05000293/LER-2011-002
| Pilgrim Nuclear Power Station | |
| Event date: | 2-20-2011 |
|---|---|
| Report date: | 4-20-2011 |
| 2932011002R00 - NRC Website | |
EVENT DESCRIPTION:
On 2/20/2011, a planned reactor shutdown/ cooldown was being performed in accordance with PNPS Procedures 2.1.5, Controlled Shutdown from Power and 2.1.7, Vessel Heatup and Cooldown to address leakage within the Reactor Building Closed Cooling Water (RBCCW) Loop 'B' heat exchanger (Reference LER 2011-001-00). During cool-down, with the startup feedwater regulating valve and reactor water cleanup (RWCU) letdown in service, Pilgrim experienced a reactor scram signal, Group II isolation, Group VI isolation and RBIS initiation on low reactor water level (reactor water level reached +10.4 inches - scram set-point is +12 inches). All control rods were fully inserted at the time of the scram. The low reactor water level was the result of reactor water level control difficulties experienced while performing a reactor cool-down using the Mechanical Pressure Regulator (MPR). This method was selected following successful performance in the simulator during Just-In-Time (JIT) training. The use of the MPR was one of two procedurally allowed options for plant cool-down, as the second method being the use of the Bypass Valve Opening Jack (BVOJ).
BACKGROUND:
During steady state and dynamic plant conditions, reactor pressure is maintained by the Mechanical Hydraulic Control (MHC) System. During plant cooldown, one of two mechanisms can be utilized to adjust main steam line, and therefore reactor pressure limiting the rate of temperature reduction to that allowed by Technical Specifications. The Mechanical Pressure Regulator (MPR) adjusts the control bypass valve (BPV) positions to control reactor pressure at an established setpoint. The BVOJ is a motor actuated linkage that can be used to directly open the bypass valves.
EVENT ANALYSIS:
During the event, the MPR was controlling reactor pressure by opening and closing the #1 BPV; the BPV movements caused reactor water level oscillations. This was due to the relatively coarse control nature of this cool-down method. Review following the event with several experienced personnel identified that the BVOJ provides a more fluid depressurization producing vessel level shrink of a much smaller magnitude than the MPR. In this case, the on-off control of the MPR resulted in the "at the controls" (ATC) operator taking action to significantly reduce the feedwater flow rate to prevent a high water level condition.
„ .
The vessel level shrink coupled with reduced feedwater flow to account for the reduced vessel inventory resulted in reactor water level lowering below the RPS actuation/RBIS and PCIS isolation setpoint of +12 inches reactor water level. The cool-down operation was stopped and operations management performed a stand-down with the operating crew.
Just In Time (JIT) training had been conducted the previous day. In attendance were the Shift Manager (SM), Control Room Supervisor (CRS), the Administrative Control Room Supervisor, Assistant Control Room Supervisor (ACRS) and two Reactor Operators (ROs). The training included review of PNPS Procedure 2.1.7 Attachment 6, Reactor Pressure Vessel Cooldown Rate Schedule and dynamic implementation in the simulator. The procedure directs, 'The cooldown will be accomplished using the MHC System (adjusting the MPR down to 150 psig then the Bypass Valve Opening Jack (BVOJ) for the last 150 psig or using the BVOJ from initial pressure all the way down)." No method was prescribed as preferred by the procedure, training, or collective experience of the participants, so the SM directed that both methods be evaluated by the team.
During the JIT training, incremental cooldown steps were performed using both mechanisms (MPR and BVOJ) starting at rated reactor pressure, 1030 psig. Adequate control of depressurization rate and reactor water level was experienced in both cases. A water level swell of approximately 4" occurred when the BPV was opened with a corresponding shrink of the same magnitude when the BPV closed. No adjustment to reactor water level makeup or reject was needed to maintain level within a narrow band. The decision was made by the SM to use the MPR method at the plant.
The cooldown was directed to be executed by the CRS and was initiated by the ATC operator both of who had participated in the JIT training. One of the additional ROs, who had not attended the JIT training, was assigned as the peer check. Because of the effect of normal steam loads, the cooldown was commenced from a lower reactor pressure of approximately 700 psig. The ATC operator began to lower the MPR setpoint in a continuous fashion as directed by the procedure to establish a target pressure to 585 psig. A prompt reactor water level swell of approximately 14" occurred causing the ATC to stop lowering the MPR setpoint and then take action to reduce feed water flow by closing down on the startup regulating valve. When water level began to lower, the operator recommenced lowering pressure using the MPR, and achieved the 585 psig MPR target setpoint. As observed on plant computer traces, level and BPV position cycled a total of 5 times with varying magnitude. The MPR operated by closing the BPV when pressure dropped to the established setpoint causing the cycling of BPV and indicated reactor level. When the BPV closed, vessel shrink combined with a very low feed water flow rate resulted in water level lowering to +12" in less than thirty seconds to a minimum of about 10.5".
A water level of +12" produced the expected plant response (reactor scram signal, primary containment system Group II, VI and reactor building isolations). Following the actuations and initiations, the control room crew verified that all automatic actions had appropriately occurred and took action to re-establish RWCU letdown flow path, reset the scram signal and restore normal ventilation. During the error review meeting, it was clear that the crew understood the fundamental concepts of reactor vessel level swell and shrink during depressurization and stabilization. The impact of nearly securing feedwater flow on indicated level when the MPR setpoint was reached was not fully appreciated by the RO or CRS.
An 8-hour Non-Emergency 10 CFR 50.72 notification was made to the USNRC.
CAUSE OF EVENT:
The Root Cause of the event was a failed opportunity to capture and up-date the reactor cooldown procedure with relevant historical pilgrim Operating Experience regarding previously attempted cooldown evolutions using the MPR.
Contributing Causes
- The crew did not apply sufficient questioning attitude and stop when unsure, when the magnitude of the initial reactor water level swell exceeded that experienced during Just-In-Time (JIT) training.
- The Simulator did not adequately model the plant's reactor level response during MPR cool-down at these low flow, high temperature conditions. This condition contributed to a high sense of confidence in performing the cooldown with the MPR evolution.
EXTENT OF CONDITION:
A review of station errors and events for the past five years failed to identify any occurrences attributed to deficiencies in any of the following areas: failure to capture pertinent operating experience in procedure or shortfalls in simulator modeling. Accuracy of simulator modeling was additionally evaluated by review of the plant simulator index and simulator design review board meeting minutes. No significant discrepancies were found and existing deficiencies were appropriately prioritized. A review of crew and plant performance where JIT training had been utilized determined that the training was effective in supporting successful performance.
Gaps in management oversight were identified in at least two of the occurrences (CR-PNP-2009-0499 and CR-PNP-2009-4036). Management engagement and correction of at risk behaviors is a significant corporate initiative through the use of Entergy Nuclear Platform 3: Set and Continuously Enforce High Standards and Fleet Procedure EN-FAP-OM-001: Leadership Forums for Continuous Improvement. These initiatives are considered sufficient to address extent of problem / condition.
FAILED COMPONENT IDENTIFICATION:
Not applicable.
CORRECTIVE ACTIONS:
Immediate corrective actions taken were to temporarily halt the cool-down operation while operations conducted a stand-down. Plant cool-down was subsequently performed successfully utilizing the BVOJ.
Corrective actions taken included the revision of the reactor heat-up / cool-down procedure to incorporate lessons learned to identify the Bypass Valve Opening Jack (BVOJ) as the preferred method for executing a reactor pressure vessel cool-down.
Corrective actions planned include the performing of an analysis of MPR/RPV and level response during plant cool-down at the plant simulator and evaluate results for disposition.
The corrective actions are being tracked in the Pilgrim Station Corrective Action Program via CR-PNP-2011- 00733.
ASSESSMENT OF SAFETY CONSEQUENCES:
The event posed no threat to public health and safety.
A low reactor water level signal (+12") with all control rods fully inserted resulted in a scram signal, PCIS Group II, Group VI and Reactor Building Isolation signals. Following verification that all automatic actions had occurred as expected, the reactor scram and isolation signals were reset, restoring system configurations to their pre-initiation status. The plant remained within the established shutdown risk evaluation of the five key safety functions (Inventory Control, Decay Heat Removal,,Power Availability, Reactivity Control and Containment). At its lowest point of +10.5", reactor water level was maintained greater than 10' above top of active fuel. There was no radiological or industrial safety impact. Based on the fact that there was no challenge to nuclear, radiological or industrial safety, the impact on safety was not significant.
SIMILAR EVENTS:
Pressure control was oscillating between the EPR, MPR and Bypass Valve Opening Jack. Pressure "Control" and "Not in Control" lights on the EPR, MPR and Bypass Valve Opening Jack were all oscillating.
During a thermal backwash on 7/10/07 while the reactor was at 50% power, the control room received a turbine trip and reactor scram on a low vacuum trip. The reactor was running at roughly 945 psig at the time of the scram. The EPIC traces show the 3 bypass valves open initially, relieve the initial post scram pressure spike and then close as the expected response. Reactor pressure dropped to roughly 840 psig. Within 10 minutes, decay heat was causing the pressure to rise. At this point, reactor pressure only increased to 928 psig when the MPR took control and the # 1 bypass valve began to oscillate. Steam supply pressure only recovered to 922 psig when the bypass valves began to oscillate. The Apparent Cause of the reactor pressure oscillation was a burr on the MPR pilot valve which most likely was caused by a piece of debris within the turbine lube oil system or age related wear to the pilot valve.
REFERENCES:
CR-PNP-2011-0773 PNPS Procedure 2.1.5 Controlled Shutdown from Power PNPS Procedure 2.1.7 Vessel Heatup and Cooldown --,