ML17312B464

From kanterella
Jump to navigation Jump to search
Insp Repts 50-528/97-05,50-529/97-05 & 50-530/97-05 on 970323-0503.Violations Noted.Major Areas Inspected: Operations,Maint & Engineering
ML17312B464
Person / Time
Site: Palo Verde  
Issue date: 05/20/1997
From:
NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV)
To:
Shared Package
ML17312B462 List:
References
50-528-97-05, 50-528-97-5, 50-529-97-05, 50-529-97-5, 50-530-97-05, 50-530-97-5, NUDOCS 9706020025
Download: ML17312B464 (40)


See also: IR 05000528/1997005

Text

ENCLOSURE 2

U.S. NUCLEAR REGULATORY COMMISSION

.REGION IV

Docket Nos.:

License Nos.:

Report No.:

Licensee:

Facility:

Location:

Dates:

Inspectors:

Approved By:

50-528

50-529

50-530

NPF-41

NPF-51

NPF-74

50-528/97-05

50-529/97-05

50-530/97-05

Arizona Public Service Company

Palo Verde Nuclear Generating Station, Units 1, 2, and 3

5951 S. Wintersburg Road

Tonopah, Arizona

March 23 through May 3, 1997

K. Johnston,

Senior Resident Inspector

D. Orsini, Resident Inspector

D. Carter, Resident Inspector

J. Russell, Resident Inspector, SONGS

D. Acker, Senior Project Engineer

D. Corporandy,

Project Engineer

Dennis F. Kirsch, Chief

Reactor Projects Branch F

Attachment:

Supplemental

Information

9706020025

970520

PDR

ADDCK 05000528

8

PDR

-2-

e

EXECUTIVE SUMMARY

Palo Verde Nuclear Generating Station, Units 1, 2, and 3

NRC Inspection Report 50-528/97-05; 50-529/97-05; 50-530/97-05

~Oererione

In general, the conduct of operations was good.

Excellent operator response

was

provided for the Unit 2 pressurizer spray valve packing leak, including the

conservative

decision to declare an Unusual Event (Section 01.4).

However,

inspectors observed inconsistent implementation of operations communications

standards:

particularly, the failure to consistently accomplish closed-loop

communications

during certain Unit 3 startup activities (Section 01.2); and

weaknesses

in the thoroughness

of logkeeping, particularly the failure to identify

changes

in major equipment status and plant problems in the logs (Section 01.3).

Operations management

responded

effectively by reemphasizing

their expectations

and assuring that procedures

clearly communicated

expectations.

Control room operations staff performance

during entry into Unit 3 midloop

operations was excellent.

Valve lineups were properly accomplished,

communications were excellent, and the nuclear assurance

organization provided

effective oversight of the evolution (Section 01.1).

Maintenance

~

Formal communications

between the scaffold team and the heating, ventilation, and

air conditioning (HVAC) maintenance

team were not established

to assure the

HVAC maintenance

team maintained adequate

clearance between

a scaffolding

platform and plant equipment

in the fuel handling building (Section M4.1).

Encnineering

e

~

Following a water hammer event in the Unit 3 Train A containment spray system,

operations

and engineering

performed thorough actions to assure the operability of

Train A. However, they did not promptly assess

the transportability of-the cause

and assess

the operability of the opposite train (Section M1.1).

~

A violation was identified by NRC wherein el'ectrical maintenance

technicians

violated electrical installation specification requirements

by using adhesive-backed

cable tie bases to anchor cable ties supporting cables in a Unit 3 vital battery

charger.

The anchors were first used in 1994 and again in March 1997, after the

first anchor failed, contributing to a charger breaker failure.

Further, the initial

repairs and subsequent

charger inspections were not documented

on the work

activity sheets

(Section E2.1).

-3-

System design weaknesses

in the automatic makeup to the volume control tank did

not allow the system to function in the last of core life and made the system

unreliable.

As a result, operator actions were required to compensate

for system

problems.

Engineering made little progress

in addressing

these issues, even though

they had, first been recognized

in 1988 and were the subject of both condition

reports and senior management

action plans.

The known system deficiencies were

not well established

in operating procedures

nor well understood

by plant operators

(Section E2.2)

~

A violation of 10 CFR Part 50, Appendix B, Criterion 3, was identified because

design engineering

did not assure that appropriate quality standards

were applied to

penetrations

in boundaries that served safety-related functions (Section E8.1).

II

1

I

t

Re ort Details

Summar

of Plant Status

Unit

1 remained at essentially 100 percent power throughout this inspection period with

the exception of a short term power reduction to 61 percent on March 25 for repairs to the

2C low pressure

heater.

Unit 2 remained at essentially 100 percent power throughout this inspection period.

On

April 21, Unit 2 pressurizer spray Valve RCE-PV-100F developed

an excessive

packing

leak.

The licensee entered

and exited an Unusual Event (UE) emergency classification

(Section 01.4)

Unit 3 began this inspection period in Mode 6.

On March 31, following completion of a

37-day refueling outage, the unit commenced

a reactor startup and power ascension.

On

April 12, the unit reduced power from 100 to 40 percent to repair a leak on the

2A condenser

hotwell.

Following the repair, the unit returned to 100 percent power and

remained there for the duration of the inspection period.

I. 0 erations

01

Conduct of Operations (71707)

01.1

Reactor Coolant S stem

RCS

Reduced Inventor

in Pre aration for Midloo

0 erations

Unit 3

a.

Ins ection Sco

e

On March 24, during the night shift, the inspectors observed the control room staff

make preparations for midloop operations.

The dedicated midloop team was in the

process of draining the RCS to the reduced inventory elevation.

The inspectors

had

discussions with the midloop team, onshift control room staff, outage management,

and other plant personnel.

In addition, the inspectors verified completion of

required prerequisites,

which included physical plant walkdowns.

b.

Observations

and Findin s

The inspectors

observed the midloop team drain the RCS from 114 feet, the reactor

vessel flange level, to 111 feet, 6 inches,

a stable level prior to entering reduced

inventory.

A decision was made to stop at this level to allow the completion of the

steam generator tube plugging activities before continuing to reduced inventory and

midloop operations.

The inspectors

noted that the Procedure 40OP-9ZZ16, "RCS Drain Operations,"

Revision 6, had been revised as part of the corrective actions for the drain path

misalignment that occurred during the last refueling outage

in Unit 1. Appendix Q

of the procedure

established

a drain path for draining of the RCS.

In two past

events, Valve CHN-V495, normally a locked closed valve, was left in the closed

position although verified to be open.

One of the changes

in the revised procedure

I

)1

-2-

removed Valve CHN-V495'from the body of the drain path alignment checklists and

converted it to a separate

action step with an added independent

verification step.

The inspectors accompanied

the auxiliary operator (AO) who performed the

independent

verification portion of Appendix Q.

The AO was directed by the

midloop control room supervisor

(CRS) to perform the checklist.

The AO performed

the checklist on the first page of Appendix Q and returned to the control room.

The

midloop CRS identified that the AO had performed only the first page of the two

page checklist and directed the AO to complete the checklist.

The AO went back to

the field and completed the second

page, which contained the independent

verification for Valve CHN-V495.

Although the midloop CRS prevented the AO from missing the second

page, the

inspectors

noted that the AO did not realize that there were two pages for valve

lineup and verification.

The inspectors verified that the AO attended the prejob

briefing that discussed

the revised procedure

and concluded that the AO did not

adequately review the contents of Appendix Q prior to performing the job. The

inspectors discussed

the observation with the unit department

leader.

The unit

department

leader stated that the AO's inattention to detail did not meet operations

management

expectations.

The inspectors verified that the makeup and gravity flow paths were aligned

properly.

Communications

between the midloop team and the control room staff

were excellent.

Nuclear assurance

was present for the evolution and provided

additional verification for the valve lineups'.

01.2

Plant Startu

Activities - Unit 3

'.

Ins ection Sco

e

The inspectors observed

portions of the Unit 3 plant startup following the sixth

refueling outage, including closing the main generator output breaker and

transferring steam generator feedwater supply from the downcomer to the

economizer.

b.

Observations

and Findin s

Both the closing of the main generator output breaker and the feedwater swapover

evolutions were conducted successfully.

System transients resulting from these

operations were within expected

levels.

The inspectors did note that the control

room staff did not consistently implement operations standards

for communications

and annunciator response

during the conduct of the main generator output breaker

closing operation.

Operations standards

provide the expectation

of closed loop

communications

(statement,

repeat-back,

acknowledgment).

However,

in some

instances,

communications

between the reactor operators

(ROs) and the CRS or

shift supervisor

(SS) did not include a repeat back and acknowledgment.

-3-

Additionally, operations standards

required ROs to announce

control room alarms

and the CRS to acknowledge the alarm announcements.

The inspectors noted that

there were instances where alarms were not announced

by ROs acknowledging the

alarm and other instances

where announced

alarms were not acknowledged

by the,

CRS.

The inspectors discussed

these observations

with the operations

standards

department

leader, who was observing the startup activities.

He acknowledged that

he had made the same observations

and planned to discuss them with the SS when

the unit was stable following the closure of the generator output breaker.

The

department

leader later confirmed that he had discussed

these communications

standards

with the SS.

The inspectors

did observe that during the subsequent

feedwater swapover activity that communications

were more consistent with

operations standards.

Review of 0 erations

Lo books

Unit 3

Ins ection Sco

e

On a routine basis, the ins'pectors reviewed RO and unit logs.

The inspectors

compared the logs to plant activities and events.

Observations

and Findin s

Section E2.2 of this report discusses

problems experienced

during the automatic

makeup to the Unit 2 volume control tank (VCT). In summary,

an automatic

makeup terminated after adding approximately 2 percent level to the VCT and the

boration flow totalizer did not count.

As a result, operators

did not have board

indication of boron added and ultimately had,to raise control rods to adjust

reactivity.

The inspectors

noted that this event was not discussed

in either the RO or unit logs.

The,issue was turned over to the oncoming shift and discussed

at the operation

director's morning meeting as a reactivity management

issue.

The inspectors

expressed

concern that this event was not logged.

In this case, logging the event

would assure that key.information was captured for an ongoing engineering

investigation.

On March 28, at approximately 4 a.m., with Unit 3 in Mode 4, operations personnel

removed the vital "C" battery charger EPKCH13 from service and placed the "AC"

swing battery charger EPKAH15 in service for corrective maintenance.

This action

complied with Technical Specification (TS) 3.8.2.1, Action B.

On April 1, the

inspectors reviewed the unit logs and determined that a logbook entry had not been

made for this evolution.

The inspectors discussed

this observation with the Unit 3

department

leader.

He recognized that major equipment status changes

should be

logged and subsequently

determined that it was an oversight of the crew.

,IJ

-4-

As discussed

in Section E2.1 of this report, maintenance

support personnel

logged

maintenance

rule unavailability of the PK C vital battery charger to the PK A vital

battery charger.

This was the result of weaknesses

in the unit logs which identified

that a Train A charger had been removed from service and did not specify which of

the two chargers

had been removed from service.

The inspectors discussed

these observations

with operations

management.

The

inspectors expressed

concern that operators may not be aware of the different

users of their logs and the importance of complete and accurate information.

Operations management

subsequently

reemphasized

their expectation for

logkeeping with operators.

In addition, they initiated a review of log keeping

procedures to assess

whether their expectations

were clearly communicated.

01.4

Hi h RCS Leaka

e - Unit 2

On April 21, 1997, at approximately 3:16 p.m., with the reactor operating at

100 percent power, operators received

a control room alarm indicating a high

reactor drain tank (RDT) pressure.

Operators observed that the temperature

and

level of the RDT were increasing

and that there was a mismatch between charging

and letdown flow. The inspectors observed

operators

assess

and eliminate

potential sources of RCS leakage to the RDT. They determined that the probable

source was a pressurizer spray valve packing.

At 3:57 p.m., the site shift manager

(SSM) declared

an UE and entered the TS

action for identified leakage greater than 10 gpm.

Emergency plan implementing

procedures

required an UE be declared with 10 gpm of unidentified RCS leakage and

25 gpm of identified RCS leakage.

TS 1.15, the definition for identified leakage,

defines identified leakage

as leakage from closed systems to collecting tanks.

At

the time they declared the UE, the SSM and SS were concerned with temperature

and level conditions in the RDT. Shortly after the declaration, the RDT conditions

were stabilized.

A team of operations,

engineering,

radiation protection, and maintenance

personnel

entered containment at approximately 5:30 p.m. to search for the leakage source.

They identified that the leak was from the packing leakoff line for pressurizer spray

Valve RC-PV-100F.

An operator closed upstream

and downstream

isolation valves

to the spray valve, terminating the leak. At 6:46 p.m., the SSM terminated the UE.

The inspectors observed

excellent operator response

and communications.

Additionally, the decision made by the SSM and the SS for event classification was

conservative.

-5-

01.5

Conclusions

on Conduct of 0 erations

In general, the conduct of operations was good.

Excellent operator response

was

provided for the Unit 2 pressurizer spray valve packing leak, including the

conservative

decision to declare the UE. However, inspectors observed inconsistent

implementation of operations communications

standards,

particularly the failure to

consistently accomplish closed-loop communications

during certain Unit 3 startup

activities, and weakness

in the thoroughness

of logkeeping,

as evidenced

by the

failure to identify a problem and changes

in major equipment status in the logs.

Operations management

responded

by reemphasizing

their expectations

and

reviewing procedures

to assure that these expectations

were clearly communicated.

08

Miscellaneous Operations Issues (92901)

08.1

Closed

Violation 50-528 96013-01 50-530 96013-01:

" ventilation boundary door

control.

This violation concerned

three examples where plant personnel

did not

properly control doors required to be closed to maintain fuel handling building and

auxiliary building ventilation boundary integrity.

The licensee responded

to this violation by letter dated November 21, ]996.

Actions completed at the time the response

was submitted were:

The maintenance

and engineering departments

had performed

a walkdown in

all three units to document field labeling on doors.

During these walkdowns,

temporary door labels were installed to clarify door numbering.

~

Bulletins were issued to inform site personnel

of the door control issues and

the consequences

of inadequate

door control.

~

A multidisciplinary design review team was established

to ensure design

requirements

were implemented for controlled doors.

Door control

Procedure 40DP-9ZZ17 was revised to incorporate their findings.

The inspectors found that each of these items had been satisfactorily completed.

The licensee also initiated the following corrective actions to prevent recurrence.

They planned to implement a design modification to install permanent door

signage

in all units.

They planned to train appropriate station personnel

on changes

to the door

control procedure.

Additionally, all personnel would receive training on

expectations

for door control and door labeling during site access training

and refresher training.

The inspectors found that these planned and completed actions were acceptable.

-6-

Closed

Licensee Event Re ort

LER 50-528 95012

Revision 0: reactor trip

caused

by a high water level in Steam Generator

1. This event was discussed

and

dispositioned

in NRC Inspection Report 50-528/95-21; 50-529/95-21;

50-530/95-21.

No new issues were revealed by the LER.

Closed

LER 50-528 95014

Revision 0: reactor trip following the degradation

of

main feedwater control system {FWCS). This event was discussed

in NRC

Inspection Report 50-528/95-21; 50-529/95-21; 50-530/95-21.

The inspection

report discussed that the 120-Vac control power Bus NNN-D11, which supplies the

FWCS, failed to complete an automatic transfer from its alternate power supply

Bus PBA-503 to the normal nonclass power supply, causing the loss of power to

Bus NNN-D11. As a result of the loss of power to the FWCS, the main feedwater

pumps went to minimum speed

and feedwater control valves closed, causing steam

generator

levels to decrease.

Steam generator

12 level decreased

below the reactor

trip setpoint resulting in a reactor trip.

The inspection report also noted that the licensee had determined that aligning

Bus NNN-D11 with its normal supply was preferable to aligning it with its alternate

supply.

Subsequent

to the inspection report, the LER identified that operating

procedures

were revised to reflect alignment of Bus NNN-D11 to the "normal"

power supply and that, prior to the action, the Plant Review Board had reviewed the

change

and concurred.

The inspectors concluded that the licensee's

actions were

acceptable.

Closed

LER 50-529/95001

Revision 0: reactor trip following the degradation

of

main feedwater flow. This event was discussed

in NRC Inspection

Report 50-528/95-14; 50-529/95-14; 50-530/95-14.

The LER also discussed

corrective actions resulting from the licensee's

event review.

Condition

Report/Disposition Request

(CRDR) 2-5-02553 added additional actions to address

other, identified problems associated

with the event.

The inspectors reviewed and verified the licensee's corrective actions and

concluded that the licensee's

completed and planned actions were acceptable.

Conduct of Maintenance (62707)

Containment

S ra

S stem Water Hammer - Unit 3

Ins ection Sco

e

On April 25, 1997, during the first routine surveillance test of the Train A

containment spray pump, operators

heard

a loud noise when the pump was started.

They subsequently

opened system high point vents and vented

a quantity of air

calculated to be less than 10 cubic feet.

The inspectors performed an initial review

of the licensee's

actions in response,to

this event.

0

-7-

Observations

and Findin s

The inspectors determined that engineering

personnel were involved in assessing

the effects of the event on April 25, and had promptly performed

a system

walkdown.

Engineering determined that the event did not damage plant equipment.

Additionally, engineering concluded that the event was less severe than a similar

water hammer event which occurred on the Unit 2 Train A containment spray

system in July 1995.

The Unit 2 event similarly occurred during the first system

test after a refueling outage.

On April 28, the inspectors inquired whether the Unit 3 Train B containment spray

pump had been tested or the system vented following the outage.

The SSM

confirmed that this had not occurred and initiated actions to vent the system.

Operators observed that although there was some air removed from high point

vents, it was considerably

less than the quantity of air removed from the opposite

train.

At the end of the inspection period, the licensee was reviewing the event to

determine root cause.

Specifically, corrective actions from the Unit 2 event were

expected to preclude

a repeat occurrence.

The inspectors will follow up the

licensee's

evaluation in a future inspection (Follow-up-Item 50-530/97005-01).

Conclusions

Following a water hammer event in the Unit 3 Train A containment spray system,

operations

and engineering thoroughly performed actions to assure the operability of

Train A. However, they did not promptly assess

the transportability of the cause

and assess

the operability of the opposite train.

Maintenance Staff Knowledge and Performance (62707)

Scaffold Platform Control in the Fuel Handlin

Buildin

Ins ection Sco

e

On April 9, 1997, the inspectors toured the Unit 2 fuel handling building with the

Office of Nuclear Reactor Regulation project manager.

A freestanding

scaffold

platform had been erected adjacent to the cask loading pit on the 140 foot level of

the fuel handling area.

The inspectors examined whether it had been erected.

according to site scaffolding requirements.

Observations

and Findin s

The scaffold platform was identified as "seismic" on an attached tracking tag.

The

inspectors noted that it was adjacent to the fuel handling building wall and in close

proximity to piping running vertically down the wall. The inspectors determined

l

-8-

that, for this construction,

a one bay, two tier scaffold platform, the construction

specification required

a clearance of 4 inches from safety related equipment.

t

The scaffold team leader walked-down the scaffold and concluded that it was closer

than 4 inches to piping.

However, none of these lines were safety related.

He also

noted that the scaffold was not in its original location.

It had been erected for the

HVAC maintenance

crew to work on air diffusers in a duct running along the west

wall of the fuel handling building.

It had originally been erected at the southwest

corner of the building and moved by the HVAC group to support their work. The

scaffold team leader was confident that, as originally constructed,

they had assured

adequate

clearance from plant equipment.

He had been aware that the HVAC group

intended to move the scaffold to support their work. It appeared that the need to

maintain these clearances

had not been adequately communicated to the HVAC

group.

The maintenance

services section leader responsible for the scaffold group initiated

a CRDR.

He noted that although they had met procedural requirements to maintain

a clearance from safety-related

equipment, this was primarily due to absence

of

such equipment along the path the scaffold was moved.

He recognized that

improved communications,

such as a precautionary step in the HVAC group's work

order, would have been necessary to assure

adequate

control.

C.

Conclusions

Formal communications

between the scaffold team and the HVAC maintenance

team were not established to assure the HVAC maintenance

team maintained

adequate

clearance between

a scaffolding platform and plant equipment in the fuel

handling building.

M6

Maintenance Organization and Administration

M6.1

Re ack of Pressurizer

S ra

Valve - Unit 2 62707

On April 24, 1997, maintenance

personnel

removed and reinstalled the packing in

pressurizer spray valve RCE-PV-100F.

The inspectors attended the prejob briefing

and reviewed the work package.

The inspectors noted that appropriate personnel

attended the prejob brief. The briefing, conducted

by the radiation protection team

leader and the Unit 2 SS, was thorough and in accordance

with the licensee's

guidance for prep'aration for risk significant work. Briefing attendees

actively

participated

in the discussion

and demonstrated

a questioning attitude.

The SS

emphasized

the need to perform the job correctly the first time and to be sensitive

to safety.

e

-9-

III. En ineerin

E2

Engineering Support of Facilities and Equipment (37551)

E2.1

Failure of Class

1E Batter

Char er C Unit 3

a

~

Ins ection Sco

e

On March 21, 1997, during Train A integrated safeguards

(ISG) testing, the PK C

battery charger AC input breaker tripped open.

A work request was written to

troubleshoot

and repair, as necessary,

and

a CRDR was written to document the

failure. The inspectors observed

portions of the corrective maintenance

performed

by electrical maintenance

and had several discussions

with the personnel involved.

Observations

and Findin s

On March 22, electrical maintenance

technicians were in the process of performing

a load test to restore the charger to service.

The inspectors discussed

with the

electrical maintenance

engineering section leader, present in the field, the cause of

the PK C battery charger AC input breaker trip. The engineering

section leader

stated that the troubleshooting

efforts were unable to duplicate the problem;

however, the as-found condition indicated that the incoming 480-Vac cables had

come into contact with the high voltage shutdown (HVS) card which may have

induced

a trip signal to the AC input Breaker, CB-1.

The engineering section leader stated that the incoming 480-Vac cables had been

supported

by a cable tie anchored to an adhesive-backed

cable tie base.

The cable

tie base subsequently

became detached

from the cabinet wall, allowing the cables

to fall closer to the HVS card.

The inspectors noted that the corrective actions

were to place the cables back to their original position, supported

by a new

adhesive-backed

cable tie base.

The load test was performed satisfactorily and the

charger was returned to operations.

The inspectors were concerned that the same problem could recur.

The engineering

section leader stated that the cables had been adjusted

in such

a manner that, if the

cable tie base were to fail, the cables would not come in contact with the HVS

card.

The inspectors were, also, concerned. that the same configuration, i.e, the

cables supported

by adhesive-backed

cable tie base, could exist in the other onsite

PK battery chargers.

The engineering section leader stated that they would inspect

the other chargers.

The licensee inspected the other units and determined that only the Unit 3 PK C

battery charger had the cables tied back with this adhesive

cable tie base.

A similar

configuration existed in the Unit 1 PK B battery charger, except that the cables

were supported

by nylon cable ties strapped to bigger cables.

The licensee had

removed the Unit 3 PK C battery charger from service to allow the electrical

-10-

maintenance

technicians to perform corrective maintenance.

The incoming 480-Vac

cables were rerouted and the adhesive

cable tie base was removed.

Similar problems had been identified in the past.

A 1986 engineering

evaluation

report and

a 1994 CRDR had been written in response

to similar events.

The 1986

engineering

evaluation report (EER) described

a problem that was discovered during

troubleshooting

efforts of the Unit

1 PK A and C battery chargers.

The 480-Vac

input cables were too close to the HVS card which caused the AC input breaker to

open from the shunt trip circuit. The corrective actions included the movement of

the incoming cables from the immediate vicinity of the HVS card for all PK battery

chargers.

The 1994 CRDR was written for the failure'of the Unit 3 PK D battery

charger.

During the performance of integrated safeguards

testing, the AC input

breaker tripped open during the simulated loss of offsite power.

As corrective

actions, electrical engineering determined that a distance of a minimum of 2.5

inches be maintanined between the 480 ac cables and the HVS card for all PK

battery chargers.

The inspectors

examined Specification 13-EN-306, Revision 5, "Installation

Specification for Cable Spacing

and Termination of Cable Systems at the Palo Verde

Nuclear Generating Station."

The specification clearly defines and lists the supports

that are allowed to be used to relieve stress on cable terminations or splices due to

cable weight and to prevent movement of cables to maintain minimum separation

and minimum bend radius requirements.

The inspectors determined that the 1994 corrective action to the Unit 3 PK C

battery charger was not in accordance

with electrical installation Specification

13-EN-306, in that adhesive-backed

cable tie bases were not allowed by the

specification.

Also, the corrective maintenance

on March 22, 1997, to the Unit 3

PK C battery charger was not in accordance

with the electrical installation

Specification 13-EN-306.

The two examples of inappropriate

use of the adhesive

cable tie base are a violation; specifically, the failure to follow approved

specifications (Violation 50-530/97005-02.)

The inspectors determined that the corrective maintenance

work package for the

initial troubleshooting

and repairs to the Unit 3 PK C battery charger did not

adequately document the work activity. The work activity sheet, which documents

the work/testing activities, did not contain the as-found condition, nor did it

describe the corrective action.

The inspectors discussed

this with the electrical

maintenance

engineering section leader.

The engineering section leader agreed with

the inspectors'bservation.

The inspectors reviewed the Unit 3 operations

logbook and determined that the

logbook did not contain an entry for the PK C battery charger that had been

removed from service for corrective maintenance

on March 28 to reroute the

incoming cables and remove the nylon cable tie adhesive

base.

This is discussed

in

Section 01.3.

-1 1-

The inspectors discussed,

with the system engineer and maintenance

support

personnel,

how the battery charger problems would be addressed

with res'pect to

maintenance

rule implementation.

They had concluded that the out-of-service time

of the PK C battery charger should be logged as unavailability time in the computer

data base.

The inspectors reviewed the unavailability data base for a PK C battery

charger and noted that no entry was logged for the March 21 occurrence.

The

maintenance

support engineering section leader stated that the unavailability was

entered,

in error, on the PK A battery charger.

The error was attributed to the

maintenance

support review of the operation unit log (see Section 01.3).

The error

was subsequently

corrected.

At the end of the inspection, the CRDR that had been initiated following the

March 21 event was still being evaluated

by the licensee.

The inspectors reviewed

the condition reporting section of the CRDR and noted that information which could

have been useful for a review had not been included.

The CRDR origination form,

did not discuss the previous occurrences

of this event.

Additionally, the CRDR did

not discuss that a walkdown of other chargers was recommended

or had been

performed, nor the immediate corrective actions to the resecure the cables.

c.

Conclusions

Electrical maintenance

technicians violated electrical installation specification

requirements

by using adhesive-backed

cable tie bases to anchor cable ties

supporting cables in a Unit 3 vital battery charger.

The anchors were first used in

1994 and again in March 1997, after the first anchor failed, contributing to a

charger breaker failure. The inital repairs and subsequent

charger inspections were

not documented

on the work activity sheets.

E2.2

Unit 2 VCT Blended Makeu

E ui ment Failure Resultin

in an Unknown Amount of

Boric Acid Addition

Ins ection Sco

e 71707

37551

On March 18, 1997, at approximately 4 a.m., the Unit 2 VCT decreased

in level,

due to identified charging system leakage,

and an automatic, blended makeup to the

VCT was initiated.

The Unit was at 100 percent power.

The makeup stopped

before the high level setpoint was reached

and the borated water totalizer had failed

to count the gallons added.

As a result, operators did not know how much boron

had been injected and waited for a system response

to adjust reactivity.

The

inspectors examined the circumstances

surrounding this situation and reviewed

procedures,

design documehts,

interviewed cognizant operations

and engineering

personnel,

and reviewed plant monitoring data.

-1 2-

Observations

and Findin s

S stem Descri tion

VCT level can be controlled automatically or manually.

In automatic, the operators

set a blend of borated water, from the refueling water tank, and pure water, from

the reactor makeup water tank, to correspond to the necessary

RCS concentration.

Operators adjusted the blend in the control room by setting the controllers for

Valve 210X (for reactor makeup water) and Valve 210Y (for borated water).

The

VCT automatic level setpoints

are 34 percent for initiation of makeup and

44 percent for termination of the makeup.

When automatic makeup occurs, the

borated water and reactor makeup water totalizers count the total amount of each,

in gallons, that are added.

Operators can, also, manually add water and/or boric

acid to the VCT, with Valves 210X and 210Y controlled either automatically (by a

totalizer counting down to a preset amount) or manually.

In automatic control a

boric acid pump will start and

a reactor makeup water pump, if not already running,

will start.

Valves 210X and 210Y will open fully, then throttle to match actual

flow to demand flow.

Event Review

Following the event on March 18, the inspectors reviewed logs, trends of reactor

power, cold leg temperature,

and makeup flow, and interviewed operators to

determine the plant impact.

The inspectors found there had been no impact on RCS

temperature.

Operators had determined,

based

on flow rates and times, that the

automatic makeup had contained too much borated water, and had withdrawn

control rods to compensate

for the negative reactivity addition.

The inspectors observed that there were no control room log entries concerning the

event, although the oncoming day SS was aware of the specifics.

The absence

of

log entries was discussed

in Section 01.3.

Subsequently,

the inspectors interviewed the cognizant system engineer.

The

system engineer noted that it appeared

that two problems had contributed to the

the event.

First, historically, Units 1, 2, and 3 had occasionally experienced

the

automatic cessation of the VCT makeup, before reaching the VCT high level

setpoint for makeup termination.

Operators

had termed this to be a "short cycle."

The system engineer did not know what caused

VCT makeup to short cycle.

The

second problem was that the totalizers would not register flow below 17 gpm.

The

circuitry used to sum the output from the flow detectors provided

a zero output if

the input was less than or equal to approximately 17 gpm.

The inspectors noted

that the operators

had set in 14 gpm for borated water flow during the activity

described

above and were not aware that the totalizer would not function with this

flow. There was a "temporary note" on the control board that indicated that the

system could short cycle.

l

I

, ~

-1 3-

I

On April 24, the inspectors met with the system engineer to discuss the results of

the engineer's

investigation into the cause of the short cycle events.

The system

engineer had determined that the system had responded

as designed.

He found

that Valve 210Y would respond quickly to a makeup demand by fully opening.

However, the valve would not then quickly throttle to a low flow position.

The system engineer also determined that system logic would terminate automatic

makeup if the mismatch between demanded flow and actual flow exceeded

5 percent when the VCT low level switch for automatic makeup reset.

The reset

occurred at approximately 36 percent VCT level, or about 82 gallons.

During an

automatic makeup, the reset typically occurred in less than 30 seconds.

Valve 210Y had not been setup to throttle from full open to low flow in this time

frame.

The inspectors reviewed Procedure 42OP-2CH01, Revision 23, "Chemical and

Volume Control System Normal Operations."

Sections 7, 8, and 9 of the procedure

discussed

automatic makeup to the VCT, reactor coolant makeup borate mode, and

reactor coolant makeup dilution mode.

Each section included a note stating that the

flow controllers were not reliable at flow rates less than 20 gpm.

However, each

section allowed a minimum flowrate of 10 gpm.

The inspectors were concerned

that the procedure

allowed operators to control the system in a region where it was

not reliable, requiring operator intervention when it failed. Although both the

procedure

and established

expectations

required that operators closely monitor

makeup evolutions,. the latitude provided by the procedure

appeared

to provide

unnecessary

challenges to operators.

Procedure

420P-2CH01

was subsequently

revised to provide a minimum flowrate of

20 gpm.

Consequently,

Unit 2 placed their VCT makeup controller in manual.

Problem Histor

The inspectors reviewed the history of problems with VCT makeup control.

In 1988, EER 88-CH-145 was initiated, documenting that the makeup flow

controllers could not reliably gain flow control in the time necessary

to

prevent

a short cycle event.

The EER proposed four possible resolutions and

some system modifications were drafted.

In 1991, EER 88-CH-145 was closed without implementing any system

changes.

The proposed

changes

were determined not to be cost effective.

In 1993, engineering initiated EER 93-CH-002, noting that Unit 3 was

experiencing flow control problems and had requested

that the resolution of

EER 88-CH-145 be revisited.

In this evaluation, engineering

discovered the

low signal cutout of the flow controllers.

-14-

~

In 1994, all open

EERs were rolled into the CRDR program when the EER

process was discontinued.

In late 1994, CRDR 9-4-800 was initiated to

address

the issue discussed

in EER 93-CH-002.

CRDR 9-4-800 included

actions to perform system testing and provide recommendations

by October

1995

~

The inspectors found that the actions for CRDR 9-4-800 were still open at

the time of the inspection.

Although the CRDR database

did not show it, the

inspectors was informed that the action plan had been rolled into a

significant (Level 1) action plan for engineering.

The inspectors was informed that initially the Level

1 action plan had

belonged to system engineering.

It was subsequently

transferred to design

engineering.

In early 1997, it became

a Vice President,

Engineering,

Level 1,

and was listed as one of the "Top 5 Customer Issues."

At the time of the March 18, 1997, event in Unit 2, both Units

1 and 2 had

"Temporary Notes" recognizing makeup problems.

Shortly following restart,

Unit 3 added

a Temporary Note reflecting similar problems.

In summary, the issues of makeup to the VCT had

a long history where

performance

issues were recognized.

However, no corrective actions had been

implemented

and at the time of the March 18, 1997, event, the causes

were not

well understood

by engineering.

However, engineering

did understand

system

limitations.

These limitations were not implemented

in operations

procedures

and

practices.

Safet

Anal sis Re ort Review

The chemical and volume control system was described

in detail in the Combustion

Engineering Standard Safety Analysis Report, Chapter 9.3.4.

The chapter discussed

all modes of VCT makeup.

However, it did not identify any limitations to the

automatic mode, other than to state that near the end of core life, it becomes

inefficient and the deborating demineralizer must be used to reduce boron

concentrations.

It did not identify that the automatic mode would be unreliable

during the last third of core life due to control system limitations.

Conclusions

System design weaknesses

in the automatic makeup to the VCT did not allow the

system to function in the last of core life and made the system unreliable.

As a

result, operator actions were required to compensate

for system problems.

Engineering

had made little progress

in addressing

these issues,

even though they,

had first been recognized

in 1988 and were the subject of both condition reports

and senior management

action plans.

The known system deficiencies were not well

established

in operating procedures

nor well understood

by plant operators.

-15-

E8

E8.1

Miscellaneous Engineering Issues (92903)

Closed

Follow-u

Item 50-529 96012-01

a

~

Backcaround

As discussed

in Inspection Report 50-528; 50-529; 50-530/96-12, the inspectors

found an unsealed

penetration

in the essential auxiliary building ventilation (EHA)

system boundary.

The EHA system was aligned with the essential fuel handling

building filtered exhaust trains during a safety injection actuation signal.

In this

mode, the EHA was required to be maintained at a negative pressure.

The

inspectors was concerned that the unsealed

penetration

could impact the ability of

the system to be maintained at a negative pressure.

The inspectors found that the licensee had previously identified a significant number

of flood, high energy line break (HELB), and EHA barrier penetration deficiencies.

These penetrations,

as well as sealed penetrations with these functions, were

identified in plant drawings as "nonquality related (NQR)." The inspectors

questioned

this classification because

the licensee had established that design basis

flood and HELB mitigation were safety-related functions.

Additionally, the EHA was

identified as a safety-related

system with the primary function of preventing

unfiltered release paths.

Civil engineering

had concluded that the nonconforming penetrations

did not impact

the safety-related functions and, therefore, did not need to be classified as other

than "nonquality related."

b.

Observations

and Findin s

In February 1997, the licensee provided the inspectors

an assessment

that

concluded that the nonconforming penetrations

with HELB, flooding, and pressure

differential boundary functions should be classified as "quality augmented

(QAG)."

They referenced

CRDR 9-1-0208 as supporting this conclusion.

Engineering

added

an action to CRDR 9-6-0691 to revise their Q-list to change

appropriate penetrations

to QAG.

The inspectors reviewed CRDR 9-1-0208.

The CRDR identified that the barrier

function list, established

as a result of the penetration

seal project, had not assigned

quality classifications.

Two memos, responding to the CRDR, dated September

1992 and March 1993, identified that penetrations

with HELB, flooding and

pressure differential functions should be classified as QAG.

Both memos indicated

that changes to appropriate documentation

should be made.

A subsequent

memo

concluded that appropriate

changes

had been made.

However, the changes

were

not made and as of March 1997, none of the penetrations,

either conforming or

nonconforming, which serve HELB, flooding, or pressure differential boundary

'I

I

-1 6-

functions had been classified as QAG.

In response to this finding the licensee

initiated CRDR 9-7-0590.

QAG was defined in UFSAR Chapter 17.2 as: "Items that do not perform a

safety-related function but which, as a result of regulatory commitment or

management

directive, require the application of certain quality assurance

program

elements."

10 CFR Part 50, Appendix B, Criterion 3, requires that appropriate

quality standards

are specified and included in design documents.

Failing to assure

that the QAG quality standard was specified for the penetrations

with HELB,

flooding, and pressure differential boundary functions was a violation of 10 CFR Part 50, Appendix B, Criterion 3 (Violation 50-528; 50-529; 50-530/97005-03).

Conclusions

'

Design engineering

did not assure that appropriate quality standards

were applied to

penetrations

in boundaries that served safety-related functions.

V. Mana ement Meetin s

X1

Exit Meeting Summary

The inspectors presented

the inspection results to members of licensee management

at the conclusion of the inspection on IVIay 7, 1997.

The licensee acknowledged

the findings presented.

The inspectors

asked the licensee whether any material examined during the

inspection should be considered

proprietary.

No proprietary information was

identified.

Ir

ATTACHMENT

PARTIAL LIST OF PERSONS CONTACTED

Licensee

P. Brandjes, Department Leader, Electrical Maintenance

Engineering

R. Flood, Department Leader, System Engineering

R. Fullmer, Director, Nuclear Assurance

M. Hypse, Section Leader, Electrical Maintenance

Engineering

W. Ide, Vice President,

Nuclear Engineering

K. Jones,

Section Leader, Design Engineering

D. Kanitz, Engineer, Nuclear Regulatory Affairs,

A. Krainik, Department Leader, Nuclear Regulatory Affairs

J. Levine, Senior Vice President,

Nuclear

D. Mauldin, Director, Maintenance

G. Overbeck, Vice President,

Nuclear Production

T. Radke, Director, Outages

F. Riedel, Department Leader, Operations Standards

C. Seaman,

Director, Emergency Services

M. Shea, Director, Radiation Protection

D. Smith, Director, Operations

J. Taylor, Unit 3 Operations Department Leader

M. Windsor, Section Leader, Mechanical Maintenance

Engineering

h

C

l

-2-

INSPECTION PROCEDURES

USED

71707

92901

62707

92902

37551

92903

Plant Operations

Plant Operations Follow up

Maintenance

Observations

Maintenance

Follow up

Onsite Engineering

Engineering Follow up

ITEMS OPENED

CLOSED

AND DISCUSSED

~Qened

50-530/97005-01

IFI

containment spray water hammer

50-530/97005-02

VIO

failure to follow approved electrical cable specifications

50-528; 50-529;

50-530/97005-03

VIO

failing to assure that quality standards

were specified for

the penetrations

boundary functions

Closed

50-528;50-

530/9601 3-01

50-528/9501 2

50-528/95014

50-529/95001

50-529/9701 2-01

VIO

ventilation boundary door control

LER

reactor trip caused

by a high steam generator water level

LER

reactor trip following degradation

of main feedwater control

LER

reactor trip following degradation of main feedwater flow

IFI

unsealed

penetrations

in the essential auxiliary building

ventilation system boundary

Ii

l

tl

k

-3-

LIST OF ACRONYMS USED

ABT

AO

CRDR

CRS

EER

EHA

FWCS

gpm

HELB

HVAC

HVS

LER

PK

PMUX

QAG

RCS

RDT

RO

SS

SSM

UE

UFSAR

VCT

automatic bus transfer

auxiliary operator

condition report/disposition

request

control room supervisor

engineeering

evaluation report

essential auxiliary building ventilation

feedwater control system

gallons per minute

high energy1ine break

heating, ventilation, and air conditioning

high voltage shutdown

licensee event report

Class 1E125-Vdc Power system

Plant Multiplex

quality augmented

reactor coola'nt system

reactor drain tank

reactor operator

shift supervisor

site shift manager

Technical Specifications

Unusual Event

Updated

Final Safety Analysis Report

volume control tank

I