ML20035F745

Insp Repts 50-498/93-04 & 50-499/93-04 on 930117 & 0227. Violations Noted.Major Areas Inspected:Onsite Followup of Events,Operational Safety Verification,Maint Observations, Followup on Previously Identified Open Items & LER Followup

ML20035F745
Person / Time
Site:	South Texas
Issue date:	04/16/1993
From:	Stetka T NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV)
To:
Shared Package
ML20035F730	List: ML20035F729 ML20035F737 ML20035F745
References
50-498-93-04, 50-498-93-4, 50-499-93-04, 50-499-93-4, NUDOCS 9304220228
Download: ML20035F745 (32)

See also: IR 05000498/1993004

Text

.

.

.

.

.*

l

APPENDIX B

U.S. NUCLEAR REGULATORY COMMISSION

.

'

REGION IV

NRC Inspcction Report:

50-498/93-04

50-499/93-04

Operating License:

NPF-76

NPF-80

.

Licensee: Houston Lighting & Power Company

P.O. Box 1700

>

Houston, Texas 77251

!

Facility Name:

South Texas Project Electric Generating Station,

Units 1 and 2

Inspection At: Matagorda County, Texas

i

Inspection Conducted:

January 17 through February 27, 1993

Inspectors:

J. I. Tapia, Senior Resident inspector

,

R. J. Evans, Resident inspector

4

3

Approved:

T. FT Stetka, Chief, Project Section D

Dat'e

Inspection Summary

I

Areas inspected:

Routine, unannounced inspection of plant status, onsite

followup of events, operational safety verification, maintenance observations,

'

preparation for refueling (Unit 2), followup on a previously identified open

item, and licensee event report followup.

j

Results:

Unit 2 experienced two automatic trips during the inspection period.

-

The first trip was caused by an electrohydraulic control-(EHC) fluid

tubing failure.

The tubing failure was determined to be'an isolated

1

incident that was caused by a defective valve feedback device

(Section 2.1).

The second Unit 2-trip was caused by a startup feedwater pump trip while

at reduced power operation. This trip could have been prevented,

however, past problems with the pump were not corrected in a timely

manner. The failure to correct the pump- problems in a: timely manner was

!

identified as.a corrective' action program weakness. A.second weakness,

!

~

involving maintenance implementation practices,-was also identified

(Section 2.2).

,

Both units were required to shut down because of continuing problems

- with the auxiliary feedwater system turbine = driven pumps . (Section 2.3).

9304220228 930416

-PDR

ADOCK 05000498

PDR

G.

.

.

.

.

.

.

-

.

.

i

-

-2-

!

.

t

The failure to place a reactor coolant system delta-temperature / average

.

temperature (delta-T/T-avg) loop instrument in the tripped condition was

a violation of Technical Specification requirements. This violation was

caused by inadequate procedure development and reviewi(Section 3.1).

The use of the incorrect measuring and test equipment on a level

!

=

transmitter resulted in an engineered safety features (ESF) actuation

!

signal. The preventive maintenance work instructions did not

specifically state the correct type of test equipment to use for the

l

application. The failure to have maintenance work instructions

appropriate to the circumstances was considered to be a violation of

Technical Specification 6.8.1 requirements- (Section 3.2).

Unit 2 entered Technical Specifications (TS) 3.0.3 when power to the

digital rod position indication was lost for 16 minutes.

Contributing

3

factors to the event included:

(1) the failure of the licensee to work

i

a service request on a defective sample pump in a timely manner,.and

l

(2) discovery of a design application error involving two pumps being

-

-

connected to the same electrical panel (Section 3.3).

,

i

A violation of TS was identified involving the failure to perform

containment pressure' channel checks while in Mode 4 operation.

This was

the second violation caused by a deficient surveillance procedure

!

(Section 3.4).

i

The failure to maintain the minimum shift crew composition during Mode 4

!

operation was a violation of TS requirements. The cause of the event

was human error (Section 3.5).

A reactivity management' issue was identified when plant operators

I

accidently diluted the reactor coolant system while they were attempting

1

_to add boron to the reactor coolant' system. The cause of the event,- in

part, was inadequate understanding of boron thermal regeneration system

operation during shutdown conditions (Section'3.6).

-f

Numerous events occurred involving secondary- plant components which had-

!

=

a negative effect on primary plant components.

Few improvements have

f

been noted in this area of plant operations despite additional

,

management oversight.

One positive action taken by.the. licensee

.

!

included the development of a steam generator power operated relief-

!

valve action _ plan'(Section 4.1).

j

The licensee's essential chiller reliability and availability rates

continue:to be a concern (Section 4.2).

~

Extensive testing of the- auxiliary -feedwater-turbine-driven pump was

,

performed to verify pump operability and' availability. During the

l

testing process, one maintenance implementation weakness was identified

1

!

.

-

h

.

.

.

-

,

-

!

!

-3-

!

that resulted in unnecessary test delays. Two Temporary Waivers of

I

Compliance were needed to complete the required testing during Mode 3

-

operation (Section 4.3).

g

The Unit 2 refueling outage scope appears Eto be well planned by the

.

licensee, however, the work scope is aggressive because of-the number of

motor operated valves (MOVs) scheduled to be tested. Shutdown risk

assessment and outage management staffing. continue to be licensee.

t

strengths (Section 5.0).

.;

-

.

Summary of Inspection Findings:

,

Violation 498/9304-01 was opened (Sections 3.1 and 3.4).

,

e

>

Violation 499/9304-02 was opened (Section 3.2).

e

,

Violation 499/9304-03 was opened (Section 3.5).

e

.

Inspection Followup Item 498;499/9304-04 was opened (Section 3.6).

.

,

Inspection Followup Item 498;499/9119-02 was closedf(Section 6.1).

i

Licensee Event Report 499/92-003 was closed (Section 7.1).

!

e

!

Attachments and/or Enclosures:

I

Attachment 1 - Persons Contacted and Exit Meeting

.

!

Attachment 2 - Attendance List for Briefing

,

t

b

i

!

,

9

-!

r

.

.

.

_

_-

_

.

-

-

-

!

,

.

.

-

?

!

-

-4-

.;

i

!

1

DETAILS

j

l

4

1 PLANT-STATUS

\\

1.1 Unit 1 Status

y

t

At the beginning of this inspection period, the unit was in Mode 3 operation

following a TS 3.0.3 required shutdown that began on January 12, 1993. The

-i

shutdown was required because of time constant setting discrepancies that were

I

identified with lead / lag amplifiers in the main steam line_ pressure instrument'

loops (refer to'NRC Inspection Report 50-498/92-36; 50-499/92-36 for a

.3

-

complete description of the shutdown). On January 17; 1993, the unit entered

l

Mode 2 operation and criticality was achieved. Mode I was reached later the-

same day. The unit reached-95 percent power 3 days later. The unit was held

!

at 95 percent power to allow xenon to equalize prior to performing a neutron

flux map. The flux map was needed to monitor and evaluate power peaking

factors prior to any additional power increases.

Power was increased in

-)

increments and 98 percent power was achieved on January 26, 1993.

!

On January ~ 27, 1993, reactor power was decreased to 65 percent to increase the

i

margin for a potential reactor trip so steam generator Feedwater

l

Pump (SGFP) 13 could be removed from service to allow for repairs to be made

on the low pressure governor valve.

Power was increased to 98 percent the

i

next. day after SGFP 13 was returned to service. On January 31, 1993, the unit

!

power was reduced to 28 percent to increase the margin to.a potential reactor

!

trip in order to transfer the source of power that. supplies 120 volt AC to the

j

steam generator feedwater pump control circuitry back to its normal supply.

The transfer was completed the same day and power _ ascension commenced. The

.

unit returned to 98 percent power the next day.

On February 4,1993, the unit declared a Notification of Unusual- Event (NOVE)

!

and a plant shutdown was commenced to comply with TS 3.7.1.2 requirements.

The shutdown was required because Auxiliary Feedwater (AFW): Pump.14 was still-

1

out of service and the allowed outage time of 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> had expired. The unit

i

-

entered Mode'4 and exited the NOUE the same day. On February 11, 1993, the-

!

unit cooled down to Mode 5 to allow for repairs to be implemented on six~

{

residual heat removal suction line valves (refer to NRC Special Inspection

"

Report 50-498/93-08; 50-499/93-08 for a complete description of the residual

.

heat removal- valve repair). The unit remained in Mode 5 through. the end of

]

the inspection period.

r

i

1.2 Unit 2-Status

.,

l

At.the beginning of this inspection period, the unit was in Mode 1, increasing

!

-power from 50 percent. The unit was in the process of returning to full power'

i

following a TS 3.0.3 required shutdown that commenced on- January- 12, 1993.

j

the unit was required to shut down becausef or similiar problems encountered

with time constant settings in the main steam line pressure' instrument loops.

l

the unit reached full power on January 17, 1993. The following day, a power

'j

l

l

t

f

.~.

<

.

-.

--

-

.-

.

-

..

.

.

.

,

i

'

.

.

-

!

'

-5-

'

!

!

I

reduction was commenced to allow repairs on the three SGFPs and the pump

control circuitry power supplies.

Power was reduced to 28 percent, the.

4

capacity of the startup feedwater pump.

The unit returned to full power on

January 21, 1993.

i

4

On January 23, 1993, the unit tripped from full power on low EHC fluid

_

pressure.

The low pressure condition was caused by a tubing failure in.the

EHC fluid supply line to SGFP 22. The unit was stabilized in Mode 3 cperation

following the trip.

Thc unit entered Mode 2, followed by Mode 1, operation.on

January 25, 1993.

Power was held temporarily at 28 percent until an SGFP

.:

could be returned to service.

Power was increased as-SGFPs were returned to

_

service and full power operation was achieved on January 30, 1993.

l

On February 3, 1993, the.startup feedwater pump was started and SGFP 21 was-

secured because of a high bearing temperature on the pump. The unit tripped

j

on low-low steam generator level after the startup feedwater pump tripped _on-

high lube oil filter differential pressure. The unit was stabilized in Mode 3

.

operation. Minutes _after the unit trip, AFW Pump 24 received a start signal

-;

and subsequently tripped on mechanical overspeed. On February 6, 1993, the

unit entered Mode 4 operation because of the continuing problems with AFW

j

Pump 24 and the associated trip and throttle valve. Refer to NRC Special

!

Inspection Report 50-498/93-07; 50-499/93-07 for a complete description of the

i

AFW system problems.

~

!

On February 17, 1993, at'10:02 a.m., the unit entered TS 3.0.3 because

"

undersized fuses were found in the solid state protection system _ actuation

1

trains. Plant cooldown from 343 F was commenced to place tho unit in Mode 5

,

operation. The cooldown was. terminated at 242of when the fuses were replaced

i -

at 3:45 p.m.

The unit remained in Mode 4 operation during the cooldown

i

process. Unit heatup began at 4:30 p.m. and the unit was stabilized at 340 F

-i

at 10:30 p.m.

Refer to NRC Special Inspection Report 50-498/93-12;

-50-499/93-12 for a complete description of the_ undersized fuse issue.

Following extensive rework of the AFW Pump 24 turbine and steam supply

.

components, the unit re-entered Mode 3 operation on February 22, 1993, so the

i

pump could be functionally tested at normal operating temperature and pressure

i

conditions. The unit ended the inspection period in Mode 3 operation.

8

1.3 _ Organizational Changes-

Several organizational changes were made during the inspection period.

The

deputy plant manager was assigned to the position of maintenance department

I

manager, following the resignation of the previous maintenance department

j

manager.

The planning-and assessment manager was assigned to the position of

deputy plant manager. Additionally, the administrator of participant services

j

was assigned to the position of planning and assessment manager.

The

licensing department general' manager was assigned to the position of project

,

manager _-(temporary), diagnostic evaluation team support. A consulting-

.

,

engineer _from the corrective action group was assigned to the position of

!

<

deputy general manager, licensing department (temporary).

}

!

l

,

.

.-

_ _

-

.

-

-

- -

-

,

-

l

.

..

.

-

!

'

-

I

l

-6-

,

!

2' ONSITE RESPONSE TO EVENTS . (93702)

i

r

2.1 Unit 2 Reactor Trip

,

!

On January 23, 1993, at 11:24 p.m., the unit control room received a SGFP 22

control oil pressure low alarm concurrent with an EHC fluid _ reservoir level

i

high/ low-low alarm, quickly followed by a turbine trip signal. ' The turbine

trip signal was generated on low EHC fluid pressure. The reactor then~

r

automatically tripped from full power on a turbine trip signal ~. An AFW-

{

actuation signal was generated, as expected, on steam generator low-low level

i

because of.a mass '" shrink" in the steam generators. The. plant operators then

.;

entered the Emergency Operating Procedure (EOP) OPOP05-E0-E000, Revision-.2,

.!

" Reactor Trip or Safety Injection." Following completion of the requirements

i

in this procedure, plant operators then transferred to E0P

Procedure OPOP05-E0-ES01, Revision 3, " Reactor Trip Response." The unit was

'

l

stabilized in Mode 3 operation.

.

About 5 minutes after the reactor trip, plant operators reset the steam

l

generator low-low level actuation signal and secured AFW Pump 24 since it was

,

no longer needed to maintain steam generator levels. When AFW Pump 24 was

,

secured, the turbine mechanical overspeed trip device actuated because of

!

misadjustment in the trip linkage (further discussion of the overspeed trip

was provided in NRC Inspection Report 50-498/93-07; 50-499/93-07). AFW

l

Pump 24 was subsequently declared inoperable. All other components operated

as designed during the plant trip'and recovery activities.

!

The cause of the trip was . low fluid pressure in the turbine EHC. system. Tha

{

low EHC fluid pressure was the result of a sheared line that led to the low

'

pressure governor valve at SGFP 22. The sheared line- allowed the EHC fluid to

drain out of the system and into the housing that surrounds SGFP'22. The'EHC

l

system was configured with an EHC reservoir common to the main turbine- and all

!'

three SGFPs.

The licensee subsequently determined that the cause-of the

sheared line was the result of extremely rapid changes in_the' low pressure

governor valve position. The governor valve position changes created

l

excessive movement and vibration within the valve _and the line. : Preliminary -

l

'

results indicated the failure,' located in the vicinity of a fillet weld,

initiated from shallow intergranular cracks on the inner surface of:the line, .

and grew radially outward and then around the circumference of the-tube by-

i

fatigue. -The licensee suspected that an exposed wire discovered on the linear

i

variable differential transformer, which provides a-feedback signal' to the-

l

~

electronic controller to confirm valve position, caused intermittent

';

electrical signals that produced rapid changes in.the valve position.

The. .

[

-

defective linear variable differential transformer was subsequently replaced.

i

Corrective actions'taken by the licensee included the performance of

..

l

. nondestructive examination of the EHC piping at SGFPs 21'and 23, as.well as on-

l

all . Unit'l SGFPs.._ One suspect tube was identified on SGFP 21. - Additionally,

.;

the. housing over SGFP'22 was lifted off the pump to provide maintenance

l

workers access to the area to clean up the spilled EHC fluid.

Unit startup

.

]j

following the automatic trip began on January 25, 1993, after_AFW Pump 24'was

q

i

. .

-

.-~

-,

-.-.

-

- - - .

.

-

--

-

.

,

..

.

.

.

-

a'

.

-7-

q

!

,

returned to service.

The repair of the EHC tubing was a main turbine roll

j

restraint, but not a startup restraint. The unit entered Mode 2, achieved

!

I

criticality, and entered Mode 1 the same day.

The SGFP 22 tubing _was replaced

and postmaintenance -testing was completed on January 25, 1993, at 7:32 a.m.

l

The tubing on SGFP 21 was also replaced and the. post maintenance testing was

!

subsequently completed on January 26, 1993. The unit returned to full power-

j

on January 30, 1993.

Previous plant transients involving EHC failures were reviewed by.the

{

inspectors. On April 14,1990, Unit 2 tripped from full power when an EHC

'

.

-supply line to a main turbine governor valve failed.

The line failed because

J

of fatigue stress of a weld.

The fatigue stress was caused by governor-valve

'

induced vibration. The valve vibration was the result of valve plug rotation.

-I

On June 28, 1990, Unit 1 tripped from 76 percent power when an EHC supply line

to a main turbine throttle valve failed. Oscillations of the Number 1

!

governor valve caused the EHC supply line to the. Number 3 throttle valve

j

{ lines were connected at a common. header) to fail.

The valve oscillations

t

were the result of a loose connection in the _ governor valve control circuitry.

1

-!

The inspectors determined that the previous EHC failures were similiar to the

January 23, 1993, incident; however, the root causes were not related.

i

2.2 Unit 2 Reactor Trip

.g

At 3:26 p.m. on February 3,1993, the unit tripped from full power on low-low

l

steam generator level. Just prior to the unit trip, at 2:48 p.m., the control

room received a high journal. bearing turbine end temperature alarm for

t

SGFP 21. Feedwater Booster Pump 23 and Startup Feedwater Pump 24 were started

1

'

4 minutes after the alarm was received to permit the plant operators to remove

the SGFP 21 from service.

SGFP 21 was secured 8 minutes after the alarm was

received.

'l

At 3:23 p.m.', Startup Feedwater Pump 24 tripped while the pump lube oil

filters were being-shifted because of high differential-pressure,

Since the

j

~

capacity of the two remaining feedwater pumps was 80' percent power, plant '

~

!

operators immediately began ramping down;the main turbine. The-reactor

I

subsequently automatically tripped on low-low level in Steam Generator:2C less

than 1 second prior to a manual reactor trip.being~ inserted.

An AFW actuation?

,

signal was generated, as expected, because of_ the steam generator low-low

i

'

level ~ condition due to the mass shrink in the steam generator.

Less than-

- si

2 minutes' after receiving a start signal, AFW- Pump 24 tripped on Lmechanical

!

overspeed (further discussion of the AFW pump 24 trip on overspeed was

provided in NRC Inspection Report 50-498/93-07; 50-499/93-07).

The control

.

room operators-then_ opened the cross connect-valves to allow the motor driven

f

AFW pumps to supply. flow to-Steam Generator 2D. Additionally,.Feedwater

Isolation Bypass Valve A2-FW-FV-7148A,' which was shut prior to the trip,

'l

indicated partially open'after the trip. At 3:31,p.m., a start was attempted

_r

on Startup feedwater Pump 24, but the pump tripped again on high lube. oil.

j]

filter differential. pressure.

-i

1

'I

!

r

w

,

,

,

<-

r-

- - - -

9

- - -

.

.-

.

-

.

.

.

-

.

. ..

'

. i

,

.

1

!

'

-8-

i

.

!

i

As a conservative measure, emergency boration of the reactor coolant

system (RCS) was initiated by operators because of the potential of a loss of-

l

~

shutdown margin (end of core life and xenon considerations). Approximately

3500 gallons of boric acid was injected into the reactor from the boric acid.

storage tanks.

The unit was stabilized'in Mode 3 following the trip.

Instrument air was

isolated to Feedwater Isolation Bypass Valve' AZ-FW-FV-7148A to shut the valve.

AFW Pump 24 mechanical overspeed mechanism was reset but the pump was not

- ;

immediately restarted.

'

The cause of the startup feedwater pump trip was low lube oil pressure,. _

.

created by a high differential pressure across the lube oil filters.

The

startup feedwater pump trip due to high-lube oil filter. differential pressure

,

and has been a recurring problem. The-failure to correct the problem

'

following previous pump trips was indicative of a weakness in the licensec's

corrective action program (further discussion of this issue is provifed in

,

Section 4.1.2 of this inspection report).

. ;

i

The loss of SGFP 21 was caused by high journal bearing temperature due to

. ,

bearing failure.

The intrusion of hard particles resulted in the destruction

l

of the babbitt bearing surface and shaft scoring. The . licensee planned to

'

modify the pump shafts during the upcoming refueling outages for each unit to

minimize the damage that can result from particle ini'rusion.

SGFP 21 bearing

rework was scheduled to be performed under Service Request (SR) 167324,

l

however, the work was incomplete prior to the end of this inspection period.

Following the:Feedwater Isolation Bypass Valve A2-FW-FV-7148A failure to close

incident, instrument air to the valve was isolated and the valve failed shut.

SR 169466 was issued to troubleshoot and repair the ' valve. During the

,

troubleshooting process, the valve positioner zero; adjustment locknut was

j

found to be loose and the positioner calibration was found to have drifted

high. The positioner was recalibrated and postmaintenance testing.was

performed on February 5, 1993. A walkdown of both units was performed and a

second loose lock nut was identified, which was indicative of a maintenance-

~3

implementation weakness. The licensee. planned to revise the applicable work-

j

instructions to ensure the lock nuts were tight following maintenance

i

i

activities.

- 1

2.3 Units-1 and 2 Forced Shutdown

'

During the inspection period,:both units were required to shut.down because of

,

continuing problems withithe turbineidriven AFW (TDAFW) pumps.

A complete

description of the AFW troubleshooting and rework activities was previded in

-

NRC Inspection Report 50-498/93-07; 50-499/93-07.

r

On February 1, 1993,'an operabi_lity test of the unit TDAFW Pump 14 was

-

.

a

performed.

The pump tripped on mechanical overspeed immediately after the-

'

start attempt. The TDAFW pump was declared inoperable at 10 a.m. that da'y.

t

According to TS 3.7.1.2,' Action b, if TDAFW Pump 14 was inoperable, then the

{

i

'

-

s

4

-

sse,--

p --

.

-

. .

,

!

'

i

.

.

-

-

!

-9-

i

!

pump must be restored to operable status within 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> or the reactor be in

l

Hot Standby within the next 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> and in Hot Shutdown within the following 6

,

hours.

.

On February 4,1993, at 9:37 a.m.,

Unit I declared an NOUE and initiated a

plant shutdown from 98 percent power to comply with TS 3.7.1.2 requirements.

The shutdown was necessary because TDAFW Pump 14 could not be returned to an

operable status within the required time interval. The unit entered Mode 2 at

2:49 p.m., Mode 3 at 3:14 p.m.,

and Mode 4 at 9:32 p.m.

The NOUE was exited

at 9:33 p.m.

As mentioned in paragraph 2.2, following the Unit 2 reactor trip on

February 3,1993, AFW Pump 24 tripped on mechanical overspeed.

4

Troubleshooting of Pump 24 was performed in Mode 3; however, the licensee

l

determined they would be unable to restore the pump to an operable condition

,

'

prior to the expiration of the TS 3.7.1.2 time limit of 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />.

On February

5, 1993, at 5:52 p.m.,

boric acid was added to the RCS to support plant

cooldown to Mode 4 operation. Mode 4 was reached at 4:21 a.m. on February 6,

j

1993, and Unit 2 exited TS 3.7.1.2 prior to the expiration of the 72-hour time

limit.

'

On February 5,1993, the NRC issued a Confirmatory Action Letter to the

licensee.

The letter discussed agreement between the NRC and the licensee

'

that the licensee would brief the Region IV staff on the results of the

i

licensee's efforts to correct the overspeed trip conditions that had affected

)

the TDAFW pumps prior to taking the reactor critical for both units.

This

status briefing was not held prior to the end of the inspection period.

j

2.4

Conclusions

I

Unit 2 experienced two reactor trips during the inspection period.- One was

i

caused by equipment failure and the second one was caused by a pump trip. The

j

failure to correct a known problem with the startup feedwater pump in a timely

manner was indicative of a weakness in the corrective action program.

The

failure to maintain the locknuts tight on the feedwater isolation bypass valve

,

positioners was indicative of a maintenance implementation weakness.

Both the

units remained shutdown at the end of the inspection period because of the

,

problems with the TDAFW pumps.

j

3 OPERATIONAL SAFETY VERIFICATION- (71707)

l

'

l

i

The purpose of this inspection was to ensure that the facility was being

operated safely.and in conformance with license and regulatory requirements.

i

i

The following paragraphe provide details of specific inspector observations

during this inspection eriod.

3.1 T_S Violation Because of an Out of Tolerance Temperature Loop (Unit 1)

{

On January 13, 1993, a review of the as-found calibration data for the main

steam pressure loops was in progress. During this review, the licensee

i

!

,

l

t

-- _

__-

- _ _

_ _ ,

.

.c

'

t-

a

.,

-

!

!

!

,

!.

,

!

-10-

l

Li

>

discovered that the curve shown in a surveillance procedure addendum, that was

l

used for lead time constant calculations for the low steam pressure safety

injection signal, was not correctly drawn.

Further investigations revealed

i

the curve' was not correctly drawn in multiple procedures, -including the

I

pressurizer pressure and the delta-T/T-avg calibration. procedures. 'The

-i

incorrect curves were generated and added to the procedures when the unit

~

specific procedures were combined into common procedures.

The curves added a'

4-second error into the time constant settings.

A review was performed to determine if the error had any detrimental effect on

!

the instrument loops that were previously calibrated. . Plant engineers

!

determined that the error caused by n a curve was in the_ conservative

i

direction for the main steam loops. ..aitionally, the' time constants were not.

!

considered a concern with the pressurizer pressure loops because the time

i

constants were not included in the plant's safety analysis. However, tne

i

delta-T/T-avg loops were adversely affected under certain scenarios involving -

'!

temperature increases.

In one case, the lead time constant was found set'in

,

!

the nonconservative direction. The lead constant of delta-T/T-avg-

1

l

Loop Instrument RC-0430 was left outside the allowed tolerance of.5 percent,

j

The setpoint was found at 37 seconds, instead of the' required 33 i l.65

!

seconds. This condition existed from January 8, 1993, when the time constants

l

were set incorrectly, until January 12, 1993, when the unit entered Mode 3

-!

-

operation.

g

,

I

l

l

On January 13, 1993, a reportability review of this incident.was initiated and

!

l

the loop was recalibrated 2 days later. On January. 21, 1993, the licensee

!

determined that the event was reportable to the NRC as a violation of

l

TS 3.3.1.

In accordance with TS 3.3.1, the inoperable _ channel must be placed

in trip while in Mode' 1. or 2 operation within 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br />.

Since.the unit

_l

L

operated during January 8-12, 1993, with the loop not in the trip position, a

M

'

violation.of TS 3.3.1 occurred. The failure to adhere to TS is considered to

be a violation of' facility license conditions (498/9304-01).

l

t

The.cause of the event was inadequate procedure development'and review.

An

-

-incorrectly drawn graph was inserted-into approximately 20 procedures a M was

l

not ' detected. in a timely manner. Corrective actions taken included

1

recalibrating the out of. tolerance time constant- setpoint anci issuing field .

l

change requests to replace the incorrect; curve'in selected procedures.

The

l

licensee committed to revising all- 20 procedures prior-to April 15,1993, or

j

.

. prior to the.next use.of the selected procedure.

';

__

_

-

i

3.2 ESF Actuation During Routine Maintenance (Unit 2)

j

The essential cooling water.(ECW) system was designed to supply, adequate-

l

cooling to selected' components during various modes of plant operation. The

]

l

heat from the ECW system was dissipated in- the essential _ cooling pond.'

A'

j

~

travelling water' screen was provided upstream of.cach ECW-pump in the'ECW

,

intake structure to minimiz'e debris entering'the system which could cause

i

damageLto the' pumps or clog heat exchanger tubes. Screen wash booster pumps,:

j

which take suction from'the ECW pump discharge piping,- provide water to wash

j

i.

!

f

(

!

-

I

.- ~ ~ = .

-

.

-

..

.

--. .- .

.-

. .

-

.-..

m

~

.

.

-

.

'

!

.

1

i

-11-

.,

!

.is

each travelling water screen. A control system was provided to automatically

i

start and stop the travelling screens during normal system operation. A high

i

differential water level sensed across any travelling screen actuates an alarm

i

in the control room and automatically. starts the screen wash booster pump and,

-l

after. reaching adequate screen wash pre.ssure, starts the travelling screen.

l

The travelling screens and booster pumps were automatically activated upon

!

receipt of a safety injection signal.

'

'

1

On January 28, 1993, preventive maintenance activities were started on the

j!

Train C ECW travelling. screen differential water level transmitter.

The

maintenance technicians calibrated the transmitter with a " wet calibrator,"

which filled the transmitter capsule assembly with water.

The technicians

l

-then began venting and draining the water from the transmitter. .The venting

1

and draining evolution did not-appear to be successful, so the technician

3

'

further opened the manifold valve to improve the venting process. When'

bubbles started appearing in the transmitter vents, the technicians realized

that the level transmitter was a dry leg type of transmitter and should _ not.

,

have been calibrated with a " wet calibrator."

t

At 10:50 p.m.,

the Unit 2 control room received the Train C travelling screen

.t

-

high differential pressure and Pump 20 bay level low alarms.

ECW Screen Wash

j

Pump 2C and Travelling Screen 2C automatically started. Due to the

.

!

transmitter venting process, the differential level switch sensed a false-

signal and initiated the ESF actuation signal.

The system was subsequently

,

restored to the normal lineup.

}

The inadvertent ESF actuation signal was indicative of a weakness in

maintenance _ implementation because the incorrect measuring and test equipment .

!

was used to calibrate the level transmitter. The causes of the ' event included

y

inadequate preventive maintenance instructions;and technician implementation.

,

The maintenance instructions did not clearly specify'which . type of test

zi

equipment to use and the-technicians.were not aware of the proper. test

O

equipment for the specific application. A contributor to the event was

.

inadequate labelling at the equipment cabinet

The equipment-being tested was-

-

!

not clearly marked as ESF related equipment. The failure' to have maintenance-

work instructions- appropriate to the circumstances was _ considered-to be a -

q

violation of TS 6.8.1 requirements (499/9304-02).

Corrective action planned

-

included revision of the maintenance procedure instructions, addition of

-

training _of maintenance personnel.

~

j

labels locally at the equipment,' review of similar maintenance activities, and

~

j

q

"

.3.3

Loss of Digital Rod Position Indication'(DRPI) (Unit 2)'

'

The unit- vent monitor' samples 'the plant vent stack L for particulates,-iodine,

'and noble gases prior to discharge to the' environment. Sample pumps'are used.

j

to provide' process air to the-two unit' vent monitors. ' Sample Pump 8010A:

<

-provides air:to the particulate / iodine monitor while Sample Pump 8010E

j

,

provides air,to the gaseous effluent monitor. .Both pumps receive single phase

i

120 volt AC power from non-Class IE Distribution Panel DP-003.

i

1

l

,

,

,

e

'

.

.

..

.

.

. - .

-

..

..

t

'

i

r

w

.

-

9

.:

-12-

.

l

,

On February 3,1993, Surveillance Procedure OPSP02-RA-8010A, Revision 0, " Unit

Vent Particulate and Iodine Effluent Monitor Digital Channel Operability

-

Test (DCOT)," was commenced at 9:06 a.m.

During the. performance of

Step 7.2.16, which provided instructions to turn on the sample pump, an

j

overcurrent condition occurred on Distribution Panel DP-003. Sample

!

I

' Pump 8010A had drawn an excessive amount of current and caused DP-003 to

automatically attempt a transfer to the emergency source.

When Panel DP-003

{

lost power, an electrical maintenance technician was dispatched to. the

t

'

distribution panel. The technician found the transfer switch stuck.in

.

midposition. The transfer switch failed to transfer to either the normal or

emergency source using the transfer push buttons.

Indications associated with

l

the panel suggested the panel was being supplied with normal and emergency

power at the same time, however, the panel was not being supplied with power

j

!

from either source because of the failure of the automatic transfer switch.

The technician manually transferred the switch to the emergency source using a

j

manual transfer handle and restored power to the panel.

Several seconds after

the manual transfer, the automatic transfer switch retransferred power to the

normal source as designed.

The loss of power to Distribution Panel DP-003 resulted in loss of power to

selected unit chart recorders and the DRPI. TS 3.1.3.2 requires the DRPIEto

be operable in Modes 1 and 2, therefore, the. failure of the DRPI caused the

unit to enter TS 3.0.3 at 10 a.m.

The unit exited TS 3.0.3 at 10:16 a.m.,

l

when power was restored to Distribution Panel DP-003 and, subsequently, the

DRPI power sources.

Immediate corrective actions taken included writing

SR VC-169498 to troubleshoot and repair the transfer switch problems and

.

!

increasing the priority level of SR RA-89143, which was written in June 1992

j

because the sample pump was noted to be drawing an excessive amount'of

j

current.

The licensee suspected the cause of the transfer. switch malfunction was either

j

mechanical binding or a defective electrical contact. During the SR'VC-169498

j

troubleshooting process, no problems were found with the distribution panel

i

transfer. switch and the event was not repeatable. During the SR RA-89143

!

~!

troubleshooting process, the licensee determined that.the.two pumps were

.

overloading the distribution panel. under certain operating conditions. 'With

!

one sample pump in service, the start of a second sample pump, which.was

j

connected to the same electrical panel, would cause- an undervoltage condition

-to occur on the panel. A pump trip, or panel power supply transfer; would

occur on the resultant undervoltage condition.

!

I

The causes of the DRPI~ power supply failure were the result,'in part, of a

design misapplication and.the failure to. work'an.SR in a timely manner. The

addition of both sample pumps on one panel was causing the panel to experience

electrical . transients, including pump trips on undervoltage. One of the

corrective actions planned was to issue a plant change form to remove. one' pump

from Distribution Panel DP-003 and install it on Distribution Panel DP-004,

)

which was located about 10 feet away.

Additionally, if SR RA-89143 had been -

j

worked in a timely manner, the event may have been prevented.

Other.

j

in

i

h

!

-

'

.

.

_

..

.

- .~

.

- - .

.

-

.

..

.

.

!

.~

]

'

.

. .

.

.

a

,

f

-13-

e

i

!

corrective actions planned included the development of a preventive

!

maintenance procedure for this transfer switch and other similar switches.

!

.

.

!

3.4 Failure to Perform Containment Pressure Channel Checks (Unit 1)

!

~

Containment instrumentation consists of pressure, radiation, temperature, .

.!

hydrogen concentration, and humidity monitoring equipment.

Pressure

j

monitoring equipment consists-of six pressure transmitters and.four pressure

switches. The pressure switches provide containment high and low pressure

i

alarm signals.

Four pressure transmitters were used to generate safety

{

injection and containment spray actuation signals.

Safety injection-

Signal HI-l was generated at a pressure of 3 psig and consists of a

j

.two-out-of-three actuation logic (Channel I was not used).

Steam Line

j

Isolation Signal HI-2 was also generated at a pressure of 3 psig and consists

i

of a two -out-of-three actuation logic.

Containment Spray Actuation Signal

!

'

H1-3 was generated at 9.5 psig and consists of a two-out-of-four actuation

!

1ogic.

These four transmitters, as well as two extended range pressure

!

transmitters, were also used for postaccident monitoring.

On February 4, 1993, the unit entered Mode 4 operation because of continuing

--

problems with the AFW system.

On February 6, 1993, during the performance.of

l

Procedure OPSP03-ZQ-0028, the plant operators determined that the shiftly'

.!

channel check requirement of TS 3.3.2 for containment pressure had not been

!

accomplished as' required in Mode 4.

Since the logsheet indicated that the

i

channel check was required only in. Modes 1 through 3, the licensee failed to -

,

perform the channel check for approximately 1 1/2 days while in Mode 4

operation.

Immediate corrective actions taken by the licensee' included

j

initiating a field change request to revise the deficient procedure,

t

performing a computer data review for the pressure channels, and notifying the

l

Unit 2 control room of the discovery.

The channel check was subsequently

.;

performed with satisfactory results.

A station problem report (SPR) was

'

written to investigate the incident.

- i

The cause of the event was determined to be a deficient procedure. -The cause-

.!

of the deficient procedure was inadequate procedure preparation and review.

{

Revision 4 of the procedure ~ had correctly reflected the channel check during

!

Mode _4 operations; however, when Revision 5 was issued, the_ Mode 4 requirement

,

was omitted. The failure to perform a' TS required channel check was.

!

determined to be a violation of TS Table 4.3-2 requirements; This failurezto

.

adhere to TS requirements was considered to be'another example of the-

!

violation discussed in Section 3.1 of this report (498/9304-01).

.j

This incident was determined to have minimal safety significance.

High

containment pressure conditions (0.3 psig'and above) would have actuated a

common " containment pressure high/ low alarm" in the control room.

l

Additionally, rising containment pressure should have been noted during

j

routine control board walkdowns of the extended range pressure recorders.or

.the computer point displays.

{

' i

?

-

.

-

q

' . .

'!

-

&

.

.)

.

-14-

i

I

l

l

,

3.5 Failure to Maintain Minimum Shift Crew Composition-(Unit 2)

On February 14, 1993, Unit 2 was in Mode 4 operation. The shift supervisor-

l

left the control room and delegated the control room command and control

function to the unit supervisor.

The solid state protection system actuation

Train A slave relay test surveillance test was commenced. At 4:34 a.m., the

unit supervisor exited the control room to observe a reactor operator perform

a portion of the surveillance procedure in the relay racks behind the control

room, in an attempt to comply with the policy manual requirements. The unit

j

supervisor immediately realized his error and returned to the control room.

The unit supervisor was determined to be out of the-control room for

l

approximately 41 seconds.

There were two licensed reactor. operators in the

j

-

control room during this time frame.

l

The cause of the event was human error. Although the unit supervisor was new

in the position (the individual had been a unit supervisor only about

j

2 months), the inspectors determined that the event was not caused by a lack

.j

of operator training in control room staffing requirements.

Corrective

j

actions taken included constructive discipline which stressed attention to

'

supervisory responsibilities. The failure to staff the control room with the

.

-required number of senior reactor operators was a failure to comply with

'

TS 6.2.2.b requirements. The failure to comply with TS.was a violation of the

facility operating license (499/9304-03).

,

During the review of this incident, the inspectors noted an apparent problem

!

with the Operations Policies and Practices Manual, Policy 0-0045.

The policy

i

'

requires that work or testing that has the potential to cause a reactor trip.

was-to be performed under direct, on-scene, continuous supervision.

It

j

appears that this policy cannot be performed verbatim if the testing was being

. performed in more than one location at the- same time, as was the case during

the performance of selected operations controlled surveillances.

Additionally, the manual did not. address- the action required if the supervisor

!

becomes unavailable during test performance. This observation was discussed

'!

with licensee personnel for their consideration.

!

t

3.6 Boration Dilution Event (Unit 2)-

i

-l

-

The chemical ~ and volume control system was designed, in part, to control the

'

changes in RCS boron' concentration to compensate.for reactivity transients

.I

during. load follow operations. This was. accomplished by the BTRS. The BTRS

'!

can also be used for other boron concentration changes, such as during startup-

_

_

or end.of core life dilution activities. The RCS letdown flow can be divert'ed

to the BTRS, where partlor all of-the letdown flow can be treated when boron

j

concentrations _ changes were desired. The storage and release of boron by the-

BTRS was determined by the temperature of the fluid entering. the' thermal .

i

regeneration 'demineralizers. To store boron in the demineralizers (and remove-

.

it.from the RCS). the BTRS lowers the fluid temperature. prior to input into

{

-

"

'

the demineralizers. Higher fluid temperatures- permit the boron 'toi be released

from the demineralizers and be discharged into the RCS. A chiller unit and a

j

group of heat exchangert were used.to provide the desired fluid temperatures;

.{

}

I

1

-

t

,+

-

,;,

,a

-

.-.

. .

.:

s

,

.

_..

.

-

_

_

_ _ _ _

_

_

_

__

, _

.

)

J

'

..

,

-

l

!

q

..

i

i

-15-

i

!

t

-On February 20, 1993, at 3:30 a.m., with the unit in Mode 4 operation, the

!

BTRS was placed in the borate mode of operation' to flush residual boron out of

!

the demineralizers and into the RCS. This practice was normal when borating

j

to the concentration needed for refueling. The plant operators were aware

j

that slight boron dilution may occur until the demineralizer effluent flow

l

temperature reached the temperature required for boron release. -A boron

'

sample was requested and the results were submitted to the control room at

7 a.m.

The sample taken indicated the effluent concentration was 240 ppm

q

. boron, which was unusually low. The plant operators speculated that the

sample results were incorrect and ordered another sample to be taken. At

.!

9:12 a.m., the backup sample indicated the effluent flow was 357 ppm, which

'

was, again, an unusually low concentration of boron. The plant. operators then

i

discussed the sample results and a third sample was requested.

The BTRS

_

operation continued despite a residual concern being voiced by a licensed'

i

reactor plant operator. ' At 10:41 a.m., the third. sample result revealed a

j

boron concentration of 433 ppm in the effluent flow. A different licensed

!

plant operator suggested the BTRS was diluting, not borating, the RCS and

'!

suggested that an RCS boron sample be taken.

The RCS. boron sample revealed.

!

that the BTRS was diluting, and not borating, the RCS, as indicated by a lower

!

concentration of boron than previously was. measured.

]

,

At 1:05 p.m., the plant operators diverted flow around the BTRS

j

demineralizers, which discontinued BTRS reactivity changes. A few minutes

I

later, a shutdown margin calculation was performed, which verified that TS -

limits were not exceeded during the event. The plant operators then added -

1

approximately 600 gallons of boric acid to the RCS in order to assure that the

TS shutdown margin was not' exceeded. The cause of the event was inadequate

-

understanding of the BTRS during plant shutdown ' operations. The plant.-

-i

operators did not expect the system to operate as it did, in part,.because of

1

the experience. level of the crew. A contributor to the event was an

-:

inadequate understanding of where the boron concentration monitoring system,

j

which was monitored throughout the event, actually measures the boron

'

concentration.

The inadvertent boron dilution of the RCS was considered to be a reactivity

i

management concern.

Corrective actions taken-(or planned) in' response to the.

'j

event included:

(1) holding management discussions with plant. operators,

!

including management expectations regarding reactivity control,-(2); contacting

other nuclear facilities to determine their methods of reactivity management,

j

(3) revising crew compositions as necessary to ensure adequate-oversight of

newly-qualified senior reactor operators,_(4) contacting the system vendor,.

j

'destinghouse, to obtain an understanding of the performance. of the BTRS .under

!

all plant conditions, (5) revising. plant procedures as appropriate,~ and

l

(6) conducting training for plant personnel as necessary on BTRS:and. boron

!

monitoring system operation.

l

!

This incident will remain as an inspection- followup item (498;499/9304-04),

i

pending.further NRC followup of the~BTRS system operability review to be..

l

performed by the licensee and the vendor, Westinghouse.

j

,

.

i

i

I

!

,

.

,_

.

,

.

_

.

-.:

.. .

-..

J

---

-.

-

,

i

.

.

-

l'

-16-

4

i

,

,

3.7 Conclusions

'

Four violations of TS requirements were identified.

Two of the four

violations were caused, in part, by inadequate surveillance procedure

development and review. The licensee continues to experience a negative trend

-

in the number of inadequate procedures being identified. The third violation

involved the failure to have maintenance work instructions appropriate to the

i

circumstances.

The fourth violation involved the failure of the licensee to'

!

maintain minimum shift crew composition because of human error.

l

The use of the wrong test equipment on the ECW level transmitter ultimately-

t

2

led to a false ESF actuation signal.

Unit 2 entered TS 3.0.3 when power was

lost to the DRPI. A contributor to the event was the failure to repair a

defective sample pump in a timely manner. The two incidents were indicative

!

of the continuing negative trend in maintenance implementation.

3

The inadvertent dilution of the RCS while attempting to borate with the BTRS

i

was considered to be a reactivity management issue.

An inspection followup

i

item was issued to review the BTRS operational characteristics in all modes of

plant operations.

4 MONTHLY MAINTENANCE OBSERVATIONS (62703)

>

Selected maintenance activities were observed to ascertain whether the

l

maintenance of safety-related systems and components was conducted in

.!

-

!

accordance with approved procedures, TS, and app _ropriate codes and standards.

The inspector verified that the activities were conducted in accordance with

"

approved work instructions and procedures,.that the test equipment was within

i

the current calibrati9n cycles, and that housekeeping was being conducted in

!

1

an acceptable manner.

Items inspected included work in progress,

-

postmaintenance testing, and review of completed work packages. All

!

.

observations made were referred to the licensee for appropriate action.

4.1

Secondary Eauipment Problems (Unit 1 and 2)

During the. inspection period, the licensee continued to experience a negative

j

trend in secondary equipment availability and reliability.

The problems

!

' encountered included two reactor trips, three power reductions, two_. secondary

[

system pump trips, one steam generator power operated relief valve (PORV)

j

failure to stroke, and one steam generator PORV stroke time test failure.

The

.j

licensee has implemented initial corrective actions and plans to implement

l

additional corrective actions to improve the reliability of the secondary

(

components, however, no improvement in secondary reliability has been

j

observed. Management oversight has improved but must continue to improve in

j

this area to reverse the negative trend in plant operations.

j

_

In response to recent PORV issues, the licensee developed an action plan in an

]

attempt .to improve the reliability of the steam generator PORVs. One of the

)

primary actions was to establish a PORV task force. Additional. actions

i

planned included:

(1) reviewing the maintenance history of the PORVs,

j

i

-

1

l

.

_

_

__

_

. _.

_

_

.

-

l.c

!

.

.

-

.,

!

.

!

>

-17-

.!

!

t

i

(2) reviewing the scope of the existing preventive maintenance activities and

{

l

maintenance procedures, (3) reviewing recent PORV problems, (4) obtaining an

!

l

independent review of the system design, maintenance, and testing procedures,

I

l

and (5) making long-term recommendations to improve PORV reliability.

l

!

4.1.1

Unit 1 Equipment Concerns

,

1

On January 27, 1993, power was reduced from 98 to 65 percent to increase the

i

margin for a potential reactor trip to allow for the removal of SGFP 13 from.

l'

l

service. The low pressure governor valve for SGFP 13 was noted to be

l.

oscillating about 20 percent of full stroke while in both automatic 'and manual

L

speed control..

This problem was initially discovered when a reactor operator

!

L

noticed that feedwater flow was also oscillating.

In addition, the licensee

'

l

noted that excessive vibration of the EHC line to the governor valve was:

{

!

present. This vibration stopped when the governor valve was isolated.

)

L

Troubleshooting was performed under SR ES-167121, The cause of the

oscillations was determined to be a defective linear variable differential

i

transformer (valve position sensor) assembly.

The new position sensor was

ir talled on January 28, 1993, and the valve was successfully postmaintenance

.l

t sted.

SGFP 13 was returned to service and the unit to 98 percent-power the

'i

some day.

I

On January 29, 1993, Feedwater Booster Pump 12 automatically started when

f

i

Feedwater Booster Pump 13 tripped after an out of service lube oil filter was.

placed in service.

The lube oil filter housing cover was not closed tightly .

enough and lube oil leaked out through the. cover seal, resulting in a low lube

.,

oil pressure trip. Approximately'30 minutes later, Feedwater Booster Pump 13

j

was started to verify operability following lube oil filter change out.

The

11

pump was subsequently returned to service.

,

L

On January 31, 1993, reactor power was decreased from 98 to 28 percent to

l-

allow for the transfer of power from Distribution. Panel DP-0048A back to the -

j

inverter supply.

This distribution panel provides power to the'SGFP master

speed control circuitry. The normal power source:to the distribution panel

i

was a rectifier inverter, while the- backup ~ source was a' voltage regulating-

transformer.

Previously on January 25, 1993, DP-0048A had automatically

transferred to its backup power source. The cause of the transfer was-

)

subsequently determined to be a blown static inverter input fuse and a tripped'

1

inverter input breaker. The inverter was determined to'have a bad-resistor

and a faulty logic control board which caused the ! inverter to overload.

The

f ailed resistor and control board were replaced- in accordance with

SR VC-170210.

The master speed controller was successfully realligned' to its-

normal power supply' and the plant subsequently returned to power.

On February 4,1993, during the unit cooldown to Mode 4 operation, the control .

.l

room operators attempted to stroke steam generator PORV 1C to assist in the

'

cooldown process 4

The PORV subsequently failed to operate from either the

l

control room or the auxiliary shutdown panel. . Plant operators were able to

stroke' the valve locally.

SR MS-179480 was issued to troubleshoot and repair.

the. valve. During the troubleshooting process, the. valve was noted to have

q

I:

LI

,

.

.-

.

.

.

- -

-

.

. - -

-

. _ .

'

-

.

!

.

. :

-

[

.-

i

-18-

!

!

t

S

noisy and erratic movement because of possible misalignment between the valve

stem and the actuator. The valve actuator, valve bonnet, and disk stack were _

-;

removed and inspected. No indications of binding or sticking were observed.

L

The piston stack and sealing rings were in good condition, and the main plug

'

had no scratches or gall marks.

Engineering concluded that the cause of the

problem was dry, tight packing. The cause of the dry packing was suspected to

!

be the result of infrequent operation during the recent outages, however, none

of the other PORVs exhibited a similar problem.

l

!

On February 16, 1992, following minor maintenance (lubrication), PORV ID

i

failed to pass the open stroke time test during the performance of

,

Surveillance Procedure OPSP03-MS-0002, Revision 0, " Main Steam System Cold

Shutdown Valve Operability Test." The valve was observed to stroke open in-

l

37 seconds, 2 seconds over the acceptance criteria limit of 35 seconds (the

!

4

valve operated within acceptance criteria in the closed direction).

'

,

SR MS-158254 was issued to troubleshoot and repair the valve. The Keane

valve, a dual solenoid valve, was found to be leaking while in the

de-energized position.

This resulted in the hydraulic pump excessively

i

cycling to maintain adequate hydraulic system pressure.

i

!

Problems were encountered during the PORV ID repair process. Spare parts were

!

obtained from the warehouse, however, the parts were determined to be

i

incompatible for use in the PORV. The spare parts problem was discovered

i

a

prior to installation of the incorrect parts.

The licensee verified that the

.!

"

Keane valves installed in the plant were the correct parts and an SPR was _

issued to investigate this incident.

Spare parts were ordered from the vendor

j

and a several day delay was experienced until the replacement parts were

-

obtained onsite.

The solenoid valve received was subsequently discovered to

j

'

have the power supply leads shorter than the previously installed valve,

.l

therefore, an electrical junction box and additional wiring had to be added

!

during the Keane valve installation. The hydraulic fluid low pressure switch

'!

was also found to be defective and was replaced.

,!

Following rework of PORV 10, the valve was stroked multiple times and_ the

!

stroke time continued to be unacceptably high. The valve stroke time was

j

measured locally and was well within the acceptance criteria limits.

However,

~;

the remote valve indications were inspected and the valve was observed to-.be

!

stroking about 15 10.17 additional seconds than was being measured locally.

-!

The valve controls were routed through the qualified display-parameter

!

system (QDPS).

The inadequate remote indications.were determined to be a

!

problem within the QDPS and additional troubleshooting was commenced. The.

l

PORV remained out of service at the end of the inspection period.

PORY ID was

j

a Mode 3 restraint, therefore, the valve has to be returned to service prior-

i

to the unit startup.

j

,

i

4.1.2 Unit 2 Equipment Concerns

-!

On January 17 and 18, 1993, the control room operators noted that feedwater

.

flow transients were occurring on all four steam generators for'no apparent

l

reason. The transients were occurring even with the feedwater system master

l

- - . -

-

.-

..

.

.- -

- - .

.

.

_ _

._

_.

_

.-. _

_

.. . _ _ . _

_

_

.

__

_.

>

.

.

-

l

!

.-

!

-19-

!

t

i

,

i

controller in the manual mode of operation. On January 18, 1993, at

4:15 p.m.. unit power reduction commenced, at management direction, from 100

.

to 28 percent to allow for repairs to be made on the SGFP speed controllers.

l

'

An SR (EH-163095) was issued in April of 1992 to troubleshoot the erratic

operation of the master and individual pump speed controllers.

This SR was

revised to replace the existing feedwater control circuit power supplies.

The

i

-

old power supplies were obsolete and were determined to be unreliable. The

l

,

power supplies were removed and replaced with new ones in accordance with

Document Change Notice JM-749. The new power supplies were tested and field

,

installed with vendor assistance.

The work was completed on January 20, 1993.

j

'

Unit power was increased the same day and Unit 2 returned to full power the

-

next day.

Because of plant conditions, the postmaintenance test was

incomplete at the end of the inspection period, although all field work was

,

complete.

On January 30, 1993, at 100 percent power, SGFP 23 unexpectedly tripped

!

offline on low lube oil pressure.

Startup Feedwater Pump 24 automatically

j

started as designed and Feedwater Booster Pump 22 was manually started to

maintain feedwater flow at full power conditions.

Initial investigation

j

-

!

showed both main oil pumps (MOP) and the 125 volt DC oil pump for SGFP 23

[

running with adequate oil pressure. After several minutes, the motor for

i

,

!

MOP-1 started smoking and tripped the motor power supply breaker.

>

MOP-1 apparently had a momentary fault while in operation, which caused a

j

decrease in oil pressure. The oil pressure decreased _to the point where

j

,

-

Standby MOP-2 and the DC powered pump automatically started.

Because of

J

system design, the low lube oil pressure signal also caused the trip header

!

q

solenoid to pick up, which tripped SGFP 23. A design change to correct this

!

problem was installed in Unit I and was scheduled to be installed in Unit 2

9

during the upcoming refueling outage.

Following the SGFP 23 trip, MOP-1

i

continued to run until the motor failed.

There was no evidence the SGFP 23

.i

4

bearings experienced a loss of oil because system temperatures appeared to be

l

normal throughout the event. An oil sample was taken and sent offsite for

4

analysis

At 5:28 p.m. the same day, the SGFP.23 was returned to service with

i

MOP-2 in service providing lube oil to the pump.

,

,

i

SR LP-169465 was issued to investigate and repair the MOP-1 motor,

The motor

!

and the power supply breaker were both replaced in accordance with

!

- SR LP-169465 instructions.

A new coupling assembly was fabricated because the

j

new motor shaft was physically different from the old motor shaf t. ' During the

work process, two " hot spots" were identified on the.line side of the feeder

l

,

breaker.

The suspect leads were lifted, cleaned, and raterminated.

i

Postmaintenance testing, which consisted of a pump operability test, vibration

j

analysis, and thermography analysis, was completed on February 3,1993.

!

!

An SPR was written to investigate the MOP-1 motor failure.

Preliminary

investigations revealed that the motor on MOP-1 was replaced in June'1992

following the identification of a rubbing sound coming from the motor.

Root

'

cause. analysis was incomplete at the end of the inspection period.

.

'

4

,

.

.

.

.

_

..

.

.

.

_.

...

'

~

.

.

,

...

i

-20-

.I

t

i

On February 3, 1993, Startup feedwater Pump 24 tripped while the lube oil

filters were being shifted because of high lube oil differential pressure.

[

The Startup Feedwater Pump 24 trip subsequently led to an automatic reactor

trip on low-low steam generator level. The startup feedwater pump trip was

j

the result of water in the oil reservoir, which was caused by mechanical seal

leakage. Vibration induced wear may have been the possible cause of the seal

.]

leakage. This was the third time the startup feedwater pump 24 had tripped

i

because of high lube oil filter differential pressure.

The pump tripped on

November 20 and December 27, 1992, for the same reason.

1

The failure to promptly identify and correct the root cause of the pump trips

resulted in a reactor trip on low steam generator level.

The failure to

!

repair the startup feedwater pump in a timely manner .was indicative of a

j

weakness in the implementation of the corrective action program. Corrective

!

actions planned included pump. seal replacement and modification to preclude

,

moisture intrusion. Additionally, the lube oil filter design will be changed

i

to include a bypass flow path around the filters to minimize the possibility

i

of a pump trip on clogged filters. These modifications were planned to be

{

implemented during the next refueling outage for each unit.

!

t

4.2 Troubleshootino and Repair of an Essential Chiller -(Unit 1)

On January 24, 1993, at approximately 6 a.m., Essential Chiller 11B tripped on

low oil pressure. A reactor plant operator was dispatched to the area and oil

was observed on the floor. The solenoid operated fill valve to the purge drum

i

had apparently failed open, which allowed the oil to spill out from the purge

[

unit and onto the floor.

SR CH-158361 was issued to troubleshoot and repair

j

the cause of the apparent valve failure.

j

During the performance of SR CH-158361, the purge unit float switch was

!

removed and inspected. The upper and lower float switches were checked for

'l

operability. The upper float switch was found to not be working properly and

-!

was subsequently replaced. The defective float- switch caused the fill valve

.to fail open, which allowed oil to be purged from the chiller. Approximately

,

5 gallons of oil was added to the chiller to replace-the oil lost because of

.j

the defective float switch. During a maintenance run of the chiller,La leak

'

was identified at the float chamber refrigerant'-filter canister. The gasket

!

-was changed on the filter and the refrigerant was charged to the proper

!

pressure. A po'stmaintenance run of the chiller was performed and no oil or

refrigerant leaks were observed. The chiller was returned to service on

=j

January 26, 1993.

During previous inspections, the inspectors reviewed the essential chilled .

.l

water system reliability and availability rates (refer also to NRC Inspection

l

Reports 50-498/92-24; 50-499/92-24. -92-26, and -92-35).

No' clear improving.

j

,or declining trend was noted.

During this inspection period, recent

unavailability rates were reviewed. -The Unit 1 essential chiller forced

unavailability rates for November and December 1992 were O percent, however,

j

-the forced unavailability rate for January 1993 was 7 percent. The. Unit 2..

forced -unavailability rates for November 1992 and January 1993 were O percent,

.j

!

<

-

-

i

.

,. .

.

l

-

l

-21-

.(

i

!

however, the rate for Decenber 1992 was about 8 percent. The licensee

j

continues to experience problems with the chillers and there was no clear

!

indication of the unavailability rates improving.

-l

)

4.3 AFW Pump 24 Postmaintenance Testing (Unit 2)

l

During the inspection period, extensive testir.g of the Unit 2 AFW Pump 24 was

!

performed following unit shutdown.

During the testing process, two

i

maintenance implementation weaknesses were identified that resulted in

j

unnecessary delays. One weakness involved the failure to check the

j

operability of a buffer compensation system spring and the second weakness

involved the failure to identify and remove a shipping cover on the governor.

Two Temporary Waivers of Compliance were generated and granted to extend TS

'

allowed outage times to permit the licensee to complete the Mode 3 testing of

,

the pump.

l

i

As part of the corrective actions for AFW Pump 24 rework, a detailed

j

postmaintenance plan was developed.

The. testing program included:

}

(1) testing of individual components, including the trip and throttle

,

Valve MOV-0514, the governor, governor valve, turbine, and drains, (2) testing

-)

!

of the mechanical overspeed trip device, manual starts, electric trip tests,

warm and cold turbine starts, and response time, (3) enhanced monitoring of

l

!

the. drainlines, steam lines, and discharge stack temperatures, (4) revising

the surveillance and testing procedures to include lessons learned, and

j

(5)' increased surveillance frequency to verify operability over the next

,

several months. As part of the postmaintenance test program, Engineering

.l

Procedure OPEP07-AF-0013, Revision 0, " Auxiliary Feedwater. Pump 14 -(24)

Special Post Maintenance Test," was developed and approved on February 16,

!

1993. This procedure described the testing that was needed to verify AFW

,

Pump 24 operability. The test consisted of baseline data measurements and.

,

eight turbine runs. The procedure also listed-the test performance sequence

l

and provided a central location for the data collected for future retention-

and review.

In addition, vendor representatives for both the AFW pump turbine

and the turbine's governor were present for the entire series-of tests.

j

On February 17, 1993, all prerequisites were verified complete and shift

'

supervisor permission was granted to allow the test activities to begin.

The temperature survey (Section 7.3 of OPEP07-AF-0013) and chart recorder

installation (Section 7.6) were completed first.

The temperature survey'was

performed to obtain a baseline temperature of the drainlines prior to the

'I

heatup and pressurization of the main steam lines to the AFW Pump 24 turbine.

!

Three temperature readings were taken at each preassigned location.

I

Additionally, Section 7.6, " Test. Equipment Installation," was completed to

i

install' a chart recorder to monitor turbine speed and governor valve' position

1

.

during test starts. The overspeed trip linkage manual local test was then

!

performed. A turbine mechanical overspeed trip was manually generated by

depressing the local' trip lever to verify the linkage trip closed MOV-0514 and

1

to ensure that no' binding or excessive. play existed in the linkage. This

'i

action vas satisf actorily completed three times. Additionally, M0V-0514 was

opened manually and closed electrically three times to ensure the mechanical

!

l

i

$

8

w

.=1

,4

.-

'

.

. - -

.;

,

,

.

j

6

-

,

'

,

-22-

!

!

overspeed trip linkage did not disengage, which was a problem that had been _

!

previously identified. No indications of a mechanical overspeed were received

l

in the control room. The test performers then verified that the MOV-0514

!

above seat drain flow paths were clear and free of obstructions.

j

Testing was delayed when Unit 2 entered TS 3.0.3 because of potentially

j

undersized fuses in the solid state protection system actuation trains (for

l

further details on this issue, refer to NRC Inspection Report 50-498/93-11:

l

50-499/93-11). Testing was delayed because the licensee was required to cool

i

down the RCS, and the steam supply to AFW Pump 24 was lost.

Following the

j

return of the steam supply, Steam Supply Vaive MOV-0143 was opened and heat-up

,

of the steam lines began in accordance with Procedure OPEP07-AF-0013,

i

Section 7.4, requirements.

Acoustic monitoring of Be piping was performed

l

before and after MOV-0143 was opened to obtain condensate leakage rates.

!

On February 18, 1993, the first turbine run was attempted. The turbine was

I

uncoupled from the pump and the licensee attempted to verify the overspeed

.;

trip setting. The turbine failed to trip on ov rspeed on two occasions,-

i

despite turbine speed being taken to 4200 rpm (the normal trip band was 4044

i

to 4144 rpm). Troubleshooting and rework was performed under SR AF-179426.

l

Inadequate axial displacement of the overspeed tappet assembly and a sticky

i

residue on the tappet and tappet guide were found. The tappet guide was

j

cleaned, the tappet was replaced, and clearances were readjusted.

The

mechanical overspeed trip test was attempted later the same day and three. trip

tests were satisfactorily completed.

The turbine tripped at 4065, 4058, and

4070 rpm, respectively. MOV-0514 was closed _and relatched, and the turbine

was released to mechanical maintenance to restore the lube oil cooler,

-

a

recouple the pump to the turbine, and correct a governor plug oil leak,

j

l

Following the overspeed trip testing, a walkdown of AFW pump and drain lines

j

was performed. Steam and condensate leakage was noted to be coming out of the

turbine casing, exhaust stack, and MOV-0514 high pressure _ leakoff drains. The

licensee suspected MOV-0514 was leaking despite the valve being fully shut.

A

temperature survey and condensate sample analysis confirmed .that the valve was

,

not leaking by.

It was concluded that the noted leakage was the result of

l

atmospheric air condensing in_the stack and casing.

l

?

On February 22, 1993, Unit 2 entered Mode 3 operation to allow for testing of_

!

AFW Pump 24 at normal operating temperature and pressure conditions. A

baseline temperature, acoustic, and drain monitoring survey was again

performed.

The second turbine run was performed on February 23, 1993,

following the coupling of the pump to the turbine. . The' turbine was manually

increased in speed by slowly opening MOV-0514 using the valve handwheel.

The

~!

governor took control of the unit at-1650 rpm. Water and steam was noted to

j

be coming from the governor valve packing.

Turbine speed was then increased

i

using the governor manual adjust knob.

Governor valve oscillations were noted

)

'

to begin with approximate 100 rpm swings at about 2000 rpm turbine speed.

Turbine speed was increased and the governor oscillations persisted.

In an

attempt to dampen the oscillations, control oil supply line- fittings at the

governor were loosened.to vent any trapped air.in the governor. No trapped

'l

1

'

.

i

,

,.

. - - , -

-

k

_

_

-

,

.

._

.

_

i

. - .

t

E.: 9 . ._ .

j

,

t

-23-

!

air was identified.

The governor compensating needle valve was adjusted to

try to eliminate the oscillations, however, all adjustments failed to stop the

l

oscillations.

The turbine was electrically tripped from the control room.

]

As part of the troubleshooting process for the governor oscillations, the

'I

droop linkage.was disconnected and the turbine was manually started a second

time. Governor oscillations began again at 1750 rpm.. The oscillations were

j

more severe than those o~oserved.during the previous tests. The oscillations

increased as speed increased and reached 300-400 rpm while at a turbine speed

i

of 3000 rpm. The turbine was electrically tripped from the control room

following unsuccessful attempts to dampen the oscillations.

Following this

,

electrical trip, MOV-0514 failed to relatch. The licensee suspected the limit

switches were actuating too soon. MOV-0514 was cycled.a second time from the-

control room and again failed to relatch. The valve was manually shut and

relatched as designed.

SR AF-170068 was issued to troubleshoot and repair the

!

-

valve.

The MOV-0514 torque switch was found with dirty _ contacts and a wire

}

was impeding switch operation. The contacts were cleaned and the wire was

rerouted. The valve was subsequently tested and the valve properly relatched

!

during all ensuing tests.

,

A decision was subsequently made to swap the governors between'the two units-

!

in an attempt to eliminate the governor valve oscillations.

This work was

performed under SRs AF-158219 and 158220. On February 24,_1993, the field

'

work was completed and a test run was attempted to determine if the work -

i

performed eliminated the oscillations.

The. turbine was manually aarted and

1

the speed was~ increased in increments until 3000 rpm was achieved. The

j

~

governor valve oscillations continued to persist at approximately'100 rpm _

_;

swings.

The turbine was tripped and further troubleshooting commenced.

The

!

licensee, with vendor assistance, subsequently. discovered that'the speed

oscillations were caused by undersized buffer springs.

]

A buffer compensation system was provided in the hydraulic section of the

>

governor control circuitry. This system functions to stabilize the governor

action by minimizing over and undersh'oot following a change in. governor speed

setting or a change in load. The buffer compensation system consists of a

_;

buffer piston, springs,- and a compensation needl_e valve:. . The force of the

buffer springs can vary, depending on the resistance of the~ interconnecting

,

linkage or. changes in turbine load conditions. The buffer springs installed

in the governor were rated at.5.2 pounds.

Following the' removal, cleaning,--

l

and reinstallation of the governor and interconnecting components,. the old

1

buffer springs did not provide sufficient resistance in the compensation-

!

system'for. proper-governor operation.

A decision was made_to. change-the

r

buffer springs to a higher value and determine the proper setpoint necessary -

to solve the hunting problem. The work was delayed because' spare parts.were

not immediately available onsite.

.On February 25,1993, at:1:45 a.m., the replacement buffer springs (10 pounds)

,

'

were.. received and subsequently installed in the governor.

A postmaintenance-

test was performed. The turbine had small lless than 50 rpm)' oscillations :

while speed was increasing, however, the governor was adjusted and the

!

,

, . -

y

7

-

. . . .

.

..

. . . .

.

l

.

i

-24-

oscillations disappeared. A second postmaintenance test run was completed

with satisfactory results.

During the second turbine run, steam and water

leakage was noted from the governor valve bon: et lower jacking bolt hole. The

licensee determined the leakage was intermittent and generally did not appear

until the unit was hot and had run for over 30 minutes. The associated gasket

was scheduled to be replaced during the upcoming Unit 2 refueling outage.

Or. February 25, 1993, at 9 a.m., Section 8.4 of Procedure OPEP07-AF-0013,

"Second Turbine Run/ Manual Roll-up," was reperformed. No oscillations were

noted as speed was increased in increments.

Temperature and drain monitoring

surveys were performed with the turbine running. After 40 minutes of

operation, the turbine was remotely tripped from the control room and MOV-0514

closed and relatched as designed.

Several hours later, the third turbine run

was attempted in accordance with Section 8.5 of the procedure following a

30-minute cooldown cycle. This run involved starting the pump from the

control room by remotely opening MOV-0514 and discharging to the AFW Storage

Tank through the long path recirculation piping.

During this third test run, steam began to fill the room from the heating,

ventilation, and air conditioning (HVAC) discharge duct. Visibility in the

room was greatly reduced, the area temperature increased, and condensation

i

formed on all surfaces.

Condensation in the governor and MOV-0514, as well as

personnel safety, were of immediate concern. The control room operators

placed the room fan in " pull to lock," which took the fan out of service in an

I

attempt to control the moisture ingress. The turbine was tripped from the

control room and the unit shut down as designed.

Condensation wa's noted to be-

" raining" in the pump cubicle onto the turbine skid and other components in

I

the room. The fourth turbine run, similar to.the' third run that was just

l

completed, was immediately performed in accordance with Section 8.6 of the

q

procedure.

Slight oscillations were noted during the speed increase.

1

Additionally, steam / water leakage was identified at the governor valve

packing, which had recently been repacked. The turbine was secured 32 minutes

later, after the temperature and drain monitoring survey were completed. The

governor valve leakage was determined to not be relevant to the test evolution

and was scheduled to be reworked during the upcoming Unit 2 refueling outage.

During the fourth run, steam ingress into the pump bay was still a problem,

and design engineering personnel were notified of the incident.

Following the completion of the fourth turbine run, a fifth run was attempted

in accordance with Section 8.7 of Procedure OPEP07-AF-0013. The fifth run was

similar to the third and fourth runs.

During this test run, the local panel

tachometer indication became erratic and indicated increases in speed without

governor valve movement.

This resulted in a manual local turbine trip because

the indication reached 4500 rpm, a value above the mechanical overspeed trip

device setpoint.

The temperature and drain monitoring surveys were not

completed as required by' the procedure Steps 8.7.6 and 8.7.7.

Following

consultations'with the shift supervisor, the test manager decided not to

perform Section 8.7, "Fifth Turbine Run," and to begin the 12-hour waiting

period required for the sixth turbine run. During the 12-hour waiting period,

_

-

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ - - - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ -

. - _ . _

_

_

.

..

._ ___ _ _

_ . _

,

.

_

_ _.

..

-

.

.

s

!

-25-

j

,

,

.the temperature, drain, and acoustic meisurement surveys were performed in

accordance with procedure Section 8.8 requirements.

-

_j

Th'e licensee determined that the local tachometer had malfunctioned because-

i

condensation had formed on the panel connections, which resulted in_ increased

l

resistance on the indicator signal, and there was no actual turbine overspeed-

condition. The connections were dried out and taped, and a calibration check

.!

was performed.

The local tachometer performed reliably during subsequent

i

tests.

i

In response to the condensation concerns, the licensee initiated several

corrective actions to mininize the steam ingress into the building.

The

exhaust stack cover was removed. This cover was recently installed to prevent

moisture ingress into the stack, however, the cover was redirecting discharge

.

!

steam flow in a horizontal direction. This steam flow was being drawn into

'

the HVAC suction duct and was being discharged into the isolation valve-

cubicle building where the AFW pumps were located. Additionally, the turbine

casing drain tail piece was extended deeper into the floor drain-piping.

MOV-0514 was inspected for moisture in accordance with SR AF-158234. The

i

governor oil was sampled for water, and-no unusual material was identified-in

the oil sample.

Further design changes were under review at the end of the

inspection period.

.j

Unit 2 entered Mode 3 operation on February 22,-1993, at 10:48lp.m.

In

accordance with TS 3.7.1.2, Action b, the unit had 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> to restore the

!

turbine-driven AFW pump to operable or a return to Mode 4 was required. On

l

February 25, 1993, tho licensee requested a Temporary Waiver-of compliance for

!

TS 3.7.1.2.

The licensee requested the allowed outage time be increased for

f

an additional 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />, on a one-time basis for Unit 2, to continue the-

,

testing of the turbine-driven pump.

This. request for a waiver.was

!

subsequently granted by the NRC on February 25, 1993, at approximately

L

2:20 p.m.

-

On' February 26, 1993, two postmaintenance pump runs were attempted following

corrective maintenance activities.

During the 'second . test run, the floor

drain was noted to be pressurized and steam was blowing up from three floor

drains. This indicated that the extension of the turbine casing drain tail

.!

piece into the floor drain piping was not effective. To correct this problem.-

l

SR AF-161275 was written to extend the casing drain-to the-room sump. -This

!

change was made to prevent _ blowing steam from the turbine casing drain tail

!

piece'into_the floor drain piping. Also during the test run, turbine ' speed

)

increased automatically to 3600 -rpm and -subsequently drifted down to _3478 rpm

i

by.the completinn of the test run. The turbine was' electrically tripped from.

!'

the control room.

On February 26, 1993, the licensee decided to perform th'e seventh pump run-

i

prior.to the sixth pump run to ensure proper-operation of the buffer springs

~l

while forward feeding to the steam generator-at. normal operating temperature

j

and pressure conditions.

The seventh' pump run was being performed prior to_.

the originally scheduled 12-hour wait period because the wait was considered

-l

!

+

t

.

.-

, , -

.%

, , _ -

. ..

--

-

-

- - -

-.- - . _~

- -

. - -

- - . - - ,

a

i

.-

u . ;,. ,

.i

-

<

.l

=.

i

-26-

!

,

1

-i

I

to be unnecessary. At 12:17 p.m.,

the pump was started in accordance with the

'j

normal operating procedure and the system was realigned to permit forward

j

-

feeding at 540 gpm to the Steam Generator 2D.

The governor responded properly -

j

with no oscillations noted and turbine speed was stable during the 9-minute

r

test run. The turbine was secured and the regulating valve was shut when the

l

steam generator reached the filled condition. The turbine was restarted about

j

35 minutes later and the requirements for the seventh run (temperature and

!

drain surveys) were completed. During the 35-minute test run, the licensee

i

noted the turbine speed had drifted down about 200 rpm..

The turbine was

-l

secured and a 12-hour cooldown cycle commenced.

1

>

On February 27, 1993, at 6:06 a.m., the licensee attempted to start the AFW

!

pump in accordance with the inservice test surveillance procedure and-

}

Procedure OPEP07-AF-0013, Section 8.9, " Sixth Turbine Run/ Inservice Test."

The pump failed to develop sufficient discharge pressure (as-found pressure

j

was 1420 psig) to meet the minimum acceptance criteria limit of 1454 psig..

'l

1

The pump discharge pressure was low because the turbine was at 3488 rpm, below

the normal speed of 3600 rpm. -The governor speed was manually readjusted,

.with the shift supervisor's concurrence, after which all parameters came

j

within the required specifications.

The turbine was locally: tripped .in

accordance with the inservice test procedure at 6:42 a.m.

[

Following the most recent test failure, the licensee held discussions with the

j

turbine vendor to determine the cause of the low turbine speeds. The licensee

i

thought the droop cam in the governor may be faulty or the-speed may have been

i

_

'

set during low flow conditions. On February 27, 1993, at 2:53 p.m., the-droop

linkage was disconnected and a TDAFW Pump 24 start was attempted. At this

point, the licensee discovered that the speed. reference bellows vent ~ cavity

,

had a shipping plug, or dust cover, which blocked the port to the bellows.

j

The operating speed of a turbine can be varied by either a local manual

!

setting (as was the case at STP) or by varying a pneumatic signal to the

l

governor from a remote control position. The speed setting'section of,the

j

governor consists, in part, of a brass bellows housed within 'a pressure

l

chamber. A change in control air pressure displaces a force-balanced

-f

-arrangement and results in a speed setting change by causing an expansion or

-l

contraction of- the bellows.

Since the turbines at South Texas Project were

controlled by local manual operation, and not remotely, the expansion bellows

~

arrangement was not used. While the governor was at-the factory.being'

,

refurbished, the dust plug was installed on the port that. was- connected' to' the

!

bellows chamber.

The governor was installed-on the turbine with the-dust:

.,

cover in place..

During a subsequent-troubleshooting test run of_ the turbine,.

!

the licensee noted the speed dropped 100 rpm when the dust cover was removed

,

and reinstalled.

Increased pressure, because of an; increase in ambient

,

temperatures, in the bellows area was causing a speed reduction. The droop

1

linkage was reconnected, the plug was removed, and a second t raubleshooting

run was attempted. The speed was manually adjusted, with the bellows chamber

!

vented to atmosphere, to 3600 rpm and the turbine was secured. ' Fo11owing plug'

'

removal.~ the turbine ran reliably during subsequent tests and speed decays

i

,

were no longer experienced. _ The failure to identify and remove the shipping-

!

l

6

!

-!

'

-

.

_.

.

. . .

-

.

.

.

i

. . - - -

_

..

-

j

-

.

. . . .

l'

.

.

L

-27-

,

cover was the second example of a maintenance implementation weakness that

resulted in -

messary delays of the AFW pump testing program. -The

surveillance it;t (test run number six) was completed with satisfactory

'

results the following day.

,

l

4.4 Conclusions

The licensee continued to experience a negative trend in secondary equipment

operability and reliability, as indicated by two reactor trips and three power

'

reductions being caused by the equipment.

The development of a PORV action

plan was a positive, yet reactive, response on the part of the licensee.

A float switch failed un an essential chiller, which resulted in the loss of

'

about 5 gallons of oil from the chiller.

The reliability and unavailability

>

rates of the chiller.s were still of concern to the inspectors.

t

The licensee performed extensive ter'ing of AFW Pump 24. One maintenance

implementation weakness was observed which negatively resulted in testing

delays.

A Waiver of Compliance was needed because of the length of time the

licensee needed to test the equipment in Mode 3 operation.

5 PREPARATION FOR REFUELING (60705)

Prior to the Unit 2 shutdown for the third refueling outage, an assessment of

.

the licensee's implementation of controls for refueling operations was

performed.

The assessment included a review of the outage scope, management

oversight, and control of plant risks during shutdown conditions.

5.1 Unit 2 Third Refuelinq Outage Scope

The Unit 2 third refueling outage was scheduled to begin on February 27, 1993.

A 78-day outage duration was planned, with a completion date of May 17, 1993.

A full core offload will be performed and roughly one third of the core will

be replaced with new fuel. An extended "No Mode" work window of 32 days will

be utilized. The activities planned during the "No Mode" work window involves

the letdown and charging subsystems, residual heat removal and safety

injection loop isolation valves, and the reactor coolant pumps. Overall,

1596 SRs,1697 preventative maintenance activities, and 948 surveillance tests-

were scoped into the outage.

'

Emergency Diesel Generator (EDG) 21 will undergo a vendor recommended 5-year

inspection and EDGs 22 and 23 will receive 18-month' inspections.

A total of

98 MOVs will be functionally tested. The low pressure Rotor 21 of the main

[

turbine will be replaced and the main turbine high pressure gland seal system

will be repaired. All- four steam generators will .be sludge lanced. The

,

unit's feedwater pumps will be reworked, including implementation of-

modifications to resolve moisture intrusion problems and replacement of the

main feedwater control system with a new solid state control system.

A full

inspection of the DRPI system will be performed.

,

.-

,

. ~ . .

. _ . _ - ~.-

-

.- . -

- .~

.

.

.-

-f

..

.

c .. .

,

-,

f

-28-

!

!

'!

l

A total of 53 modifications were scheduled to be implemented during the

outage, including:

(1). removal of the negative flux rate trip hardware from.

_'

the nuclear instrument system, (2) upgrading the feedwater isolation valve

a

hydraulic system, (3) adding pipe spools and supports to eliminate water

!

,

hr.,mer events on the EDGs, (4) replacement of the residual heat removal system

)

circuit breakers, (5) modification of the standby diesel generator shutdown

air system to eliminate premature cooldown cycle trips, (6) replacement of. the

' i

toxic gas analyzers, and (7) modification of the EDG fuel oil nozzle and

delivery valve holders.

Because of continuing problems with the'AFW system, the licensee decided not

to restart the unit following the reactor trip on February 3, 1993.

Because

l

of manpower and vendor availability, the original outage schedule could not be

!

totally implenented prior to the originally scheduled start date. Key

j

activities were started early in an effort to expedite completion of

i

potentially time consuming activities. Key activities that were scheduled'to

r

begin early included Train B maintenance, including the emergency diesel

generator inspections, refueling machine maintenance, and testing of selected

secondary valves.

.;

t

5.2 Shutdown Safety

q

Shutdown risk assessment will be a licensee priority during the outage. A

shutdown risk assessment team was developed ~that consists of one chairman and

four additional members. The' chairman was a representative.from plant

operations who was a licensed senior reactor operator. The other.four

-

!

representatives were from the independent safety review, integrated planning

1

and scheduling, and plant analysis groups, as well as a shift technical'

-

advisor. The group was tasked with performing a pre-outage review and

)

L

-developing a shutdown risk assessment report. The group will also review

daily activities, meet twice a week, and make recommendations during the

outage.

Lessons learned from the risk assessment of the recent Unit 1 fourth

y

refueling outage were applied to the shutdown risk assessment program for the

i

jl

Unit 2 outage.

These lessons learned included group quorum requirements,

overall responsibilities, and meeting frequencies.

i

5.3 Outage Management

j

The management arrangement for the refueling outage was essentially similar to

{

the organizational arrangement used during the recent Unit 1 outage. The

j

plant manager has overall responsibility for the outage.

The outage manager.

1

reports to' the plant manager through the integrated planning and scheduling-

manager.

Four rotating outage shif t managers will be available around the

clock and will report to the outage manager. Outage coordinators and .

~

schedulers will report to. the outage shift managers. The position of: outage

~

'

director, developed during the recent Unit 1 outage, was deleted. ;The

_i

licensee determined that the outage director position, which was manned by

j

rotating department managers, was no longer needed.

l

t

I

!

!

,

,

-

+

.

1:'a

,o'.

-

.

p

,

L-

1

>

1

-29-

,

5.4 Outage Scope and' Shutdown Risk Meetinq

!

On February 11, 1993, the' licensee conducted a meeting _ with members of the

i

-

Region IV staff to present an outage scope and shutdown risk assessment of the

.,

'

third refueling of Unit 2.

A list of attendees and the meeting handout are

contained in Attachment 2 of this report.

The staff found this presentation to be beneficial and provided additional'

'

'

insights to the efforts that the licensee has taken to manage the upcoming

outage and address shutdown risk.

'

5.5 Conclusions

The Unit 2 third refueling outage scope appears to be well-planned by the

licensee.

The schedule appears aggressive because of' the number of MOVs

scheduled to be tested.

Selected refueling outage activities were started

'

early because the plant was not restarted following auxiliary feedwater system

rework.

This was an initiative that was implemented by the licensee to help

,

reduce the scheduled outage time. Several long-standing equipment problems,

>

such -as feedwater control system malfunctions and EDG cooldown-cycle trips,

were planned to be resolved during the outage.

Shutdown risk assessment and

outage management staffing continue to be licensee strengths.

6 FOLLOWUP (92701)

6.1

(Closed) Inspection followup Item 498:499/9119-02:

Reactor Vessel Head

Vent System Analysis

In December 1990, the-licensee received a Westinghouse plant specific analysis

for the reactor vessel head vent system (RVHVS).

In this analysis,

Westinghouse determined that the sum of thermal expansion and hydraulic

stresses would exceed the American Society of Mechanical Engineers (ASME)

Code,Section III, Class 1 allowable stress values when the RVHVS was operated

with only-a single vent path open.

The licensee reviewed this analysis and

concluded, in Justification for Continued Operation (JCO) 90-0521, that the

,

.RVHVS would still be . capable of initiating and terminating venting, letdown,

-

and head cooling via remote manual operations from the control room ~in the

overstressed condition.

The NRC has since reviewed the JC0 and concurred with

the licensee's conclusions.

In response to the analysis, the licensee decided

to modify the RVHVS to conform to the original design and licensing basis.

Inspection followup Item 498:499/9119-02 was issued to track the licensee's

corrective actions.

.Short-term corrective actions taken by the licensee'in 1991 included issua,'ce

'

-

of a JCO, revising plant emergency operating procedures to provide :

instructions to open both vent path isolation valves, and initiation of:a-

design change request to modify the RVHVS piping.

In 1991, the RVHVS piping

,

merged and became a-single' pipe downstream of.the isolation valves. A design

'

change request was approved which separated the paths up to the' pressurizer-

relief tank header piping. Modification 91-13 was implemented in Unit 1

L

l

l

.

, . ,

,

'

'

-30-

!

during the unit's fourth refueling outage, while Modification 91-14 was

!

implemented in Unit 2 during the unit's second refueling outage.

'

7 ONSITE FOLLOWUP OF WRITTEN REPORTS OF NONROUTINE EVENTS (92700)

l

7.1

(Closed) LER 499/92-003: Manual Reactor Trip

On February 24, 1992, Unit 2 was manually tripped from 100 percent power to

i

prevent an automatic low-level steam generator reactor trip. The

,

turbine-driven feedwater pumps had experienced several speed control

fluctuations earlier in the day and, at 6:10 p.m.,

Feedwater Pump 21 was

observed to have a decreasing speed. Manual speed control was attempted with

no success and the reactor was manually tripped with steam generator levels at

47 percent and decreasing.

The licensee determined that tne event was caused by rain water leaking

through expansion joints in the turbine building roof and into the

electrohydraulic control cabinet.

The cabinet contains the common controls

,

'

for the three fecdwater pumps. Modifications 89007 (Unit 1) and 89008

(Unit 2) were implemented to provide watertight sealing of the turbine deck

roof.

The inspectors verified the installation and performance of the

watertight seals. The inspectors verified that the licensee had completed all

identified corrective actions.

During this inspection period, the site experienced heavy rains. The Unit 2

turbine building was noted to have several leaks,'however, the room that

'

housed the feedwater pump speed controls was dry. Additionally, sealant was

observed around the base of the room, which prevented moisture intrusion into

the room.

The feedwater pump control circuitry in Unit 2 was scheduled to be

replaced during the upcoming refueling outage.

The implementation of the new

state of the art equipment should improve the reliability of the system and

'

reduce the number of incidents that were caused by feedwater control system

4

malfunctions.

.

t

9

._

_ _

,<

y

. a. < > - . .

,

'

,

f

ATTACHMENT 1

i

'

.j

1 PERSONS CONTACTED

,

.

]

1.1

Licensee Personnel

- R. Balcom, Director, Nuclear Security

H. Bergendahl, Manager, Technical Services

!

C. Bowman, Corrective Action Group Administrator

!

M. Chakravorty, Executive Director, Nuclear Safety Review Board

[

i

K. Christian, Manager, Unit Operations

M. Coughlin, Senior Engineer, licensing

R.. Dally-Piggott, Engineering Specialist, Licensing

i

D. Denver, General Manager, Nuclear Assurance

D. Hall, Group Vice President, Nuclear

{

S. Head, Deputy License Manager

T. Jordan, General Manager, Nuclear Engineering

W. Kinsey, Vice President, Nuclear Generation

'

F. Mallen, Manager, Planning and Assessments

t

!

T. Meinicke, Deputy Plant Manager

M. Pacy, Manager, Design Engineering

l

G. Parkey, Plant Manager

!

R. Rehkugler, Director, Quality Assurance

5. Rosen, Vice President, Nuclear Engineering

j

T. Underwood, Maintenance Manager

!

1.2 Other Personnel

B. McLaughlin, Owner's Representative, Central Power and Light

The personnel listed above attended the exit meeting.

In addition to the

personnel listed above, the inspectors contacted other personnel during this .

inspection period.

2 EXIT MEETING

j

i

An exit meeting was conducted on March 2, 1993. During this meeting, the

inspectors reviewed the scope and findings of the report. The l_icensee did

not identify as proprietary any information provided to, or reviewed by, the

j

inspectors.

.

>

1

.

.

_

-

.

. _ - -

. .

1

ty;s.

.1

l

-~

-l

4

ATTACHMENT 2

!

!

ATTENDANCE AT SD RISK /0UTAGE SCOPE BRIEFING

l

HL&P

-

!

R. Dally-Piggott,-Engineering Specialist

i

K. Richards, Outage Manager

T. Miller, Outage

,

R. Graham, Shift Supervisor

,

W. Dowdy, Unit 2 Operations Manager

S. Rosen, Vice President, Nuclear Engineering

F. Comeaux, ISEG

'

M. McBurnett, Manager, Integrated Planning and Scheduling

!

NRC

T. Stetka, Chief, Project Section D, Division of Reactor Projects.(PSD, DRP)

,

M. Satorius, Project Engineer, PSD, DRP

!

R. Evans, Resident Inspector, PSD, DRP

!

!

.

t

i

!

)

a

r

'[

t

,