ML20128B828

Power Reactor EVENTS.November-December 1984

ML20128B828
Person / Time
Issue date:	04/30/1985
From:	Massaro S NRC OFFICE FOR ANALYSIS & EVALUATION OF OPERATIONAL DATA (AEOD)
To:
References
NUREG-BR-0051, NUREG-BR-0051-V06-N6, NUREG-BR-51, NUREG-BR-51-V6-N6, NUDOCS 8505280074
Download: ML20128B828 (45)

Text

r - - - - - - - - - - -

NUREG/BR-0051 Vol. 6, No. 6 8..,T POWER REACTOR EVE NTS

//

United States Nuclear Regulatory Commission Date Published: April 1985 Pow:r Reactor Events is a bi monthly newsletter that compiles operating experience information about commercial nucint power plants. This includes summaries of noteworthy events and listings and/or abstracts of USNRC and other docum:nts that discuss safety related or possible generic issues. It is intended to feed back some of the lessons learned from operational experience to the various plant personnel, i.e., managers, licensed reactor operators, training coor-dinitors. and support personnel. Referenced documents are available from the USNRC Public Document Room at 1717 H Street, Washington. DC 20555 for a copying fee. Subscriptions and additional or back issues of Power Reactor Events m'.y be requested from the Superintendent of Documents. U.S. Government Printing Office, (202) 257-2060 or -2171, or et P.O. Box 37082. Washington. D.C. 20013-7982.

Table of Contents Page 1.0 SUMMARIES OF EVENTS...

I 1.1 Explosion and Fire in Auxiliary Transformer Results in Loss of Startup Transformer at Duane Amold I

1.2 Reactor Vessel Water Level Transient at Grand Gulf..

4 1.3 Extraction Steam Line Break Due to Wet Steam Erosion at Calvert Cliffs...

6 1.4 Reactor Trips Due to Air Line Failures Causedby Fatigue C"acking at Callaway:

7 1.S Inadvertent Trip from FullPower Due to Human Errorat Haddam Neck..

8 1.6 Owrshoot on Spring Retum Handswitch Causes inadvertent Safety injections at Sequoyah...

9 1.7 Circuit Breaker Failures Due to Dirt and Dust Accumulation at Haddam Neck..

10 1.6 Repeated Trips of Emergency Diesel Generators at North Anna..

12 1.9 Stress Corrosion Crxking in Nonsensitized316 Stainless Steelat Peach Bottom 14 1.10 References -

18 2.0 EXCERPTS OF SELECTED LICENSEE EVENT REPOR TS.

19 3.0 ABSTRACTS / LISTINGS OF O THER NRC OPERA TING EXPERIENCE DOCUMENTS...

37 3.1 Abnormal Occurrence Reports (NUREG-0090)....

37 3.2 Bulletins and Information Notices..

38 3.3 Case Studies and Engineering Evaluations.

40 3.4 Generic Letters 42 3.5 Operating Reactor Event Memoranda.-

43 3.6 NRC Document Compilations.

44 Editor: Sheryl A. Massaro Office for Analysis and Evaluation of Operational Data U.S. Nuclear Regulatory Commission Period Covered: November-Deccmber 1984 Washington, D.C. 20555 8505280074 850430 PDR NUREG PDR BR-0051 R

1. 0 SUMMARIES OF EVENTS 1.1 Explosion and Fire in Auxiliary Transformer Results in Loss of Startup Transformer On November 4, 1984, Duane Arnold

• experienced a failure of the auxiliary trans-former, resulting in a fireball that damaged the startup transformer. The failure experienced was not preventable by presently known or used maintenance and inspection techniques.

The preliminary finding by the transformer manu-facturer (Westinghouse) is that the failure resulted from a short circuit between turns in the high voltage winding of the C phase coil.

Damage to the startup transformer was limited because of a concrete shield wall, not required by regulations, which the licensee had constructed between the auxiliary and startup transformer.

The event is detailed below.

On November 4,1984 at 1:34 a.m., with Duane Arnold operating at approximately 56% power, the auxiliary transformer failed.

As a result, the auxiliary trans-former differential power relay operated, tripping the unit backup lock out relay.

This caused the main turbine to trip and the reactor to scram from turbine stop valve closure, as designed.

A resulting fireball from the trans-former failure reached as high (about 150 feet) as the electrical high voltage lines (161 kV) connected to the startup transformer, causing damage and carbon deposits on two line insulator strings for the startup transformer primary.

(See Figure 1.)

In addition, the 161 kV primary busings on the startup transformer received heavy carbon deposits.

The deluge systems for both the startup and auxiliary transformers initiated.

The carbon deposits and insulator damage on the startup transformer caused the startup transformer to trip on a phase-to ground fault. When the startup transformer tripped, the essential loads switched to the standby transformers.

Both diesel generators automatically started, but were not loaded due to the successful transfer from the startup to standby transformer.

The diesel generators were secured within 30 minutes.

The transfer of essential loads from the startup to the standby transformer was accomplished in approximately four to six cycles, so that the reactor protec-tion system motor generator set flywheel maintained stability long enough to prevent the electrical protection assembly breakers from tripping.

All non-essential loads were lost when the startup transformer tripped.

The scram pro-ceeded as expected and the equipment performed consistent with design except for some minor difficulties.

The emergency notification system (ENS /NRC) hot-line was inoperable due to the loss of nonessential power.

(An onsite " direct line" phone was used to make NRC and other offsite notifications.) Other effects of the loss of nonessential power are detailed below.

CDuane Arnold is a 515 MWe (net) General Electric BWR located 8 miles northwest of Cedar Rapids, Iowa, and is operated by Iowa Electric Power and Light.

1

i 35 KV WEST BUS

=_____

i o,

4.

6 4

9 Ip _ _ _ _ _ _ _ _,)

,, A,

AUTO XFMR 346 KV

. __ ____.A. _ _ EAST B US 1

I l

M, l

6 s 161 KV WEST BUS T

_ _ _ _. T -'- --- - - --- - -- - - - - - -- - - j - - - !-- - - - - - - *2 i

i._ _ _l_. _.______ __ _ 7.

I l

161 KV EAST BUS I

CD

___________;_________3___

h E

r3 g

STANDBY Y

}

~~

XFMR mm STARTUP MAIN XFMR eg cg XFMR ca l

AUXILIARY j

XFMR psp 3

t i

l MAIN l

GENERATOR N

(STATION TURBINE GENERATOR) h W

@r h

4160 V BUS 1A1 4160 V BUS 1A3 4160 V BUS 1A4 4160 V BUS 1A2 i

a NON-SAFETY RELATED SAFETY RELATED LOADS SAFETY RELATED NON-SAFETY RELATED LOADS LOADS LOADS l

1 Figure 1. Duane Arnold Electrical System

}

1 1

Immediately after the main turbine trip, the bypass valves to the main condenser opened as designed.

Reactor pressure peaked at approximately 1009 psig immedi-ately after the turbine trip and scram.

Reactor water level began dropping from a collapse of steam voids, as expected, and reached a minimum level of approximately 140 inches above the top of active fuel within 1 minute (normal operating level is 193 inches).

High pressure coolant injection (HPCI) was

)

manually started and injected to recover coolant inventory.

The residual heat removal system was manually started and placed in the torus cooling mode.

In 1

addition, several spurious reactor water cleanup isolation signals were received during the-course of the transient.

j Approximately 4 minutes after the scram, operators attempted to reset the startup transformer power to nonessential busses.

The attempt was unsuccessful because of the insulator damage and carbon deposits.

Operators closed the main steam isolation valves at 1:52 a.m., in anticipation of automatic isolation on loss of condenser vacuum.

At 1:58 a.m., the reactor core isolation cooling (RCIC) system was manually started and then the high pressure coolant injection system (HPCI) was secured.

HPCI and/or RCIC were used manually by operators throughout the rest of the event for vessel level and pressure control.

Operators secured the control rod drive pump at 3:19 a.m. in an effort to control the reactor vessel bottom head cooldown rate.

(This pump had not been secured earlier because it is normally kept running to maintain control rod drive seal cooling and prevent degradation of the seals.) The vessel bottom head cooldown rate (less than 200 degrees F/hr) exceeded the normal cooldown rate of 100 degrees F/hr during the transient.

Recovery operations proceeded as expected, although problems were noted as follows:

(1) Normal radiation air sampling capability within the power block was lost.

Battery powered samplers were placed at two locations, one each in the turbine and reactor buildings.

Samples were taken hourly and all were normal.

Upon loss of their power sources, area radiation monitors alarmed.

Health Physics technicians conducted surveys to determine actual conditions in the areas of the alarmed monitors.

These monitor alarms were indicative of momentary loss of power and not abnormal radiation.

Included in these monitors were the high range accident monitors which tripped on the momen-tary loss of vital power.

These monitors could have been reset quickly by Health Physics technicians if necessary.

(2) Per design, normal control room reactor conductivity indication and radwaste panel indications were. lost until 10:15 a.m., when equipment was moved to the post-accident sampling system lab and grab samples were taken every hour thereafter.

(Normal conductivity monitoring was restored when non-essential power was restored.)

Fire resulting from the transformer failure was reported at the transformer and at the turbine building, where metal siding in the area was damaged. While reactor recovery actions were being performed,-the fire brigade was dispatched to the area.

The deluge system at the transformer operated as designed and the fire brigade activities were mostly of a clean-up nature. A fire wall between auxiliary and startup transformers limited damage to the startup transformer.

(This wall was not required by regulations.) Repairs to the startup transformer 3

insulators were completed at approximately 4:00 p.m. and nonessential loads were brought on line.

A preliminary study was conducted by the transformer manufacturer (Westinghouse),

with the preliminary finding that the failure resulted from a short circuit between turns in the high voltage winding of the C phase coil.

The exact cause of the short circuit is still under investigation.

When the short circuit occurred, arcing was introduced which generated gas at a rapid rate.

Pressure built up in the transformer tank too quickly for the pressure relief device to release enough gas to prevent tank rupture.

Transformer test records were reviewed.

Westinghouse ~noted that the 6 month test schedule followed at Duane Arnold was more than adequate for proper pre-ventive maintenance.

They stated that the most informative test is gas-in-oil chromatography analysis and oil condition testing.

Samples had been taken on October 31, 1984. When these results were compared with the results of the November 1983 and March 1984 tests, it was found that levels of methane, carbon dioxide, ethane and ethylene had increased slightly.

Westinghouse feels this could be an indication of possible insulation breakdown.

However, the amount by which dissolved gas levels had increased would not normally be considered a precursor to the type of failure encountered during this event.

Corrective actions by the licensee include the following:

Transformer testing was reviewed in order to ensure use of the best applicable oil and gas-in-oil analysis in an attempt to anticipate transformer problems.

In addition, other transformer maintenance procedures were reviewed for adequacy.

Auxiliary transformer repairs have not yet been completed.

A complete checkout of remaining transformers and the main generator was per-formed to ensure no damage occurred during the event.

The reliability of the various telephone systems during a loss of offsite power is being evaluated.

Operations personnel were briefed in means of making offsite emergency notifications and restoring communications given the present systems.

Power supplies to several administrative areas are being evaluated for possible improvement in power supply reliability.

An engineering evaluation was conducted on the effect of the reactor vessel bottom head cooldown rate.

The conclusion of the licensee was that the 100 degrees F per hour cooldown rate applies to normal cooldowns and not transients.

General Electric stated that the cooldown rate was acceptable.

(Refs. 1-3.)

1. 2 Reactor Vessel Water Level Transient On November 29, 1984, during startup testing of Grand Gulf Unit 1,* a reactor vessel level transient occurred which resulted in a challenge to the high

• Grand Gulf Unit 1 is a 1250 MWe (net) General Electric BWR located about 25 miles south of Vicksburg, Mississippi, and is operated by Mississippi Power and Light.

4

s i

pressure core spray'(HPCS) system, actuation of various other safety systems,

~

'and an erroneous-switchover'of the HPCS system suction from the condensate storage tank.(CST) to the. suppression pool.

The licensee declared an unusual avent.and_ notified offsite agencies as required.

L

_The plant'was operating at about 53% power.with the reactor vessel level at 36 inches. DThe condensate and feedwater (FW) systems were in operation for FW system-testing with the A FW pump in manual control (with test _ equipment con--

nected to the automatic control circuit) and the B FW pump control.in automatic.

L

.Anerroneous; input (causedbyinadvertentgrovadingoftheinput)to-theBFW

' pump' speed controller resulted in a large FW ilow decrease and a rapid decrease l

-in the reactor _ vessel water level.

An operator quickly took manual control of the B'FW pump?and_ stopped the level decrease by increasing FW flow.

The minimum level-reached approximately +19 inches.

As the vessel. level reached normal, the operator began matching FW flow with the steam flow..It is suspected that the' rapid increase in FW flow caused the condenser hotwell level to decrease,.

and the condensate pumps tripped on low suction pressure.

i With a loss of all condensate flow, the Shift Superintendent ordered that the reactor be manually' scrammed.

The vessel level decreased rapidly.

The reactor core isolation cooling (RCIC) system was manually initiated near its automatic actuation setpoint (-41.6 inches) approximately 12 seconds after the scram.

The HPCS pump. automatically started about 4 seconds later.

Other automatic actions.which took place were:

the auxiliary building and containment _ isolated; standby gas treatment trains A and B initiated; the control room ventilation system shifted to the isolation mode; the reactor recirculation pumps tripped

~

~ to the-low frequency motor generator; and the reactor water cleanup system

- i sol ated.'.

- The level reached a minimum of -49 inches. When the level was restored to

-20 inches:and rising, approximately 30 seconds after HPCS began injecting, the

. HPCS' injection valve was throttled closed to reduce the rapid rise in vessel level and to' limit the cooldown rate, yet allowing the HPCS-pump to continue running. This action is specified in procedures.

Shortly thereafter, the

HPCS/RCIC. suction shifted from the condensate storage tank (CST) to the sup-fpressionpool. This erroneous automatic shift was due to a sensed low CST level caused by pressure' oscillation when the injection valve was closed.

About 4 minutes into the event, the RCIC turbine tripped for no apparent reason, but was immediately restarted.

The condensate pumps and condensate

. booster pumps were restarted a few minutes later, to maintain normal vessel level. LThe plant systems were placed in normal configuration about 20~ min,utes later.

r Investigation of the event showed that a combination of defective special test

. equipment and an incorrectly connected ground lead from the test equipment resulted in grounding the input to the B FW pump-speed controller; The B FW pump slowed'down, causing the decrease in FW flow.

Investigation failed to determine the cause of the RCIC trip.

A review of computer data recorded during the transient showed RCIC system parameters well below trip setpoints.

L No' system abnormalities were found.- The trip is considered to be an isolated o

i ncident.-

i-5

~..

-Corrective actions taken by the licensee included (1) making modifications and

repairs to the special' test equipment to eliminate grounding; (2) removing the condensate pump startup suction strainers and the low suction pressure trip to enhance the' condensate pump availability; (3) increasing the time constant on

~

the pressure transmitter for the HPCS/RCIC suction path transfer to help elimi-nate spurious transfers; and (4) holding a training session for engineers involved in startup testing on the installation, operation and use of the special test equipment.

In addition, plant management issued memoranda to all supervisors / superintendents-reiterating that independent verifications be performed in series for modifications -

that require verifications..In this regard, the startup test procedure was mod-ified to provide better control of independent verification during connection and removal of test equipment.

(Ref. 4.)

1.3 Extraction Steam Line Break Due to Wet Steam Erosion On November 20, 1984, Calvert Cliffs Unit 1* was manually tripped due to high differential pressure on the traveling screens caused by an influx of fish.

Immediately following the trip an extraction steam line ruptured, filling the turbine building with steam and causing first degree burns to one individual.

Repairs were made and the unit returned to operation on November 26, 1984.

The event is detailed below.

At 5:16 p.m. on November 20, 1984, Unit 1 was manually tripped from 100% power due to the accumulation of dead fish on the circulating water intake screen.

(The large influx of fish is believed to have resulted from a relatively rapid drop in Chesapeake Bay water temperature.) At 5:18 p.m. a steam break occurred and the main steam isolation valves were closed.

Investigation revealed a break in the extraction steam line supply to the No. ISA feedwater heater.

This steam line carries low pressure, nonradioactive steam from the main steam system to feedwater heaters in order to preheat feedwater being pumped back into the steam generators. A worker approximately 50 feet away from the location of the break suffered first degree burns of the face and left hand.

He was taken to the hospital and was subsequently released.

Analysis of the break revealed a 30-inch long rupture at an elbow in the extraction steam line. The original 0.375-inch steel wall was severely eroded due to wet steam erosion.

The licensee performed nondestructive examination on similar' elbows in the higher pressure extraction steam lines to determine wall thicknesses.

Extraction steam line repairs were completed on November 25, 1984, and the unit was returned to power operation.

A total of five' elbows on the No. 15 and 16 A and B extraction lines were replaced, and an additional ten had patch overlays to strengthen thinned areas.

The original piping was carbon steel A106, grade B, schedule 40; the replacement piping is low alloy steel A234WP11 (1.25% chromium,'0.5% molybdenum).

~*Calvert Cliffs Units 1 and 2 are 825 MWe (net) Combustion Engineering PWRs located 40 miles south of Annapolis, Maryland, and'are operated by Baltimore Gas and Electric.

j 6

Both Calvert Cliffs Units 1 and 2 experienced previous failures of extraction steam /feedwater heater' piping (at least three failures).

A large scale program consisting of nondestructive examination and replacement of the more severely eroded piping was begun on Unit 2 during the spring 1984 refueling outage.

Some piping was al_so_ replaced in 1984 on Unit 1 during a short outage.

That program was scheduled to continue on Unit 1 during its next refueling outage during spring 1985.

The licensee estimates that they are only about 15% into the overall program.

Main steam piping (including inside containment) is also scheduled to be checked.

Some main steam piping locations in the turbine building were checked on Unit 2 during spring 1984, and no significant prob-lems were identified.

(Refs. 5-7.)

~

1. 4 Reactor Trips Due to Air Line Failures Caused by Fatigue Cracking On November 5 and 6, 1984 at Callaway Unit 1,* air lines supplying the steam generator (SG) feedwater regulating valves failed, resulting in actuations of c;ngineered safety features and reactor trips.

The air line failures were due to improper material applications'and resulting fatigue cracking caused by the vibrations imposed on the air lines during feedwater system operation.

The events and corrective actions are discussed below.

On November 5, 1984, a reactor trip occurred when a low low level signal was received on SG C.

The plant was operating at 45% power.

The service air line manifold common to the SG feedwater control valves (FCVs) sheared at the connection which channels the service air supply for SG C FCV-530, causing it to' fail. closed.

Approximately 50 seconds after FCV-530 closed,

.a low-low leyel signal on SG C caused a reactor trip accompanied by a turbine trip, feedwater isolation signal (FWIS), auxiliary feedwater actuation signal (AFAS), and steam generator blowdown isolation signal (SGBIS).

The service air line was repaired, the feedwater isolation valves were reopened, auxiliary feedwater was secured, and SG blowdown was reinstated.

On November 6, 1984, about 1-1/2 hours after resuming power operation (16%) power), a high level signal on SG C resulted in a turbine trip, FWIS, AFAS and SGBIS.

A low-low level signal on SG 0 caused a reactor trip.

Subsequent investigations revealed that the service air line to feedwater recirculation valve FV-28 had ruptured, causing FV-28 to fail l

initially.in an intermittent fashion, which caused the recirculation valve to oscillate and finally to open.

The bypassing of feedwater through l

FV-28 caused the FCVs to open. Operators also compensated for the lost feedwater flow by increasing the flow rate through the main feedwater pump, which was in manual control.

The resultant feedwater flow oscil-1 lations caused the high level signal on SG C, producing a turbine trip, l

FWIS, AFAS and SGBIS.

Due to the FWIS and AFAS, SG levels began to shrink, and a low-low level on SG D caused the reactor trip.

Both air line failures were due to improper material application, which resulted in fatigue cracking caused by the vibration imposed on the air lines during feedwater system operation.

To prevent future failures, the following changes were incorporated:

• Callaway Unit 1 is a 1120 NWe (net) Westinghouse PWR located 10 miles southeast of Fulton, Missouri, and is operated by Union Electric.

7

(1) The bypass tubing nipples previously installed on the air line manifold to the FCVs were replaced with copper tubing.

(2) -Stainless steel tubing was installed from the air line manifold to the FCV current / pneumatic converters.

(3) A design change was evaluated which would add a support to the air line

.upsteam of-the manifold for the FCVs.

(4) Temporary modifications.were installed for hangers and flexible air lines for the feedwater recirculation valves.

A design change was. submitted to make these modifications permanent.

(Ref. 8.)

1.5 Inadvertent Trip'from Full Power Due to Human Error At-10:03 a.m. on November 20, 1984, a Haddam Neck

• Operator notified the NRC via the Emergency Notification system of a reactor trip that had occurred

.approximately 1/2 hour earlier.

The plant tripped from full power when a Reactor.0perator inadvertently secured a reactor coolant pump (RCP), causing a loss of flow scram.

All systems functioned normally during the event and the plant was subsequently stabilized in the hot shutdown mode.

This event chal-lenged the reactor protection system, and subjected the plant to an unnecessary reactor scram from full power.

The event occurred when a Reactor Operator erroneously secured RCP No. 3 during a routine surveillance of the containment air recirculation (CAR) fans, at which time he intended to secure CAR Fan No. 3.

The control switches for CAR fan No. 3 and RCP No. 3 are located in adjacent panels and in the.same relative position on'the main control board (MCB).

The' actual appearances of the panels, however, are substantially different.

A typical label for the CAR fan is shown below:

CONT. RECIRC. FAN 1 P-17-1 BUS 4 A typical label for the RCP is shown below:

REACT. COOL. PUMP 1 P-17-1 BUS 1-1A Additional labels on the switch handles were also in place at the time of the trip.

It appears that this event was due to a lack of attention by the operator, and not to poor human engineering.

Furthermore, the event could have been prevented by a general administrative procedure requiring a " verify before executing" routine.

(NRC Information Notice 84-51, " Independent Verification," issued June 26, 1984, provided guidance on the subject.)

• Haddam Neck is'a 569 MWe (net) Westinghouse PWR located 13 miles east of Meriden, Connecticut, and is operated by Connecticut Yankee Atomic Power.

8

Corrective action by the licensee included (1) discussing the event with the Operator who tripped the RCP; (2) attaching additional labels to the RCP switches on the MCB; (3) changing the component identifier code for the CAR fans from P-17 to F-17 and attaching new labels to the MCB; and (4) initiating a design change to cover the pump switch handles.

This event emphasizes the importance of adequate procedures, planning, labeling, awareness and training of personnel, as well as an independent verification program to eliminate or at least reduce events involving human error.

Somewhat similar events are described in NRC's IE Information Notice 84-58, " Inadvertent Defeat of Safety Function Caused by Human Error Involving Wrong Unit, Wrong Train, or Wrong System," issued July 25, 1984.

The events described in that information notice involved human errors that defeated safety functions and thus degraded a plant's ability to mitigate the effects of postulated accidents.

(Refs. 9 and 10.)

1.6 Overshoot on Spring-Return Handswitch Causes Inadvertent Safety Injections A followup review was performed by an NRC inspector for an inadvertent safety injection (SI) which occurred on December 16, 1984, while Sequoyah Unit 2* was in hot standby and in the process of increasing pressure.

An SI actuation occurred on low pressurizer pressure.

Prior to the event, the operators identi-fied a primary safety relief valve weeping, based on elevated tailpipe tempera-ture.

Primary pressure was reduced using pressurizer spray to reseat the valve with the low pressurizer pressure SI channels blocked.

To assure SI blockage while performing the evolution, prior to reaching 1870 psig (SI setpoint) the operator cycled the SI block spring-return-to-center switch for both trains.

Upon release of the A train switch, the switch passed through the center posi-tion and momentarily actuated the contacts on the reset side, thereby unblocking the low pressurizer pressure signal.

A partial emergency core cooling system (ECCS) actuation and boron injection tank (BIT) injection of approximately 570 gallons occurred.

The plant was stabilized at 1850 psig and the SI was reset.

Previous similar inadvertent sis had occurred on Unit 1 in September 1979, and were also attributed to problems with the handswitch.

A design change to replace the existing switches with new ones was implemented for both units, but was ineffective due to inadequate root cause determination.

The switches were replaced with the same type of design and vendor source.

On January 11, 1985, the licensee initiated another design change to replace these switches with a different type in order to prevent recurrence of the problem.

Background

information related to inadequate and excessive time for corrective action of inadvertent sis due to switch malfunction follows.

There are eight Westinghouse Type OT-2 handswitches in use at Sequoyah, four on each unit.

The handswitches are spring-return-to-neutral (center) from either the block position or the unblock (reset) position.

These switches tend to spring back past the neutral position to the reset position when released in the block position.

• Sequoyah Units 1 and 2 are 1148 MWe (net) Westinghouse PWRs located 10 miles northeast of Chattanooga, Tennessee, and are operated by Tennessee Valley Authority.

9 i

k_

p As mentioned previously, the handswitch problem was first identified during i

startup testing for Unit 1 in September 1979.

Investigations of spurious sis which occurred at that time indicated that the reset contact could be opened by a very slight movement (approximately a 10 millisecond overshoot) toward the reset position.

Subsequently, as part of the corrective action the licensee placed administrative controls in the appropriate procedures'which required

. reactor operators to release the switch slowly to prevent the switch from springing back to the reset position.

Additionally, cautionary placards were placed above the applicable switches.

l The licensee also initiated a design change request in June 1980 to replace the existing switches, and requested that Westinghouse replace the switches in December 1981.

In response, Westinghouse initiated field deficiency reports for Units 1 and 2 on January 11, 1982 to replace the switches.

The field deficiency reports, which were signed off in February 1982, indicated that the switches were replaced, but there was no need for a design change of existing equipment.

The licensee's Unreviewed Safety Question Determination (USQD),

dated November 5, 1982, for the design change request, indicated that the original handswitches were defective and that the new handswitches would stop in the normal position as required.

The USQD stated that the new handswitches are the same design as the original ones.

There was no indication in the USQD of testing or other investigations to support a determination that the original handswitches were defective versus a determination that the switch design caused

.the problem.

The handswitches were not replaced until February 1983, for Unit 1, and December 1982 for Unit 2.

The inspector's review of the licensee's post-modification testing of the replaced handswitches identified insufficient testing.

Testing verified that the switches would perform the block function; however, testing to detect the original problem, i.e., the " bounce" back to reset after release, was not performed.

On January 11 and 25, 1985, the inspectors discussed their findings with the licensee pertaining to these handswitches.

And, as a result, the licensee initiated another design change to replace these switches with a different l

type.

On January 15, 1985, Westinghouse indicated to the licensee that the reported i

problem was analyzed and that the resulting " bounce" was not due to defective l

parts but was characteristic of the particular switch. Westinghouse also l

claimed that the application of the OT-2 switches did not present a safety n

problem.

(Ref. 11.)

1.7 Circuit Breaker Failures Due to Dirt and Dust Accumulation Between June and October 1984, six incidents of Westinghouse DB and DHP circuit breakers failing to close on demand occurred at Haddam Neck.* Intermittent failure of the closing coil cutoff X-relay to properly return to its deenergized position prevented this-relay from energizing the breakers' closing coil upon receipt-of a close signal.

Dirt and dust accumulation on the moveable parts of the relay caused the faulty operation.

• Haddam Neck is a 569 MWe (net) Westinghouse PWR located 13 miles east of Meriden, Connecticut, and is operated by Connecticut Yankee Atomic Power.

10

The symptom of the X-relay malfunction was failure of the breaker to close upon receiving a close signal, although in most cases the breaker closed upon receiv-ing a second close signal.

This failure mode can cause equipment and/or systems to be inoperable without detection until that equipment is called upon to oper-ate. At Haddam Neck, the X-relays on all safety-related DB and DHP breakers were inspected and cleaned.

Procedures were revised to include maintenance (cleaning) of the X-relays during scheduled preventive maintenance.

(Mainte-nance of the X-relays was not addressed in the vendor's maintenance procedures.)

No other relays were found to have the dust or dirt build-up found on the failed relays mentioned above.

These types of failures were noted previously in IE Information Notice No. 83-50,

" Failures of Class IE Safety-Related Switchgear Circuit Breakers to Close on Demand." An NRC study of over 100 LERs on similar problems with malfunctioning circuit breaker closing control circuitry was issued in April 1983 (" Failures of Class IE Safety-Related Switchgear Circuit Breakers to Close on Demand,"

AE00/C301).

According to the licensee, five of the breaker failure problems occurring between June and October 1984 were positively linked to the closing coil cutoff X-relay on the breaker.

One breaker failure may have involved an X-relay malfunction.

The closing coil cutoff X-relay is an electromechanical device used to energize the closing coil upon receipt of a close signal.

A close signal energizes the X-relay coil, pulling a slug and latch arm assembly into the coil.

The slug and latch arm pull the X-relay contacts closed to energize the closing coil.

The closing coil is deenergized when the breaker is closed by the physical unlatching of the X-relay contact block from the slug and latch arm. When the close signal is removed, the X-relay is deenergized.

The slug and latch arm assembly is free to slide back to its original position and latch to the con-tact block.

The X-relay is then ready to accept another close signal.

If the latching assembly does not slide back to its deenergized position due to dirt or dust binding its movement, the X relay will not operate the closing coil when energized by a close signal.

This malfunction occurred on three nonsafety-related breakers between July 1 and August 13, 1984.

The X-relays on two of the breakers were cleaned.

A third X-relay which exhibited mechanical wear was replaced.

On August 24, 1984, with the plant in a refueling mode, the output breaker from emergency diesel generator (DG) EG2A failed to close automatically during a total loss of offsite power.

The breaker remained open for 20 minutes before it finally closed.

This left one train of Class 1E medium voltage emergency power inoperable.

At the instant of breaker closure, an operator was in the process of resetting an under/oltage lockout relay which has one of its contacts in the breaker closing circuit.

At that time, the output voltage permissive relays delayed automatic closure of the DG output breaker due to relatively high frequency and low voltage observed as the generator was operating.

It was thought that a rise in voltage, a drop in frequency, or vibrations from resetting the lockedt relay finally caused the output voltage permissive relays to pick up and close the breaker.

During a refueling outage on October 21, 1984, the DG output breaker again failed to close automatically during loss of offsite power testing.

Immediate 11 r

investigation and further testing proved the failed component to be the breaker itself.

It was observed during this testing that removing the automatic close signal from the X-relay and then reapplying it caused the breaker to close.

This breaker was removed from service and replaced with a spare.

Several days later,.the X-relay was replaced by Westinghouse and the breaker was returned to service.

While still in the refueling outage, on October 29 a reactor coolant pump breaker failed to close on demand during performance of a surveillance test.

The breaker was in the test position at the time.

Westinghouse investigated and found that the X-relay on this DHP type breaker was malfunctioning due to dirt and dust accumulation on the slug and latch arm assembly.

The plant learned at this time that maintenance of the moveable parts of the X-relay is not addressed by Westinghouse in their maintenance program for.these particular circuit breakers.

The X-relay malfunctions were thus identified as a generic problem in the plant and additional maintenance guidance was requested from Westinghouse.

Immediate corrective action was to inspect all Westinghouse DB and DHP breakers in safety-related systems. All these X-relays were inspected and cleaned to remove surface rust.

Dust or dirt accumulation to the extent of what was found on the five failed relays was not found on any other relays.

Lubrication was not used on the relays after cleaning.

Upon discovery of this generic problem, it was realized that the breaker failure on August 24, 1984 described above may also have involved an X-relay malfunction.

It was previously stated that an operator was resetting a relay which has a contact in the automatic closing circuit when the breaker closed, also, it was proven during the incident on October 21, 1984 that deenergizing and immediately reenergizing of the X-relay closed the breaker.

Since the operator was manipu-lating a contact in the closing circuit when the breaker closed, X-relay mal-function is a possible cause for the breaker failure.

That is, the breaker closure could have been caused by this contact opening (operator resetting the lockout relay), and then momentarily closing long enough to close the breaker with the operator's hand still on the reset handle.

The equipment mentioned above is approaching 20 years of service life at the Haddam Neck plant.

The malfunction described above could cause safety systems to be inoperable without detection until the equipment is called upon to operate either by test or when actually required.

The licensee is confident that the immediate corrective action taken to inspect all safety-related breakers is adequate to insure reliable operation of the equipment.

As a long-term cor-rective action, the cleaning of these X relays will be incorporated in the plant's preventive maintenance program.

(Ref. 12.)

1.8 Repeated Trips of Emergency Diesel Generators Between October 19 and March 15, 1985, both emergency diesel generators (EDGs) at North Anna Unit 2* tripped once on high jacket coolant temperature, six more

• North Anna Unit 2 is an 890 MWe (net) Westinghouse PWR located 40 miles north-west of Richmond, Virginia, and is operated by Virginia Electric and Power.

12

~

times on high crankcase pressure, and failed to start once because of a problem with the air start distributor.

The failures of these diesels (Fairbanks-Morse, 3840 HP, Model 38TD 81/8), resulting in three valid failures to provide emer-gency electricity on demand, and 10 days of lost generation, are detailed below.

On October 19, 1984, while performing the 24-hour surveillance run required by technical specifications on the 2H EDG, the diesel tripped on high jacket coolant temperature.

The temperature switch was subsequently calibrated and the diesel was retested.

On October 20, 1984, during the retest, the diesel tripped on high crankcase pressure.

In an effort to troubleshoot the diesel, a technical representative from Fairbanks-Morse was called in.

The upper and lower crankcase covers were removed and the diesel was inspected for mechanical problems and debris.

No debris was found.

However, while the diesel was being turned over with starting air, one of the air start valve gaskets was found to be leaking.

The gasket was replaced.

No cylinder leakage was detected during this test and the crankcase covers were replaced.

The diesel was started again on October 22, 1984, for the 24-hour surveillance run.

The diesel ran for approximately 11 hours1.273148e-4 days <br />0.00306 hours <br />1.818783e-5 weeks <br />4.1855e-6 months <br /> and tripped on high crankcase pressure.

The diesel technical representative had the lube oil strainer cleaned and reinstalled.

The diesel was'then started for the 24-hour surveillance and completed the run without incident.

On November 2, 1984, during the normally scheduled monthly surveillance, the 2H EDG tripped on high crankcase pressure af ter approximately 20 minutes.

The crankcase air ejector was cleaned, and the diesel successfully passed the sur-veillance test.

Up to this point no clear cause for the crankcase pressure trips or the higher than normal crankcase pressures could be determined.

It was decided to instrument the 2H EDG with continuous monitors during its next run.

On December 3, 1984, during the next monthly surveillance test, additional instrumentation was installed to continuously monitor the 2H EDG crankcase pressure as well as other parameters.

The diesel failed to start in the required 10 seconds due to a problem with the air start distributor.

During a second test on December 3, the diesel successfully passed the 2-hour surveil-lance; however, crankcase pressure was slightly positive during part of the test.

Just prior to this test, the crankcase pressure switch had been replaced.

Bench testing of the old crankcase pressure switch chowed it could not be cali-brated, and that it would trip randomly at various points.

The malfunctioning crankcase pressure switch could explain earlier diesel trips which appear to have occurred below the trip setpoint.

On December 7, 1984, the 2H EDG was removed from service for preventive mainte-nance due to the higher than normal crankcase pressure observed during December 3 testing, and to inspect the air start system.

During the inspection of the 2H EDG air start system, the 2J EDG tripped on high crankcase pressure.

The plant technical specifications require an operability test on the 2J EDG at least once per 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> while the 2H EDG was tagged out for maintenance.

The 2H EDG maintenance was terminated, and the diesel was immediately returned to service since the 2J EDG was now inoperable.

13

The technical representative for Fairbanks-Morse began troubleshooting the 2J EDG.

The No. 2 and No. 3 upper pistons were found to be leaking and the No. 11 cylinder liner seal was leaking.

The Nos. 2, 3, and 11 upper pistons and the No.11 cylinder liner were replaced.

During this maintenance on December 9, 1984, the 2H EDG tripped on high crankcase pressure.

This was the sixth run of the 2H diesel within 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br />.

The crankcase pressure of the 2H EDG had been increasing and had reached the trip setpoint.

The unit commenced a reactor shutdown down since both diesels were now inoperable.

On December 11, 1984, following a 12-hour break-in run of the 2J EDG, the diesel was. started for the 2-hour surveillance test.

The diesel tripped after 6 minutes due to high crankcase pressure.

The air ejector orifice was found to be clogged by a small piece of rubber hose.

The hose was replaced and the diesel was tested successfully.

The maintenance effort on the 2H EDG found the No. 10 lower piston rings shattered.

The lower piston, complete with new rings, was replaced, and the diesel was returned to service.

The plant was then heated up and the unit placed on line on December 16, 1984.

Since a review of the above events indicated three valid start failures had occurred, the EDGs were placed on a weekly testing frequency as required by the technical specifications.

On January 13, 1985, the 2J diesel tripped on high crankcase pressure during the weekly surveillance test.

The No. 4 upper piston was believed to be leaking.

The No. 4 upper piston and piston rings and the No. 4 lower piston rings were replaced.

Further examination of the No. 4 upper piston found no indication of cracking.

The diesel was subsequently returned to service satisfactorily on January 15, 1985. A review of the failures indicated that four valid failures have occurred in the previous 100 valid tests; therefore, the Unit 2 EDGs were placed on a 3-day surveillance interval.

On March 15, 1985, during surveillance testing at 12:21 a.m., the 2J EDG tripped on high crankcase pressure and was subsequently removed from service.

Unit 2 was operating at 100% steady state power when this occurred.

Since the 2J diesel required extensive inspection and maintenance, it could not be declared operable within the 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> allowed by the Action Statement, and a unit rampdown from 100% power began at 11:45 a.m. on March 18, 1985.

The repairs of the 2J diesel were completed satisfactorily on March 19, 1985.

The unit was placed on line the following day.

(Ref. 13.)

1.9 Stress Corrosion Crackino in Nonsensitized 316 Stainless Steel On December 7, 1984, NRC issued Inspection and Enforcement Information Notice 84-89, " Stress Corrosion Cracking in Nonsensitized 316 Stainless Steel," alert-ing reactor licensees to a potentially significant problem pertaining to crack-ing of Type 316L, low carbon stainless steel jet pump inlet riser safe ends, especially in the safe end to thermal sleeve weld area.

One concern, based on the finding at Peach Bottom Unit 2* discussed below, is that cracking has never l

• Peach Bottom Unit 2 is a 1051 MWe (not) General Electric BWR located 19 miles south of Lancaster, Pennsylvania, and is operated by Philadelphia Electric.

I 14

before been reported as occurring in this location, where the use of low carbon stainless steel has been assumed to be more resistant to cracking than high carbon steel.

l On July 27, 1984, the licensee for Peach Bottom Unit 2 reported that indications of cracks had been identified on the inside diameter of the 12-inch diameter i

jet pump inlet riser (recirculation) safe ends during pipe replacement activi-ties at the plant.

Liquid-dye penetrant examination of three of the safe ends revealed circumferential indications in the safe end near the thermal sleeve attachment weld in two of the nozzles (see Figure 2).

These indications were i

about 0.25 to 0.70 inches in length. A boat sample (a scoop) containing the tip of one of the indications was removed for examination.

The safe ends are 12-inch diameter Type 316L, low carbon stainless steel, with a pipe-end wall thickness of 0.83 inches and a vessel-end wall thickness of 1.20 inches.

Ultrasonic examinations of the safe ends inboard of the thermal sleeve attach-ment weld were deferred until decontamination, because radiation levels were high in this area.

The configuration required the use of special automatic ultrasonic examination equipment.

Several probes with different scanning angles were used.

On August 14, 1984, General Electric and the licensee provided pre-liminary ultrasonic examination results for the ten jet pump inlet riser safe ends.

By August 20, it was ascertained by ultrasonic examination of these safe ends and both 28-inch recirculation suction safe ends that there were 14 shallow circumferential indications in five riser safe ends.

These shallow indications were present on the non-creviced side of the weld on all five affected safe ends, and on the creviced side of the weld on two of the five safe ends.

The majority of the indications on the non-creviced side were about 1/2 inch long and 1/16 inch deep.

The maximum reported indication on the creviced side was about 3 inches long by less than 1/8 inch deep.

One additional axial indication of undetermined length also was identified in one of these safe ends.

Ultra-sonic examinations did not identify any indications in the recirculation suction safe ends.

Independent ultrasonic examination of the jet pump inlet riser safe ends con-firmed eight indication (seven circumferential, one axial).

Additional radio-graphic testing and liquid-dye-penetrant examinations have been planned for the riser safe ends. General Electric reported that metallurgical examination of the boat sample showed indications of intergranular stress corrosion cracking, both on the non-creviced side and the creviced side of the weld.

In addition, the cracking on the non-creviced side of the weld was associated with a surface that had been upset or cold worked.

It was confirmed that the material was of low carbon content and nonsensitized.

This event raised a concern that may prove to have generic implications for the following reasons:

1.

The cracks are in low carbon stainless steel both in creviced and non-creviced locations.

2.

This represents the first reported field experience where cracking has occurred in a low carbon grade austenitic stainless steel.

3.

The design configuration at the safe end/ thermal sleeve weld location requires the use of special ultrasonic equipment and techniques.

15 t

Figure 2. Sketch Showing Riser Safe End/ Thermal Sleeve Attachment ULTRASONIC EXAMINATION PROBE TO NOZZLE SAFE END/

4 TO PIPING

NssssN\\\\\\k\\

\\

CIRCUMFERENTIAL\\

\\

INDICATIONS

\\

7.\\

,v CREVICE HEAT AFFECTED ZONE THERMAL SLEEVE WELD REACTOR COOLANT FLOW 16

4.

Laboratory test data has shown that cracks could occur at creviced or cold worked locations in low carbon grades of austenitic stainless steel.

The NRC is continuing to review relevant design and material information for all boiling water reactors.to aid in assessing generic implications.

(Refs. 14 and 15.)

l l

l l

l 17

1.10 References (1.1)

1. NRC, Preliminary Notifications PNO-III-84-102 (November 23, 1984) and 84-102A (November 26, 1984).

2. Iowa Electric Light and Power, Docket 50-331, Licensee Event Report 84-40, December 4, 1984.

3. NRC, Region III Inspection Report, 50-331/84-14, December 10, 1984.

-(1.2)

4. Mississippi Power & Light Company, Docket 50-416, Licensee Event Reports 84-53 (December 21, 1984) and 84-53-1 (February 28, 1985).

(1.3)

5. NRC, Preliminary Notification PNO-I-84-99, November 21, 1984.

6. Baltimore Gas and Electric, Docket 50-317, Licensee Event Report 84-15, December 10, 1984.

7. NRC, Region I Inspection Report 50-317/84-31, February 1, 1985.

(1.4)

8. Union Electric, Docket 50-483, Licensee Event Report 84-59, December 5, 1984.

(1.5)

9. Connecticut Yankee Atomic Power, Docket 50-213, Licensee Event Report 84-25, December 20, 1984.

10. NRC memorandum from E. Jordan, IE, to C. Heltemes, AE00, re:

Potential Appendix C Item - Inadvertent Reactor Trip from Full Power at Haddam Neck, December 28, 1984.

(1.6) 11. NRC' Region II Inspection Reports 50/327/84-38 and 50-328/84-38, February 1985.

(1.7)'12. Connecticut Yankee Atomic Power, Docket 50-213, Licensee Event Report 84-23, November 28, 1984.

(1.8) 13.. Virginia Electric Power, Docket 50-339, Licensee Event Reports 84-11-02, April 19, 1985, and 85-04, April 11, 1985.

(1.9) 14. Philadelphia Electric, Docket 50-277, Licensee Event report 84-16, August 27, 1984.

15. NRC, Inspection and Enforcement Information Notice 84-89, December 7, 1984.

These referenced documents are available in the NRC Public Document Room at 1717 H Street, Washington, DC 20555, for inspection and/or copying for a fee.

(AE00 reports may also be obtained by contacting AE00 directly at 301-<3r-4484 or:by letter to USNRC, AEOD, EWS-263A, Washington, DC 20555.)

18 k1

2. 0 EXCERPTS OF SELECTED LICENSEE EVENT REPORTS On January 1, 1984, 10 CFR 50.73, " Licensee. Event Report System" became effec-

.tive.

This new rule, which made significant changes to the requirements for licensee event reports (LERs), requires more detailed narrative descriptions of the reportable events.

Many of these descriptions are well written, frank, and informative, and should be of interest to others involved with the feedback of operational experience.

This section of. Power Reactor Events includes direct excerpts from LERs.

In general, the information describes conditions-or events that are somewhat unusual or complex, or that demonstrate a problem or condition that may not be obvious.

The. plant name and docket number, the LER number', type of reactor, and nuclear steam supply system vendor are provided for each event.

Further information may be obtained by contacting the Editor at 301-492-4499, or at U. S. Nuclear Regulatory Commission, EWS-263A, Washington, DC 20555, 2.1 Switchyard Computer Design Deficiency Causes Loss of Normal Offsite Power McGuire Unit 1; Docket 50-369; LER 84-24; Westinghouse PWR On August 21, 1984, a reactor trip was initiated by a nuclear instrumentation power range high flux rate signal.

The trip signal was caused by an electrical disturbance. induced into the station when two transmission line power circuit.

breakers (PCBs) tripped on overcurrent.

All of the Unit 1 generator outputs had been concentrated on the two lines 17 seconds earlier, when all but eight 230 kV and 525 kV switchyard PCBs and associated disconnects simultaneously opened.

The opening of 30 switchyard PCBs occurred when the switchyard operator re-enabled computer control outputs by pressing the ENABLE pushbutton on the switchyard control board.

[The control output relays should have been open before the control outputs were enabled.] The switchyard computer was being returned to service, following satisfactory operability checks which were per-formed at the completion of computer maintenance.

The incident:is classified as a' component malfunction / failure because the con-trol circuits were changed to an undesirable state without a command from the computer (a General Electric 4010 process control digital computer), apparently during computer and inverter maintenance.

A design deficiency also contributed because the computer program did not include a function to reset the computer output control circuits to a predetermined state when the computer is restarted.

Corrective actions initiated to prevent recurrence of this event include addi-tion of a control output relay test circuit to provide computer maintenance and operating personnel with co W rmation that all control output relays are open prior to re-enabling control outputs.

Also, the computer's initialization soft-ware has been modified to reset all control output relays to the "open" position on re-initialization (manual and auto-restart).

Relay settings on the two lines which tripped on overcurrent during the event have been increased to allow these lines to carry full Unit 1 output. Other related corrective actions include:

.3 1

f (1) Relay house emergency lighting will be improved.

(Low-level emergency lighting in the switchyard relay house hampered restoration of power to

-the switchyard.)

(2) Operating'and maintenance activities will be reviewed for potential impact on offsite power.

(3) Consideration is being given to the replacement of existing " latching" type control outputs with " momentary" type outputs.

(4) Independent of this incident, plans had been made to replace the existing switchyard computer by December 1986.

This schedule has been accelerated.

These corrective actions will be completed by December 1985.

2.2 Procedural Inadequacies and Equipment Failure Contribute to Reactor Scram on High Reactor Pressure While in Hot Standby Quad-Cities Unit 2; Docket 50-265; LER 84-10; General Electric BWR On the evening of October 24, 1984, the Unit 2 night shift Control Room Operator was assigned the task of bringing the unit to the hot standby condition from power operation, per procedure QGP 2-4.

At 5:46 a.m. on October 25, this pro-cedure was completed through step 0.35, and all outboard main steam isolation valves (MSIVs) were closed.

Once the outboard MSIVs were closed, it was observed that reactor pressure started to increase.

In an attempt to control pressure, additional control rods were inserted into the core.

Since the reactor core was already subcritical, the insertion of additional rods had no effect on the increasing pressure.

A second attempt was made to control pres-sure by starting the reactor core isolation cooling (RCIC) system.

RCIC tripped on overspeed and had to be manually and locally reset at the turbine.

Concur-rently, the high pressure coolant injection (HPCI) system was started by another operator, but the reactor scrammed from a high pressure of 1044 psig at 6:41 a.m.

When the reactor scrammed, control rod K-13 remained at position 48.

Rod K-13 was successfully inserted from the individual rod scram back panel at 7:12 a.m.

One minute later, it was confirmed that all control rods were fully inserted.

The root cause of this scram was procedural inadequacy.

A contributing cause was equipment failure.

Procedure QGP 2-4 failed to anticipate a possible pres-sure increase while in hot standby, due to decay heat from the core.

This resulted in the lack of specific instructions dealing with such a condition.

' An inspection of the hydraulic control unit (HCU) 38-51, which drives control

-rod K-13, revealed that the scram discharge riser valve, EP 305-112, was closed, preventing the drive from venting properly when the scram took place.

A thor-ough investigation was conducted to determine the root cause of the manual-valve 112 being closed by re examining Work Requests, Out of Service requests subsequent to the refueling outage, the September 18 hot scram timing surveil-lance tests, accumulator alarms logged in the Unit Operator's log book subse-quent to the September 18 hot scram surveillance tests, and work performed on control rod drive modules in the vicinity of that for the K-13 HCU.

Despite all of the above measures, the reason for valve EP 305-112 being in the improper position could not be determined.

20 f

I i

.The overspeed trip of'RCIC is being investigated by Station Nuclear Engineering l

Department as per Action Item Record 4-84-27.

Personnel performance concerns centered around operator attentiveness at Unit 1 i,

during the transient on Unit 2, and Shift Supervisor responsibilities during a h

transient. Between 6:30 a.m. and 6:45 a.m., on October 25, the Unit 1 Shift Supervisor evidently had left-the controls of Unit 1 to assist with the Unit 2

' transient, although he-did return to Unit 1 three times during this period.

Corrective actions includeo revising Procedure QGP 2-4 to include necessary steps to control. reactor pressure while in hot standby.

All control room per-sonnel were instructed to make a thorough review of the_ computer rod position scan as soon as possible after a scram to detect any unscrammed control rods as well as rods at intermediate position.

The Shift Foremen were instructed to survey the position of all manual valves on the HCUs once per day, in the evening, for both units.

This surveillance will continue until the six valves that control the major lines from the HCU are lockwired and valves associated with the accumulator charging equipment are relabeled to eliminate possible confusion.

The following additional corrective actions were taken in response to some of the NRC's concerns, as mentioned in the Confirmatory Action Letter dated October 26, 1984:

(1) Friction, timing, scram and stall flow tests were performed on control rod drive 38-51 for control rod K-13 prior to its removal.

All results obtained from these tests were found acceptable.

Control rod drive 38-51 was then replaced by a new drive.

(2) The internals of the removed control rod drive were examined.

No exces-sive wear or damage was found on the seals of the drive piston, on the bushing of the stop and drive pistons, on the index tubes, and on the spring washers of the stop piston.

The seal on the stop piston was found to be damaged;_however, this has been considered to be normal wear.

(3) The Control Room Operators were instructed not to leave an operating unit unless ' control room evacuation is necessary.

If one of the units requires the help of an additional operator, either the Center Desk Operator or the Operator from the stable unit may assist, but not both at the same time.

There must always be a qualified Operator at the control, who is aware of

his responsibility, left in charge of the stable unit.

Also, all operating personnel were instructed not to attempt a shift change when a unit is experiencing a transient.

The' shift change must wait until the unit is stabilized.

The apparent reason for control rod K-13 failure to scram was discussed with all Equipment Attendants and Operators so that they can be on the alert for other incidents of a similar nature.

(4) As a result of the deviation, the licensee is reevaluating its Nuclear Stations Division Directives, and will issue any necessary revisions.

21 m.

2.3 Manual Reactor Trip on Loss of Main Feedwater Pump Due to Inadequate Surveillance of Oil Filter McGuire Unit 2; Docket 50-370; LER 84-27; Westinghouse PWR On 0ctober 25, 1984, during startup with main feedwater (MFW) pump 2A operating and pump 28 not in service, operators began changing flow from the upper steam generator main / auxiliary feedwater flow nozzles to the lower main flow nozzles.

As the MFW isolation valves were opened, MFW header pressure dropped about 30 psi.

Operators then increased MFW pump speed to restore header pressure.

Pump speed increased rapidly, beyond the controller setpoint.

MFW pressure rose proportionately, and pump 2A tripped on high discharge pressure.

(1435 psi).

With pump 2B not operating, the turbine generator tripped on loss of both MFW pumps.

The operators then manually tripped the reactor.

The main turbine feedwater pump speed control system uses small orifices to admit control oil to the speed control servomotors.

The pressure of the con-trol oil is governed by the amount of leakage through four " cup" valves.

As control signals vary, the seating force on these cup valves changes, allowing more variance in the leakage and pressure in the control oil to the servomotors.

The feedwater pump trip was caused by water in the control oil system.

The water could have caused the overspeed condition (1) as it entered any of the orifices causing pressure surges, or (2) as it entered a cup valve causing its back pressure to change suddenly.

The water was discovered in the recently installed Nugent filters on the con-trol oil system.

Previous problems with the orifices being clogged with trash had resulted in the installation of the filters.

Instrument and Electrical (IAE) personnel drained approximately 2 to 2.5 gallons of water from the in-service filter and more than 2 gallons from the standby filter.

If more than 1.25 gallons of water settles in the bottom of the filter, the water will flow.

with the oil into the control systems.

It is believed that the water got into the lube / control oil system by leaking past the turbine shaft seals.

This leakage is higher during periods when the turbine pump is tripped and cooled down, as it was during the period prior to this startup event.

The lube / control system is periodically checked by knowl-edgeable operations personnel.

There were no procedures of administrative con-trols to do this.

The only provision for checking water in the filter is by drawing a sample from the drain.

An indication such as a sight glass may have been helpful in determining the water problem and also preventing future prob-lems.

Operations personnel were aware of the water problem and had been peri-odically checking the filter and draining the water as necessary.

After the water was drained from the filters, the turbine pump was restarted without problems and McGuire Unit 2 resumed startup.

Water samples now are being taken at regular intervals from the filters, to determine how often the water should be drained.

When this frequency is determined, the filter surveil-lance will be added to the routine surveillance checklist.

Also, a determina-tion will be made whether a sight glass is viable for this system.

22

2.4 Sulfur Dioxide Gas' Release Near Site Crystal River Unit 3; Docket 50-302; LER 84-21; Babcock & Wilcox PWR On October 27, 1984, Unit 1 and 2 personnel informed the Unit 3 Nuclear Shift Supervisor of a leaking sulfur dioxide tank on the Unit 1 and 2 premises (two fossil units located just west of the Unit 3 site).

An unusual event was declared due to the proximity of the leaking tank to the Unit 3 site, coupled with the hazard presented by sulfur dioxide gas.

(Sulfur dioxide is technically classified as an irritant and not a toxin.)

The Unit 3 control, complex ventilation system was immediately placed in the emergency recirculation mode as a precautionary measure.

Personnel access to the west side of the Unit 3 site was restricted even though the wind was from the southeast (140 degrees), tending to blow the sulfur dioxide gas northwest and away from Unit 3.

Additionally, the turbine building ventilation system fans located on the west side of the turbine building were secured.

About five. hours after the initial notification from Unit 1 and 2 personnel, Unit 3 was notified that the sulfur dioxide leak had been stopped, and the unusual event was terminated. All ventilation systems were restored to normal status and normal plant access was restored.

The peak sulfur dioxide reading during the event was 3.45 ppm, which is below the short-term threshold of toxicity for sulfur dioxide.

In response to this event the licensee has initiated a task force to investi-gate this release and to determine appropriate measures necessary to preclude

. future occurrences of this and similar events.

A supplement to this report will be made when the results of the task force evaluation are available.

These are expected to.be available by June 30, 1985.

2.5 Incorrect Valve Lineup Following Surveillance Leads to' False Steam Flow Indication in Reactor Protection System Haddam Neck; 50-213; LER 84-24; Westinghouse PWR On November 10, 1984, while the reactor was at nominal 25% of full power for a steam generator chemistry hold, it was discovered that one portion of the reactor protection system (RPS) was inoperable.

This portion was the steam /

feedwater flow mismatch coincident with low steam generator level reactor trip logic.

The problem was observed as indicating less than normal steam flow.

It was determined that the steam pressure density compensation line to each of the four flow transmitters was valved out.

The pressure compensation circuit within the main steam transmitter corrects

-the transmitter output for variations in the steam density (pressure).

As the steam pressure increases, the density of the steam increases, thereby increasing the mass flow rate for a given differential pressure.

As plant load increases, the steam generator outlet pressure decreases from 920 psia at 0% power to 690 psia at 65% of full power.

From 65% to 100%, the steam pressure remains constant.

The density compensation circuit uses the measured steam pressure at the flow element and the differential pressure to determine the correct mass flow rate.

23 i

^

Closure of this pressure compensation line during the event caused the trans-mitter to "see" 300 psig as the value of compensation (the lowest point of the compensation range). When the plant was operating at nominally 25% power, the steam generator outlet pressure was 770 psig.

The transmitter compensation circuit calculated the density as though the pressure was 300 psig.

Since the density of the steam at 300 psig is less than at the actual pressure of 770 psig, the indicated steam mass flow rate was less than the actual.

Prior to the power ascension on November 10, both steam flow and feed flow read normally at approximately zero.

During the power ascension to 25% of full power, the steam flow followed the feed flow but steam flow was less than feed flow (an abnormal relationship).

Plant management interviewed ten operators who looked at the~ steam flow / feed flow trace.

They failed to notice the signifi-cance of the relative difference of the steam and feed flow rates.

During the August 1984 refueling outage, the steam flow transmitters had been recalibrated.

The isolation valves for the compensating circuits must be closed

.for calibration'of the transmitters.

Although the I&C surveillance procedure calls for a return to operational positions for all valves, it does not specify valves-by number (no number exists) or by function.

Subsequent to the trans-mitters' calibrations, the valve lineups were verified as correct by one I&C Foreman and three I&C specialists.

No department approved work orders had been authorized since the calibration on the steam flow transmitters.

The I&C Department personnel have been interviewed, and no one had operated these valves since August 14, 1984.

The failure cause is personnel error, in that the I&C Department personnel failed to identify that the valves were actually closed instead of open.

The immediate corrective action taken at the time of discovery was to maintain continuous surveillance of the steam /feedwater/ steam generator level control panel and valve in the pressure (density) compensation line.

Additional cor-rective action planned is to develop a procedure to check the functional opera-bility of all RPS inputs after an extended outage prior to startup.

2.6 High Reactor Water Conductivity Due to Resin or Air Intrusion Near End of Extended Operating Cycle Duane Arnold; Docket 50-331; LER 84-41; General Electric BWR After_ putting a freshly precoated condensate /demineralizer on line on November 11, 1984,~ reactor water conductivity increased to 14 micromho/cm.

As required by

~ technical specifications, a plant shutdown was initiated.

This was cancelled when conductivity was restored to below 10 micromho/cm.

Later disassembly and inspection of the demineralizer found some septums caked with resin.

After subsequent cleanup and unit startup incorporating venting and more gradual flow increases, demineralizer performance returned to normal.

This conductivity excursion was the result of resin or air intrusion while

. putting a condensate /demineralizer (C/0) in service.

Duane Arnold has in tM past experienced milder resin intrusions.

The frequency of these intrusion; was greatly reduced following a 1983 study of the system which resulted in hardware and operational changes.

However, mild resin intrusions were aga#n noted during the month preceding this occurrence.

During this event, the t.'/D 24 m

a

c

- was operated in accordance with the updated operating instructions, i.e., flow through the C/D was increased to an interim setting and the system was allowed to equalize.

Operational testing of one C/D was unable to identify a specific problem.

The unit was backwashed, disassembled, and inspected.

Resin was still found caked on the septums and some septums were bowed.

The unit was cleaned and the bowed septums were replaced.

Outlet conductivity has been excellent through subse-quent startup and full flow operation of the unit.

Disassembly and inspection of the remaining four units has been initiated on an individual basis.

Inspec-tion of another C/D has found some septums with breaks in the mesh and one septum split open along a seal weld.

The plant is presently nearing the end of an extended operating cycle.

This extended run is considered to have contributed to the resin caking problem.

It is noted that the septums are not rigid and are susceptible to some flexing which can crack the resin cake or even result in damage to the septum.

Either condition can lead to resin intrusion.

The possibility of air in the C/D system which then enters the reactor vessel was raised because main steam radiation was higher than would be expected due to resin breakdown alone.

This is con-sidered likely because of a transient experience prior to this startup due to loss of the auxiliary transformer (see LER 84-40).

The C/D operating instruc-tions will be revised to augment venting during system startup.

Consideration will be given to using even smaller flow increments when starting up or shutting down a C/D unit.

2.7 Manual Reactor Scram Due To Main Generator Rotor Collector Ring Arcing Yankee-Rowe; Docket 50-029; LER 84-17; Westinghouse PWR On November 12, 1984, while the plant operating at 61% reactor power, and during load escalation following main condenser tube cleaning, a manual reactor scram and turbine trip were initiated due to main generator rotor collector ring arcing.

The boiler feed pumps (BFPs), required to automatically trip upon reactor scram at power levels of 15 MWe and greater, were manually tripped approximately 5-to 10 seconds after i.atiation of reactor scram and turbine trip.

Deterioration of a turbine steam pressure sensing line and pressure switch, later determined to have failed, were the cause for failure of BFP automatic trip initiation.

The pressure switch was a J6 Model 9575 manufactured by United Electric Controls Company.

Shortly after initiation of plant shutdown, breakers for the 2400 V bus No. 1 and 480 V bus 4-1 (energized through bus No. 1) were opened, as required.

Loss of bus 4-1 initiated automatic start of diesel generator No. 2, as required.

Upon stabilization of plant system parameters,-busses 1 and 4-1 were cross-tied to respective sections of the 2400 V and and 480 V busses and diesel generator No.-2 secured from operation.

- The post trip review of this event determined that, with the exception of steam pressure sensing line deterioration and pressure switch failure (making manual

. trip of the BFP necessary), all systems, components, and structures performed as intended.

25 m

A collector ring brush problem possibly related to brush material, spring pres-sure, or faulty pigtail was determined to have induced rapid heating and deteri-oration of the collector ring surface and some melting of the brush holders.

Repair of the collector ring surface and brush housing included resurfacing of the collector ritg and replacement of damaged brushes and brush holders.

Fur-ther corrective action included detailed training for electrical maintenance personnel of the collector assembly; procedural revision providing enhanced detail for records of the collector ring, brushes, and assembly; and increased surveillance frequency of the collector assembly.

Elimination of the failed pressure switch from the BFP circuitry was accomplished via design change. Additional corrective action included procedural revision regarding the BFP circuitry modification, a proposed design change for BFP trip permissive to be initiated by main generator electrical output, and repair of the steam pressure sensing line during the next high pressure turbine maintenance.

2.8 Reactor Trip on MSIV Closure Due to Seal Degradation of Air Operator Cylinder Millstone Unit 2; Docket 50-336; LER 84-11; Combustion Engineering PWR On November 15, 1984, while the plant was operating at 100% power, the Control Room Operators noticed the No. 1 steam generator main steam isolation valve (MSIV) 2-MS-64A had moved from its fully open position.

This was indicated by dual lights (red-open and green-closed) on the control board.

A plant equip-ment operator was dispatched to determine the cause of the dual indication.

Just before the operator entered the room containing the MSIV, the valve indi-cated fully shut, and the reactor tripped due to thermal margin / low pressure.

The MSIV is a 36-inch diameter, swing check valve manufactured by Atwood and Morrill.

Investigation revealed seal degradation in the air operator cylinder (Miller, Mode 1 No. 748) of the check valve.

This caused the disc to move down into the steam flow, stopping the flow.

The air operator cylinder and test cylinder were replaced.

The unaffected MSIV air operator cylinder and test cylinder were also replaced.

Currently the air operator cylinder seals are replaced every refueling outage.

The cause of the seal degradation is not readily apparent.

The seals appeared

-dry and brittle.

A contributing factor to the degradation may have been the high area temperature near the cylinder.

Further examinations of ambient con-ditions and possible corrective actions will be explored.

2.9 Potential Failure of Safety-Related Battery Racks Due to Inadequate Vendor Specifications LaSalle Units 1 and 2; Dockets 50-373 and 50-374; LER 84-80; General Electric BWRs On November 20, 1984, the seismic analysis of Units 1 and 2, Div. I and II, 125 V de and 250 V dc Gould Batteries was declared invalid.

Gaps between the battery cells and the battery rack end stringers of greater than 1/4 inch, could not be concluded to be acceptable or unacceptable based on vendor seismic testing.

Calculations performed by the architect engineer, Sargent and Lundy, 26

p

[

L I

-indicated.that the batteries could slide in the rack if not restrained by the

(:

end stringers.

The batteries were declared inoperable, and a Generating Station

~ Emergency Plan Site Alert was initiated.

Unit 2 was already in the process of reducing power, and this shutdown was continued.

The cause of.the excessive gaps was initial construction.

The battery rack dimensions per the vendor drawings did not specify a maximum gap between the battery cells and rack end stringers.

Seismic testing by the vendor with 1/4 inch gaps used a higher acceleration value than LaSalle's seismic requirement.

Therefore, the vendor could not conclude acceptability or unacceptability of gaps greater than 1/4 inch.

The gaps were reduced to the specified tolerance using sheets of fire retardant plywood.

Station Nuclear Engineering Department and the vendor, Gould, Inc.,

determined that this would restore the batteries' operability with respect to seismic requirements.

This configuration was actually used in the vendor's seismic test.

The following modifications were issued to devise a long-term solution:

MODEL NO.

BATTERY EPN SERVICE M-1-1-84-124 1DC14E Unit 1, Div. II, 125 VDC M-1-1-84-125 1DC07E Unit 1, Div. I, 125 VDC M-1-1-84-126 1DC01E Unit 1, 250 VDC M-1-2-84-174 20C14E Unit 2, Div. II, 125 VDC M-1-2-84-175 2DC07E Unit 2, Div. I, 125 VDC

.M-1-2-84-176 2DC01E Unit 2, 250 VDC 2.10 Manual Reactor Scram on Failure of Condensate Booster Pumps During Unit Startup Browns Ferry Unit 3; Docket 50-296; LER 84-12; General Electric BWR On November 20, 1984, during preparation for performance of relief valve (RV) functional surveillance testing, the licensed reactor operator manually scrammed the unit.from about 4.5% power. Master Refueling Test Instruction requires performance of the functional surveillance instruction.

In preparation for the test, the pressure setpoint controller was lowered to 270 psig, thereby opening approximately 1.5 bypass valves.

The reactor vessel water level began decreasing due to the increased steam flow through the bypass valves.

The licensee tried to start condensate booster pumps 3B and 3C to increase feedwater flow, but was not successful.

The pumps' local start switches were found to be set in the SAFE /STOP position.

Plant procedure did not have a checkoff verification for proper position for the start switches.

The Operator felt that raising the pressure setpoint by closing the bypass valves would cause a pressure spike and drive the reactor water level lower. When the reactor pressure vessel level indication by the control room level instruments decreased below plus 11 inches without an automatic scram, the Shift Engineer directed the Unit Operator to manually scram the reactor.

The control room level instruments do not input to the reactor protection sys-tem.

Independent scram switches are used for the reactor protection system automatic scram for reactor vessel low level.

Reactor level instrumentation 27

isicalibrated for rated temperature and pressure.

At less than rated condi-

'tions,' indicated level and actual level may not agree exactly.

Nuclear Engi-neering calculations show that the actual level at the time of the scram was 12 inches and not 8 inches as indicated by the control room instruments.

The scram instrumentation had been calibrated the morning prior to.the event and would have actuated to provide the required low water level scram.

These instruments'were again checked following the event and verified to be within calibration.

-Procedural inadequacy relating to standby lineup of the condensate booster pumps was the root cause of the event.

Necessary procedures for startup have been revised.

2.11. Reactor Trip on Overtemperature Differential Temperature Due to Faulty Loop Power Supply Card McGuire Unit 2; Docket 50-370; LER 84-31; Westinghouse.PWR On November 24, 1984, the Unit 2 tripped when a 2-out-of-4 overtemperature

-delta T (0 TDT) reactor trip signal was generated.

The OTDT reactor trip signal occurred when channel one OTOT was in its test position, and a spike occurred in the channel four OTDT setpoint, satisfying the 2-out-of-4 trip logic.

Chan-nel one was in its test position because of earlier problems with power range (PR) nuclear instrumentation (NI) 41.

The channel four OTOT spike is believed to have been caused by pressurizer pressure channel four failing, due to a faulty loop power supply card in the 7300 process control system (PCS).

A: component failure / malfunction was the cause of the event because the loop power supply card (NLP) for pressurizer pressure channel four failed.

A con-tributing factor to this event was the failure of.PR NI 41.

The problem with PR NI 41 resulted in placing channel one of OTDT in its trip condition at the time.of the event.

It is believed that a major contributing factor in the NLP card failures is overheating in the 7300 PCS cabinets.

In June of 1984, maintena'nce personnel rebalanced the air flow in the control room ventilation system to provide addi-

.tional cooling to the 7300 PCS cabinets.

In the five months prior to the air

' balancing, 35 card failures occurred.

In the five months since the air balanc-ing, 13 card. failures have occurred.

Initial indications are that the improved cooling has increased the 7300 PCS card _ reliability.

When the control room ventilation system has failed in the past, signals from the 7300 PCS have been erratic. An increased rate of erratic signals occurs for over a month after the ventilation is restored.

This observation ~ emphasizes the importance of adequate cooling for 7300 PCS card reliability.

McGuire Unit 2 has had problems with NI PR 41 since unit startup.

The prob-lems with this instrument loop usually disappear by themselves.

The licensee believes the problems result from a poor connection in a cable, connector, or the detector.

The cables, connections, and the detector in this instrument loop will be repaired or replaced during the upcoming Unit 2 refueling outage.

It should be noted that during the post trip review, it was discovered that a setpoint limiting jumper was not installed on the process control lead lag cards 28

- for the Unit 2 channel A and D overpower delta T (0PDT).

The OPDT problem is detailed in LER 50-370/84-30, below.

2.12 Overtemperature Delta T Instrumentation Channels Inoperable Due to Absence of Jumpers on Lead / Lag Cards

.McGuire Unit.2; Docket 50-370;-LER 84-30; Westinghouse PWR

- On November 26, 1984, following a Unit 2 reactor trip on November 24 (see LER 50-370/84-31, above), it was discovered that jumpers on two lead / lag cards which determine overpower delta T (0PDT) setpoints in two of four instrumentation chan-nels were not installed, thus. degrading the ability of the channels to perform their protective function. The type of cards involved have many applications in.the instrumentation at McGuire; none of~which besides OPDT requires that the jumpers be installed.

Deficient vendor drawings and documents are considered to have been the cause of the condition, in that explicit guidance was not given regarding the need for the jumpers in that application.

Corrective actions include verification that similar jumpers on Unit 1 were installed properly, and checks to ensure that current testing procedures are adequate to detect omission of jumpers in other applications.

2.13 Loss of Feedwater Heating Procedure Nonconservative with Analysis LaSalle Units 1 and 2; Dockets 50-373 and 50-374; LER 84-85; General Electric BWRs On November 29, 1984, it was determined via a telephone conversation between LaSalle, Nuclear Fuel Services, and General Electric (GE)-personnel that the present limit of a 150*F temperature change due to a loss in feedwater heating prior to initiating a manual scram, as dictated by procedure LOA-FW-01 (Loss of-Feedwater Heater), was non-conservative with respect to a recently received analysis.

Although such a temperature change has not occurred at high power levels, the potential existed for operation outside a bounding analysis.

In October 1983, LOA-FW-01 had been revised to procedurally allow reactor opera-tion to continue with feedwater temperature decreases of up to 150 degrees F during transients.

This change was based on a GE response to an FSAR question.

The discrepancy between the current analysis and the previous analysis is that the earlier analysis is only applicable to events occurring from 100% flow and power conditions.

The current analysis covers events initiated from 100% flow, and less than 100% power conditions.

A revision was made to LOA-FW-01 on Nnvember 29, 1984.

The feedwater tempera-ture change (due to a loss in feedwater heating) at which the manual scram must be initiated was changed from 150 F to 100 F.

Since GE has stated that the computer model used to perform their analysis is known to be conservative for the. loss of-feedwater heating event, they are presently performing this analy-sis using a three-dimensional steady state model.

The results of this new anal-ysis and any subsequent station procedure revisions based on the results will be labeled by AIR 1-84-67192.

29 m

2.14 Containment Penetration Leak Rate Tests Improperly Performed Due.to Procedural Deficiency Catawba Unit 1; Docket 50-413; LER 84-30; Westinghouse PWR On November 29,'1984, with Unit 1 in hot standby, it was discovered that cer-tain containment penetration valve leak rate tests might have been performed without the penetration piping being completely drained.

(These tests had been completed on July 6, 1984.)

If the associated penetration's piping was par-tially_ filleo when the valve leak rate test was performed, inaccurate leak rate data could have been obtained.

In order to accurately verify the sealing integrity of the containment penetra-tions :in question, the associated piping was completely drained and all but one of the containment penetrations in question were retested.

The remaining con-tainment penetration was retested prior to Unit I reentering hot shutdown.

This incident.is classified as an administrative / procedural deficiency.

The procedure used for draini.ng the applicable penetration's piping was not adequate in that it did not' allow the associated piping to be completely drained prior to performing the applicable valve leak rate test.

Corrective procedural changes were made.

2.15 CCW Surge Tank Relief Valve Inoperable Due to Blank Flange on-Outlet, Possibly Present Since Construction Davis-Besse Unit 1; Docket 50-346; LER 84-19; Babcock & Wilcox PWR On November 8, 1984, with the unit in cold shutdown, Operations Personnel reported,that the pressure in the component cooling water (CCW) surge tank had increased to 26 psig while they were adding water to the system.

The tank is supposed.to have relief protection at 5 psig (even though the tank is rated to 100 psig).

The plant was in a defueled mode at the time.

An initial check of the relief valve system found that the three-way pressure control valve CC1412

_ was-aligned to relieve through valve CC3602 to the miscellaneous waste drain tank.

Further investigation found a blank flange installed on the outlet side of-CC3602.

On November 30, 1984, the licensee determined that the surge tank was not ade-quately protected against overpressurization failure in this condition.

Since.

both CCW loops share a common surge tank structure, which has internal separa-tion, it is possible to have a failure that affects both CCW loops.

The CCd system's safety function is to supply cooling water to reactor auxiliaries and emergency core cooling system components.

Therefore, this event was being 1

reported as a condition that alone could have prevented'the fulfillment of safety functions of systems needed to removed residual heat.-

Valve CC3602 was not passing flow because of the blank flange installed on the outlet of the valve.

A search of past maintenance and testing did not find a record of an activity that would have installed the flange.

It is probable therefore, that this condition could have existed since plant construction.

. (Initial criticality for Unit 1 was August 12, 1977.) Previous ASME Section XI Pump and Valve Testing Programs have not included either of the relief valves 30 m

on the surge tank.

This was noted during the review of the recently approved pump and valve test program, and both valvas now require ASME valve testing.

On November 8, 1984 when the overpressurization was noted, the three-way control valve was manually selected to the normal relief path through CC1643 to the atmosphere, which relieved the excess pressure.

On November 11, the blank flange was removed.

Both relief valves are now part of the ASME Pump and Valve Test Program.

Both were also successfully tested on December 21, 1984.

Any valve not tested in accordance with the ASME Pump Valve Test Program will be inspected.

2.16 Anticipatory Reactor Trip on Generator Field Breaker Trip Caused by Wire in Amphenol Connector Oconee Unit 1; Docket 50-269; LER 84-06; Babcock & Wilcox PWR On December.2, 1984, the unit tripped from 43% full power when wire EHC6 became loose and opened relay 41MXa.

The opening of relay 41MXa caused generator lockout relays 86GA and 86GB to open.

The generator field breaker also opened, causing a subsequent turbine trip.

The tripping of the turbine caused an anticipatory reactor trip.

During the trip recovery, main steam relief valves (MSRVs) 2 and 10 did not reset properly. As a result, main steam pressure had to be dropped to approxi-mately 850 psi to reseat these valves.

After the reseating of MSRVs 2 and 10, the main steam pressure was raised to approximately 900 psi.

These MSRVs were tested and their setpoint was reset.

Attempts were made to find the reason for the opening of the generator field breaker.

No problem was found with the breaker but as an extra precaution, a decision was made to replace it.

The reactor was taken critical.

After the critical data had been taken, power was increased to 15% at less than 10% per hour. While power was increasing, the new generator field breaker was installed.

The breaker was then closed, but the control room indication still showed that the breaker was open.

Further investigation revealed that wire EHC6 was loose at an Amphenol Connec-tor in the electrohydraulic (EH) junction box.

As this wire was tightened, the generator field breaker indication in the control room showed that the breaker was closed.

Wire EHC6 being loose at an Amphenol Connector in the EH junction box had caused relay 41MXa.to open. When this happened, relay 62X/G1 and generator lockout

' relay 86Ga opened. When relay 62X/G1 opened, a timer circuit to open printed circuit boards (PCBs) 20 and 21 was started.

PCBs 20 and 21 connect the Unit 1 generator to the switchyard.

After the timer circuit timed out, PCBs 20 and 21 opened and isolated the Unit 1 generator from the switchyard.

The generator field breaker tripped when relay 86GA opened. When PCBs 20 and 21 opened and the generator field breaker opened,-the turbine trip coil was energized and the turbine tripped.

This caused an anticipatory reactor trip.

The MSRV behavior observed in this event has also been observed in several pre-vious trips.

The cause of the delayed reseating is not fully known at this time but is under continuing evaluation.

Possible contributors to this behavior 31 1

L

are uncertainties in lift setpoint (lifting at low pressures) and blowdown set-point settings and the proximity of the reclose pressure and the post-trip tur-bine header pressure control setpoint.

This behavior is being monitored and evaluated to identify needed corrective action.

Improved MSRV lift setpoint adjustment procedures were recently implemented.

Their effect on MSRV perfor-mance will be evaluated.

2.17 Potential Loosening of Control Rod Drive Mechanism Guide Screws Due to Fabrication Error Catawba Unit 1; Docket 50-413; LER 84-29; Westinghouse PWR McGuire Unit 2; Docket 50-370; LER 84-32; Westinghouse PWR On December 5, 1984, Duke Power was notified by Westinghouse of an event at a foreign reactor concerning the control rod drive mechanism (CRDM) design simi-lar to the one installed at Catawba Unit 1 and McGuire Unit 2.

Based on initial information, it was considered by Westinghouse to be an isolated event.

On the af ternoon of December 6,1984, Westinghouse notified Duke of the results of inspections of several plants.

It was determined that the potential existed for a guide screw in the CRDMs to work loose and possibly jam in the gripper coil mechanism.

This could cause it to not drop the control rod in the event of the reactor trip.

The guide screw is 0.52 inch in length and 0.433 inch in diameter.

If this guide screw unscrews from the drive rod assembly, then it will-fall down the annulus between the external breech and the rod travel housing.

This annulus is nominally 3/8-inch

. ide, but the drive rod is not rigidly held in the rod travel housing, and its w

' flexibility allows the screw to fall down the annulus where it would land on top of the latch assembly guide tube.

The screw cannot pass this point because the. clearance between the guide tube and the drive rod assembly is reduced to 0.055 inch at this point.

However, the screw can cause binding and misstepping of the rod if it becomes wedged in this annulus.

In_ order to lock the guide screw into position, a pin is intended to intersect the mating threads.

This pin is welded to the guide screw to hold it in place.

The problem occurs when the pin fails to intersect the threads and the screw is free to rotate.

On December 15, 1984, an inspection was performed on all of the Catawba Unit I drive rods.

There were 14 unacceptable drive rods found.

The 14 drive rods have been replaced and were returned to Westinghouse for repair.

These drive rods were removed from Unit 1 and replaced with good drive rods from Unit 2.

The rods will be repaired by Westinghouse.

The proposed repair for the rods that the screw did not come completely ort of is to drill another hole in the screw a minimum of 90 degrees from the original hole and insert another pin.

The screws that can be removed will be replaced with a new screw and pin that will provide proper thread engagement.

McGuire Unit 2 has completed preoperational testing, startup testing and more than one year of power operation with no occurrence of rod drive interference from loose guide pins.

At the time of the Westinghouse notification, the unit had less than 2 months of power operation left before the end of the current fuel: cycle.

Duke increased performance of rod movement checks required by tech-nical specifications from a monthly to a weekly basis throughout the remainder of the Unit 2 cycle.

32

2.18 ' Procedural Inadequacy and Faulty Equipment Result in Automatic Reactor Scrams on' Spurious Upscale Trips of Intermediate Range Monitors

Brunswick Unit-1; Docket 50-325; LER 84-34; General Electric BWR On December 7, 1984, an automatic' reactor scram and reactor protection system actuation trip signal occurred, and~was followed 1 minute later by another one,-

during a reactor.startup of'the unit which followed an inspection outage for

~intergranular' stress corrosion cracking.

During this outage, reactor. main con-denser shutdown. cooling had been used twice to supplement reactor core heat removal and reactor vessel water level control.

Following closure of the reac-tor head. vents, reactor recirculation loop temperatures increased to approxi-mately 220 degrees F.-

At the time, it was unknown to the Unit 1 Control Operators that residual reac-tor water from main condenser cooling still remained within'the reactor main steam' lines.

As-a result of the water in the main steam lines, oscillations occurred,'.as steam flow was initiated, on the following reactor instrumentation:

reactor steam flow, reactor vessel level, main steam line radiation monitoring, reactor core' delta pressure, reactor core flow, and source range period indica-

. tion.'.The oscillations caused signal noise-spikes to be induced into reactor power level intermediate range monitors (IRMs) and resultant instrument spiking on IRMs-E and H, which caused a reactor scram on 120% of instrument scale trip.

A unit reactor scram recovery was carried out in accordance with procedures, and normal plant operating _ parameters were maintained.

Following the unit scram recovery, an investigation determined that no single electrical-malfunction could have caused the reactor instrumentation oscilla-tions observed immediately. prior to the reactor; scram.

Immediately prior to the scram,-noises indicative of water hammer were heard emanating from the unit

.. reactor building main. steam line pit area.

Approximately 3,000 gallons of. water were drained from the main steam lines.

Plant auxiliary operators were then dispatched to visually inspect the main steam lines for possible damage result-ing from the incurred water hammer.

This-inspection did not reveal any damage to the. main. steam lines or their respective pipe. supports and pipe-hangers. -At the time,'it was determined the scram resulted solely from the water in the reactor main steam lines and the resultant noise spikes.

A review of the appli-cable operating procedure showed it was inadequate to ensure complete draining

-of the main steam lines following main. condenser cooling.

An appropriate-revi-sion was made to the subject procedure and actions begun to recommence.startup of'the unit.

During the subsequent unit reactor.startup on December 8,1984 while at approxi-

--mately-150 psig reactor pressure and approximately 4% reactor power,--an auto-matic. reactor scram occurred.due to a spurious instrument upscale trip of IRMs E-and H.

At the time of the scram, an operability test of the unit high pres-

-sure coolant injection system was in progress.

Reactor steam pressure was

< decreasing due to the subject testing and, in an effort to increase reactor pressure, the reactor manual control system (RMCS) was energized in order to h

withdraw appropriate reactor control rods.

Following the scram, a unit reactor scram recovery.was carried out in accordance with plant procedures and normal-

plant operating parameters-were maintained.

33

An investigation determined that spurious noise signals generated when the RMCS was energized had been picked up by IRMs E and H.

The root cause of both IRMs' E and H spurious instrument upscale trips and resultant reactor scrams is attrib-uted to a defective signal cable on IRM E and insufficiently tensioned signal j

cable connections on IRM H, which allowed spurious signal noises to be induced into their circuitry.

The investigation also revealed various problems involv-ing signal cable connectors, power supplies, and signal cables on the other IRMs.

The problems affecting the IRMs (General Electric Company Part No. 194X672G8) were resolved and the IRMs were returned to service.

Following the second event, plant engineering group personnel performed a walkdown of the unit reactor main steam lines and no problems were identified.

2.19 Flow Imbalance on HPSI Leg Due to Improperly Set Limit Switches Calvert Cliffs Unit 1; Docket 50-316; LER 84-16; Combustion Engineering PWR On December 15, 1984, with Unit 1 in cold shutdown, the high pressure safety injection (HPSI) system individual injection leg flow rates were found to be outside the allowable technical specification limits during performance of a surveillance test.

Six of the eight flow rates were outside of the 170 1 5 specification.

Additionally, on Unit 2 on June 14, 1984, seven injection leg flow rates were found to be outside the specification.

Other similar events were reported under LERs 50-317/83-64 and 50-318/83-64.

The throttle valves' limit switches were adjusted to obtain the proper flow rates on both units.

Improperly set limit switches caused the flow imbalance.

Limit switches were improperly set because of the identified difference in throttle valve stem travel when the valves open without line flow (no-flow condition) as opposed to opening the valves coincident with HPSI pump actuation (flow condition).

This difference was observed when the valves' stem travel was measured, with a micrometer, in the flow and no-flow conditions.

In the no-flow condition, the valves do not repeatedly return to the same position and travel further than in the flow condition.

Limit switch adjustments made over the past year based on no-flow stem travel measurements had induced errors in valve position setting.

These no-flow measurements will no longer be relied on to perform limit switch adjustments.

Refinement of measurement technique was developed to reduce the variance in the stem travel data.

This variance in the past may have also induced errors in valve setting.

Additional testing was performed on Unit 1 to determine what effect valve pack-ing gland tightness has on each valve in the flow and no-flow conditions.

It was determined from this testing that packing gland tightness has no effect on valve stem travel in the flow condition but does affect travel in the no-flow condition.

For purposes of normal HPSI system operation, the flow condition is the only concern since a safety injection actuation signal (SIAS) alone would cause the valves to open and the HPSI pump to start simultaneously.

If there was a loss of offsite power coincident with an SIAS, the pump would be sequenced on 5 seconds after the valves began to open.

This case presented the need for further testing to determine the effects of the " sequenced" condition.

The outcome of the test showed no significant difference between the sequenced and the flow conditions relative to final injection leg flows.

Nevertheless, a 34

design change to place the throttle valves' open signal on the same sequence step as the HPSI pump start signal has been initiated to eliminate any concern

- about system performance during the sequenced condition.

When this design change is implemented, the flow condition will be the only mode of system oper-ation. -Testing has shown that under flow conditions, valve stem travel is repeatable and, as previously mentioned, packing tightness has no effect on valve stem travel. Therefore, as packing loosens with time and valve operation, it will not affect valve travel and hence, injection leg flow rates.

Since the only maintenance performed on the Unit 2 throttle valves was to tighten packing, as a maintenance record search showed, and the valves were adjusted under flow conditions, there is reasonable assurance the valves will return to the correct position when required.

Concerning the safety implications of this event, given the limiting small break loss of coolant accident (LOCA), assuming the worst single failure and the lowest combination of three injection leg flows, and crediting no flow from the chemical volume and control system (CVCS), the flow reaching the core would have fallen short of the 495 gpm assumed in the small break LOCA analysis.

However, the hot full power moderator temperature coefficient, the peak linear heat rate, and the axial shape index have been significantly less adverse than those assumed in the accident analysis, and it is likely that some CVCS flow would exist.

A specific small break LOCA calculation using the most adverse conditions that existed throughout cycle life might, therefore, show acceptable results for peak clad temperature.

Such a calculation was not performed.

Finally, NRC test programs have shown that significant conservatism exists in the mandated LOCA methodology.

The licensee is seeking a change to the speci-fication such that a minimum additive flow rate for the lowest three injection leg flows and a maximum additive flow rate for all four legs will be specified.

~

l 2.20 Emergency Diesel Generator Fire Caused by Leaking Fitting on Fuel Injector Line Surry Unit 1; Docket 50-280; LER 84-27; Westinghouse PWR On December 18, 1984, Surry Unit 1 was preparing for startup following a refueling outage, and Unit 2 was at 89% power when at 7:05 a.m. the No. 3 emer-gency diesel generator (EDG) was started for surveillance testing.

At 7:36 a.m.,

EDG output indication went to 0%, the generator output breaker was manually opened, and the engine was shut down.

About 7 minutes later, the Operator who was sent to the EDG room to investigate reported a fire in the vicinity of the turbocharger.

The fire brigade responded and attempted to extinguish the fire

- with portable extinguishers, but several reflashes occurred.

The room was evacuated and the fixed low pressure carbon dioxide system was activated.

The fire was declared out at 8:05 a.m.

A leaking fitting on a fuel injector line allowed fuel oil to leak into the lube oil.

The lubricating oil became diluted to approximately 15% fuel oil.

The fuel oil changed the viscosity of the lube oil, causing failure of the turbocharger thrust bearings.

A small crankcase explosion and a fire in the turbocharger ensued.

An Electro-Motive Division factor representative and technical consultants were called to the plant to evaluate the cause of the fire and recommend corrective 35 L_

s.

actions.

Repairs.were made and the No. 3 EDG was returned to service on Decem-ber 20,- 1984.

Lube oil samples from the Nos. 1 and 2 EDGs were analyzed, and no fuel oil was detected.

All.EDG fuel and injector. lines will be inpsected more frequently; the lubricating oil levels and viscosities will be monitored-more closely.

4

.e 36

-3.0 ABSTRACTS OF OTHER NRC OPERATING EXPERIENCE DOCUMENTS 3.1 : Abnormal Occurrence Reports (NUREG-0090) Issued in November-December 1984 LAn-abnormal occurrence is defined in Section 208 of the' Energy Reorganization Tact of 1974 as an unscheduled' incident or event which the NRC determines is lsignificant. from the-standpoint of public health or safety.

Under the pro-visions of Section 208, the Office for Analysis and Evaluation of Operational

-Data reports abnormal occurrences to the public by publishing notices in the Federal Register, and issues quarterly reports of these occurrences-to Congress 4

in the NUREG-0090 series of documents.' Also included in the quarterly _ reports

-are' updates of some previously _ reported abnormal occurrences, and summaries-of-certain events that may be perceived by the public as significant but do not

. meet the Section 208 abnormal occurrence criteria.

No Abnormal Occurrence Reports were issued during November-December 1984.

37-

.-_n_

T 3.2 Bulletins and Information Notices Issued in November-December 1984 The Office of Inspection and Enforcement periodically issues bulletins and information notices to licensees and holders of construction permits.

During the period, 16 information notices and one supplement were issued.

Bulletins are used primarily to communicate with industry on matters of generic importance or serious safety significance (i.e., if an event at one reactor raises the possibility of a serious generic problem, an NRC bulletin may be issued requesting licensees to take specific actions, and requiring them to submit a written report describing actions taken and other information NRC should have to assess the need for further actions).

A prompt response by affected licensees is required and failure to respond appropriately may result in an enforcement action. When appropriate, prior to issuing a bulletin, the NRC may seek comments on the matter from the industry (Atomic Industrial Forum, Institute of Nuclear Power Operations, nuclear steam suppliers, vendors, etc.),

a technique which has proved effective in bringing faster and better responses from licensees.

Bulletins generally require one-time _ action and reporting.

They are not intended as substitutes for revised license conditions or new requirements.

Information Notices are rapid transmittals of information which may not have been completely analyzed by NRC, but which licensees should know.

They require no acknowledgement or response, but recipients are advised to consider the applicability of the information to their facility.

Information Date Notice Issued Title 84-48 11/16/84 FAILURES OF ROCKWELL INTERNATIONAL GLOBE VALVES Suppl. 1 (Issued to all nuclear power reactor facilities holding an operating license or a construction permit) 84-78 11/2/84 UNDERRATED TERMINAL BLOCKS THAT MAY ADVERSELY AFFECT OPERATION OF ESSENTIAL ELECTRICAL EQUIPMENT (Issued to all nuclear power reactor facilities holding an operating license or a construction permit) 84-79 11/5/84 FAILURE TO PROPERLY INSTALL STEAM SEPARATOR AT VERMONT YANKEE (Issued to all boiling water reactor facilities holding an operating license or a construction permit) 84-80 11/2/84 PLANT TRANSIENTS INDUCED BY FAILURE OF NON-NUCLEAR INSTRUMENTATION POWER (Issued to all Babcock &

Wilcox designed power reactor facilities holding an operating license or construction permit) 81 11/2/84 INADVERTENT REDUCTION IN PRIMARY COOLANT INVENTORY IN BOILING WATER REACTORS DURING SHUTDOWN AND STARTUP (Issued to all boiling water reactor facilities holding an operating license or construction permit) 38

L i

Information Date Notice Issued Title j

84-82 11/19/84 GUIDANCE FOR POSTING RADIATION AREAS (Issued to all power plant facilities holding an operating license or construction permit and research and test reactors)

)

(

84-83 11/19/84 VARIOUS BATTERY PROBLEMS (Issued to all nuclear power reactor facilities holding an operating license or construction permit)

{

)

l 84-84 11/27/84 DEFICIENCIES IN FERRO-RESONANT TRANSFOR_MERS (Issued j

l to all nuclear power reactor facilities holding an

)

operating license or a construction permit) 84-85 11/30/84 MOLYBDENUM DREAKTHROUGH FROM TECHNETIUM-99m GENERATORS (Issued to all NRC medical licensees, and radio j

pharmaceutical suppliers)

I 84-86 11/30/84 ISOLATION BETWEEN SIGNALS OF THE PROTECTION SYSTEM AND NON-SAFETY-RELATED EQUIPMENT (Issued to all nuclear power reactor facilities holding an operating license or construction permit) 84-87 12/3/84 PIPING THERMAL DEFLECTION INDUCED BY STRATIFIED FLOW (Issued to all power reactor facilities holding an I

operating license or construction permit) 84-88 12/3/84 STANDBY GAS TREATMENT SYSTEM PROBLEMS (Issued to all l

boiling water reactor facilities holding an operating license or construction permit) 84-89 12/7/84 STRESS CORROSION CRACKING IN NONSENSITIZED 316 STAINLESS STEEL (Issued to all boiling water reactor facilities holding an operating license or construction permit) 84-90 12/7/84 MAIN STEAM LINE BREAK EFFECT ON ENVIRONMENTAL QUALIFICATION OF EQUIPMENT (Issued to all pressurized water reactor and gas cooled nuclear power plants holding an operating license or construction permit) 84-91 12/10/84 QUALITY CONTROL PROBLEMS OF METEOROLOGICAL MEASUREMENTS PROGRAMS (Issued to all nuclear power reactor facilities holding an operating license or a construction permit) 84-92 12/17/84 CRACKING OF FLYWHEELS ON CUMMINS FIRE PUMP DIESEL ENGINES (Issued _to all nuclear power reactor facilities holding an operating license or construction permit, research and test reactors, and fuel facilities) 84-93 12/17/84 POTENTIAL FOR LOSS OF WATER FROM THE REFUELING CAVITY (Issued to all holders of a nuclear power reactor operating license or construction permit except for Fort St. Vrain) 39

3.3 Case Studies and Engineering Evaluations Issued in November-December 1984 The Office for Analysis and Evaluation of Operational Date (AE0D) has as a primary responsibility the task of reviewing the operational experience reported by NRC nuclear power plant licensees.

As part of fulfilling this task, it selects events of apparent interest to safety for further review as either an

(

engineering evaluation or a case study.

An engineering evaluation is usually an immediate, general consideration to assess whether or not a more detailed protracted case study is needed.

The results are generally short reports, and the effort involved usually is a few staffweeks of investigate time.

Case studies are in-depth investigations of apparently significant' events or situations.

They involve several staffmanths of engineering effort, and result in a formal report identifying the specific safety problems (actual or potential) illustrated by the event and recommending actions to improve safety and prevent recurrence of the event.

Before issuance, this report is sent for peer review

.and comment to at'least the applicable utility and appropriate NRC offices.

These AE0D reports are made available for information purposes and do not impose any requirements on licensees.

The findings and recommendations contained in these reports are provided in support of other ongoing NRC activities concerning the operational event (s) discussed, and do not represent the position or require-ments of the responsible NRC program office.

Engineering Date Evaluation Issued Subject E427 11/6/84 LICENSEE EVENT REPORTS THAT ADDRESS SITUATIONS WHICH POTENTIALLY COULD RESULT IN OVERLOADING ELECTRICAL EQUIPMENT IN THE EMERGENCY POWER SYSTEM OR PREVENT OPERATION OF THE-ONSITE POWER SYSTEM SEQUENCER This engineering evaluation provided information concerning situations involving onsite emergency power systems which could result in degradation or loss of safety-related electrical equipment when needed.

These situations were identified in referenced licensee event reports (LERs) covering events considered to have significant safety implications.

The events involved problems with (1) engineered safety loads on Class IE safety busses and their feeds at Three Mile Island Unit 1; (2) overloading of emergency diesel generators at Davis-Besse Unit 1; (3) emergency diesel generator sets at Sequoyah Unit 1; (4) overloading of diesel generators at Three Mile Island Unit 1; (5) sequencer logic circuits for the onsite power system at Palisades; (6) selected motor control centers, feeder breakers, and cables at Palisades; and (7) diesel generator loading sequences under certain accident conditions at St. Lucie Unit 1.

40

- Engineering Date

- Evaluation Issued Subject E427 Se' arches were conducted to identify additional,

. (cont'd)-

similar LERs which addressed related situations.

1 The result of these searches was that no such addi-tional recent reports were. identified..This result, along with previous NRC actions and industry reports which related to similar situations and safety con-cerns suggests that these items have been adequately addressed previously. -Although these situations and resulting concerns may exist at specific plants, this engineering evaluation concluded that they generally are-not a concern for most nuclear plants.

6-4 41

3.4 Generic Letters Issued in November-December 1984 Generic letters are issued by the Office of Nuclear Reactor Regulations, Division of Licensing.

They are similar to IE Bulletins (see Section 3.2) in that they transmit information to, and obtain information from, reactor licensees, applicants, and/or equipment suppliers regarding matters of safety, safeguards, or environ-mental significance.

During November-December 1984, one letter was issued.

Generic letters usually either (1) provide information thought to be important in assuring continued safe operation of facilities, or (2) request information on a specific schedule that would enable regulatory decisions to be made regarding the continued safe operation of facilities.

They have been a significant means of communicating with licensees on a number of important issues, the resolutions of which have contributed to improved quality of design and operation.

Generic Date Letter Issued Title 84-24*

CERTIFICATION OF COMPLIANCE TO 10 CFR 50.49, ENVIRONMENTAL QUALIFICATION OF ELECTRICAL EQUIPMENT IMPORTANT TO SAFETY FOR NUCLEAR POWER PLANTS (Issued to all licensees of operating reactors and applicants for an operating license)

\\

)

l

• Generic Letters 84-21 and 84-22 had not yet been issued during the report period.

42 t

3.5 Operating Reactor Event Memoranda Issued in November-December 1984 The Director, Division of Licensing, Office of Nuclear Reactor Regulation (NRR),

disseminates information to the directors of the other divisions and program offices within NRR via the operating reactor event memorandum (0 REM) system.

The OREM documents a statement of the problem, background information, the safety significance, and short and long term actions (taken and planned).

Copies of OREM's are also sent to the Office for Analysis and Evaluation of Operational Data, and of Inspection and Enforcement for their information.

No OREMs were issued during November-December 1984.

43

3.6 NRC Document Compilations The Office'of Administration is' sues two publications that list documents made

. publicly available through the NRC.

The quarterly Regulatory and Technical Reports (NUREG-0304) compiles bibliographic data and. abstracts for the formal regulatory and technical reports issued by the NRC Staff and its contractors.

-The monthly Title List of Documents Made Publicly Available (NUREG-0540) contains descriptions of.information received and generated by the NRC.

This-information includes (1) docketed material associated with civilian nuclear power plants and other uses of radioactive materials, and (2) non-docketed material received and generated by NRC pertinent to its role as a regulatory agency.

This series of documents is indexed by Personal Author, Corporate Source, and Report Number.

The monthly License Event Report (LER) Compilation (NUREG/CR-2000) contains Licensee Event Report (LER) operational information that was processed into-the LER data file of the Nuclear Safety Information Center at Oak Ridge during the monthly period identified on the cover of the document..The-LER summaries'in this report are arranged alphabetically by facility name

'and then chronologically by event date for each facility.

Component, system, keyword, and component vendor indexes follow the summaries.

Copies and' subscriptions of these documents are available from the Superintendent

' of Documents, U.S. Government Printing Office, (202) 257-2060 or -2171, or at P.O. Box 37082, Washington, DC 20013-7982.

l l

l 44 l

t

~ UNITED STATES rinst etass man

" NUCLEAR REGULATORY COMMISSION

• T*** *g'5 "*

- WASHINGTON, D.C. 20566 wasu o c.

• PERMIT No. G 87 OFFICIAL BUSINESS ~

PENA'LTY FOR PRIVATE USE. 4300 dg0gg078877-

~ ~

1-1ANIN411S11M1-ADM-DIV 0F TIOC

- {0 y.:&JPUB MGT.iBR-POR SU' REG WASHINGTON DC -.20555' i.

n

,rw 9

vw g

v,

-M7 y--.--a

-e-7

- - - -. - +