Headquarters Daily report OCTOBER 13, 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 OCTOBER 12, 1994 Licensee/Facility: Notification: Public Service Electric & Gas Co. MR Number: 1-94-0112 Hope Creek 1 Date: 10/12/94 Hancocks Bridge,New Jersey SRI PC Dockets: 50-354 BWR/GE-4 Subject: HOPE CREEK REACTOR SCRAM Reportable Event Number: 27876 Discussion: PSE&G investigation into the October 7 turbine trip and reactor scram during the startup determined that a capacitor failure caused the speed regulating card in the EHC system to fail. Troubleshooting efforts were hindered by the non-repeating failure in the card. The licensee suspects that the capacitor may have been slightly damaged during original manufacture and then prematurely failed (after about five years service). The capacitor failure initially resulted in a short circuit causing the turbine control valves to respond erroneously during the startup activities on October 7. Subsequently, during troubleshooting activities, the shorted capacitor failed open, resulting in the non-repeatable failure. PSE&G replaced the faulty circuit card in the EHC system and revised operating procedures so that operators could identify this type of failure prior to turbine rolling so as to avoid future similar reactor scrams. Following resolution of the October 7 EHC system failure, and a broad review of recent events (five automatic scrams since May 1994) and proposed actions by station management, Hope Creek commenced a reactor startup at 12:44 p.m., on October 11 (Mode switch in STARTUP). Criticality was achieved at 4:13 p.m., and the mode switch was placed in RUN at 12:16 a.m., on October 12. The turbine roll commenced at 2:35 a.m., and subsequently reached 1800 rpm without further EHC difficulties. Prior to synchronizing the turbine generator to the grid, operators determined that 1AFHV-1357C (extraction steam inlet isolation valve to the #3C feedwater heater) could not be properly operated remotely from the control room. Inspection of the valve indicated severe galling on the valve stem. PSE&G determined that valve binding was occurring due to a cocked lantern ring inside the packing area of the valve yoke. The main turbine was tripped to support repair of the valve. Current conditions as of 8:00 a.m., October 12: Unit holding at 20% power with turbine tripped. Turbine will be restarted and synchronized to the grid following successful completion of packing replacement and stem repairs to 1AFHV-1357C. It is estimated that repairs to the valve should be completed later this morning and by mid-afternoon the generator will be placed on-line. Regional Action: Residents monitored licensee investigation, recovery actions and the subsequent reactor startup. Contact: John White (610)337-5114 Robert Summers (610)337-5189 PRIORITY ATTENTION REQUIRED MORNING REPORT - REGION IV OCTOBER 12, 1994 Licensee/Facility: Notification: Entergy Operations, Inc. MR Number: 4-94-0118 Arkansas Nuclear 1 2 Date: 10/12/94 Russelville,Arkansas Senior Resident Inspector Dockets: 50-313,50-368 PWR/B&W-L-LP,PWR/CE Subject: INADEQUATE ENGINEERED SAFETY FEATURES ACTUATION TESTING Reportable Event Number: N/A Discussion: On October 7, 1994, the NRC senior resident inspector identified during an independent verification effort that the licensee had not adequately tested the Unit 2 emergency diesel generator (EDG) loading sequence to confirm that swing High Pressure Safety Injection (HPSI) Pump 2P-89C would automatically load on an emergency bus in the following circumstance: - the swing pump was being relied on while the primary pump was out of service, and - a loss of offsite power (LOOP) occurs, and - the EDG is supplying power to the emergency bus, and - a safety injection actuation signal (SIAS) has initiated. In response to a recent operational concerns at the Cooper and Waterford stations concerning the adequacy of integrated EDG load testing, the Senior Resident Inspector performed an independent verification of the testing methodology used at Arkansas Nuclear One (ANO). The ANO design uses two engineered safety features (ESF) trains with swing components for reliability. The components include a swing high pressure injection (HPI) pump on Unit 1, a swing charging pump and a swing HPSI pump on Unit 2, and a swing service water (SW) pump on Units 1 and 2. These swing components can be aligned to either ESF bus to replace a pump removed for maintenance or testing. The inspector found that the licensee had not tested the Unit 2 swing HPSI pump during the integrated (ESF) system 18-month time response test, which was last performed in April 1994, to confirm it would start following a LOOP in conjunction with an SIAS. The licensee had only confirmed that the pump would start on an SIAS actuation when aligned to offsite power. Following notification by the inspector, the licensee entered Technical Specification 3.5.2 at 3:40 p.m., declared HPSI Pump 2P-89C inoperable, and aligned HPSI Pump 2P-89B as the operable pump. The licensee performed additional testing of the HPSI Pump 2P-89C starting circuitry and verified the proper functioning of the portion of the circuit that had not previously been tested (i.e., contacts from the EDG output breakers). In response to the inspector's concern, the licensee performed additional reviews of the testing associated with swing pumps on both units and identified the following additional features which had not been adequately tested: Unit 2 - The EDG loading sequence had not been verified to confirm that more than one SW pump or more than one HPSI pump would not automatically load on an ESF bus. As immediate corrective action, the licensee issued night orders and caution cards to ensure only one SW pump and one HPSI pump hand switch would be out of pull-to-lock on an ESF bus. This assured that only one pump would sequence on each ESF bus during EDG loading. Unit 1 - The EDG loading sequence had not been verified to confirm that more than one HPI pump would not automatically load on an ESF bus. As immediate corrective action, the licensee placed swing HPI Pump P-36B in pull-to-lock until additional testing confirmed proper contact functioning. This assured that only one pump would sequence on each ESF bus during EDG loading. Unit 1 - The starting circuitry for swing HPI Pump P-36B had not been verified to confirm that the pump would automatically start during an ESF actuation if it was powered from offsite power. As immediate corrective action, the licensee placed swing HPI Pump P-36B in pull-to-lock until additional testing confirmed the proper functioning of the contacts. Regional Action: Continued followup by the resident inspection staff. Contact: C. A. VanDenburgh (817)860-8161 PRIORITY ATTENTION REQUIRED MORNING REPORT - REGION OCTOBER 13, 1994 MR Number: H-94-0094 NRR DAILY REPORT ITEM SIGNIFICANT EVENTS EVENTS ASSESSMENT BRANCH DIVISION OF OPERATING REACTOR SUPPORT Subject: Inadequate Load Shed and Logic Testing The NRR/AEOD Events Assessment Panel on September 27, 1994, classified the inadequate load shed and logic testing at Cooper Nuclear Station as a Significant Event for the NRC Performance Indicator Program. On May 16, 1994, a contractor engineer performing fuse configuration verification in a nonsafety-related 480 VAC motor control center (MCC N), found the feeder breaker undervoltage trip device mechanically defeated. A plastic cable tie was tied around the trip arm, preventing it from tripping the breaker on a loss of offsite power. The feeder breaker for MCC N is required to open on a loss of offsite power so that the EDG is not overloaded while powering safety-related loads. The cable tie had been installed as part of a breaker surveillance test, but the procedure did not include a step to remove the cable tie after the test was completed. No tie wraps were found in other 480 VAC breakers. An evaluation by the licensee determined that the affected EDG would remain operable, even if MCC N failed to load shed. NRC inspectors reviewed the load shed testing procedure to determine why the MCC N breaker failure to load shed was not previously identified during performance of the test. They found that there was no step to verify any of the 480 VAC loads (vital and non-vital) shed within a specified time (8.5 sec) and some 480 VAC loads (the control rod drive pumps and the station air compressors) which were required to shed, were not included in the procedure. The test procedure for the four 4160 VAC RHR service water booster pumps was reviewed and found to be inadequate because the load shed terminals were jumpered to simulate a load shed signal and the relays themselves were not tested. Subsequent testing of the relays proved that the booster pumps would load shed as required. A review of the test results from the July 1993 outage found that all the 480 VAC loads except the MCC N breaker shed from the bus. However, some of the loads did not shed within the required 8.5 seconds. The licensee decided to perform a load shed test to verify all loads would properly shed. Four nonsafety-related 480 VAC breakers (Westinghouse model DB-50) failed to trip on an undervoltage signal. Bench testing of the breakers identified mechanical binding and sticking of the undervoltage trip devices. This problem with DB-50 breakers was the subject of NRC Bulletin 83-08 and 85-02, and Information Notice 93-009, but the licensee had not reviewed any of them. The licensee declared both EDGs inoperable, an Unusual Event was declared, and the plant was subsequently shut down on May 26, 1994. The plant remained in the Unusual Event for 56 days due to the many deficiencies that were found as the investigation unfolded. Calculations were subsequently performed by the licensee to determine if the EDGs would have performed their safety function with the nonsafety 480 VAC loads failing to shed from the vital bus. It was determined that EDG 1 would have slightly exceeded (4508 KW) its capacity of 4505 KW, and EDG 2, with a capacity of 4490 KW would have experienced a peak load of 4414 KW. While EDG1 would have slightly exceeded its continuous load capacity, the calculation noted that the 2 hour peak load capacity of 4950 KW would not have been exceeded. The licensee stated that the EDG would not be operated over 4000 KW for more than 2 hours because operators are required by procedure to monitor the EDG output and limit it to 4000 KW by manually shedding nonsafety loads if necessary. The Electrical Systems Branch (EELB) of NRR is currently reviewing the licensee calculations. An Enforcement Conference was held in RIV on September 16, 1994. This event has been classified as a Significant Event due to programmatic breakdown. A probabilistic safety analysis of this event will be performed by the Probabilistic Safety Assessment Branch (SPSB) of NRR after review of the licensee calculations is complete. CONTACT: David L. Skeen, NRR/DOPS/OECB (301) 504-1174 PRIORITY ATTENTION REQUIRED MORNING REPORT - HEADQUARTERS OCTOBER 13, 1994 MR Number: H-94-0095 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 94-73, "Clarification of Criticality Reporting Criteria," was issued October 12, 1994. The NRC is issuing this information notice to alert addressees to a clarification of the criticality reporting criteria previously provided in Bulletin 91-01, Supplement 1. Technical contact: Jerry Roth, NMSS (301) 415-7156 PRIORITY ATTENTION REQUIRED MORNING REPORT - REGION II OCTOBER 13, 1994 Licensee/Facility: Notification: Tennessee Valley Authority MR Number: 2-94-0085 Browns Ferry 2 Date: 10/13/94 Decatur,Alabama Dockets: 50-260 BWR/GE-4 Subject: FOREIGN MATERIAL FOUND IN TORUS Reportable Event Number: N/A Discussion: On October 10, during inspection of the underwater surfaces in the Unit 2 torus, divers identified numerous pieces of material in the torus. Unit 2 is in a scheduled refueling outage. The resident inspectors viewed a videotape of the "as found" conditions, toured the torus area, and discussed the issue with the divers and other involved personnel. The material appears to be a cloth or paper type material that contains small fibers. The pieces are typically four square inches in size, but smaller pieces are common. The videotape showed pieces resting on the bottom of the torus and on the ECCS strainer surfaces. Some of the material was protruding from the holes on the safety relief valve tailpipe T-quenchers. The material appeared to have been forced out of the T-quencher by the steam from lifting of the safety relief valves. The inspectors noted that approximately twenty percent of the surface area of one of the ECCS strainers was covered with the material. The other strainer in the video tape did not have as much material on it. (Browns Ferry has four ECCS suction strainers in the torus which supply a path to an exterior ring header.) Licensee management has directed that the entire interior of the torus be inspected and cleaned. An incident investigation has been initiated. Management has indicated to the inspectors that the total amount of material will be quantified. Regional Action: The resident inspectors will continue to monitor the licensee's investigation and corrective actions. Contact: Mark S. Lesser (404)331-0342 PRIORITY ATTENTION REQUIRED MORNING REPORT - REGION III OCTOBER 13, 1994 Licensee/Facility: Notification: Consumers Power Co. MR Number: 3-94-0181 Big Rock Point 1 Date: 10/12/94 Charlevoix,Michigan FAX FROM RI Dockets: 50-155 BWR/GE-1 Subject: IMPROPER HANDLING OF SPENT FUEL ASSEMBLY Reportable Event Number: N/A Discussion: On October 12, 1994, at 6:25 a.m. (CDT), a spent fuel assembly was dropped during handling. Big Rock Point is presently in a refueling outage and was in the process of handling spent fuel for inspections prior to reload. The fuel assembly had just been returned to the spent fuel pool rack when an operator failed to fully disengage the lifting grapple from the fuel assembly prior to lifting. The fuel assembly was lifted with the partially engaged grapple approximately four inches before dropping back into the rack. The operator stopped the lift upon realizing that the fuel assembly was not fully disengaged. Currently, the licensee halted all further fuel movements pending a root cause evaluation and corrective actions to prevent recurrence. The fuel assembly is undergoing inspection for damage. Regional Action: The inspectors will followup on the licensee's root cause evaluation and corrective actions. Contact: M.P. PHILLIPS (708)829-9637 PRIORITY ATTENTION REQUIRED MORNING REPORT - REGION III OCTOBER 13, 1994 Licensee/Facility: Notification: Commonwealth Edison Co. MR Number: 3-94-0182 Byron 1 Date: 10/13/94 Byron,Illinois SRI Dockets: 50-454 PWR/W-4-LP Subject: WATER LEAK INTO THE CONTAINMENT THROUGH AN OPENED S.G TUBE Reportable Event Number: N/A Discussion: Unit 1 is currently in a refueling outage which commenced September 8, 1994. On October 13, 1994, at approximately 12:15 a.m. (CST), a Babcock & Wilcox (B&W) representative informed the licensee that water was draining from an unplugged steam generator (S/G) cold leg tube. Two S/G tubes had been pulled on the "C" S/G hot leg side. When the tubes are pulled, both the cold and hot leg sides must be plugged. The licensee was placing the "C" S/G into wet layup when secondary water was found flowing out of the primary manway onto the Unit 1 containment floor. The leak from the unplugged tube was approximately 13 gpm and released an estimated total of 2000 gallons of secondary side water into the containment sump. The licensee has determined that the "C" S/G cold leg penetration was not plugged before the responsible B&W supervisor released the S/G to operations for wet layup. Regional Action: A regional inspector discovered this incident during routine observation of the control room shift turnover. The resident inspectors are following the licensee corrective actions. Contact: L.F.MILLER (708)829-9629 PRIORITY ATTENTION REQUIRED MORNING REPORT - REGION IV OCTOBER 13, 1994 Licensee/Facility: Notification: Entergy Operations, Inc. MR Number: 4-94-0119 River Bend 1 Date: 10/13/94 St Francisville,Louisiana Senior Resident Inspector Dockets: 50-458 BWR/GE-6 Subject: CRACKS FOUND ON CONTROL ROD DRIVE PIPING Reportable Event Number: N/A Discussion: On October 10, 1994, while performing a pre-startup walkdown of the drywell, the operators found a hairline longitudinal crack in Control Rod Drive (CRD) 44-33 insert piping. The crack was weeping water from the CRD system. At the time, the plant was in cold shutdown, near the end of a 32-day forced outage. The CRD piping is 304 stainless steel, schedule 80 pipe with a 1 1/4 inch diameter on the insert line and a 1 inch diameter on the withdraw line. The licensee found a brown substance on this and 15 adjacent withdraw and insert pipes, which was analyzed to contain about 75 percent chlorides. The licensee found a discontinuous throughwall crack of up to 1 1/2 inch on one pipe and an indication of cracks on six adjacent pipes directly under the area where the brown substance was deposited. The licensee is currently liquid penetrant and ultrasonic testing the remaining pipes. In addition, the licensee found some of the brown substance splattered on the reference leg of the reactor pressure vessel level indication system. The licensee stated that trans-granular stress corrosion cracking has occurred on these pipes, caused by the presence of the brown substance. As of yet, the licensee has not determined the source of the substance, nor how long it has been present. The substance appears to be isolated to one bundle of pipes, as if something was spilled from above. The leaking pipe was replaced and the licensee plans to replace four more pipes; however, the licensee is still in the process of characterizing the condition of the remaining pipes. Regional Action: Routine followup by the resident inspectors. Contact: C. A. VanDenburgh (817)860-8161