05000361/LER-2004-002

From kanterella
Jump to navigation Jump to search
LER-2004-002, Manual Reactor Trip and Initiation of Auxiliary Feedwater in Response to Loss of Main Feedwater
Docket Number
Event date: 04-10-2004
Report date: 06-04-2004
Reporting criterion: 10 CFR 50.73(a)(2)(iv), System Actuation
3612004002R00 - NRC Website

1. FACIUTY NMIE 2. DOCKET NUMBER 6, LER NUMBER PAGE (3) Event Date: April 10, 2004 Reactor Vendor. Combustion Engineering Mode: Mode 1 — Power Operation Power. 97 percent

Description of Event

On April 10, 2004, at 1147 PDT, Unit 2 was operating at about 97 percent power when both main feedwater [SJ] pump [P] turbines (K005/K006) tripped on high feedwater pump discharge pressure. Plant operators (utility, licensed) manually tripped the reactor and the Auxiliary Feedwater System [BA] started automatically.

Plant Operators reported this event to the NRC Operations Center (Log No. 40664) at 1354 PDT in accordance with 10CFR50.72(b)(2)(iv) and 10CFR50.72(b)(3)(iv) for actuation of the Reactor Protection System (RPS) [JC] and the Auxiliary Feedwater system. This follow-up Licensee Event Report is provided in accordance with 10CFR50.73(a)(2)(iv).

Cause of Event

In accordance with their design, both main feedwater pump turbines (K005/K006) tripped on high discharge pressure when main feedwater regulating valve 2FV1111 [FCV] and associated bypass valve 2HV1105 began closing. SCE determined these valves began closing due to two concurrent electrical grounds in the feedwater control system [JB].

The first ground (Ground 1) was caused by a pinched interconnect wire on one of the two feedwater regulating control systems (FWCS). The wire was pinched under the corner of a horizontal terminal strip and has most likely been pinched since initial plant startup. SCE concluded that the insulation just recently separated at the pinch point exposing the copper conductors, and shorting to ground when it contacted the mounting rail.

SCE was unable to locate the second ground (Ground 2) but concluded it was an intermittent ground located in the FWCS (an ungrounded system). SCE verified that 2FV1111 and 2HV1105 would begin closing (absent a close signal) when Ground 1 was present and the FWCS power supply common was grounded. That is, a single ground by itself, either the pinched wire or a ground in the power supply, would not cause the valves to begin closing.

Ground 2 cleared before it could be located.

1. FACLITY NAME 2. DOCKET NUMBER S. LER NUMBER PAGE (3)

Corrective Actions

1. The damaged wire (Ground 1) was replaced. The system was retested and returned to service.

2. SCE examined both FWCS cabinets and did not locate any other wiring near terminal blocks that might be crimped or grounded.

3. SCE also inspected and tested accessible electrical wires and components in the FWCS, particularly those closer to metal surfaces, to locate Ground 2. Ground 2 cleared before it was located. (Note, however, that because two grounds must be present to cause 2FV1111 and 2HV1105 to close when a close signal is not present, repair of Ground 1 will preclude a similar occurrence).

4. For other critical control systems with floating power supplies, SCE will review and revise applicable maintenance and test procedures for plant personnel to test for power supply grounds.

5. SCE is planning to replace the current analog feedwater control system with a fault tolerant digital control system. The digital FWCS will reduce the vulnerability for single failure.

Safety Significance

An assessment of the conditional core damage probability (CCDP) and the conditional large early release probability (CLERP) for the April 10, 2004 event determined that the Unit 2 CCDP and CLERP were 2.8E-6 and 1E-7, respectively. The assessment was based on the reported actual component unavailability, system alignments and operating conditions that existed at the time of the event.

The Main Feedwater pumps are not essential for safe shutdown of the plant. The SONGS 2/3 UFSAR, Sections 15.2.2.5 and 15.10.2.2.5, credits the AFW system for maintaining an adequate heatsink during a transient in which an instantaneous and complete loss of feedwater occurs. The event reported herein, where main feedwater was lost and all safety systems functioned correctly is bounded by the UFSAR safety analysis identified above.

Additional Information

SCE has reported two recent events involving the FWCS:

main feedwater controller card 3FC1121 failed due to a manufacturing defect (pinhole oxide defect) in an operational amplifier. This controller card had been in service for many years and when it failed, it failed at the location of the manufacturing defect.

Corrective actions for the November 21, 1999 event focused on cards with Fairchild operational amplifiers.

2FC1111 main feedwater controller card had a shorted low limit operational amplifier.

This was believed to be an age-related failure. Corrective actions for this event focused on the replacement of Unit 2 main feedwater controller cards with new cards.

Corresponding cards in Unit 3 were also proactively replaced.

As noted, corrective actions that have been completed for these two prior events focused on replacement of the failed controller cards. SCE concludes the two grounds reported herein were not present concurrently when the previous repairs were completed. Therefore, previous completed corrective actions could not have prevented this event.