05000331/LER-2009-001
| Duane Arnold Energy Center | |
| Event date: | |
|---|---|
| Report date: | |
| Reporting criterion: | 10 CFR 50.73(a)(2)(iv)(A), System Actuation |
| Initial Reporting | |
| 3312009001R00 - NRC Website | |
I.�Description of Event:
On February 1, 2009, the DAEC was reducing reactor power in preparation for entering Refueling Outage (RFO) 21. As part of a normal shutdown, Operating Instruction (OI) 442, Circulating Water System, provides direction for removing one circulating water pump and cooling tower from service.
At the time of the event, the operations crew was in the process of performing this activity.
The pre-job brief for this evolution was conducted at 1300 hours0.015 days <br />0.361 hours <br />0.00215 weeks <br />4.9465e-4 months <br />. At approximately 1730 hours0.02 days <br />0.481 hours <br />0.00286 weeks <br />6.58265e-4 months <br />, the operators assigned to perform this activity arrived at their respective stations: two licensed operators in the control room, two in-plant operators at the 'A' cooling tower breaker house (to throttle the 'A' cooling tower (CT) riser valves closed), and one in-plant operator at the pump house to monitor the securing of the 'A' circulating water pump.
The sequence of removing one circulating water pump and one cooling tower from service per CH 442 is to throttle the cooling tower riser valves on the tower to be removed from service until circulating water discharge pressure is about 35 psig, and to then secure the desired pump. This is a historical pressure limit to prevent the circulating water pump that is to remain in service from running out. The CT riser valves are motor-operated (MO) butterfly valves controlled from hand switches located at the breaker cubicles inside the CT breaker house.
At approximately 1740 hours0.0201 days <br />0.483 hours <br />0.00288 weeks <br />6.6207e-4 months <br />, the direction was given from the control room to commence throttling the 'A' CT riser valves in the closed direction. The direction given to the operators at the cooling tower was to apply a 10-second closed signal to both riser valves, and then stop and wait for the next throttle closed order to be given from the control room. During this process, control room operators were monitoring computer point F015, circulating water discharge pressure (note this is the only indication in the control room for this parameter.) The F015 reading prior to the start of the evolution was 29.8 psig. This computer point is fed from PT4205, which had an open Work Request Card (WRC) (A84341) written on November 10, 2008 for erratic indications. This WRC was scheduled to be worked during RFO 21.
Upon the application of the first throttle closed signal, it was discovered that CT riser valve M04250 did not reposition. This was ascertained when the cooling tower operators did not see dual indication at the breaker. After consulting with the control room, the cooling tower operators proceeded to the sparging valve to check the condition of the limit switch which provides the interlock between the sparging valve and a close permissive for M04250. It was determined that the limit switch had hung up when the valve was opened. (Note that the sparging valve had been opened several hours prior.) The operator then tapped the switch and it positioned into place. The time required leaving the breaker house, traveling to the sparging valve, tapping the limit switch, and returning to the breaker house was approximately 10 minutes.
At approximately 1750 hours0.0203 days <br />0.486 hours <br />0.00289 weeks <br />6.65875e-4 months <br />, the cooling tower operators gave a 10-second closed signal to M04250 and saw dual valve indication. From this point forward, the cooling tower operators continued to close both risers when directed by the control room using 10-second closure signals. Between each bump closed, approximately 1 minute elapsed until the next bump close while the operators in the control room monitored system pressure on F015 between each order. A total of four to five 10 second closure signals were given to the riser valves.
While the operators at the cooling tower were throttling the riser isolation valves, the pump house operator observed the local discharge pressure increase from 29 psig at the start of the evolution to 33 psig after the 4th or Stn bump closed. Pressure indication at this time on computer point F015 was 30.3 psig. The pump house operator reported this to the control room and the direction was given that the local pressure reading would be used as the 01 target since it was reading more conservatively. The pump house operator also stated that both local pressure indicators responded together.
At approximately 1755 hours0.0203 days <br />0.488 hours <br />0.0029 weeks <br />6.677775e-4 months <br />, the order was given to throttle the riser valves closed by applying a 5 second signal. At the end of this 5-second signal, both riser valves lost the dual indication and showed green only (full-closed). This was reported to the control room and the control room operators decided to secure the circulating water pump per procedure.
After the last 5-second bump closed on the risers, the pump house operator noted no pressure change on the local pressure indicators as compared to before the last 5-second bump. Shortly after this point, the pump house operator observed indications of circulating pump cavitation. Specifically, the operator heard the pumps rumbling and noticed both circulating water pump local discharge pressure indicators oscillating around 10 psig. The operator also noted circulating pit level lowering as it passed through 10 feet.
Between 1754 hours0.0203 days <br />0.487 hours <br />0.0029 weeks <br />6.67397e-4 months <br /> and 1800 hours0.0208 days <br />0.5 hours <br />0.00298 weeks <br />6.849e-4 months <br />, circulating pit level lowered from 19 feet to 6 feet. Annunciator Response Procedure (ARP) 1C06A, D-11, Circ Water Pit Lo Level, directs a manual reactor scram if circulating pit level cannot be restored and maintained above 8 feet. At 1800 hours0.0208 days <br />0.5 hours <br />0.00298 weeks <br />6.849e-4 months <br />, with the reactor at 53% power, a reactor manual scram was initiated.
II. Assessment of Safety Consequences:
There was no direct impact on nuclear safety as a result of this event. The event resulted in the initiation of a manual scram in anticipation of a loss of circulating water. The On Shift Analysis completed for the scram concluded that all safety systems operated as necessary.
The plant was stabilized per abnormal operating procedures. Containment isolation signals operated per design with Groups 2, 3, and 4 received.
An unexpected loss of one 161 KV bus in the switchyard also occurred during this event. This was caused by a breaker failure lockout of the bus caused by signal from the output breaker. This output breaker did open.
The power loss of the 161 KV bus had minor plant effect. One power supply to essential service offsite transformer 1X3 was lost due to the bus lockout. The transformer and offsite circuit remained available. The plant did sustain a temporary loss of spent fuel pool cooling system.
Cooling was restored with no level or temperature complications.
Essential power remained available with both onsite SBDG's and offsite circuits current source.
All ECCS were available. All safety related equipment operated as designed, except Reactor Core Isolation Cooling (RCIC) turbine which had difficulty in automatic control. Manual control of RCIC was effective. This behavior is not unexpected during the low flow conditions in which RCIC was operating.
Therefore, the plant shutdown did not result in any radiological or nuclear concern which would impact the health and safety of the public.
This event did not result in a Safety System Functional Failure.
III. Cause of Event:
A root cause evaluation was completed (RCE 1079). The RCE determined that operating practices on the day of the event placed the plant in a configuration that the cooling towers were not designed for. This event revealed that the cooling towers were not designed to have both circ water pumps discharging over a single tower. The metallurgical report completed as a result of the failure indicated that the 'EV tower west riser failed catastrophically by separating at the slip joint between the riser and the distribution header at the top of the cooling tower. Based on the fact that the failure was from the forces exerted from the increased flow seen on the '13' cooling tower as a result of isolating the risers on the 'A' cooling tower, the RCE concluded that the root cause of the event was that ()I 442 was inadequate to prevent an inappropriate operational configuration. This inappropriate operational configuration revealed the fact that the cooling towers were never designed to individually receive rated discharge flow from both circ water pumps. The team identified one contributing factor associated with the plugged pressure transmitter, PT4205, which provided a computer point to the control room for circ water pump discharge pressure. The plugging delayed the expected circ pump discharge pressure increase that should have been seen while throttling the 'A' cooling tower riser isolation valves. This resulted in the control room allowing for further throttling of the 'A' cooling tower riser until the riser isolation valves were fully closed.
IV. Corrective Actions:
Immediate Actions Taken Engineering Change Package (ECP) 1884 restored the failed riser of the 1E069B cooling tower with a more robust design, and enhanced the design of the remaining risers on both the cooling tower risers. This also addressed the extent of condition. The modified riser design now conservatively supports corresponding flow for two pump operation over a single tower.
CH 442, Circulating Water System, was revised by Procedure Work Request (PWR) 45894 to specify the sources to be used for circulating water pump discharge pressure wherever it is called upon to be read.
Long Term Corrective Actions 1. Conduct a design review to determine the equipment manipulations needed for the transition between one- and two-pump operations. The review should determine whether pump runout is a concern. If so, determine the correct riser valve position to prevent pump runout while securing the first cooling tower. This design review shall include a consideration with regards to pipe pressure limits and hydraulic forces.
2. Based on conclusions of long term corrective action #1, revise OI 442 to provide cooling tower riser isolation valve closure guidance. This will address the extent of cause associated with the process of transitioning from two to one cooling tower operations.
3. Create a preventive maintenance action to inspect PT4205, Circ Water Pump Supply Pressure sensing line, for blockage on a periodicity that supports the current operating cycle such that the inspection occurs prior to the start of an RFO.
4. Establish a means to identify degraded computer points that are used in the manipulation of plant equipment and to track their resolution.
2. DOCKET
05000 - 331 2009� 001 0 OF 6
V. Additional Information:
Previous Similar Occurrences:
A review of LERs over the previous 5 years revealed the following similar occurrences:
HIS System and Component Codes:
Cooling Tower - CTW Reporting Requirements:
This event is reportable under 10 CFR 50.73(a)(2)(iv)(A).