Event Notification Report for February 24, 1999

MALFUNCTION OF DOOR INTERLOCK FOR TELETHERAPY UNIT

On 02/24/99, while conducting a test of the treatment room door interlock associated with a Co-60 teletherapy unit, the treatment timer stopped and the source retracted when the door was opened (expected), but the timer then restarted and the source became exposed again after the door was closed (not expected). The source then retracted once again after the treatment timer expired. The treatment timer should not have restarted until manually directed to do so from the operator console. The teletherapy unit is a Theratronix model T-780,
containing 3,500 Ci of Co-60. A vendor technician responded to the site but was unable to reproduce the event. However, the licensee reported that the technician was able to reproduce a similar problem by bumping the door very slightly and allowing it to quickly reclose. The licensee also reported that the hand pendant associated with the treatment couch recently malfunctioned (in a reproducible way) until the machine was switched off and then back on, when the malfunction ceased. The technician was unable to reproduce this malfunction.

The unit appears to be working properly at this time although the licensee has locked the treatment room until all radiation therapists can be briefed on the potential for this malfunction.

No patients or personnel were in the room at the time of the test.

(Call the NRC Operations Center for contact names and telephone numbers.)

DISCOVERY OF MINOR UF6 LEAKAGE IN THE X-330 BUILDING AND CONFIRMED VALID CASCADE AUTOMATIC DATA PROCESSING (CADP) ACTUATION (24-HOUR REPORT)

The following text is a portion of a facsimile received from Portsmouth:

"At approximately 1205 hours on 02/24/99, operations personnel investigating a smokehead alarm that occurred at 1005 found visible evidence of minor UF6 leakage on a valve in the evacuation header for position 3 of the liquid withdrawal station at tails. Finding this visible evidence of UF6 leakage confirmed that the actuation of the CADP UF6 smoke detection safety system was valid. This discovery concludes the investigation that was started on 02/22/99 to determine if three smokehead alarm actuations that occurred on 02/19/99 were actual or invalid signals. When the alarms were initially received, operations personnel responded and performed an investigation in accordance with response procedures. Their initial inspections did not reveal any evidence of UF6 leakage which would indicate the alarm actuations were valid. On 02/22/99, additional actions were taken to investigate the alarms which included isolating sections of piping near the smokeheads. Visual inspections and leak tests performed on this section of piping did not reveal any evidence of UF6 leakage. On 02/24/99, with the suspected section of piping isolated, another alarm was received. This led the operations personnel to re-inspect equipment which did not previously show any evidence of leakage. This inspection revealed evidence of minor leakage which was not previously visible. This evidence confirmed that a valid CADP actuation had occurred."

"This is reportable to the NRC as a valid actuation of a 'Q' safety system in accordance with the Safety Analysis Report, Section 6.9."

"There was no loss of hazardous/radioactive material or radioactive/radiological contamination exposure as a result of this event."

Portsmouth personnel notified the NRC resident inspector and the Department of Energy site representative.

MEDICAL MISADMINISTRATION INVOLVING RECEIPT OF A BRACHYTHERAPY DOSE THAT WAS 24% LOW

On 10/29/98, a patient at Indiana University Medical Center located in Indianapolis, IN, was being treated with a three-channel, low-dose-rate, brachytherapy device containing Cesium-137 sources in the form of small pellets. When nurses entered the room to attend the patient, the sources were retracted, and there were problems resuming the treatment. The problems were initially believed to be the result of a power problem. The on-call medical physicist was contacted, power problem recovery steps were followed, and the treatment was resumed. The following morning, a resident physician noticed that only one of the three channels on the brachytherapy device was actually operating (treating the patient) and that the sources in the other two channels were still in the shielded position. Therefore, the problem was the result of a pellet problem rather than a power problem and different recovery steps should have been followed. The medical physicist was contacted, and the treatment resumed. When the patient was informed that the treatment time would need to be lengthened to compensate for the delay in reactivating two of the three channels, the patient refused the additional treatment. The prescribed dose was 2,500 centigray, and the patient actually received 1,900 centigray.

On 11/09/98, the licensee sent an Incident Report to NRC Region III (Bob Gattone). NRC Region III sent a Technical Assistance Request to NRC Headquarters regarding reportability, and a response stating that this incident was reportable was received by licensee via facsimile at 1217 CST on 02/24/99.

(Call the NRC Operations Center for a site contact telephone number.)

TORUS WATER LEVEL DETERMINED TO BE OUTSIDE DESIGN BASIS DUE TO INSTRUMENT ERROR

The following text is a portion of a facsimile received from the licensee:

"During the design installation of a more accurate narrow range torus level instrument, as part of our [basis for maintaining operations] process, it was identified that the current level indication was 1.5 inches lower than actual level. Under these conditions we had the potential to be slightly outside the acceptable torus operating level band high. At this time, we are within the Tech Spec limit but outside our administrative limit. With the 1.5-inch discrepancy, it is likely that in the past we have unknowingly operated outside the Tech Spec limit for more than 24 hours.

"We don't feel that this is of a significant safety concern since even with the 1.5-inch error, the torus level would only have been approximately 1/2 inch outside the Tech Spec limit. This would still be greater than three feet below a level where emergency depressurization would be required by the emergency operating procedures.

"The new indication installed by the design process has been declared operable, and the instrument in error
has been declared inoperable. The plant has entered a 24-hour shutdown [limiting condition for operation (LCO)] for being outside our administrative limit as required by plant procedures. Plans are in place to lower torus water level to within administrative limits and exit the 24-hour LCO."

The licensee notified the NRC Resident Inspector.

* * * RETRACTION 1011 3/29/1999 FROM SALTWELL TAKEN BY STRANSKY * * *

"VY has been operating since its 05/98 outage with an administrative limit that credited in-place torus narrow-range level instrument uncertainties as <= 1.0 inches. On 02/24/99, VY installed a new torus water level indication system. Upon energizing the new level indication system, it was noted that there existed a disparity between the in-place/operable level indication instrumentation and the new instrument. The difference between the original and operable level indication, and the newly installed instrument was approximately 1.5 inches. It was believed at that time that the new instrument indication was very close to actual torus level.

"Upon observing the variance between the new and old indications, VY made the conservative determination that the new instrument was indicating actual torus level, completed the necessary testing, declared the new system operable, and began controlling level based upon that indication.

"It was concluded that the instrument uncertainty must have been significantly higher than the 1 inch credited for operation following the 05/98 outage, indicating up to 1.5 inches below actual level. Were this true VY would have been operating with a torus water volume greater than allowed by its Technical Specifications for periods greater than those Technical Specifications allow. It was further concluded that such operation may have invalidated the conclusions of several VY calculations and analyses which demonstrate the adequacy of the VY Primary Containment System. VY therefore made an ENS notification for operation outside of the design basis of the plant.

"On 02/25/99, VY installed, and placed in service, a second high accuracy torus level indication instrument. The two new instruments unexpectedly showed a 0.5 inch difference between one another. The 1 inch administrative margin to TS maximum and minimum levels were still in place; therefore, there was no immediate operability concern.

"An evaluation was performed on 03/11/99 using temporary stand-pipes to determine the actual torus water level and thereby determine the accuracy of the newly installed instrumentation. That evaluation concluded that the first new instrument (activated on 02/24/99) had been consistently indicating approximately 0.5 inches higher than actual level.

"This evaluation revealed that the report made on 02/24/99 was not required. The actual instrument inaccuracy identified on 03/11/99 indicated that the administrative limits in place since the 05/98 outage accounted for all but 0.08 inches of the maximum error observed in the instrumentation used to control torus water volume.

"Plant operating logs taken since the 05/98 plant start up were reviewed. The logs show that VY operated consistent with its Technical Specifications.

"The maximum error observed (0.08 inches) was evaluated in light of the analyses and calculations used to verify the adequacy of the Vermont Yankee Primary Containment. In no case did this minor variance challenge the conclusions of any of the analyses/calculations.

"Therefore, it was concluded that the conditions were not inconsistent with the design bases of the VY plant, and the report made on 02/24/99 is being retracted."

The licensee will inform the NRC resident inspector of this retraction. Notified R1DO (K. Modes).

TRANSPORTED POTENTIALLY CONTAMINATED INJURED PERSON

The following text is a portion of a facsimile received from the licensee:

"At 1607 hrs on 02/24/99, a diver qualified maintenance contractor was injured and transported to Salem Memorial Hospital offsite as a potentially contaminated individual. The diver was performing maintenance activities on plant equipment located underwater and inside the plant suppression pool. While setting up a high torque machine, the diver severed his right index finger. The diver was surveyed and found to be clean except for a small portion of his right hand, which could not be surveyed due to the extent of the injury. Radiation Protection personnel accompanied the diver to the hospital, surveyed the diver upon arrival, and determined that the diver was not contaminated. The severed finger tip was subsequently located, determined to be contaminated, and transported to the Salem Hospital. [ ... After an evaluation by hospital medical personnel, the severed finger was returned to the station. ... ] The diver is currently in stable condition and being monitored by hospital personnel. Diving operations and associated maintenance have been suspended pending further investigation."

The licensee notified NRC Resident Inspector and the local township officials.

BOTH TRAINS OF HPCI DECLARED INOPERABLE

The following text is a portion of a facsimile received from the licensee:

"During the performance of a routine surveillance test for high pressure coolant injection (HPCI) operability, feedwater booster pump 12 was secured and observed to be rotating backwards. The discharge check valve failed open causing reverse flow through the pump [which] resulted in both HPCI trains [being declared] inoperable. When both HPCI systems are declared INOP, Tech Spec 3.1.8.c requires an initiation of a shutdown within 1 hr and reactor coolant pressure and temperature [to] be reduced to less [than] 110 psig and saturation temp with 24 hrs."

The initial problem was noted at 1357 EST, the determination of inoperability was made at 1420 EST, and actual power reduction was started at 1447 EST. The licensee is investigating the event and plans to isolate the line containing the failed check valve, which will allow the HPCI system to be returned to operable condition and allow maintenance on the check valve.

The licensee notified the NRC Resident Inspector.

PARTIAL LOSS OF PLANT EMERGENCY ASSESSMENT CAPABILITY

The following text is a portion of a facsimile received from the licensee:

"On 02/22/99 at 2040 [CST], the plant computer's [safety display systems] in the [emergency operations facility (EOF)] and [emergency response data system] datalink to the NRC stopped communicating with the plant computer. These components were unavailable for use if required due to an activation of the EOF as per the Emergency Plan. Investigation found that the bridge which is used to link the EOF components with the plant computer had 'locked up.' There was no apparent cause for the bridge to be 'locked up.' "

Technicians had performed an outage on 02/21/99, which lasted about 5 minutes. The components were working properly after the outage. The bridge has been reset as of 1323 CST on 02/24/99 and is functioning properly.

The licensee informed the NRC Resident Inspector.