Headquarters Daily report MARCH 15, 1995 *************************************************************************** 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 - HEADQUARTERS MARCH 15, 1995 MR Number: H-95-0075 NRR DAILY REPORT ITEM GENERIC COMMUNICATIONS BRANCH/EVENTS ASSESSMENT BRANCH DIVISION OF OPERATING REACTOR SUPPORT OFFICE OF NUCLEAR REACTOR REGULATION Subject: N/A NRC Information Notice 95-18, "Potential Pressure-Locking of Safety-Related Power-Operated Gate Valves," issued March 15, 1995. The NRC is issuing this information notice to alert addressees to a recent analysis demonstrating the potential susceptibility of safety injection valves to pressure-locking. Technical contacts: Thomas Scarbrough, NRR (301) 415-2794 Donald Kirkpatrick, NRR (301) 415-1849 PRIORITY ATTENTION REQUIRED MORNING REPORT - REGION I MARCH 15, 1995 Licensee/Facility: Notification: Peco Energy Company. MR Number: 1-95-0037 Limerick 1 2 Date: 03/14/95 Philadelphia,Pennsylvania SRI PC Dockets: 50-352,50-353 BWR/GE-4,BWR/GE-4 Subject: UPDATE TO THE LOSS OF UNITS 1 AND 2 DUE TO DISTURBANCE IN THE OFFSITE POWER GRID Reportable Event Number: N/A Discussion: On February 21, 1995, Limerick Units 1 and 2 experienced simultaneous automatic reactor scrams when a disturbance in the offsite power grid caused generator lockouts and turbine trips on both units. The grid disturbance was caused by a failure of a lightning arrester in the Whitpain substation during switching of voltage control capacitors. The event was experienced by both 220 (Unit 1) and 500 kV (Unit 2) lines, which are connected via transformers at the Whitpain substation. Offsite power remained energized throughout the event, including power to the vital safeguards buses in both units. The units were placed in a stable condition shortly after the scrams with all safety systems responding as designed. During a routine switching evolution, involving deenergizing a capacitor bank to lower the voltage on the transmission system, a voltage spike was caused when a 220 kV breaker (No. 245) experienced a restrike while opening. (A restrike is loosely defined as the breaker temporarily clearing current, but then allowing the current to reestablish within the next half cycle.) The voltage spike damaged the 'B' phase lightning arrestor on the 220-16 line in the Whitpain substation. The resulting electrical fault was cleared in five cycles by protective relay action, but reappeared on the system when the 220-16 line was reenergized by the automatic reclose system at the North Wales substation. The Limerick units tripped, thus isolating them from the system during the second fault due to the duration of the fault and the failure of the protective relays at North Wales to function and clear the fault. All protective systems at Limerick worked correctly, safely isolating the station from the transmission system during the disturbance. The PECO Energy's initial investigation determined that at approximately 1.2 seconds after the first fault was cleared, 220 kV breaker No. 905 automatically reclosed at North Wales to reenergize the 220-16 line. This is a standard practice for PECO Energy aerial 220 kV lines to reclose one second after an interruption, because most faults are transients caused by lightning. However, the reclosing of breaker No. 905 reinitiated the fault at the failed lightning arrestor, at which point the 905 primary relays and backup relays should have opened the breaker again. This did not happen. The cause of the primary relay failure is still under investigation at this time. The failure of the backup relay was caused by the failure of the SC ground relay to activate the auxiliary relay and open the 905 breaker in a timely manner. (The backup relay system finally activated, opening the breaker after approximately 114 cycles.) The reason for the delay in relay actuation was the improper application of the SC relay in series with a solid state timer which stops whenever current flow drops. PECO Energy's investigation revealed the manufacturer sent out a notice to all users in 1978 warning against the use of SC relays in this type of configuration. Apparently, this type of installation is very common on line- and stuck-breaker protective systems for PECO Energy. An effort is underway to count the actual number of similar installations in the system. The restriking of the 245 breaker at Whitpain was also reviewed by PECO Energy. The breaker was installed in 1986. The present maintenance policies indicate it should have been overhauled in 1994 with preventive diagnostics performed in 1990. There was no record of any maintenance being performed on the breaker. Additionally, the breaker was close to the manufacturer's published limit of 2000 operations, having been operated approximately 1950 times. Regional Action: The resident inspectors and the Region will continue to review the licensee's current investigation. The inspectors will review the formal root cause analysis and resulting recommendations when the licensee has completed its investigation, expected by the end of March 1995. Contact: Clifford Anderson (610)337-5227 Neil Perry (610)327-1344 PRIORITY ATTENTION REQUIRED MORNING REPORT - REGION II MARCH 15, 1995 Licensee/Facility: Notification: University Of Florida MR Number: 2-95-0029 Univ Of Florida Date: 03/14/95 Gainesville,Florida Dockets: 05000083 License No: R-56 100 KW ARGONAUT Subject: SUSPECTED HEAT EXCHANGER LEAK Reportable Event Number: N/A Discussion: On March 14, 1995 the University of Florida reported a probable leak in the primary coolant heat exchanger based on observation of decreasing resistivity of the primary coolant when secondary coolant was flowing. Based on preliminary radioactivity measurements and surveys, the licensee believes that there has not been a release of radioactivity to the environment. The licensee noted the decrease in primary resistivity during shutdown on March 9 but did not confirm that the indications suggested a potential leak until March 13 when additional testing was performed. The reactor has not operated since March 9. The licensee has isolated the heat exchanger and plans to continue testing to confirm the leak. The reactor is a 100 kW Argonaut type with light water primary cooling. The secondary side water runs at a higher pressure than the primary, comes from either well or city water, and goes to the University sewage treatment plant. The licensee submitted a written notification on March 14 concerning the situation and will follow with a 14-day report. Regional Action: The region will continue to monitor the licensee actions by telephone and will perform inspection follow during the next routine inspection. Contact: Edward McAlpine (404)331-5547 PRIORITY ATTENTION REQUIRED MORNING REPORT - REGION III MARCH 15, 1995 Licensee/Facility: Notification: Commonwealth Edison Co. MR Number: 3-95-0035 La Salle 1 2 Date: 03/15/95 Marseilles,Illinois SRI VIA MEETING Dockets: 50-373,50-374 BWR/GE-5,BWR/GE-5 Subject: UNANALYZED FLOODING SCENARIO IDENTIFIED POTENTIAL (50.9 ISSUE) Reportable Event Number: N/A Discussion: On March 14, 1995, ComEd identified a potentially significant flooding scenario. This scenario could result in a loss of all ECCS pumps, RCIC, and offsite power for both units, if the flooding were not mitigated. ComEd discovered that two pipes in the basement of the turbine building were not included in the flooding analysis contained in the FSAR. They were the service water discharge header (48 inch dia) and the intake bay de-icing line (120 inch dia). Breaks in either of these could cause flooding, with an unlimited supply of water (the cooling lake). ComEd has postulated the maximum possible break as 18 sq. inches, using a moderate energy line break methodology. ComEd's preliminary estimate was that it would take at least 15 hours to cause flooding in the ECCS pump rooms in the reactor building. ComEd considers that this flooding scenario could be mitigated. ComEd is determining whether this condition is reportable under 50.72 and 50.9. Regional Action: The resident staff is following the issue, and NRR has been briefed on the issue. Contact: L. MILLER (708)829-9629 PRIORITY ATTENTION REQUIRED MORNING REPORT - REGION III MARCH 15, 1995 Licensee/Facility: Notification: Commonwealth Edison Co. MR Number: 3-95-0036 Byron 2 Date: 03/15/95 Byron,Illinois E-MAIL Dockets: 50-455 PWR/W-4-LP Subject: APPARENT REACTOR VESSEL LEVEL TRANSIENT Reportable Event Number: N/A Discussion: Unit 2 was in cold shutdown with the reactor vessel head on. On March 13, 1995, the licensee performed motor operated valve testing on the loop D Safety Injection (SI) Accumulator Isolation Valve. When the licensee stroked opened SI8808D, an apparent level transient occurred. The three reactor vessel level indications rapidly increased from approximately 399.8 feet to 409 feet. The 399.8 feet level indication corresponds to the top of the reactor vessel flange. The reactor operator immediately shut the isolation valve to stop the level transient. Subsequently, the level indication slowly decreased to approximately the original indication of 399 feet within 2 hours. The licensee considers that residual pressure in the SI accumulator caused a pressure transient in the coolant system, and an apparent level increase indication. The loop D SI accumulator had been earlier drained and depressurized to approximately 10 psig. Both vent paths from the reactor vessel head and the pressurizer were available. The licensee dispatched equipment attendants to the vent valves located inside the containment. The operators found gas venting from the vents. The licensee also noted the residual heat removal (RH) pump suction pressure had increased 5 psig. Shutdown cooling flow rate was not affected. Boron concentration of the reactor coolant was not affected. The licensee continues to investigate the event. Calculations of the volume of water displaced, vessel level changes, and pressure changes were being developed. Initially, the licensee considers that the level changes were an erroneous level increase, and not an actual water displacement equivalent to 9 feet. Regional Action: The resident inspectors are monitoring the licensee's investigation. The licensee briefed the region and NRR on February 14, 1995. The region plans to conduct a special followup inspection. Contact: LEWIS F. MILLER, JR. (708)829-9629