Event Notification Report for April 24, 2001

DECLARATION OF AN UNUSUAL EVENT DUE TO A SMALL DEBRIS FIRE ON A STEAM GENERATOR PLATFORM DURING THE PERFORMANCE OF EDDY CURRENT TESTING

At 0148 CDT, the control room was notified of a fire in the Unit 1 containment. It is currently believed that a problem with eddy current testing equipment (possibly an electrical fault) resulted in a fire involving a small pile of debris on the 8-foot elevation steam generator 'A' platform. The fire brigade was dispatched immediately, and the fire was extinguished within 2 to 3 minutes of the time the extinguishing agent (water) was applied. Offsite assistance was not required. At 0211 CDT, a second team provided the control room with a confirmatory report that the fire had been successfully extinguished. There were no reports of injuries or the spread of contamination as a result of the fire.

The exact time the extinguishing agent was applied to the fire was not known at the time of this event notification. Because the exact duration of the fire (i.e., less than or greater than 10 minutes) was not known, the licensee declared an Unusual Event at 0206 CDT as a conservative measure.

At the time of this event, Unit 1 was defueled, and Unit 2 was operating at 100% power.

The licensee notified Manitowac and Kewaunee Counties, the Sate of Wisconsin, and the NRC resident inspector.

***** UPDATE AT 0343 EDT ON 04/24/01 FROM MIKE MEYER TO LEIGH TROCINE *****

The licensee terminated the Unusual Event at 0243 CDT.

The licensee plans to notify Manitowac and Kewaunee Counties, the Sate of Wisconsin, and the NRC resident inspector. The NRC operations officer notified the R3DO (Kozak), NRR EO (Richards), IRO (Mamish), and FEMA (Canupp).

STOLEN CPN MOISTURE DENSITY GAUGE


Geo Technical Exploration Inc., on 04/24/01 at 1600 PT reported to the State of California Radiation Control Program that a CPN Model 3 moisture density gauge had been stolen from the back of a truck probably on the evening of 04/19/01. The CPN gauge was in its manufacturer case and chained to the truck. The CPN gauge was discovered missing on the morning of 04/24/01 by the licensee. The San Diego Police were notified of the theft and a reward has been offered for the recovery of the CPN gauge. The serial number for the stolen moisture density gauge is M30069611. The gauge contained between 8 to 10 millicuries of Cesium-137 and between 20 to 30 millicuries of Americium-241.

The State of California case assignment number is 042401.

AGREEMENT STATE REPORT

On 4/24/2001, an irradiator facility experienced a loss of electrical power to a programmable logic controller (PLC) which resulted in the inability to automatically lower the source racks. The source racks were manually lowered to a safe condition upon discovery of the failure.

The following information was provided by the licensee to the California Radiation Control Program on 5/15/2001:

"Description of Events:

"The first indication that there was a problem came at 9:00 p.m. 4/24/01, when the in-line water monitor signaled a failure. The Shift Leader took appropriate action per the Emergency Procedures and determined that there was no radiation present in the water system. He notified the QA Technician, who had performed a calibration of the monitor that afternoon.

"After reviewing the procedure followed for calibration routine and determining that no problems occurred with the calibration that would trip the alarm, the QA Technician, notified the facility Radiation Protection Office (RPO). She determined that this was a false alarm, probably caused by air bubbles in the system, as has previously occurred with the water monitor. [... Permission was given] to allow the system to continue running with the water alarm disarmed until more investigation could be performed in the morning.

"Starting at about 7:00 a.m., additional water counts, using the monitor, were taken and resulted in normal background readings. Since the routine counts showed expected background levels and the alarm did not activate again, the concluded that a pocket of water bubbles from a filter change had worked its way through the system and caused spurious readings on the monitor, which had occurred on previous occasions. All of the events to this point were consistent with this determination.

"At 8:20 am., the Operator notified the RPO that the in-line alarm was sounding again. On reviewing the PLC control panel, she noted that none of the indicators on the panel were lit, as they should have been, even though the computer monitor (PLC user interface) was operating. In concert with the Plant Manager, they determined that the audible alarm that the Operator heard was not the in-line monitor, but was an alarm indicating that the PLC was off-line.

"Further investigation revealed that the system conveyor had stopped moving (i.e., product was stationary within the cell), but that the source racks bad not automatically returned to the shielded position, as they should have, the source racks were manually lowered from the roof by 8:40 a.m. During this time, the door interlock continued to function properly, prohibiting access to the cell through the personnel access door.

"In determining the probable cause of the event, the first evaluation was that the power supply had malfunctioned. However, upon further investigation, it was determined that the most probable cause was an electrical short in the system. After extensive trouble-shooting and investigation, the electrical short was finally located in the line going to one of the emergency pull cords in the cell. The cable had actually melted at the point of the failure.

"That part of the systems was rewired and the system restarted at approximately 4:00 p.m. The safety system was checked for proper operation and routine processing resumed at 4:45 p.m.

"Evaluation of Event and Root Cause:

"Upon Engineering review of the electrical drawings, it was determined that a short circuit on the pull cords or other devices could have tripped one of the circuit breakers, power from which feeds the PLC and other modules in the PLC rack. The audible alarm was the PLC Off Line Sonalert, which, as intended, served as a warning the PLC was not operating. With the PLC off, there was no power control to lower the source racks. In normal circumstances of power failure, the uninterruptible power supply (UPS) provides adequate emergency power to lower the source racks by releasing the hoist brakes in a pulsed mode. However, with the PLC not operating, this power was not supplied to the brakes, which then had to be released manually.

"The water monitor alarm activation was probably caused by shorting line voltage to the grounding circuit. This momentary surge in current, particularly on the ground path, could cause an erroneous indication at the monitor. Other facilities have had spurious water monitor alarms resulting from ground fault conditions.

"Corrective Actions and Additional Considerations:

"Corrective actions to the event are:

"1. The circuit will be modified to ensure the PLC does not lose power if a device or device wiring causes a short circuit.

"2. Additional training will be provided to operators to be more cognizant of the system response to a PLC off-line fault. While the PLC off-line alarm is a local alarm, meaning that it sounds at the control console and does not active general alarms throughout the warehouse, all system operations are stopped, including overhead conveyors and the 4-shelf elevator (i.e., device that shifts totes among positions in the carrier). The absence of movement in these systems should have alerted the operator to a systemic failure of the controls. In this instance, the tune period between the equipment failure and initial resolution (manually lowering the source racks) was only a few minutes. Because the door interlock continues to function under these circumstances, the situation did not pose a radiation safety hazard to the operator or other personnel. Although, operator training currently includes instructions for determining console power status and the proper procedure for lowering the source racks under circumstances such as occurred here, the training will be reinforced and repeated

"3. To avoid further problems with the in-line water monitor alarm, an evaluation is being conducted to determine whether the water monitor can be connected to an isolated-ground receptacle and circuit. This would have the effect of making the monitor less affected by stray currents, and other sources of 'noise' on the power line."

* * * UPDATE 0945EDT ON 5/29/01 FROM CA RAD CONTROL PRGM TO S. SANDIN * * *

California Radiation Control Prgm update to identify NMED Report Number XCA52 for this incident.