IR 05000409/1986009

Insp Rept 50-409/86-09 on 860614-0926.No Violations Noted. Major Areas Inspected:Operational Safety Verification,Maint Activities,Licensee Event Repts Followup,Bulletins,Plant Trips & Open Items

ML20215G272
Person / Time
Site:	La Crosse File:Dairyland Power Cooperative icon.png
Issue date:	10/14/1986
From:	Boyd D NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III)
To:
Shared Package
ML20215G252	List: IR 05000409/1986009 ML20215G249
References
50-409-86-09, 50-409-86-9, IEB-84-02, IEB-84-2, IEB-86-002, IEB-86-2, NUDOCS 8610200302
Download: ML20215G272 (16)

Text

s U. S. NUCLEAR REGULATORY COMMISSION

REGION III

Report No. 50-409/86009(DRP)

Docket No.;50-409 License No. DPR-45 Licensee: Dairyland Power Cooperative 2615 East Avenue - South La Crosse, WI 54601 Facility Name: La Crosse Boiling Water Reactor Inspection At: La Crosse Site, Genoa, WI Inspection Conducted: June 14 through September 26, 1986 Inspectors: I. Vi11alva J. Wiebe Approved By: D. '# #~

Reactor Projects Section 20 Date

,

Inspection Summary Inspection from June 14 through September 26, 1986 (Re: ort No. 50-409/86009(DRP))

Areas Inspected: Routine, unannounced inspection by tie resident inspector of Operational Safety Verification; Maintenance Activities; Licensee Event Reports Followup; Bulletins; Plant Trips; and Open Item Results: No violations of NRC requirements were note DR 861015 ADOCK 05000409 PDR

.

'. -

DETAILS Persons Contacted

• J. Parkyn, Plant Superintendent

• G. Boyd. 0perations Supervisor

• L. Kelley, Assistant to Operations Supervisor L. Nelson, Health and Safety Supervisor R. Wery, Quality Assurance Supervisor S. Raffety, Reactor Engineer P. Bronk, Nuclear Engineer

• L. Goodman, Operations Engineer R. Brimer, Electrical Engineer D. Rybarik, Mechanical Engineer The inspector also interviewed other licensee personnel during the course of the inspectio * Denotes those attending exit interviews during the inspection perio . Operational Safety Verification The inspector observed control room operations, reviewed applicable logs and conducted discussions with . control room operators during the period of this report. The inspector verified the operability of selected emergency systems, reviewed tagout records and verified proper return to service of affected components. Tours of the reactor building, turbine

~ building, and cribhouse were conducted to observe plant equipment conditions, including ~ potential fire hazards, fluid leaks, and excessive vibrations and to verify that maintenance requests had been initiated for equipment in need of maintenance. The inspector by observation and direct interview verified that the physical security plan was being implemented in accordance with the station security pla The inspector observed plant housekeeping / cleanliness conditions and verified implementation of radiation protection controls. The inspecte walked down the accessible portions of the emergency core spray systems to verify operab.ilit These reviews and observations were conducted to verify that facility i operations were in conformance with the requirements established under

,

technical specifications, 10 CFR, and administrative procedure.

l l Maintenance Activities i Station maintenance activities, especially those conducted during the outage to repair the decay heat suction piping, of safety related systems and components listed below, were observed / reviewed.to ascertain that they were conducted in accordance with approved procedures, regulatory guides and industry codes or standards and in conformance with technical specification ,

-

Items considered during this review included verifying that: limiting conditions for operation were met while components or systems were removed from service; approvals'were obtained prior to initiating the work; activities were accomplished using approved procedures and were inspected as applicable; functional testing and/or calibrations were performed prior to returning components or systems to service; quality control records were maintained; activities were accomplished by qualified personnel; parts and materials used were prnperly certified; radiological controls were implemented; and, fire prevention controls were implemente Work requests were reviewed to determine the status of outstanding jobs and to assure that priority was assigned to safety related equipment maintenance which could affect system performance. Maintenance activities associated with the repair of the decay heat suction piping and the replacement of 1A Static Inverter were observed and reviewe Following completion of the maintenance of these items, the inspector verified that they had been, returned to service properl . Licensee Event Reports Followup Through direct observations, discussion with licensee personnel, and review of records, the following event reports were reviewed to determine that reportability requirements were fulfilled, immediate corrective action was accomplished, and corrective action to prevent recurrence had been accomplished in accordance with technical specification (Closed) LER 86-011: High Pressure Core Spray Bundle. This LER addresses the damage done to the high pressure core spray (HPCS)

bundle while it was being installed in the reactor vessel during the 1986 refueling outage. This event has been discussed previously in two inspection reports: Report No. 50-409/86005(DRS) and Report No. 50-409/86006(DRP). Both reports indicated that there were no concerns regarding the technical aspects of the event. Nevertheless, this LER was held open in Report No. 50-409/86006 pending improvement of the alignment markings for the crane. The licensee has improved the crane's alignment markings; therefore, this LER is closed.

I (Closed) LER 86-011 Revision 1: High Pressure Core Spray Bundl This LER revises the original LER by: (1) correcting previous assumptions as to why the HPCS bundle would not seat properly, and (2) revealing that a new fuel assenbly had not been properly

installed during the 1986~ refueling outage. The original LER attributed the problem experienced in seating the HPCS bundle to several factors, including the postulation that the HPCS bundle

-

had been slightly skewed. In contrast, the revised LER attributes the problems in seating the HPCS bundle to the protruding handle on a fuel assembly that had not been properly installe On August 3, while the reactor was being defueled in preparation for a pipe replacement (LER 86-022), Fuel Assembly 5-04 in. core position H-4 could not be grappled with the fuel grapple. By using

_ . _ _ _

.. - _ _ .

,

s .

binoculars and a remote camera, it was determined that the center of the fuel assembly handle was bowed downward about 1/2 inc (The bend was not. apparent looking straight down at'the handle.)'

On August 5, the assembly was removed using a specially fabricated hook on the fuel grappl Each fuel' assembly.is seated in a shroud can, but the shroud can locking ring for Fuel Assembly 5-04 was found sitting on top of the locking pins, rather than engaging the pins. Consequently, Fuel Assembly 5-04 protruded approximately 1/2 inch higher than norma After finding ~this error,.all,the locking rings were inspected, with two being found partially unlocked, but fully. seated. These locking rings were subsequently locked in place. The licensee has concluded that the protruding fuel assembly's handle obstructed the seating of the HPCS bundle.during the refueling outage, and that a HPCS nozzle eventually. bent the fuel assembly's handle, thereby allowing the HPCS bundle to sea As a result of this incider.t, personnel involved in locking ring verification.are ~being better trained in what to look fce, and a better camera monitor was installed on the refueling floor. The new camera allows personnel handling the camera to clearly see the picture-being taped. In addition, an independent verification of the locking ring. position for each shroud can that has been replaced during a refueling outage will be conducte Fuel Assembly 5-04 was visually inspected and no damage to the fuel pins were noted. The nozzle above position H-4 was also visually inspected and no damage was noted. An air test was performed which verified flow through the nozzle. ' Exxon, the fuel manufacturer, performed an evaluation on the effects of this event on Fuel

. Assembly 5-04, and did not identify any significant concerns regarding its continued use. Rather, Exxon recommended that the licensee consider replacing the upper tie plate at the next

. refueling outage to reduce the risk-of its bail failing at a higher exposur The actions taken by the licensee have been reviewed and found to be acceptable. This LER, therefore, is close (Closed)LER86-016: . lA Static Inverter Failures. On May 10, 1986, with the reactor in hot shutdown, the 1A Static Inverter failed and transferred the 1A Noninterruptible Bus to its alternate source.

3 During the transfer, Reactor Water Level Safety Channel No. 2 tripped i which caused both high pressure core spray pumps. to start, both

emergency diesel generators to start, and the containment building to isolate. The inverter was replaced with a spare.

t On May 13, the reactor scrammed from low power when the 1A Static Inverter again failed and the 1A Noninterruptible Bus transferred i to its alternate source. Again, as in the previous case, both HPCS

pumps and the emergency diesel generators started, the containment building isolated, and, additionally, the shutdown condenser was

_- .

. .

activated. The IC Static Inverter shifted to its alternate sourc The fuse which supplies the emergency phone circuits from the 1C Noninterruptible Bus was found blown. The 1A Static Inverter failure was due to an overheated resistor on a circuit board installed by the manufacturer in 1977. Troubleshooting determined that interaction between the 1A and 1C Noninterruptible Buses was associated with the recent change of the power supply for the

-

emergency phone circuit from the Turbine-Building Motor Control Center 1A to the Bus. Consequently ,the emergency phone circuit'

was rewired to its original power suppl The immediate corrective actions taken by the licensee were reviewed and found acceptable. In addition, the ultimate corrective actions-taken in response to a subsequent failure of this inverter should preclude similar events in the futur (See Item 4. e., below, addressing LER 86-021). This LER, therefore, is closed.

, (0 pen) LER 86-018: Reactor Scram due to Relay Malfunction /Line Separation at Reactor Water Level Transmitter. On June 22, 1986, the reactor scrammed because an erroneous signal indicated that the main steam isolation valve (MSIV) was not fully open. In addition, the shutdown condenser (SDC) automatically initiated'at the time of the scram. Subsequent investigation revealed that the scram and activation of the SDC resulted from a failed rela Following the scram the SDC reduced pressure such that the MSIV shut' automatically when pressure reached about 1000 pound About two minutes after the scram, severe level oscillations were noted on Safety Channels Nos. 1 and Safety Channel No. 3 probably exhibited the same oscillations, but since its indicator is'on a different panel, its oscillations were not observed. When the oscillations on Safety Channel No. 2 reached the low level setpoint, both emergency diesel generators and the core spray pumps started. Upon verifying levels, the operators noted that the feedwater level control recorder and the reactor vessel wide range level indicator were off scale high. The operators allowed the core spray pumps to add water to the reactor vessel until an increasing water level was indicated on the safety channels and the water level indicated high in the operating ban Investigation inside containment revealed that a fitting on the reference leg of a feedwater level control transmitter had parted and was leaking. The transmitter was isolated and the leak stoppe When the transmitter was isolated, Safety Channel No. I spiked low momentarily, causing the emergency diesel generators and the core spray pumps to start. The reactor vessel's wide range level indicator came on scale (from high off scale) and indicated that the level was a few inches above the top of the core. The core

'

.

spray pcmps were operated until a level increase was noted and the indicated level on the three safety channels was high in the operating band. The core spray pumps ano diesel generators were then secure .

- -- - ,. , , , - - - - - - - em --m..-, ,.

. .

-Recalibration of the reactor vessel's wide range level transmitter and the three safety channels indicated an anomaly. Namely, the zero shift on the reactor vessel's wide range transmitter caused the indicators to read lower than the actual level, but the zero shift on the three safety channels caused the indicators to read 3 to 12 inches higher than the actual level. The manufacturer of the instruments was contacted and stated that these zero shifts are to be expected following a pressure transient such as occurred when the fitting parte The inspector verified by review of graphs, calibration data, and discussions with the operators that actual water level did not drop below the technical specification limit. The inspector also evaluated the indications received during the transient and determined that the indications were as expected given the transient that occurred, the configuration of the water level transmitters, and the unexpected zero shift of the transmitter Following a reactor shutdown on July 9, the five reactor water level transmitters were recalibrated. Safety Channels 1 and 3 had been recalibrated subsequent to the post-transient calibration and required no adjustment. Neither did the wide range transmitte The zero on the transmitter for Safety Channel 2 was adjusted less than 2 inches. The zero on the narrow range recorder / reactor water level controller-transmitter was adjusted by almost 5 inches. The output of both these transmitters was found to be low. The four transmitters which were re-adjusted had shifted back, rather than shifting an additional amount higher. The net zero shifts were approximately: Safety Channel 1, less than an inch; Channel 2, 1-1/2 inches; Channel 3, 9-1/2 inches;. narrow range recorder / reactor water level controller, 19 inches; and wide range, 30 inches. No additional shifts are expected since the Channel 1 and 3 transmitters did not require further adjustments on July 9, 198 The safety significance of this incident was reviewed and discussed with NRC personnel. Only one of the safety-related water level channeis experienced a significant zero shift, but all did experience some shift. The zero shift can occur on differential pressure transmitters if they experience a high differential pressure transient. .The licensee has informed the operators of this occurrence to alert them of the possible zero shift following level indicator oscillations which may be indicative of pressure transients in the sensing lines. In addition,-training will be

'

conducted on this incident for the operations departmen In addition to the normal primary sample analysis performed following a power change greater than 15% per hour, a gamma isotopic analysis was performed on a sample collected on June 23, 1986. The results indicated that no fuel degradation occurred as a result of this inciden __ _- - . - _ - ..

. _ - . _ - _ - - _

. .

The actions taken to date by the licensee in resolving NRC concerns are acceptable; however, the proposed training on this event has not yet been conducted. Therefore, this action is being held as an open item (50-409/86009-1) pending completion of training on this even (Closed) LER 86-019: Reactor Scram on High Power / Flow Due to Reactor Feed Pump Malfunction. ~ 0n June 27, while the reactor was at approximately 81% power and escalating, the reactor scrammed on high power / flow. A few minutes prior to the scram, a " Reactor Level Lo" alarm was received. At that time, Reactor Feed Pump 1B (RFP-1B) was operating and the flow contrF tzr was demanding increased flow; however, the flow recorder indicated that flos had started decreasing several minutes earlier. .As a resul.t of the decreased flow, reactor water level was decreasing at a rate of approximately 1.4 in/ min. To prevent a scram, RFP-1A was started and brought up to speed as quickly as possible, (reactor water level was approaching the reactor scram setpoint of 11 inches below normal). RFP-1A started feeding when water level was at -1 inches. Feedwater flow increased rapidly because RFP-1B did not back down automatically. The reactor operator-attempted to' decrease RFP-1A flow; however, because of a lag in pump response to the

controller and because the controller had been demanding increased flow, feedwater flow and water level increased rapidly. The power increase asscciated with the excess cold feedwater being injected into the reactor resulted in a high power / flow scra Following the scram, both RFPs were manually tripped because of the high water leve The power / flow trip signal is based on the relationship between power and recirculation flow. Since the reactor had been operating at 81% power, the forced circulation pumps were at less than full speed. (Recirculation flow was approximately 70% of full flow.)

Therefore, the power / flow trip occurred at a lower power than the high power trip setpoint. If the power / flow signal. scram had not occurred, the reactor would have tripped on high reactor water level. The plant responded to the scram normall Approximately 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> later, Seal Inject Pump-1A tripped. Seal Inject Pump 1-B started , but the forced circulation pumps (FCP)

tripped before sufficient seal injection flow was achieve The seal inject system supplies seal water to the FCP seals and the control rod drive mechanism seals. The FCPs were restarted in approximately 5 minutes. The Seal Inject Pump 1A speed controller had failed. The speed controller was later reworked with the assistance of the manufacturer's representative Each RFP has a constant speed motor with a variable speed couplin The controller adjusts the amount of scoop tube penetration into the coupling and hence the amount of coupling and the pump speed. There is an electrical stop which limits the maximum scoop tube trave Following the scram, the technician observed that the stop had been reached on RFP-1B but it should not have been reached at 81% powe s ...

The stop was adjusted up, and the coritroller motor was also changed because some arcing had been note On July 1, it was determined that the scoop tube linkage to the controller was misadjusted, causing.RFP-1B's zero reference and minimum _ speed to be set low. Since the upper _stop was set at an amount above minimum scoop-tube travel (minimum speed), the maximum travel (speed) was also set too low. In brief, the improper setting _ of the upper stop was responsible for the decrease in feedwater flow on June 27, The actions taken by the licensee have been reviewed and found acceptable, therefore this LER is close f. _(Closed) LER 86-020: Reactor Scram due to Failed Scram Solenoid on Control Rod Drive No 12. On July 8, 1986, the reactor scrammed when one of the two scram solenoids on Control Rod Drive No. 12 failed open. Opening of the solenoid caused the insertion of Control Rod No. 12 which, in turn, re'sulted in low gas pressure for the affected control rod drive mechanism. As a result of the low gas pressure on the control rod drive mechanism for Control Rod.No.12, a partial scram signal was generate The first out alarm for~this incident was " Control Rod Accumulator Gas Pressure Low". Low gas. pressure in any control rod drive mechanism's accumulator causes a partial scram. A partial scram inserts the center 13 control rods, rendering the reactor subcritical. The IB Static Inverter Trouble alarm also annunciated at the time of the scram. Approximately 4 minutes after the partial scram, while the reactor operator was manually inserting the remaining control rods, a full scram occurred when reactor water level reached the high level scram setpoin While the IB Static Inverter was supplying the load, the " Inverter in Process of Synchronizing" light was on. At 2002, the IB Static Inverter trouble alarm cleared, and.at 2038, the operator tried to reset the scram condition. When the pushbutton was reset, the IB Static Inverter amperage decreased and the trouble alarm-re-annunciated, but the scram-condition did not reset. At 2102, the trouble alarm cleared, after which a fuse in'the circuit which supplies power to some of the control rod drive mechanism scram solenoids was found blown. Power to this circuit is provided by the IB Non-interruptible Bus, which is normally fed by the 18 Static Inverter. At 2126, the blown fuse was replaced and the reactor operator attempted to reset the scram. When he pushed the reset button, the 18 Static Inverter's amperage increased, the same fuse blew again and the inverter's amperage returned to its previous level. Troubleshooting disclosed a short in one of the scram solenoids in Control Rod Drive Mechanism No. 1 .

The lower control rod drive machanism in position 12 was replaced with the spare. The scram solenoid in the mechanism which had been in position 12 was replaced and the mechanism was bench teste The licensee could not determine whether the solenoid failure was random or due to water intrusion from the flange on rod 12 which was leaking price to the scra The actions taken by the licensee on this event have been reviewed and found acceptable. This LER, therefore, is close (Closed) LER 86-021: Reactor Scram Due to 1A Static Inverter Transfer to Alternate Source With Fast Cooldown. In brief, this transient was very convoluted, es evidenced by the number of safety-related systems that were activated. For example, the reactor scrammed when the 1A Static Inverter transferred to its alternate source, momentarily de-energizing one of the scram train After the inverter was checked, the 1A Non-interruptible Bus was transferred bcck to the inverter. Later, the inverter load transferred again and two fuses blew, de-energizing some instrumentation. The high pressure core spray (HPCS) pumps started, the 1A Shutdown Condenser (SDC) train initiated and the containment building isolated. The HPCS pumps' control switches were placed in

" PULLOUT" to prevent the pumps from running. The 1A SDC Lteam Inlet Valve was manually isolated. The fuses were replaced and equipment returned to normal. In addition, while the SDC was in service, the reactor vessel cooled down at a rate in excess of the technical specification limit. A sequential description of the more salient aspects of this event follow At 1810 on July 16, while the reactor was operating at 97.5% power, Alarm F2-9, "Hi Pressure-Lo Level ATWS Signal" annunciated and cleared immediately. This alarm actuates when certain contacts in the Anticipated Transient Without a Scram (ATWS) circuit are close The alarm is used for ATWS circuit testing. A different alarm annunciates if the complete ATWS logic is met and both forced circulation pumps are tripped automaticall At 2321 and 2326 on July 16, Alarm F2-9 again annunciated and cleared immediately. At 2328, Alarm F2-9 again annunciated and cleared and the reactor scrammed. The first-out scram alarm was

"All Rod Scram". The "1A Static Inverter Low Voltage" alarm was also received. Investigation disclosed that the 1A Non-interruptible Bus had transferred from its normal source (the 1A Static Inverter)

to its alternate source (the Turbine Building Regulated Bus). The licensee, therefore, has surmised that the scram was caused by poor transfer action by the inverter. This assumption is based on the fact that the 1A Static Inverter is not synchronized with its alternate source, and thus, the break-before-make transfer from the inverter to the alternate source is not always bumples Consequently, since one of the scram trains is supplied by the 1A Non-interruptible Bus, the reactor will scram if power is momentarily interrupted during a transfe .

' ,

The standby seal inject pump started following 'the scram and both forced circulation pumps tripped. It is believed that the forced circulation pumps tripped and the standby seal inject pump started because of a temporary loss of seal inject flow during the transfer from station power to offsite power. Both forced circulation pumps were restarted within 15 minute An electrician checked the 1A Static Inverter and indicated that it appeared to be in satisfactory condition. At 0148 on July 17, the emergency diesel generator's (EDG) shorting plugs were removed and Reactor Water Level Channel No. 2 and the High Pressure Core Spray Pumps (HPCS) were placed in " PULLOUT" prior to transferring the 1A Non-interruptible Bus back to the 1A Static Inverter. This action was taken because the transfer from the alternate source to the inverter usually causes perturbations on the instrumentation supplied by the 1A Non-interruptible Bus, which, in turn, may cause the EDG's to start. Removing the EDG's shorting plugs had no adverse safety implications in this instance because they affect only the low water level start, not the undervoltage start. The 1A Non-interruptible Bus was transferred back to the 1A Static Inverter without problems, and the HPCS pumps' control switches were returned to "AUT0".

At 0240, Alarm F2-9 annunciated and cleared, and returned momentarily at 0245, 0247, 0329, 0430 and 0440. The inverter and ATWS relays were monitored to see if any correlation existed between the alarms and the inverte At 0515, the 1A Static Inverter transferred to its alternate sourc Both HPCS pumps started, the containment building isolated and the 1A SDC steam inlet and condensate valves opened. The HPCS pumps were placed in " PULLOUT" to prevent injection of excess water into the primary system, and were run manually to maintain normal water level. This action was taken because Reactor Water Level. Safety Channel No. 2 had previously failed downscale and caused the HPCS pumps to start and containment to isolat The drawer for Reactor Pressure Safety Channel No. I had lost power, causing 1A SDC train initiation. The SDC could not be removed from service from the control room; therefore, the 1A SDC steam inlet manual isolation vahe was closed at 0534. At 0538, two fuses in the output of the 1A Static Inverter were found to be blown. As a result of the blown fuses, the following instruments lost power:

Reactor Water Level Safety Channel 2; Reactor Pressure Safety Channel 1; Nuclear Instrumentation Intermediate Range Channel 3; Nuclear Instrumentation Source Range Channel 1; ATWS Channel 1; 1 of 2 channels of Containment Building Pressure; and 1 of 2 channels of Containment Building Level. The fuses were replaced at 0545, and the 1A SDC steam inlet and condensate building isolation valves then closed. The HPCS pumps were returned to "AUT0", the containment building isolation valves which are normally open during plant shutdown were reopened, and the EDG shorting plugs were replace At 0705, the 1A SDC steam inlet manual isolation valve was returned to its normal locked open positio _ . .__ _

s ,

The inverter was replaced with a unit that had recently been rebuilt by the manufacturer. The unit removed from service was

' bench tested and found to have a. sluggish mechanical transfer relay, and a failure to transfer on a loss of DC input power. In addition, arcing was noticed at~one of the relay contacts. A load simulating actual operating conditions was applied to the inverter. A cycling load of 1 ampere, simulating the operation of the ATWS alarm relay, and a steady static load of approximately 4.5 amps were applie The inverter transferred to the alternate source only about~30% of the tim Because of the inadvertent activation of the shutdown condenser previously described, the reactor vessel was cooled at a rate exceeding the technical specification limit of 150F/ hour. The maximum vessel cooldown rate experienced during this event was 264F/ hour. This cooldown rate-involved a 55F drop in temperature from 540 to 485F in approximately 12.5 minutes, with the maximum temperature change in a one hour time period being approximately 100F. The licensee conducted a detailed analysis of the effects of this cooldown rate and verified that the resultant vessel stresses did not exceed allowable limit The licensee had planned to replace the 1A Static Inverter with a state-of-the-art unit having a synchronizing circuit to assure a bumpless transfer during its next refueling outage. However, as indicated in the review of LER 86-016 (Item 4.c., above), the licensee took advantage of the outage that followed this event to procure and install a new 1A Static Inverte ~

In addition, the actions taken-by the licensee regarding the cooldown rate of the reactor during this event were reviewed and found to be acceptabl This LER, therefore, is close (0 pen)LER86-022: Decay Heat Line Cracks. On July 17, while the reactor was shutdown because of the July 16th inverter failure (LER 86-021), water was' detected leaking from the decay heat suction line. .On the~18th of July, insulation around the leaking line was

. removed and weepage was observed from two smal1 cracks The cracks were adjacent to a purge port plug that was installed during the 1970 safe end replacement project. After detecting these cracks, other purge port plugs on the decay heat line were ultrasonically examined as were four butt welds and a 12 inch band of the line in the vicinity of the cracks. Except for the two cracks, no indications were observed. In addition to these examinations, the licensee had ultrasonically examined 19 welds in the decay heat discharge line during the-1986 refueling outage with acceptable results.

'

The decay heat suction line containing the cracks is an 8-inch, 304 stainless steel line. This section of the line was original LACBWR pipe manufactured by Capitol Pipe Company which could not be isolated from the reactor vessel. The reactor vessel was therefore defueled and-drained, after which the affected pipe section was replaced and the cracks examined in detai s .

The two small cracks in the decay heat suction line were approximately 1/8 inch long each on the pipe surface. Ultrasonic examination showed the cracks to be approximately 1/2 inch long at

'the pipe inner surface. The cracks were circumferential, one on each side of a a slightly misaligned purge port plug that was installed when the safe end at the decay heat to the forced circulation pumps' line nozzle was replace Southwest Research Institute performed the ultrasonic testing (UT)

and the licensee's Quality Assurance / Quality Control Department performed the dye penetrant testing (PT) of the following area The other purge port on the decay heat suction line was UT'd and PT'd. Two purge port plugs similar to the one rear the cracks were

.UT'd and PT'd on the decay heat discharge line. Four butt welds in the vicinity of the cracks were UT'd, the dissimilar weld was also PT'd. The decay heat line was UT'd all the way around the pipe for six inches up and downstream of the purge port plug (12 inch total band). Except for the two cracks, no indications were observed during these examinations of the discharge lin The affected pipe section was replaced using the guidelines of NUREG 0313 " Technical Report on Material Selection and Processing Guidelines for BWR Coolant Pressure Boundary Piping," Revisions 1 and 2, for partial replacement. (Welding was performed by the " Heat Sink" method.) Observation of the replaced pipe showed that the cracks on the inner surface extended approximately 3/4 inch from the base of the 1/2 inch purge port hol The licensee's actions on this event have been closely coordinated with NRR and Region III and found acceptable, with no stops having been placed on the actions taken by the licensee. In addition, a section of the piping near the cracks has been set aside for examination by NRR consultants. Nevertneless, this event is being held as an open item (50-409/86009-2), pending final evaluation by NR (Closed) LER 86-023: Loss of Offsite Power Due to Lightning While Shutdown. On July 19, at 0630 while the LACBWR facility was in a cold shutdown mode of operation and a severe thunderstorm was in progress, offsite power was lost. The loss of offsite power was caused by a lightning strike at the adjacent coal plant's auxiliary switchyard, causing the switchyard's breaker and LACBWR offsite power supply breakers to open. Both emergency diesel generators started and supplied power to the essential 480 volt buses. In addition, the containment buildir.g isolated and 1A High Pressure Service Water Diesel Pump started when the high pressure service water system pressure dropped to the low pressure setpoin At 0642, offsite power was restored. Shortly thereafter, the electrical system was aligned such that station power was being supplied from the offsite source. At 0659, the 1A High Pressure Service Water Diesel Pump was secured and returned to AUT0. At

i ..

0702, the emer'gency diesel generators were secured and returned to AUTO. All safety systems operated as designed during this even The actions taken by the licensee to restore offsite power were timely and in accordance with procedure This event occurred early on a Saturday morning while the plant was in the cold shutdown mode. Thus, a normal complement of day shift staff personnel was not present. Since the plant was in a cold shutdown mode of operation, the shift supervisor erroneously believed that the States of Wisconsin and Minnesota, and the Counties of Vernon, Wisconsin and Houston, Minnesota need not be notified of this event. On the following Monday, when the normal complement of day shift staff personnel was present, the notification error was detected and actions were taken to rectify the mistake, including notification of the event to the required partie The notification deficiency notwithstanding, this LER is being closed in this report to preclude duplication of effort. (the failure to notify all the required parties has been cited as a i Severity Level IV violation in Inspection Report No. 50-409/86008 and is being tracked as Open Item No. 409/86008-01 in said report).

J. (Closed) LER-024: Initiation of Safety Systems Due to Shorting Out Solenoid While Deconning. On July 19, while the plant was in cold shutdown, water that was being used for decontamination splashed off a pipe surface and onto a solenoid valve.. The water shorted out the solenoid and caused a fuse to blow. As a result of the shorted solenoid and blown fuse, the 1A High Pressure Core Spray Pump and the 1A Alternate Core Spray /High Pressure Service Water Diesel Pump started, and the containment building isolate All safety-related systems responded as designed for this malfunction. Corrective actions included securing the pumps, replacing the blown fuse, and replacing the shorted solenoid, thereby restoring the vent header to its normal ventilation exhaust pathway. Information on this event and its effects has been disseminated to all plant management and operations personne The response and corrective actions taken by the licensee for this event have been reviewed and found to be acceptable and timel This LER, therefore, is close k. (Closed) LER 86-025: Containment Building Isolation Due to Weldin At 1045, on August 21, while the plant was in a maintenance outage with the reactor vessel defueled and drained, the containment building (CB) gaseous activity monitor alarmed and caused the CB ventilation to isolate. At the time of the alarm and isolation, gas tungsten arc welding (GTAW) was being performed on the decay heat suction lin Following the alarm, the personnel in the 1A Forced Circulation Pump cubicle were instructed to evacuate. Prior to evacuating they secured the welding machine, which they had started

- - _ - -. . -

. - _ - _ -- .- . _ .

- -

.

> - 1 shortly before. When the monitor indication returned to normal, the workers were asked to turn the argon purge gas back on. -When they turned on the welder, the gas activity monitor again alarme It was.therefore concluded that.the high frequency from the GTAW welding machine or the argon gas was causing the CB gas-activity monitor to alar The alarm setpoint was increased and the containment building's

- ventilation dampers opened. Indications on other activity monitors were normal. The welding was completed and the gas monitor returned to operable, status at approximately 130 Containment building ventilation isolation has occurred before during welding; however, on this occasion it was considered unexpected by the crew on watch. This event, therefore, was reported to the NRC via the ENS. The LACBWR containment is normally continuously ventilated to maintain personnel accessibility and to prevent pressure buildup, but short periods of isolation have negligible effects. Since this event was terminated in a timely manner, and since the initiating event was not radiological, per se, it is close . IE Bulletins For the IE Bulletins listed below the inspector verified that th bulletin was received by licensee management and reviewed for its applicability to the facilit If the bulletin _was applicable the inspe'ctor verified that.the written response was within the time period stated in the bulletin, that the written response included the information required to be reported, that the written response included adequate corrective action commitments based on information presented in the bulletin ~and the licensee's response, that the licensee management forwarded copies' of the written response to the appropriate onsite management representatives, that information discussed in the licensee's written response was accurate, and that corrective action taken by the licensee was as described in the written respons (Closed) IEB 409/84-02, " Failure of General Electric Type HFA Relays in Use in Class IE Safety Systems." By letter dated July 17, 1984,

!

the licensee indicated that six such relays are used in Class IE l

'

safety systems. Because test circuit logic is not provided for these relays, they cannot be functionally tested during operation.

i The relays, however, have been inspected on a periodic basis since 1976 with no deterioration or malfunctions having ever been note Further, the licensee ordered new relays from General Electric i Company and installed them during the 1986 refueling outage. This l Bulletin, therefore, is closed.

i (Closed)IEB 409/86-02, " Static "0" Ring Differential Pressure i_ Switches." By letter dated July 24, 1986, the licensee informed the.

! NRC that the subject differential pressure switches are not used at l LACBWR in safety-related systems or in systems important to safety.

l This bulletin, therefore, is closed.

l'

.

a- Plant Trips In addition to the five scrams described in the section of this report entitled,'" Licensee Events Reports Followup", four other scrams occurred during the period covered by this report. The more salient aspects of these-scrams are: (1) three of the' scrams occurred during reactor startup and one while~the plant was operating at approximately 73% power; (ii)

two of the startu scrams were effected by the. plant's one-out-of-two scram logic; (iii two of the scrams were the direct result of equipment malfunctions; (iv one scram was the direct result of an operator error; and (v) one scram can be attributed to an operator error that was partly induced by equipment malfunction. Subsequent to these scrams, the

.

inspector _ reviewed the corrective actions taken by the licensee, and ascertained the status.of the reactor and safety systems by observation of control room indicators and discussions with licensee personnel regarding plant parameters, emergency system status and reactor coolant chemistry. Additional details on these four scrams follo At 1450 on September 10, shortly after the reactor had been taken critical, the reactor operator upscaled the range selector switch for Nuclear Instrumentation (NI) Channel No. 5 from the 60 E-5 to the 150 E-5 power scale. Upon upscaling the range switch, a flux spike of sufficient magnitude occurred on NI Channel.No. 5 to cause a scram (Note:

-0n startup, the trip logic is one-out-of-two such that a scram will occur upon a high flux signal on either NI Channel No. 5 or No. 6, but at power levels 'above 15%, the high flux scram logic is two-out-of-four such that

- a high flux spike on any one channel will not result in a scram.) The short-term troubleshooting and corrective actions determined that the balance screw on the potentiometer on NI Channel No. 5 was loose, and included the cleaning of the range switches for NI Channels Nos. 5 and 6 and replacing the NI Channel No. 5 drawer with a spare. The long term corrective actions include the replacement of the NI system with a new system that is scheduled for' delivery to the site in time for installation during the 1987 refueling outage. The new instrumentation system eliminates the need for manual upscaling, thereby reducing, if not eliminating, the likelihood of similar inadvertent scrams in the future, including the scram described belo At 1918 on September 10, the reactor was again taken critical. Shortly af terwards, the reactor operator was distracted such that he failed to upscale the range selector switch for NI Channel No. 6 in a timely manner. As a result, the reactor scramed when the high flux trip setpoint was reached on NI Channel No. 6. The scenario leading to this event follows. Just prior to the scram, while the reactor operator was watching for a one-half decade overlap between the source range and the intermediate and wide range NI channels, he noticed that the count rate indication on source range NI Channel No. 2 started decreasing and that the period meter was reading less than infinity. While he was checking

.the other instrumentation, the wide range channels (NI Channels Nos. 5 and 6) were starting to respond. The source range scaler, a digital readout of the counts per a selected time period, had been selected to NI Channel 2, the channel providing suspect data; therefore, the operator

- - - - . - - - .-.- - - - - - - - . .

.

transferred the scaler to NI Channel No. 1. When he transferred the scaler to NI Channel No. 1, a short period warning alarm annunciated on NI Channel No. 2. The operator acknowledged the alarm and it cleared; however, immediately afterwards, a high flux scram occurred on NI Channel No. 6. Subsequent troubleshooting on NI Channel No. 2 drawer could not duplicate the malfunction; however, since the pins on the channel output tube were found to be undersized, it has been speculated that the undersized pins could have caused intermittent losses of outpu During a reactor startup_on September 13, the turbine stop valve was erroneously opened later than required by procedure. (Reactorsteam pressure was at 800 psig when the stop valve was opened.) The design of the steam control system is such that when the turbine stop valve is opened, the reheater inlet valve also opens. Consequently, the delayed opening of the turbine stop valve caused the rapid filling of the reheater, causing reactor pressure to decrease. The pressure drop resulted in a swell which, in turn, resulted in a high water level scra During this transient, the water level in the reactor went through shrink and swell oscillations before stabilizing, such that the high pressure core spray pumps and emergency diesel generators started, and containment isolated on a low water level signa The last of the four scrams occurred on September 18, while the plant was operating at approximately 73% power. This scram was caused by a malfunctioning control system for the seal injection system of the 1B Forced Circulation Pump. The malfunction, apparently due to debris in the control air system, caused the supply valve for the IB Forced Circulation Pump to close. As a result, since the forced circulation pumps are not operated at rated speed at this power level, the reactor scrammed on' power to flow mismatc These events will be covered in greater detail in the next inspection report when their respective LER's are issue . Open Items Open items are matters which have been discussed with the licensee, which will be reviewed further by the inspector, and which involve some action on the part of the NRC or licensee or both. Two new open items are described in Paragraphs 4.d. and . Exit Interview The inspector met with licensee representatives (denoted in Paragraph 1)

throughout the month and at the conclusion of the inspection and summarized the scope and findings of the inspection activities. The licensee acknowledged the findings as reported herein and did not identify such documents or processes as proprietar