IR 05000313/1990005

Insp Repts 50-313/90-05 & 50-368/90-05 on 900201-0315. Violations Noted.Major Areas Inspected:Previously Identified Items,Followup of Events,Operational Safety Verification,Surveillance & Maint

ML20012F476
Person / Time
Site:	Arkansas Nuclear
Issue date:	03/30/1990
From:	Westerman T NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV)
To:
Shared Package
ML20012F473	List: IR 05000313/1990005 ML20012F472 ML20012F475
References
50-313-90-05, 50-313-90-5, 50-368-90-05, 50-368-90-5, NUDOCS 9004130269
Download: ML20012F476 (23)

Text

y

.

,

.

.

. APPENDIX B i

'

U.$. NUCLEAR REGULATORY COMMIS$10N r

REGION IV

Inspection Report:

50-313/90-05 Licenses:

OPR-51 50-368/90-05 NPF-6 Dockets: 50-313 50-368 Licensee: Arkansas Power & Light Company (AP&L)

i P.O. Box 551 Little Rock, Arkansas 72203 Facility Name: Arkansas Nuclear One (ANO), Units 1 and 2 Inspection At: AND Site, Russellville, Arkansas Inspection Conducted:

February 1 through March 15, 1990 Inspectors:

C. C. Warren, Senior Resident Inspector Project Section A. Division of Reactor Projects R. C. Haag, Resident Inspector Project Section A, Division of Reactor Projects

Approved:

N f, 9-WM T. F. Westerman,~ Chief, Project Section A, Division Date of Reactor Projects Inspection Summary

.

Inspection Conducted February 1 through March 15, 1990 (Report 50-313/90-05:

50-368/90-05)

~

Areas Inspected: Areas examined during the inspection included followup on previously identified items, followup of events, operational safety verification, surveillance, and maintenance.

Results: One apparent violation of NRC requirements was identified involving

.

licensee failure to perform a Technical Specification required surveillance

'

within the required time limit (Section 4.7).

'

strenoths: The licensee's development and use of forced outage work lists to prioritize and plan necessary maintenance appeared to be successful in insuring that high priority items were completed during the most recent forced outages of both units (Section 7.0),

hk[$$j

-

g

.

.

..

.

.

i

L I

,

Weaknesses: The licensee's program for the addition and control of lubricants appears unstructured and not consistent with other material control programs

,

($ection4.6).

The licensee's program to control equipment configuration

during maintenance continues to exhibit breakdowns.

Licensee management

attention to insure personnel and equipment safety during maintenance should.

-

' continue (Section4.8).

[

,

'

.

Observation:

Licensee management's decision to shut down Unit 2 to pursue

.

problems with "D" core protection calculator was conservative and indicative of

,

[.

an operational philosophy which stresses safety and equipment operability as a

i high priority.

i

4

!

1..

.

!

>

,

'

e

-

.

,

.

'

<

'

!

'

DETAILS

,

1.0 Persons Contacted

'N. Carns, Vice President, Nuclear Operations

'T. Baker, Technical Assistant to Plant Manager, Central

'D. Boyd, Nuclear Safety and Licensing Specialist

-

K. Coates, Unit 2 Maintenance Manager R. Eddington, Unit 2 Operations Manager

'E. Ewing, General Manager, Technical Support and Assessment R. Fenech, Unit 2 Plant Manager

'J. Fisicaro, Licensing Manager C. Fite, Supervisor In-House Events Assessment

'L. Humphrey, General Manager, Nuclear Quality J. Jacks, Nuclear Safety and Licensing Specialist

'

R. King Plant Licensing Supervisor J. Kowalewski, Mechanical Engineer R. Lane, Manager. Engineering Standards and Programs D. Mims, System Engineering Superintendent J. Mueller, Unit 1, Maintenance Manager A. Sessoms, Plant Manager, Central

,

'J. Vandergrift. Unit 1 Plant Manager C. Zimmerman, Unit 1 Operations Menager

'W. Perks, Nuclear Operation Standards Manager

• A. Jacobs Technical Surveillance Supervisor

• Present at exit interview.

The NRC inspectors also contacted other plant personnel, including l

operators, engineers, technicians, and administrative personnel.

'

2.0 Plant Status Unit 1 operated at 80 percent power from February 1-28, 1990, with the exception of a power reduction to 24 percent on February 17, 1990, for governor and throttle valve testing. On February 28, 1990, a Notification I

of Unusual Event was declared due to a shutdown required by Technical

[

Specifications. A loss of containment integrity caused the plant shutdown.

Following repairs of a reactor bui Ming cooling unit, the unit returned to l

80 percent power operation on March $. 1990, and remained at that power

!

level through the remainder of the inspection period.

i Unit 2 operated at or near 100 percent power from February 1,1990, until the unit was shut down on March 4, 1990, due to electromagnetic noise that

>

interfered with plant instrumentation and control systems. After

!

replacement of the surge capacitors on the reactor coolant pumps, the unit i

went critical on March 7, 1990. The unit operated at 70 percent power from March 8-13, 1990.

The unit reached 100 percent power on March 14,

,

1990, and maintained that power level through the remainder of the t

inspection period.

• i

!

I

7.p-w

-

~

j

.-

'

'

.

1 3.0 Followtg of Previously Identified Item (Units 1 and 2) (92701)

(0 pen) Itspector F 11owup item (368/8945-03):

Unit 2 dilution event 3,1

The licensee experienced an unplanned moderator dilution of the RCS when a flow cont.rol valve in a supply itne from the demineralized water system failed to fully reclose after a planned dilution activity and allowed continued dilution of the RCS. After reviewing this event, the licensee's initial corrective action was to reemphasize the need for close operation attention when performing dilution activities.

The operator for Flow Control Valve 2CV-4927, which did not fully reclose and started the unplanned dilution event, was recently adjusted to allow

-

better seating of the valve.

Seat leakage was reduced from 1/2 gpm to 60 drops per minute. While the amount of flow through the valve during the unplanned dilution event was much greater than the original seat leakage and involved the valve not reclosing, the decrease in seat leakage will improve the operator's ability to control dilution activity. A similar adjustment of downstream Veive 2CV-4941 was also performed to decrease seat leakage. While the e rall leakage through 2CV-4941 was significantly reduced, the exact leakage could not be measured due to

.

parallel path Check Valve 2CVC-217 and the inability to distinguish

1eakage between the two parallel valves. The check valve has been addad to the forced outape work list for seat leakage repairs.

One aspect of the licensee's corrective action that does not appear to be resolved involves the positioning of downstream Valve 2CV-4941.

Prior to the dilution event, the ge n ral practice was to keep 2CV-4941 open, which was based on the ratitn.a1 that frequent dilution made it impractical to close 2CV-4941 af ter each dilution. The licensee has not made a decision on the positioning of 2CV-4941.

Individual operators are allowed to determine if the valve should be closed or remain open after dilutions.

-.

During recent control room tours, the inspector has observed that 2CV-4941 remains open. While the inspector realizes that various considerations ascociated with closing 2CV-4941 need to be addressed by the licensee, the practice of closing 2CV-4941 would provide a positive means of preventing an unplanned dilution event.

Other areas that the licensee is reviewing for preventing an unplanned moderator dilution event involve:

Modifying the limit switches on Flow Control Valve 2CV-4927 to

-

provide operators with more accurate valve position indication Improvement of the flow indication in the control room to allow

-

detection of low flow t.onditions through 2CV-4927, i.e., 5-10 gpm flow rates Provide simulator training that models the leakage past 2CV-4927

-

to ensure operators are proficient in detecting an unplanned moderator dilution event based on changing RCS parameter

_ _ _...

_..__

-

,

,

'

-

,

.

'

-

,

Due to several licensee action,tems remaining open, the inspector will

-

continue to follow up on this item.

Inspector Followup Item 368/8945-03 remains open.

3.2 [ClosodInspectorFollowupItem(368/8930-01}:

Failure of a high pressure safety injection (EFSI) stop check. valve to reseat

~

During a previous review of this followup item, all aspects of this issue i

had been resolved with the exception of an evaluation of the reverse rotation experienced by the "B" HPSI pun.p and any potential damage to the suction piping due to overpressurization. The licensee has completed

_

their evaluation which concluded that the pump and suction piping did not experience any damage from this event. The inspector noted during the previous review that the condition report (CR) system had a large nanber

,

of late corrc.tive action (CA) items. The inspector expressed his concern

"

L that, with suCi a large backlog, the identification and prioritizing of

_

critical late CA items would be difficult.

Recently, the licensee has been successfui in reducing the backlog of

'

late CA items in the CR system. A historical rev$ew of the CR system identified that the number of late CA items averaged in the 20 percent range, with a 1-month high of 36 percent.

For February 1990, the average of late CA items was 7 percent. The current goal of licensee managen:ent is to have no late CA items by the end of April 1990. The i

inspector will continue to review the licensee s progress in reducing

the backlog of late CA items, however, this item is closed.

3.3 (Closed) Unresolved Item 313/8918-01: Missing caulking in a penetration

-

room seisoic gap During a plant tour the inspector noted sections of missing caulking in a

-

seismic gap which separates the Unit 1 lower south electrical penetration

-

room (LSEPR) from the upper south piping penetration room (USPPR). The

-

inspector questioned the licensee to determine if the joint integrity was required to separate the LSEPR from the USPPR, which is a high energy line break area.

,

The licensee evaluation of this concern included a review of safety-related equipment in the LSEPR and an estimate of the energy flow through the seismic gap. The conclusion was that the environmental qualification of

--

equipment'in the LSERR wovid not be affected by a high energy line break

--

in the USPPR.

In addition, an action item was initiated to periodically 7=

review seismic gap filler material to ensure 'that the gap material is y

properly maintained.

This item is closed.

-

ll" 4.0 Followup of fvents (Units 1 and 2) (93702)

,

s'

g;L 4.1 Unit 2 Emergency Diesel Generator (EDG) Beariqp Inspection

.

While reviewing vibration reading for the "B" EDG, the licensee noted et

-

an upward trend in the axial vibration for the outboard generator bearing.

t.

=

=-

_. _. _..

.

'

.

.

.

The trend involves data taken with a handheld vibration probe during the monthly surveillance runs from October 1989 to January 1990.

In response to the upward trend, a vibration signature and an oil samp'.e of both generator bearings were taken on January 30, 1990.

A comparison of the vibration signature with data taken in October 1989, identified some minor change in vibration amplitude for the outboard bearing, however, the changes were not significant enough to affect EDG operability.

Results from the oil sample identified a high concentration of ferrous particles with a recommendation from the laboratory that the bearing be inspected.

On February 6, 1990, the oil in both generator bearings was changed.

During a subsequent EDG run, vibration signatures and oil samples were taken for the bearings. Again the vibration data was similar to past readings, with the oil sample indicating a high concentration of ferrous particles.

Based on their review of bearing data and the laboratory's analysis, the licensee concluded that the bearings were experiencing higher than normal wear, however, the bearings were fully functional and no immediate action was required.

On February 21, 1990, the covers for both brarings were pulled to allow cleaning and inspection of the bearings.

A large amount of small metal particles and sludge was flushed from the bearings.

The licensee stated that a review of records indicated that this was the first time the bearings had been cleaned since they were installed in 1985.

Visual inspection revealed only moderate wear of the bearings.

After the bearing covers were installed and the oil reservoirs were refilled, the EDGs were run to verify operability.

The oil sample was very dark in color and appeared to be heavily contaminated.

An analysis of the sample identified a high concentration of waar particles in the oil.

The licensee attributed this to an increase in the oil level for the inboard bearing by approximately 5/8 inch, which caused increased agitation of the oil while running the EDG.

This increased oil novement then loo.ened sludge and particles that were not accessible when claaning the bearing.

The licensee also stated that while the cil level was lower than the optimum level, it was suff$cient for adequate oil lubrication.

Subsequently, the licensee has completed two evolutions where the bearing oil was drained and filled six to eight times with the EDGs run afterward to prove operability. Oil samples taken after each run indicated that the particle concentrations have decreased, bLt were still higher than normal levels.

The licensee's current plans are to take vibration signatures of both bearings for all future EDG runs for detection of any bearing

"dation.

Also, the oil for the inboard bearing will be drained and refilled with new oil in an attempt to oecrease oil contamination levels.

The licensee is plenning to continue this pattern of vibration signatures and oil flushes for 3 months, then all the data will be reevaluated to determine it any chance in the monitoring prograni or repairs are needed.

The inspecter will follga up on the licensee's review of individual surveillance data ano on the litersee's eassessment of EDG performance in 3 months.

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

.

p

.

L

.

0 t

I

I

,

i 4.2' Unit 2' Power Increase Transient f

n

.

.

!'

At 5:35 a.m. on February 22, 1990, the control room received an annunciator alarm for high vibration on "A" Main Feedwater Pump. At the same time, the operator noticed a 5 percent step increase in steam generator levels

!

and an increase in feedwater flow.

Reactor power, as indicated by the excore instrumentation, was 103 percent. A 38 megawatt increase in the t

main generator output load indicated to the operator that the increase in i

reactor power was a result of a change in the main turbine / generator load.

'

When operators made a small reduction in the main turbine load demand, the

-

turbine / generator made a step change in load back to the original step point.

Subsequent troubleshooting by the licensee of the power increase I

transient identified that the main generator load limit potentiometer

'

was the source of the load increase signal.

The potentiometer was replaced

'

when the unit was shut down for replacement of the reactor coolant pump motor surge capacitors.

The decrease in generator load back to the original value caused an

.i increase in pressurizer pressure to the actuation setpoint for the i

pressurizer spray valves.

This caused Spray Valve 2CV-4651 to open to

'

allow a reduction in pressurizer pressure. After the high pressure condit1on cleared. ?CV-4651 failed to fully reclose. The operator

recognized that the valve closure light did not actuate and that pressurizer pressure was continuing to decrease. The spray valve-

handswitch was then taken to manual and the valve was closed. The overall

,

reduction in pressure was approximately 50 psi below normal operating L

pressure.

Subsequent testing of 2CV-4651 identified no unusual operational conditions.

During the test, pressurizer pressure was increased by the backup heater until the spray valve open setpoint was reached.

The valve opened in response to the automatic signal and then reclosed after the high pressure

I condition was cleared. A second test of cycling the valve was successfully performed.

The other spray valve, 2CV-4652, was also cycled to verify

'

I proper operation for that portion of the pressurizer pressure control

,

system.

l During the plant shutdown to replace the reactor coolant pump surge capacitor, Spray Valve 2CV-4651 opened again in response to high reactor

'

i.

l coolant system (RCS) pressure. After the high pressure condition cleared,

-

the valve stroked to the closed position, however, it failed to fully l.~

reciose. The observation that the valve had not completely closed was based on RCS pressure continuing to decrease.

The operator then placed L

the spray valve handswitch to manual and tha valve fully closed. After L

review of the two events involving spray valve actuation, the licensee stated that the amount of valve closure differed in the two events.

During the first event, the 40 percent open limit switch was actuated when the operator took manual control, but during the second event the valve stroked closed past the 40 percent limit switch while in the " auto"

'

!

l

.. vow of operation. The licensee informed the inspector that historically l

l l

!

>

F a

.

'

}

a

).

..

.

,

!

the spray valves have experienced periodic problems in fully closing and l

that operators take manual control and " torque" the valves closed by placing the handswitch in the closed position.

)

Several areas of review, such as adjustment of torque switch and limit

!

switch settings and periodic cycling of the. valves, have been identified

by the licensee for possible resolution of the spray valve closing problems.

,

The licensee also stated that, while this issue needs to be resolved, the overall safety concern is reduced by prompt operator action that closed

,

the spray valve. While the inspector agrees that operators have been

.

responsive to the spray valve closures and that operator training

emphasizes the need to ensure that spray valves are closed, the practice

.

of using operator action to compensate for known equipment inadequacies is

,

not desirable. This issue will be tracked as an inspector followup item

.

(368/9005-01) pending review of the licensee's action regarding the l

L pressurizer spray valve closures.

4.3 Leve! Switch Failures Recently the licensee has experienced failures of four level switches for the Unit 1 EDG jacket water coolant expansion tank.

In October 1989,

'

during the refueling outage, both level switches (one per expansion tank)

were replaced due to calibration difficulties and end-of-life i

considerations.

Following replacement of the level switches, numerous DC ground alarms were received.

Investigation of the alarms identified failure of the "B" EDG 1evel switch on December 29, 1989, and failure of

'

the "A" EDG level switch on January 29, 1990.

Both level switches were l

replaced with subsequent failures occurring on February 16 and 26, % C.

L In response to the failed level switches, the licensee initiated a surveillance log to verify correct level in the EDG expansion tanks ev n 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />.

'

l Inspections by the licensee and an independent laboratory of the failed level switches revealed accelerated corrosion of a soldered joint as the

'

l-cause of the failure. A " soft" solder was used in this joint. A review

L of the chemistry control program of the EDG jacket water system identified

!

i that soft solder will experience accelerated corrosion when in contact I

with chemicals used in the treatment of the jacket water. The licensee stated that no other components in the EDG jacket water system contain

" soft" solder.

All of the replacement level switches were supplied by the EDG vendor, Colt-Fairbank Morris, for use in the specific application of the jacket water expansion tank.

Through communications with the licensee, the vendor maintains that the new level switches are authorized replacement

'

for the original level switches and that they are suitable for installation in the expansion tank, Currently, the licensee is pursuing the procurement and dedication of an alternate limit switch.

The licensee also stated

.

!

that they are continuing discussions with the EDG vendor to determine if any manufacturing differences exist between the replacement limit switches and the original limit switches. This issue will be tracked as an s

-

=

-

.

,

.

I

inspector followup item (313/9005-01) pending the licensee's actions regarding replacement of the level switch for the EDG jacket water expansion tanks.

4.4 Inoperable Floor Drain Check Valves While reviewing the flood protection status for saf6ty-related equipment rooms, the licensee identified discrepancies in the Unit 2 emergency feedwater (EFW) pump rooms. The floor drain check valve in the turbine driven pump (2P-7A) room was missing while the check valve in the motor driven pump (2P-78) room was inoperable due to a deteriorated condition caused by an apparent lack of maintenance. Based on the condition of the check valves, the EFW pump rooms would be susceptible to flooding caused by flooding in the turbine building or the adjacent EFW room.

The licensee's immediate action was to install a plug above the check valve in the 2P-7B room to remove the source of external flooding for that room and ensure the operability of 2P-76. 'At the same time, 2P-7A was declared inoperable due to inadequate flood protection.

Further review by the licensee identified an additional open ended drain line in the 2P-7A room that was connected to the turbine building drain system.

This drain line was plugged and an operable floor drain check valve was installed in the 2P-7A room which allowed 2P-7A to be returned to operable status.

The cause of discrepancies associated with the EFW pump room floor drain check valve was a lack of preventive maintenance (PM).

The licensee's current PM program does not include any floor drain check valves. The Unit 2 EDG rooms were the only other safety-related locations that were identified as having check valves in the floor drains. These valves utilize an in-line swing check design while the EFW pump room drain valves are a ball float design.

In addition, the elevation of the EDG rooms precludes the concern of external flooding into the EDG rooms. The remaining concern of an inoperable check valve involves the consequence of internal flooding between the two EDG rooms. Action has been initiated to include the EFW and EDG room floor drain check valves in the PM program. The licensee informed the inspector that sufficient dissimilarities exist between the EFW and EDG room check valve to warrant no immediate action on the EDG room check valves.

4.5 Leak in Unit 1 Containment Cooling coil On February 20, 1990, during the performance of reactor building (RB)

cooler operability test, Procedure 1104.33, the licensee noted that the service water (SW) return valve from the loop one coolers, CV-3814, failed to open on the first attempt. At the time of the failure, the shift supervisor declared the affected train of the RB coolers inoperable and entered the appropriate Technical Specification action statement. After opening the motor operated inlet valve (MOV), CV-3812, a second attempt to open Air Operated Valve (A0V) CV-3814 was successful.

CV-3814 was successfully cycled three times without incident and declared operable, I

!

e*.

.

,

however, the licensee documented the malfunction with CR 1-90-61 and began

,

efforts to determine the cause of the initial valve failure.

!

It was initially postulated that a hydraulic hek between Valves CV-3812

-

,

and CV-3814 was preventing CV-3814, the A0V, f 'm opening. A hydraulic i

lock between the two valves could be set up if both valves were leaktight and the cool SW (approximately 45'F) between the closed isolation valves

,

was heated to reactor building ambient temperature.

On February 21, 1990,

'

while investigating CR-1-90-061, under Job Request (JR) 788458, a temporgry

'

pressure gage was installed between CV-3812 and CV-3814.

Upon placing the gage in service, licensee personnel noted a reading of 259 psig. With

'

system pressure at 259 psig, the operations staff attempted to open CV-3814, but the valve would not open. After reducing system pressure by opening CV-3812, the operators were able to cycle CV-3814 successfully.

,

Although the hydraulic lock on the system addressed the earlier failure of

CV-3814 to open, it raised additional questions of system operability j

based on potential overstressing of components in the containment cooler loops. The licensee conducted walkdowns of the containment cooler loops to inspect for signs of leakage or system damage, and none were found, t

During conference calls between Region IV, NRR, and the licensee, the

potential overstressing of components was discussed, including the possibility of cyclic fatigue of the system.

The licensee performed calculations using assumed worst case conditions which indicate the system t

to be well within the allowable stresses for the system piping and components.

In addition to the pressure stresses, the licensee concluded that less than 450 thermal cycles had been induced in the system. Since

,

the B31.1 piping code requires no correction in the piping allowable stress range for less than 7000 cycles, the number of actual cycles was of

no consequence, f

In parallel with the engineering review of allowable stresses, the licensee I

conducted an evaluation, for both units, of equipment configuration where

thermal expansion of trapped fluids could cause overpressurization.

• Although other heat exchanges with no thermal relief capability were t

identified, no further examples, where the inlet and outlet valves are isolated, were identified.

It should be recognized that the response to this issue by the Unit 2 operations staff was timely and aggressive.

By quickly recognizing the potential problems they were able to quickly dispel any concern that similar conditions existed on Unit 2.

To preclude the overpressurization of the containment coolers in the future, the licensee revisod the SW valve lineup to have the inlet valves, l

CV-3812 and CV-3813, norme11y open.

On February 17, 1990, after CV-3812 and CV-3813 had been opened, the licensee noted an increase in the RB sump fill rate.

Because the increase in sump fill rate was coincident with the opening of CV-3812 and CV-3813, a containment sump sample was drawn. Analysis of the sample revealed that some of the leakage was SW.

A walkdown of the Loop 2 containment coolers l

l.

'

,

a

-

[

i v

i l

'

iL,

o l

'

l

i revealed a leak on one of eight coils for Cooler VCC-2D.

The leak was at the coil to piping interface nipple, and was approximately 0.1 gpm.

.

Bscause the SW loop inside containment serves as one containment boundary

'

i under the criteria in General Design Criteria 57 (GDC-57), the licensee

closed CV-3813 and the SW return insolation Valve CV-3815 to ensure

'

containment integrity. With both valves closed, the leakage from VCC-20

!

continued which indicated that CV-3815 was not leaktight. With a leak

.

path through the containment cooler into the SW loop and through leaking

!

.

Valve CV-3815, a leak path to the environment existed and containment

integrity could not be demonstrated.

Loss of containment integrity required the licensee to enter Technical Specification Action i

Statement 3.6.1, which required the licensee to place the facility in hot

'

standby. The licensee declared an unusual event, shutdown required by'

l Technical Specification, and began a shutdown at 8:15 p.m. (CST), on February 28, 1990. Without RB integrity the licensee was required to

.:

place the unit in hot shutdown within 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> and cold shutdown within 36-

!

hours. To prevent having to cool down the unit, the licensee processed a temporary modification to replace a flow instrument orifice in the SW supply line with a blirid flange. After leak testing the flange to 60 psig,

,

the licensee determined that containment integrity had been restored and

!

exited the Technical Specification action statement.

'

To return the Loop 2 containment coolers to operable status, the licensee

';

decided to isolate the leaking coil in VCC 2D.

The removal of one cooling coil from the cooler was evaluated by engineering and determined to not be an operability issue provided that the required service water flow of 1200 gpm could be maintained through the cooling loop. After completion

'

of the temporary modification, which isolated the leaking coil, the licensee successfully conducted a flow test to insure adequate service

,

water flow. The leaking coil in VCC 20 will be repaired and returned to service prior to the Unit I restart from the next refueling outage, IR9.

Complation of the temporary modification, which removed the leaking coil l

from VCC-20, allowed the licensee to return the Loop 2 cooler lineup to

its standby configuration, however, members of the NRC staff questioned I

the licensee's plans concerning the repair of CV-3815.

In conference calls between NRC and the licensee on March 1 and 2, 1990, the licensee's plan of action for addressing the known leakage through CV-3815 was

,

discussed. Although CV-3815 acts as a containment isolation valve, the

primary boundary for this configuration is the SW system piping inside the

'

l containment.

Under the valve testing requirements of 10 CFR 50, Appendix J, valves in configurations like CV-3815 are not required to be leak rate tested. Although no additional repair or monitoring was required prior to i.

restarting the unit, the licensee performed an analysis of the risk associated with operation of the facility with CV-3815 known to leak.

L The licensee concluded, and the NRC staff concurred, that any additional

'

'

risk was minimal and that operation of the unit until repairs can be completed in 10.9 was, justified. The licensee's evaluation was documented and submittec NRC for information in AP&L Letter ICAN039006, dated March 2.,

199U.

,

r-

'

.

.

j l

.

i

,

w

-

!

,

L

'

i

4.6 Personnel Errors in the Lubrication Program

]

Following the addition of oil into the gearbox of the Sodium Hydroxide

Addition Pump 2P-136B and a surveillance run of the pump, a nonlicensed t

operator identified that the wrong oil was added to the gearbox. The pump was declared inoperabie until the oil could be drained and the correct oil added to the gearbox. The following sequence took place when the initial oil was added:

.

When checking the oil level prior to the surveillance, the operator

'

noticed that the gearbox oil level was low, i

The opterator called the control room to obtain the correct type of oil from the station information management system (SIMS)

database.

,

lhe control room informed the operator to use R&D 100 oil, which is for the pump, in lieu of Chevron 220 oil, which is for the

gearbox.

'

While reviewing this event, the inspector learned that there is no procedure to control the audition of oil to safety-related components.

Procedure 1092.186, " Preparation, Review, and Approval of Component Lube

'

Information Packages," has been used to establish an approved list of i

lubricants for safety-related (Q) components. The inspector reviewed a portion of the list which is located on SIMS The database is grouped by

,

component identification number with the approved lubricants being listed for each component. One aspect'of the database that has caused confusion involves the listing of lubricants for a component that has subassemblies.

For example, Pump 2P-136B is grouped by the motor, pump, and gearbox on SIMS with different-lubricants given for each subassembly.

The inspector noted that personnel from several groups associated with the lubrication program had difficulty obtaining lubrication information for component

subassemblies.

Lubrication information is also located on the materials t

management information system (MMIS). While reviewing 2P-136B lubricants with planning personnel, the inspector noted that the incorrect lubricant is specified for the motor on the MMIS database. The inspector was unable to determine if MMIS information is used in critical applications.

,

.

Approximately 2 weeks later, when attempting to add oil to the gearbox for i

"A" sodium hydroxide pump, the wrong oil was referenced on the job order (J0).

The shift supervisor recognized this and stopped the job i

prior to the addition of the wrong oil.

The addition of oil in 2P-136A involved maintenance personnel working to instructions on a JO that had been prepared by a planner, while the oil addition for 2P-136B only

involved operations personnel. The storage cabinet used by Unit 2 operations personnel for storage of various lubricants was inspected I

during followup of this issue. The labeling on the lubricant storage cans varied and the inspector observed no measures taken to segregate qualified and nonqualified lubricants. Based on the inspector's observations, it

.

appears that the licensee was not complying with plant procedure for the l

storage of lubricants.

N

.

!

.

,

.

7-c

..

'

F

,

The inspu: tor informed l'

+.nanagement of his observations and concerns

'

regarding the fragmen+p vsed for addition o' lubricants for a.

safety-related component.

i on the inspector's comments and the CR

.

written on the addition of

. w g oil in 2P-136B, the licensee initiated a l

review of the oil addition program. The following actions resulted from

>

this review:

The content of the two Unit 2 lubricant storage cabinets were upgraded

,

to comply with material storage requirements. A proposal was made to f

include inspection of the storage cabinets on the weekly surveillance

,

that checks the lubricant / oil level of plant components to ensure the

,

cabinets are properly maintained.

• Include on the SIMS database the lubrication requirement for non-Q components that routinely have the lubricant level checked. This

will allow standardization of the process for adding lubricants to

plant components.

!

Issue a new administrative procedure to define the program for adding lubdcants to components and to provide general instructions for adding lubricants.

• Operation's personnel were informed of the process used to nbtaining

'

correct lubrication information from the SIMS database.

Formal

training would be performed at a later date.

'

f I

Update the weekly surveillance procedure that ched.s lubricant levels to provide greater details on the process and techniques used to-

verify proper lubricant levels.

If the above actions are completed, the inspector's concerns regarding l-the lubricant additions process should be resolved.

The inspector contacted the Unit 1 plant manager and discussed-the need to review the current practices used by Unit 1 personnel to determine if similar actions

'

are' warranted. The need to complete the above actions in a timely manner

'

was discussed at the exit meeting to prevent any future additions of incorrect lubricants.

This issue will be tracked by inspector followup item (313:368/9005-02) pending the inspector's review of action taken to u

improve the lubrication program, t,. 7 Failure to Perform Technical Specification Surveillaace (Unit 2)

On March 11, 1990, at 7 p.m., the licensee recognized that the Technical Specification surveillance requirement to perform a local leak rate test (LLRT) on the containment personnel air lock doors had not been l

completed within 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> of the last door opening, The time of the last door opening was at 9:15 a.m on March 7, 1990. Based on the 72-hour requirerent, the doors should have received a LLRT prior to 8:15 a.m. on March 10, 1990. After the missed surveillance was recognized, the L.

applicable Technical Specification action statement was entered and the LLRT of.,the doors was initiated.

Both doors passed the subsequent LLRT.

L l

u

'

s

.

.

,

.

g

,

g.

.

,

.

I

!

,

Technical Specification Surveillance Requirement 4.6.1.3a states that after each opening, except when the air lock is being used for multiple

,

entries, an LLRT of the door seals shall be performed.

No time limit is given in the Technical Specification for completion of the LLRT.

The licensee has used 10 CFR Part 50, Appendix J, for guidance in establishing

,

a time requirement for completion of the LLRT. Appendix J, Paragraph III.D 2(b)(iii) states, in part, " Air lock opened during periods

when containment integrity is required by the plant's Technical Specifications shall be tested within 3 days after being opened." The

,

inspector agrees with the use of 10 CFR Part 50 as guidance for i

establishing an LLRT time requirement.

'

Failure to perform the LLRT surveillance on the containment air lock door within the required period is an apparent violation (368/9005-03) of Technical Specification 4.6.1.3a.

While the safety significance of this

event is minor due to a successful LLRT of the doors having been completed 4 days earlier, it is another example of the failure to perform Technical Specification surveillance within the required time.

The staff is concerned that, while the licensee has initiated the

corrective action, it has not been successful in preventing additional examples of surveillances performed outside the required interval.

4.8 Errors in Equipment Control During maintenance on the Unit 2 EDG, two errors in equipment tag-out

'

occurred. The first event involved the unplanned start of the EDG while air rolling / rotating the EDG after cle. ring the generator bearing. The intent of the air roll was to rotate the EDG several revolutions to

>

agitate the oil in the generator bearings. An operator who was in the EDG room promptly secured the EDG after the unplanned start.

No equipment damage or personnel injury resulted from this event. The licensee identified that an incorrect tag-out of the EDG caused the unplanned start. The tag-out of the EDG included isolating the DC control power even though there was no electrical work planned. Operators stated that the current tag-out was based on a previous tag-out which included

i~

isolation of the DC control power for electrical work.

'

The second event involved the failure to assure that the EDG was properly tagged out and secured prior to changing the inboard generator bearing oil

,

during a subsequent maintenance activity. While electricians were working in the vicinity of the EDG flywheel to drain and refill the bearing oil, no controls were provided that would have prevented an EDG auto start. A normal tag-out of the EDG, including isolation of the air start system and placing the local handswitch in " Maintenance Lockout," had been planned by operating personael; however, electricians did not notify the control room prior to starting the bearing oil change out.

The licensee has reviewed these events and identified several areas of the equipment control program that needs imnovement. These areas of needed improvement, the short-term correcti m action that had been completed and the proposed long-term corrective actions, were discussed b

-

.

.

,

,

..

t ~

.'

g

,

!

!

t

.

'

with the inspector and licensee management.

Due to the pattern of recent equipment tag-out problems, the staff is concerned that sufficient controls may not exist or are nnt currently being implemented. To allow the staff

.

to better understand the licensee equipment control program and proposed

improvement to the program, this area will be a specific item for discussion during the next quarterly performance meeting.

<

4.9 Unit Shutdown to Repair "D" Reactor Coolant Pump Surge Capacitors On March 4, 1990, the licensee conducted a shutdown of Unit 2 to

!

investigate a potential cause for the numerous problems associated with

,

the "D" core protection calculator (CPC). Test equipment received on site

.

in late February had enabled the licensee to identify that a source of

,

electromagnetic interference inside the RB was the most probable cause of the numerous trips and sensor failure that had been experienced on "D" CPC since the restart from refueling outage 2R6.

e

'

During 2R6, the licensee replaced all four core protection calculators with a more advanced model which has greater capability than the model installed during initial construction. The modification which installed the new units, as well as the postinstallatior, testing, was highly

.'

successful. Although testing revealed no problems with the "0" CPC or its numerous sensors, the facility has been plagued with numerous trips and sensor failure indications on "D" CPC since restart from 2R6.

i Efforts by licensee personnel to identify and repair the cause of the trips and failures initially centered on the individual sensors involved.

Speed sensor failure on the "D" reactor coolant pump were initially thought to be caused by an inadequate cable grounding configuration, however, modifications to the grounding circuit proved unsuccessful in preventing speed sensor failure indications.

The sensor failures and CPC trips were transitory in nature until one hot leg temperature element (TE)

input to "D" CPC, 2TE-4634-4, began oscillating on February 12, 1990.

The

!

oscillations in the affected TE caused numerous spurious protective trips

'

until operators placed the CPC in bypass, as allowed by the Unit 2 Technical Specifications.

The licensee commenced troubleshooting the affected circuitry under Job Order 806493 and initial indications wer:

that the compensating lead for 2TE-4634-4 was faulted.

The licenree conductec e detailed review of the perceived failure mechanism and develoned a temporary modification, 90-2-07, which removed 2TE-4634-4 frc,.i service. The temporary modification was installed on February 21, 1990,

,

and was successful in terminating the spurious trips to "D" CpC.

"D" CPC was subsequently declared operable and returned to service.

On February 25, 1990, indications of oscillations in the output from 2TE-4610-4 began to manifest themselves as CPC trips.

In addition to the oscillations, speed sensor failures also began to occur.

At this point

"D" CPC was again declared inoperable and placed in bypass.

On February 28, 1990, the licensee initiated troubleshooting efforts to determine what single fault could be causing the multiple failure

,

j'

.

-

v

-

i s:

-

[

-

,

.

l i

,

indications in "D" CPC.

Troubleshooting was directed at eliminating i

possible electromagnetic emission sources.

The license' also decided

'

that should troubleshooting fail to reveal the source of t.he enissions, b)

'

'

March 4, 1990, the unit would be shut down.

Although no positive source was determined by March 4, 1990, licensee personnel atd outside contractors working on the problem strongly believed

-

that the electromagnetic emissions were being produced by breakdown of the

!

!

power supply surge capacitors for the "D" reactor coolant pump.

'

The licensee began shutting down the facility at noon on March 4, 1990.

During the shutdown, the operations staff noted three of the four log-

.

power nuclear instrumentation (NI) channels not trending down even though

'

all other indications, startup NIs, boron concentration and rod position,

,

indicated that the reactor was shut down.

The shift operations supervisor

declared Channels B, C, and D of the log power NIs inoperable. At that

point, the facility could not meet the action statement for the required number of operable channels of log power instruments and the decision was made to trip the plant and perform the requirements of Technical

'

Specification 3.3.1.1, Action Statement 3.

The licensee declared an

'

unusual e rent, as required by their emergency plan ard performed a shutdown margin verification as required by Action Statement 3 of Technical

.

Specification 3.3.1.1.

During the reactor trip, Control Element Assembly (CEA) 61 did not fully insert into the core but remained at an

.

indicated height of eleven inches.

Discussion of the licensee's resolution

-

of this problem can be found in Section 7 cf this report. Although source

..:

range channels continued to trend down and all CEAs, except CEA 61, were fully inserted into the core, log Channels B, C, and D continued to

indicate high which indicated that some outside source was influencing the j

'

readings. Approximately 1/2 hour after the reactor was tripped, the

operations staff secured reactor coolant Pump 2P320. At the time 2P32D

'

,

L was secured, readings on log Channels B, C, and D immediately dropped to

'

i levels appropriate to reactor condition.

. ;

y i

,

Inspection of the surge capacitor assemblies on 2P320 showed the electrical connection between the capacitor and incoming line to be loose with obvious evidence of arcing damage.

Licensee management

decided to replace all raactor coolant pump (RCP) surge capacitors, three L

per pump, prior to restart.

Restart of 2P32D, after capacitor replacement, had no effect on log channel indications.

L The licensee removed the temporary modifications to the RCP speed sensing

,

circuit and 2TE-4634-4, and the unit was restarted on March 7, 1990.

,

L Since the unit restart there have been no additional sensor trips or temperature element oscillations.

Although no license requirement mandated a unit shutdown, licensee management made a conservative decision to shut down and make repairs.

The licensee's decision to shut down the unit to pursue this problem is

consistent with an operational philosophy which places plant safety and performance ahead of unit capacity goals.

L l

li

,

t

,;-

p i

s

.

,

y

,

5.

Operational Safety Verification (Units 1 and 2) (71707)

'

The inspectors routinely toured the facility during normal and backshift hours to assess general plant and equipment conditions, housekeeping, and s

adherence to fire protection, security, and radiological control measures.

Ongoing work activities were monitored to verity that they were being conducted in accordance with approved administrative technical procedures and that proper communications with the. control room staff had been established.' The inspector observed valve, instrument, and electrical equipment lineups in the field to ensure that they were consistent with system operability requirenents and operating procedures.

i During tours of the control room, the inspe: tors verified proper staffing, access control, and operator attentiveness. Adherence to procedures and limiting conditions for operations were evaluated.

The inspectors examined

..

equipment lineup and operability, instrument traces, and status of control

'N p

room annunciators. Various control room 1:gs and other available licensee documentation were reviewed.

While touring in Unit 2 auxiliary building, the inspector observed erratic discharge pressure readings for Boric Acid Makeup Pump 2P-39B as indicated on Pressure Indicator Switch 2PIS-4917.

Both the "A" and "B" Pumps wue operating at the time for recirculation of the boric acid tank.

The discharge pressure for 2P-39A was steady at 100 psi while 2P-39B pressure fluctuated between 95 and 215 psi.

The inspector-identified this condition to the shift supervisor. JR 845285 was written to' investigate the erratic indication on 2PIS-4917.

The operators stated that other parameters for 2P-398 did not indicate any pump performance problems.

After observing maintenance work on Unit 1 "B" EDG, the inspector toured the "A" EDG room and inspected for conditions that could affect the remaining operable EDG, The " Low Level in Day Tank" annunciator on the local. panel was illuminated, however, the EDG day tank level gage indicated i

that-the tank was full.

In the control room, the EDG critical alarm

'

annunciator was not 111aminated.

The operators were unaware that the

- <

annunciator light was on at the local EDG panel. After reviewing this condition the licensee informed the inspector that the circuitry for the i

,

'

annunciators on the EDG local alarm panel allowed the control room

'

annunciator to be cleared while the local annunciator light is still on.

The licensee did note that the abnormal condition that brought in the j

alarm must be cleared prior to acknowledging the control room annunciator.

After additional review of the design change package that installed the EDG annunciator panel, the licensee concluded that the annunciator circuitry was correct.

The inspector noted that the logic for

. '

"

acknowledging EDG alar.ns appeared consistent with other local alarm panels in the plant, however, responsive operator action is needed to acknowledge

both control room and local alarms to avoid potea,1al confusion.

The licensee informed the inspector that additional training would be provided to operators to ensure they understand the annunciator acknowledge 1cgic.

No violations or deviations wera identified.

i

!

Nj

/

-

t

-x i

!

L h

,'

V e

f

r'

6.0 Monthly Surveillance Observations (Units 1 and 2) (61726)

,

The NRC inspector observed the Technical Specification required surveillance testing on the various components listed below and verified testing was performed in accordance with Wequate prcouures, test instrumentation wss calibrated, limiting conditions for operation were met, removal and restoration of the affected components were accomplished, test results conformed with Technical Specifications and procedure requirements, test results were reviewed by personnel other than the individual directing the test, and any deficiencies identified during the testing were properly reviewed and resolved by appropriate management personnel.

The NRC inspector witnessed portions of-the following test activities:

'

Quarterly test of Red Channel Battery Bank 2011 (Procedure 2403.024, Job Order 806213).

During the test, battery specific gravity, voltage, temperature, and electrolyte levels were checked.

• Operability run of EDG 2K4A (Procedure 2104.36, Supplement 1). The surveillance was performed to comply with actiun statement of Technical Specification 3.8.1.1 which requires the EDG to be proven operable every 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> when the redundant EDG is out of service. The

"B" EDG had been taken out of service earlier to allow inspection of tha generator bearings. While the "A" EDG was running, the inspector

<

noticed a small leak at a fuel oil waste line connection at a fuel I

injector. The amount of leakage was one drop every 20 seconds. The

inspector informed the operator who was in the EDG room of the leak.

!

l1 The operator verified this particular fuel oil leak and identified

l two additional small fuel oil leaks while performing a thorough inspection of the EDG. A.iob request was written to repair these

,

l 1eaks.

b-During a monthly surveillance run of this EDG in November 1989, the i

inspector observed a fuel oil leek which was large eaough to require immediate attention. The licensee was able to tighten the connection while the EDG was running and the surveillance was completed. Again the operator in the EDG room was unaware of the fuel oil leak.

In I

response to that leak, the licensee informed the inspector that the

.

need for early identification and correction of fuel oil leaks during L

futurc EDG operation would be emphasized to operators, Based on the i

W 1eaks identified during the recent EDG run, it does not appear the l'

emphasis of operators' responsibilities has been successful.

-

,

L

,

i

'

Monthly surveillance test of Channel

"A" excore instrumentation

(Procedure 2304.100, Job Order 804332).

Prior to performing the

!c su:,reillance, corrective maintenance to resolder three loose test connections on a signal amplifier was performed.

The monthly L

surveillance then provided the required retest of the test connections.

lg

'

.

\\l W) k

.

J W

,

.

.

f..

d g

e

!

>

f

!

h Channel functional test of the logarithmic power instrumentation _

!

'

(Procedure 2105.15). Technicel Specification 4.3.1.1.1 requires a

-

<

H channel function test of the log power instrumentation prior to a

plant startup unless the test was performed in the previous 7 days.

The inspector observed the operators performing the functional test

on the "B" log channel. The test involved taking the log power o

'

output meter readings and comparing the actual readings with the i

required procedural readings for the six calibrate positions. The

operators were having difficulty reading the log power meters to the

'

accuracy of the readings given in the procedure.

The difficulty in

-

verifying compliance with the procedural requirement and questions

{

regarding the completeness of the functional test caused the licensee

'

to review this surveillance for compliance with TS.

Items included in the review were the functional procedure that is performed by operations, the monthly excore surveillance procedure

'

that is perfcrmed by I&C technicians and applicable portions of TS, i

1.e., definitions and surveillance requirements. The licensee concluded from this review that the functional test per Procedure 2105.15 does not fully comply with TS since operability of the trip function associated wtth the log channel is not tested.

It was noted that the log channel crip function is tested by the

menthly excore and plant protection system (PPS) surveillance i

procedures. To resolve this issue a temporary change to the startup

,

procedure was made to require applicable portions of the instrumentation and control surveillance procedures be performed to

~'

complete the functional testing of the log channel instrumentation.

.

The licensee issued LER 50-368/90-006-00 concerning the failure to

perform a complete functional test of logarithmic power l

instrumentation as required by TS. Additional NRC review of this

'

issue will be performed during the closeout inspection of the LER.

No violations or deviations were identified.

7.0 Monthly Maintenance Observations (Units 1 and 2) (62703)

{

7.1 Station maintenance activities for the safety-related systems and components listed below were observed to ascertain that they were conducted in accordance with approved procedures, regulatory guides, and I.

industry codes or standards, and in conforn.ance with the Technical

.

'

Specifications.

The following items were considered during this review:

the limiting i

conditions for operation were met while components or systems were removed

'

w from service, approvels were cbtained prior to initiating the wore,

f activities were accomplished using approved procedures and were inspected as applicable, functional testing and/or calibrations were performed prior t

to returning components or systems to service, quality control records were maintained, i n ivities were accomplished by qualified personnel, parts and materials used were properly certified, radiological cuntrols were implemented, and fire prevention controls were implemented.

',

i tt

_, g b n

i,,j k i-tf

'

, ' fi' )\\ J f d !

.

.[

g l

_.

p

.g N

.

O g

'

'

p

,

.

!

Work requests were reviewed to determine the status of outstanding

,

jobs and to ensure that priority is assigned to safety-related equipment maintenance which may affect system performance.

~

The following maintenance activities were observed:

Re.> r of back leakage on SW pump discharge check Valve SW-1A (Job l

Orcer843899).

Inspection of the valve internals revealed excessive i

clearance between the disc and swing arm.

This allowed " cocking" of

,

the disc which prevented proper seating of the disc when Sd flow

!

s through the valve was secured.

The valve manufacturer stated that I

the original diameter of the disc hub which' mates with the swing arm

'

was 1.500 plus/minus 0.050-inches.

The measured diameter of the i

.,

'

existing hub varied from 1.170 to 1.230 inches.

To correct this

'

condition, a new disc was installed in the valve.

,

In June 1989, SW-1A was partially disassembled and visually inspected.

At that time, no problems were identified.

The licensee currently has

'

plans to replace all three c' the carbon steel SW pump discharge check. valves with stainless steel valves during the next refueling

,

outage.

The inspector has questioned, in light of the degraded condition of SW-1A and the large number of previous backleakage

'

,

problems for these check valves, whether licensee management should

'

'

give replacement of the valves a higher priority.

• Preventive and corrective maintenance on EDG K4B.

PM on the lube oil

'

and fuel oil systems along with repair of various oil and air leaks were scheduled during the EDG outage.

The PM work is an ongoing

effort by the licensee to complete all portions of the 18-month

surveillance involving the manufacturer's recommended inspect.lons.

While inspecting the scavenging oil pump strainer, a paper towel was found in the bottom of the the housing. The scope of the inspection was expanded to include inspection of the crankcase oil sump and

'

.

additional strainers.

No additional paper material was found.

'

The licensee reviewed previous work associated with the EDG lube oil system and identified three activities that could have introduced

"

the paper cloth in the lube oil system. Two of the activities dealt with routine maintenance, i.e., changing oil and filters which included quality control (QC) cleanliness inspection. The third

,

'

activity involved oil additions to maintain proper oil level.

m l

Removal of the strainer cover is required for these oil additions and QC inspection is not required for this task.

The licensee stated that the most likely source of the paper towel in the strainer housing involved the addition of oil in the EDG. The licensee is evaluating the need for QC inspection holdpoints in the oil addition procedure.

Reassembly of SW Pump 2P-4C (Procedure 2402.034, Job Order 728898).

The SW Pump 2P-4C was the last of the three pumps to be disassembled,

"

inspected, and have new impellar snap rings installed.

In 1988,

,

it f.

g g

..

o

,

..

-

,

after experiencing impellar damage due to the failure of a snap ring,

'

the SW pumps had been given a service life of 14 months, at which

time the snap ring would require replacement. All three SW pumps,

'

were disassenLM in 1988 and new carbon steel snap rings were

!

installed.

In December 1989, the snap rins were replaced in 2P-4B-l when the end of the 14-month service life was mehed.

Similar work i

was completed on 2P-4A in January 1990.

The inspector viewed the old snap rings from all three pun -

While

<!

obvious corrosion of the snap rings had occurred, it appeare, that the snap rings were capable of maintaining proper impellar or W tation.

l The licensee assessment and evaluation of the old snap rings to

maintain various loading conditions, including loads imposed during

the 30-day operability requirement, were reviewed by the inspector.

Adequate conservatism and manufacturer's performance data was t

contained in the evaluation.

Based on the inspection results and the i

results o' the evaluation, the licensee is continuing the limited l

14-month service life for the SW pumps.

Review of the alternate i

snap ring material is continuing with the possible use of stainless

-

steel material being a serious consideration.

On February 27, 1990, the "A" EDG was tagged out of service for corrective maintenance and various preventive maintenance items associated with the 18-month surveillance ir.spection extension.

Work was started on the morning of the 27th and was expected to

'..

be completed in 12-16 hours.

Due to the concern of an inoperable train of reactor building coolers, the operations manager directed that maintenance work be completed on the "A" EDG by 4 p.m.

The inspector was observing the completion of the maintenance activities

,

and noted that maintenance supervisors were not present to direct work to ensure timely completion.

Mechanics were making decisions on th9 amount of work to be performed. The inspector observed that

!

direct supervision was apparently not present to make the critical

decisions that co"1d affect the restoration of safety-related l

equipment.

i

The inspector witnessed portions of the CEA drop test that was performed per Special Work Plan 2409.248. The test involved raising

each group of CEAs to the fully withdrawn position, then opening

-

individual CEA circuit breakers and verifying that each CEA fully inserted into the core.

This successful test provided additional assurance to the licensee that the CEA would insert into the core following a reactor trip and would not experience problems similar to

CEA 61 when it stopped at 11 inches follow ng the manual reactor l

trip.

l 7.2 Development of Forced Outage Work Lists Since the fourth calendar quarter of 1989, members of operations and maintenance on both units have been working to develop, refine, and implement a prioritized ready-to-work maintenance schedule for short 1*

M wb

-

L

..

c

,n I

u /

' ).

r

i

,

duration forced outages. Although some progress in this area was evident in the past, the most recent forced outages on both units showed strong evidence that this system of preplenning forced outages has been

!

successfully integrated into the licensee's material condition management

'

scheme,

,

i Currently, Unit 1 management maintains a work list for 3, 5, or 7-day

'

outages. All job tasks are prioritized by the operations department and have complete work packages prepared prior to being placed on the list.

i Whenever a forced outage occurs, jobs on the list are immediately started jl based on expected outage duration.

Using this methodology, the licensee was able to complete the 26 highest priority jobs from the outage list in r

addition to the 31 jobs required to repair equipment required to address the shutdown issues.

l During the Unit 2 forced outage to make repairs to the "D" RCP, members of

!

the Unit 2 maintenance staff completed 21 job orders which required a plant

outage, of which 14 were in the autage schedule.

Development and updating a forced outage list is an efficient way of improving facility material conditions. Although the programs are still in the development and refinement stage they have already proved beneficial.

No violations or deviations were identified.

I 8.0 In-Office Review of Licensee peport (90712)

The following LERs and licensee special reports were reviewed and closed.

The inspector verified that reporting requirements had been

'

met, causes have been identified, corrective action appeared

,

'

appropriate, reactive NRC inspection is not warranted, generic

,

applicability had been considered, and that the LER forms were i

l complete. The inspector confirmed that violations of Technical i

Specifications, license conditions, or other reoulatory requirements had

'

been adequately discribed.

i l.

LER 50-313/89-006 i

LER 50-313/89-009

-

LER 50-313/89-011 LER 50-313/89-014 l

LER 50-313/89-020

-

l LER 50-313/89-025 LER 50-313/89-031 LER 50-313/89-035 LER 50-313/89-036 LER 50-368/89-005

,

LER 50-368/89-008 LER 50-368/89-013 l

LER 50-368/89-016 6'

-

.

i

,;

.

,

,,

,

,

n

,

f jd Q' =,, " '-

,

,

,

7...

,

i

+

,

/

!.

j

!

$

.-

' o,r 9.0 Exit Interview The-inspectors met with the Director, Nuclear Operations and other members

,

of. the AP&L staff at the end of the inspection. At this meeting, the

,

inspectors summarized the scope of the inspection and the findings. The-

-

,

., 'i

.

licensee did not identify as proprietary any of.the material provided to, i

' -

.,.

. or reviewed by, the. inspectors during.this 1.nspection.-

{

,

. 3..

<

l

.o

'

l

i

,

j..i s, '$

,

'

<

r

.

I4,

I

'

.1

. <

s

_I E

i

-l t

l

t.'

t

'l

,

e '

..) 's

.

,

s,

,1 () 3

',j e

,

I 1: q L

.y

,

.b

,

I.-

L.

i

1 r

'?

?/ :

'

f

!

l.,

1,

,

l '.

i

+

I

'N:L,

,

,

,

}l

'\\

l

.. -

f

v.

,

', Nf it

.., n :

,. _

.

...,.,...

... -,,,,, --

r

,--

-,n-

- - -

,-