Headquarters Daily report DECEMBER 23, 1994 *************************************************************************** REPORT NEGATIVE NO INPUT ATTACHED INPUT RECEIVED RECEIVED HEADQUARTERS X REGION I X REGION II X REGION III X REGION IV X *************************************************************************** PRIORITY ATTENTION REQUIRED MORNING REPORT - REGION I DECEMBER 23, 1994 Licensee/Facility: Notification: Philadelphia Electric Co. MR Number: 1-94-0134 Peach Bottom 2 Date: 12/23/94 Philadelphia,Pennsylvania SRI PC Dockets: 50-277 BWR/GE-4 Subject: STEAM/WATER LEAK DURING REACTOR WATER CLEANUP SYSTEM TESTING Reportable Event Number: N/A Discussion: On December 22, 1994, while lining up the 2B reactor water cleanup (RWCU) system pump at Peach Bottom Unit 2 for a modification acceptance test (MAT), a steam/water leak was initiated from two instrument vent valves. The leak contaminated the shoes of three plant workers, and sections of the reactor building. The leak was quickly isolated and did not result in any increase in off-site release levels. At approximately 10:14 a.m. the 2B RWCU pump suction isolation valve was opened in preparation for a MAT. Opening this valve allowed reactor system pressure (approximately 1000 psig) to be transmitted to a flow sensing element located on the discharge side of the pump. The sensing lines (1/4 inch stainless steel tubing) to the flow element contained two air removal vent valves which were in the open position and provided a flowpath from the reactor vessel through the RWCU system to the reactor building atmosphere. A water and steam fluid stream issued from the vent valves for approximately three minutes before plant workers could isolate the leak. At 10:46 a.m. PECO evacuated the Unit 2 reactor building as a precautionary measure due to the potential radiological concerns associated with the leak. The shoes of three personnel in the vicinity of the vent valves were contaminated to approximately 300 counts/minute above background. The leak did not cause any increase in the reactor building airborne or general radiation levels. Sections of three elevations in the reactor building were contaminated up to levels of approximately 200 mrad/100 square centimeters. PECO commenced decontamination of the reactor building and restored normal access to most of the reactor building by 12:08 p.m. The apparent cause for this event was that the RWCU lineup specified by the MAT procedure did not address positioning of the instrument line vent valves. PECO initiated a performance enhancement program review to investigate this event and determine the root cause(s) and necessary corrective actions. Regional Action: The resident inspector will review the results of PECO's investigation. Contact: Raymond Lorson (717)456-7614 Clifford Anderson (610)337-5227 PRIORITY ATTENTION REQUIRED MORNING REPORT - REGION IV DECEMBER 23, 1994 Licensee/Facility: Notification: Pacific Gas & Electric Co. MR Number: 4-94-0147 Diablo Canyon 1 2 Date: 12/22/94 Avila Beach,California Resident Inspector Dockets: 50-275,50-323 PWR/W-4-LP,PWR/W-4-LP Subject: SEISMIC ANALYSIS OF DIESEL FUEL OIL TRANSFER SYSTEM SUCTION LINE Reportable Event Number: N/A Discussion: On December 22, 1994, at approximately 11 a.m. (PST), licensee engineering personnel determined, during a design review, that both trains of the diesel fuel oil transfer system may not meet FSAR seismic design criteria for a pipe support in the suction line from the fuel oil tanks. The licensee's preliminary analysis indicates that the suction line pipe support of concern may not meet the original seismic design basis. However, the licensee considers the supports to be currently operable, using previous experience with piping analysis and the criteria of Appendix F of Section III of the ASME Code recommended as operability guidance by Generic Letter 91-18. The licensee is currently utilizing finite element analysis techniques to determine actual potential stresses on this support. The licensee anticipates this analysis to take several days. Regional Action: Region IV will review the licensee's analysis when completed. No other action is anticipated. Contact: D. Kirsch (510)975-0290 M. Tschiltz (805)595-2354 D. Corporandy (510)975-0319 PRIORITY ATTENTION REQUIRED MORNING REPORT - REGION IV DECEMBER 23, 1994 Licensee/Facility: Notification: Entergy Operations, Inc. MR Number: 4-94-0148 River Bend 1 Date: 12/22/94 St Francisville,Louisiana Senior Resident Inspector Dockets: 50-458 BWR/GE-6 Subject: DIESEL GENERATOR EXHAUST MANIFOLD BOLT FAILURE Reportable Event Number: N/A Discussion: On December 21, 1994, at 11:10 a.m. (CST), during an operational surveillance test of the Division II diesel generator (DG), an operator noticed an increase in noise levels and located a broken fastener on the exhaust manifold connection to Cylinder 8. System engineering was notified and responded to the DG room. While observing the broken fastener, a second fastener broke. The DG was immediately secured. As the DG was secured, the operator heard noises from the circuit breaker for the room exhaust fan. The operator noted that the circuit breaker charging spring motor was still energized and running, even though the spring was charged. An operator racked out the breaker to ensure that the circuit breaker would not be damaged. As a result of the broken fasteners and the exhaust fan breaker not being energized, the licensee declared the DG inoperable. The Division I and II DGs are eight-cylinder, four-cycle Transamerica Delaval engines. There are four fasteners used to connect the exhaust manifold to each cylinder. As immediate corrective action, the licensee replaced all four fasteners and a thermocouple, which was damaged by vibration during the event, on Cylinder 8 of the Division II DG. The other 28 fasteners were checked for wrench tightness and the presence of lockwashers. The Division I DG was also visually inspected to verify that all the fasteners and lock washers were in place. In addition, the licensee initiated a maintenance work order to check the wrench tightness of these fasteners during the next scheduled outage on January 4, 1995. The licensee's current maintenance procedures required that the fasteners be verified wrench tight within 4 hours of a DG run and did not specify a specific torque value. The licensee has preliminarily concluded that fasteners failed due to cycling fatigue and has sent them to a laboratory for further analysis. Preliminary results indicate that the fasteners were the correct material with no material flaws and that the fasteners failed from mechanical fatigue. A potential cause of the failure was the lack of uniform tightness on the fasteners because they are wrench tight vice a specified torque value. A final report will be issued by the laboratory sometime during the week of December 26-30, 1994. The DG vendor, Cooper Energy Services, indicated that the loss of all exhaust manifold flange bolting at one cylinder would not adversely effect the DG ability to carry the sites rated load. However, the vendor indicated that Service Information Memo (SIM) 383, dated October 12, 1992, called for an upgrade of the fasteners to a higher tensile strength material (i.e., 160,000 psi) and specified a specific torque value (i.e., 100-120 ft-lbs). Although the licensee believed they never received this SIM, the licensee indicated that they had replaced the original fasteners in 1984 because of similar problems with the fasteners breaking during preoperational testing. The fasteners installed have a tensile strength of 125,000 psi. The licensee is currently evaluating the vendor recommendations. Troubleshooting of the General Electric 480 Volt AKR-4BE-30-1 circuit breaker for the DG exhaust fan determined that the ratchet pawl assembly for the charging motor became worn, causing an improper alignment. The breaker was replaced and tested satisfactory. The DG was declared operable after the DG satisfactory completed its surveillance test. Regional Action: A conference call was held between the Regional Office and the licensee on the cause of the fastener's failure and their corrective actions. The resident inspectors will continue to follow the licensee's actions. Contact: C. A. VanDenburgh (817)860-8161 PRIORITY ATTENTION REQUIRED MORNING REPORT - REGION IV DECEMBER 23, 1994 Licensee/Facility: Notification: Entergy Operations, Inc. MR Number: 4-94-0149 Arkansas Nuclear 1 Date: 12/23/94 Russelville,Arkansas Senior Resident Inspector Dockets: 50-313 PWR/B&W-L-LP Subject: DEGRADED SAFETY-RELATED INVERTER Reportable Event Number: N/A Discussion: On December 20, 1994, the licensee noticed that a trouble alarm for safety-related Inverter Y-13 occurred coincident with a start of Circulating Water Pump P-3C. In addition, Inverter Y-13 had automatically transferred vital AC loads to the alternate (maintenance) power source. While this transfer was designed to be a bumpless transfer, vital AC power was momentarily lost, resulting in a trip of the Channel C reactor protection system. The licensee declared the inverter inoperable at 5:30 p.m. on December 20, 1994. During subsequent troubleshooting activities, the licensee determined that protective devices (i.e., fuses and breakers) had actuated in both the input and output circuitry of the inverter. The licensee believed the input protective device actuations were caused by a momentary short circuit within the inverter. External electrical noise induced timing errors in controls for the silicon control rectifiers (SCRs) used in the inverter. Misfiring of an SCR most likely resulted in a momentary short circuit which caused the input protective devices to actuate. The licensee theorized that external noise was caused by the starting of the circulating water pump. The licensee believed the output protective device actuated because the static transfer switch did not function properly during the transfer to the alternate power source. The licensee reset the inverter's protective devices and retested the inverter to ensure the resulting transients had not caused any internal damage to the inverter. The maintenance history for all similar inverters indicated that two other failures had resulted in actuation of the input circuit protective devices within the last 2 years. These failures both occurred on Inverter Y-22; however, the Inverter Y-22 static transfer switch functioned properly and the reactor protection system was previously unaffected. The licensee evaluated the frequency of inverter failures related to external noise and declared the invertor operable based on resetting the protective devices, a successful test which determined there was no internal damage, and the low frequency of noise-induced failures. Nevertheless, the licensee opened the AC input breakers for Inverters Y-13 and Y-22 to limit susceptibility to external noise. Operating these inverters in the DC mode was within the capabilities of the DC power supply. Night orders were also issued to ensure the operators were aware of this issue. As long-term corrective action, the licensee had previously planned to replace two inverters and refurbish the remaining two during the next refueling outage, which will start February 14, 1995. Regional Action: The resident inspectors will continue to follow the licensee's actions. Contact: C. A. VanDenburgh (817)860-8161