharge check valve, LO-79, when failure of a weld within the valve internals caused the valve disc to separate from the hinge. The disc lodged in the valve body and momentarily blocked adequate flow of bearing lube oil within the system, creating the low pressure condition which caused the Main Turbine to trip. Backup motor driven pumps started immediately to supply bearing lube oil, but not before a turbine trip.
The RPS performed as designed in response to the turbine trip resulting in an automatic shutdown of the reactor from approximately 95 percent power. Following the reactor trip, an invalid actuation of Emergency Feedwater (EFW)[BA] occurred in response to a spurious low "A" Once Through Steam Generator (OTSG) [AB] level signal which was generated by the Emergency Feedwater Initiation and Control System (EFIC)[JB]. Using alternate steam generator level instrumentation, operators verified that an actual low level condition was not present in "A" steam generator. Accordingly, operator action was taken in accordance with procedures to restore normal steam generator level control. The plant was promptly stabilized in Hot Standby (Mode 3) conditions.
C. Root Cause
The reactor trip was initiated by a trip of the Main Turbine which was caused by low turbine bearing lube oil pressure. Analysis, which included radiography testing and ultimately the disassembly of the turbine lube oil ejector discharge check valve, LO-79, determined that the valve disc had separated from the hinge due to a failure of the attachment weld. The disc then lodged in the valve body obstructing lube oil flow to the extent that oil pressure was reduced below the Main Turbine trip setpoint. Examination of the failed welds indicated failure from overload stresses, meaning that the welds were undersized for the applied load. The most recent maintenance on LO-79 prior to this event was performed to repair the hinge sleeve and disc wear pad, which were discovered to be cracked. This weld repair was performed in October, 2005, during refueling outage 1R19.
� Subsequent evaluation of the weld and design engineering processes employed during this repair activity determined that these processes did not adequately facilitate the implementation of the intended like-for-like repair of the valve.
C. Root Cause (continued) Consequently, adequate welding design information for reconstruction of the hinge-to-disc configuration was not developed or applied during maintenance. These factors resulted in a failure to accomplish a like-for-like repair of LO-79, and the weld repair made during the 1R19 refueling outage was structurally inadequate for the application.
The OTSGs were replaced during the fall 2005 refueling outage. Both the original and replacement OTSGs contain an adjustable_flow orifice in the feedwater inlet downcomer region to provide flow / level stability during power operation. Flow orifice settings for the new OTSGs were adjusted to achieve these conditions. However, during normal operations, the setting position of the adjustable flow orifice also impacts the EFIC low range level indicated value. Due to effects associated with the location of the EFIC low level instrument taps (one tap on each side of the orifice), indicated level from these instruments trends downward, deviating from actual OTSG level as Main Feedwater [SJ] flow rates increase during power ascension. It is important to note that these offset effects are only present with Main Feedwater in operation at moderate to high power levels and disappear immediately upon a loss of normal feedwater. Therefore, the EFIC low level instrumentation remains capable of performing its specified safety functions. Nevertheless, the flow orifice setting associated with the installation of the replacement steam generators resulted in lower than anticipated indicated levels from these instruments at normal operating conditions. Thermal hydraulic conditions in the OTSG feedwater inlet downcomer are difficult to model and are sensitive to small changes in OTSG conditions (density, fouling, etc.) as well as the orifice setting and dimensions themselves. The analysis methodology used by the manufacturer to assess the impact of these conditions in the new steam generators did not adequately account for the sensitivity of this indication to small changes. Although the low level initiate function included a time delay to prevent spurious actuation due to the back pressure wave phenomenon through the OTSG after the turbine stop valves close on a turbine trip, the risk of spurious invalid actuation of EFW was increased due to the closer proximity of indicated level from the EFIC low level instruments to the actuation setpoint.
D. Corrective Actions
Repairs to Check Valve LO-79 have been completed using original equipment manufacturer supplied parts. Actions are also in progress to strengthen the engineering and welding program processes to ensure that the appropriate level of rigor is applied when design documentation is not available.
To reduce the risk of spurious actuation of Emergency Feedwater, a reduction in the steam generator low-level actuation setpoint has been made and the time delay was increased.
These changes were implemented following receipt of an NRC approved Safety Evaluation Report which permitted a reduction of the Technical Specification allowable value for low OTSG level actuation along with an extension of the actuation time delay. These changes provide for additional protection from invalid actuations.
D. Corrective Actions (continued) In the interim, while the Technical Specification change was undergoing NRC review and approval, reactor power was limited to approximately 98 percent in order to provide margin between the EFIC low steam generator indicated level and the Emergency Feedwater actuation setpoint.
E. Safety Significance
The Reactor Protection System operated as designed to initiate the automatic reactor trip in response to the trip of the Main Turbine. Although there was an invalid actuation of Emergency Feedwater following the reactor trip, post-trip response was not significantly complicated by this occurrence, and the ERV system actuated and functioned as designed in response to the spurious low steam generator level signal. The post-trip plant response was normal with stable Hot Standby conditions (Mode 3) promptly established. Therefore, this event had minimal safety significance.
F. Basis for Reportability The automatic actuation of the Reactor Protection System is reported in accordance with 10CFR50.73(a)(2)(iv)(A). A report of this event was made to the NRC Operations Center at 1435 CST on December 26, 2005, in accordance with 10CFR50.72(b)(2)(iv)(B). The invalid actuation of Emergency Feedwater is reported in accordance with 10CFR50.73(a)(2)(iv)(A).
G. Additional Information
There have been no previous similar events reported by ANO as Licensee Event Reports.
Energy Industry Identification System (HS) codes are identified in the text as [XX].