Event Notification Report for April 11, 2003

MANUAL REACTOR TRIP DUE TO VIBRATIONS ON THE MAIN TURBINE

"A manual scram was initiated at 0145 on April 11, 2003, due to vibrations on the Main Turbine trending up to the trip setpoint. A scheduled plant shutdown was in progress for a maintenance outage [of the 'B' recirculation flow control valve sensor]. All plant systems operated normally on the scram. The plant is shutdown at 0% power in Mode 3. The turbine vibrations returned to normal values after the turbine tripped."

All rods inserted normally. All safety and electrical systems operated as designed during and after the reactor trip. The plant is stable and using normal feedwater. There was nothing unusual or not understood.

The NRC Resident Inspector has been notified.

UNANALYZED CONDITION COULD CAUSE HIGH PRESSURE INJECTION PUMP DAMAGE

"As a result of an in-depth design and performance capability review, a non-conforming condition was identified whereby, utilizing only safety grade equipment, long term cyclic repressurizations of the reactor coolant system (RCS) may occur following a subset of postulated very small Loss of Coolant Accidents (LOCAs) with effective break sizes in a range between 0.0021 ft2 to 0.0045 ft2. The repressurization cycles were not previously analyzed, but are predicted by a new application of the license basis 10 CFR 50.46 Evaluation Model. Although non-safety grade equipment would be available to prevent repressurizatons, if only safety grade LOCA mitigating equipment is credited, these repressurizazion cycles could be postulated to damage both High Pressure Injection (HPI) pumps. This could occur due to pump deadheading after HPI recirculation flow back to the borated water storage tank is procedurally isolated upon tank low level and pump suction has been manually transferred to the containment emergency sump at minimum of approximately 20 hours into the postulated event. Minimum recirculation flow back to the borated water storage tank initially provided protection against deadheading the pump and previously assumed reactor coolant system pressures would have allowed continued HPI pump flow. During part of the newly predicted repressurization cycle, RCS pressure would exceed the shutoff head of the HPI pumps. Without minimum flow, the pumps would be damaged.

"This issue is currently evaluated by Condition Report 02-06702. Davis-Besse has determined this condition is reportable under § 50.72(b)(3)(ii)(B) 'Any event or condition that results in ..'The nuclear power plant being in an unanalyzed condition that significantly degrades plant safety.' Although the plant is currently in cold shutdown and the HPI pumps are not required to be operable per the Technical Specifications, this issue represents a historical condition that existed within the last three years."

The licensee has notified the NRC Resident Inspector.

LOSS OF DOUBLE CONTINGENCY PROTECTION

NRC BULLETIN 91-01 24 HOUR NOTIFICATION

Westinghouse Electric Company, Commercial Fuel Fabrication Facility, Columbia SC, low enriched (less than or equal to 5.0 wt. % U-235) PWR fuel fabricator for commercial light water reactors. License: SNM-1107.

Time and Date of Event: 18:30 hours, April 11, 2003.

Reason for Notification:

On March 19, 2003, six UF6 cylinders were placed on hold because Westinghouse questioned if the cylinders were properly tested following repair. The ANSI N14.1 nameplate had been removed from supporting "feet" and-welded directly onto the pressure vessel dome. The "U" Stamp had been replaced with an "R" Stamp and documentation from the shipper indicated that the fillet weld on the dome had undergone dye-penetrant testing. The UF6 cylinder pressure vessel, however, had not undergone hydrostatic testing.

The "hold" consisted of a flag in the UF6 cylinder tracking computer database, which was inserted as a manual edit. The hold flag should have prevented the cylinders from being transferred into work in process (WIP) to allow processing. On March 31, 2003 one of the six cylinders was allowed by the tracking database to be processed. A second cylinder with the hold flag was allowed to be processed on April 1, 2003.

Unaware of the failure of the hold flags, the safety analysis proceeded. On April 10, 2003 NCS and process engineering completed the safety review begun on March 19, 2003 of the nameplate welding using applicable pressure vessel standards including an on-site interview with a certified boiler code inspector. The conclusion of the safety review was that the "R" Stamp cylinders met the ANSI N 14.1 requirements and were acceptable for processing.

On April 11, 2003, NCS began a review of the sequence of events leading to the processing of the two hold tagged cylinders. At approximately 18:30 hours April 11, 2003 NCS completed its final interview. Shortly thereafter, it was determined that there had been a loss of previously documented double contingency protection. NCS immediately informed the EH&S manager of the event.

Double Contingency Protection:

The parameters that directly affect neutron multiplication for the vaporizers are mass (density) and geometry (level control). A criticality could be possible in a vaporizer under the following conditions:

Sufficient material is discharged from the cylinder into the vaporizer in order to form a critical UO2F2 H2O density (optimum moderation), and water slab height increases to a critical height.

Cylinder integrity maintains mass control. The U235 mass control depends upon maintaining the structural integrity of the cylinder to ensure that no material is released due to a sudden uncontrollable rupture.

The geometry control consists of ensuring that condensate drains properly from the vaporizer, and detecting water accumulation should it occur.

It has been determined that less than previously documented double contingency protection remained for the system and that greater than a safe mass was involved. Double Contingency protection was restored within 4 hours. In accordance with Westinghouse Operating License (SNM-1107), paragraph 3.7.3 (c.5b), this event satisfies the criterion for a 24-hour notification.

As Found Condition:

See "Reason for Notification." As detailed above, the Investigation found that the cylinders were safe to process. In fact, there never was an actual safety issue. The event did point out a weakness in our control of UF6 cylinders that will be addressed.

Summary of Activity:

1) The four remaining cylinders were physically tagged out.
2) A complete inventory and inspection of all cylinders on-site was performed.
3) It was verified that no movement of UF6 cylinders onto the site or into processing will occur for the next week.

Conclusions:

1) Loss of double contingency protection occurred. Greater than a safe mass was involved.
2) At no time was the health or safety to any employee or member of the public in jeopardy. No exposure to hazardous material was involved.
3)The Incident Review Committee (IRC) determined that this is a safety significant incident in accordance with governing procedures.
4) Notification was the result of an event, not a deficient NCS analysis.
5) A causal analysis will be performed.

TEXAS AGREEMENT STATE REPORT FOR FIRE DAMAGED THREE SOURCE HOLDERS

"Fire detectors activated in the multi-Product Unit 1 on Friday afternoon, insulation and sheeting on the unit were damaged by the fire. Three Ohmart SH-F1 source holders were installed on the unit. Each source holder contains a Cesium-137 source with an estimated activity of 120 millicuries each. The on-site fire department was notified and extinguished the fire. It was suspect that the gauges had been engulfed in the fire. However, after notification to Techstar, an on-site visit for inspection and verification determined that none of the gauges were burned in the fire but had been affected by the heat of the fire. It has been explained that the detectors were damaged. The Techstar firm confirms that all labels are intact and legible. A leak test was performed on each unit for gross contamination and a radiation survey and occupancy evaluation was performed. No contamination was detected. The ion chamber detectors will have to be replaced, as the greater part of the fire was on the detector side of the vessels. Techstar placed the three units shutters in the closed position until repairs to the facility can be made. Estimated time to repair the facility is approximately 10 working days. After repair the ion detectors will be replaced and the gauges will be tested by Techstar."

Notified NMSS EO (F. Brown)